Logo

Connection, Consciousness, Wisdom

Here is my attempt to answer the question of the meaning of life. You can read this as a series of ten parts here with accompanying videos.

I’m a person who’s had a lot of time to think and reflect on life and my experience. I’ve searched for meaning my entire life. Any talk of meaning has focused my attention.

If you think the answer will bring you peace, tranquility, or happiness, then I will tell you this:

Love. Love yourself, love other people, love your work.

Accept that you are a human being with needs. Humans are driven by a lack of needs on different levels. Maslow’s hierarchy of needs provides a list.

Which needs dominate our lives depends on our development over the life span.

Reflect on what need is missing from your life and work to satisfy it.

If you feel your needs are satisfied you should search for a state of flow.

“In positive psychology, a flow state, also known colloquially as being in the zone, is the mental state in which a person performing an activity is fully immersed in a feeling of energized focus, full involvement, and enjoyment in the process of the activity. In essence, flow is characterized by the complete absorption in what one does, and a resulting transformation in one’s sense of time.”

Components of flow:

Flow theory postulates three conditions that have to be met to achieve a flow state:

  1. One must be involved in an activity with a clear set of goals and progress. This adds direction and structure to the task.
  2. The task at hand must have clear and immediate feedback. This helps the person negotiate any changing demands and allows them to adjust their performance to maintain the flow state.
  3. One must have a good balance between the perceived challenges of the task at hand and their own perceived skills. One must have confidence in one’s ability to complete the task at hand.

Schaffer (2013) proposed seven flow conditions:

  1. Knowing what to do
  2. Knowing how to do it
  3. Knowing how well you are doing
  4. Knowing where to go (if navigation is involved)
  5. High perceived challenges
  6. High perceived skills
  7. Freedom from distractions

I strongly recommend the book The Happiness Hypothesis by Jonathan Haidt. The author compares the findings of modern science with the wisdom of ancient philosophies and cultures. Haidt proposes the equation for happiness as: H = S + C + V

Set point - “The hedonic treadmill, also known as hedonic adaptation, is the observed tendency of humans to quickly return to a relatively stable level of happiness despite major positive or negative events or life changes,” hedonic adaptation - Wikipedia, “Hedonic adaptation is a process or mechanism that reduces the effective impact of emotional events. Generally, hedonic adaptation involves a happiness “set point”, whereby humans generally maintain a constant level of happiness throughout their lives, despite events that occur in their environment. The process of hedonic adaptation is often conceptualized as a treadmill, since one must continually work to maintain a certain level of happiness.”

Many of the things we think will make us happy will only lead to a fleeting, temporary happiness which we adapt to. Anything you are searching for that will be static and unchanging will eventually not bring you happiness. We evolved this way because content creatures often become stagnant and don’t grow.

If you want to know how happy you will be you should reflect on how happy you have been. The average is likely your set point.

If your average happiness makes you dissatisfied, don’t give up yet!

You may not have reached your true set point. Your set point is heavily influenced by your health. Focus first on posture, breathing, sleep, nutrition, and exercise, in that order. Some will focus on exercise as a priority. Attaching yourself to the highest level helps to prioritize the lower levels. When you exercise well you want to eat and sleep well to support your training. Similarly we can attach ourselves to meaning and purpose which orders our actions on lower level needs.

Your circumstances influence happiness and your ability to maintain your health. Some environments are toxic or don’t support growth. We must make a voluntary action to change to a different set of circumstances as best we can.

We mostly need to invest work and energy into our voluntary actions to make ourselves as happy as we desire. We must choose to prioritize our health with a focus on balancing our needs in life.

Recognize and accept that happiness and well-being are not a destination to reach and rest at but you must continually strive for and renew.

Grow to accept all your emotions. There are 5 emotions thought to be like primary colors that mix to give subtler feelings. Most people will have these feelings across cultures. I remember them as “emotion FADES.”

I invite you to explore the Atlas of Emotions to learn more. I recommend associating activities with the 5 basic emotions. What is a healthy outlet for each of your feelings? Once you recognize cues, how can you channel your emotional energy into something positive?

A short story on controlling emotions:

A monk asked their master how to control their own emotions. The master told the monk to go to the ocean and stop the waves from crashing. Accustomed to strange challenges, the monk faced the open ocean and tried to stop the waves by punching and kicking. The waves did break in some parts, but the monk was still pushed over and tumbled by the strong waves. Nothing the monk tried stopped the waves.

Emotions are like waves. Sometimes they are pleasant and gentle. Sometimes we can ignore them. Sometimes the waves are so powerful they threaten to destroy us.

We all have different levels of sensitivity. Some of us are in the ocean of emotion with water at our waist. Some are deeper in the ocean with turbulent and frightening currents. Some of us are not even in the water but stand far away.

In all cases we must accept the waves will come. The waves will hit us and they will recede.

With this reflection, the monk sits in the ocean and no longer attempts to break the waves. The waves wash over the monk, each one unique and different from the last. Some waves push the monk out of balance, but the monk in time returns to their position and peace. The monk cannot control the waves, but the monk has the most influence over their reaction.

Consider how many of the 5 emotions are positive or negative.

Would you say 4 out of 5 are negative and enjoyment is positive? Many of us will understand that much of life is suffering. A small part, and if we want to give a proportion we can say 1 in 5 events like the Pareto principle, will be absent of suffering. If you want to be resilient to suffering and grow stronger, you should ask yourself and answer, “why is my suffering worth enduring?”

The work of Shaolin Master Shi Heng Yi provides a guide on self-mastery:

5 Hindrances to Self-Mastery

  1. Sensual desire (sight/touch/taste/smell/hearing) = Am I distracted or addicted?
  2. Ill will/aversion = Am I feeling a negative emotion?
  3. Dullness/heaviness = Am I unmotivated?
  4. Restlessness = Is my mind jumping from thought to thought?
  5. Skeptical doubt = Am I indecisive?

RAIN

Your Reason for Being

Some say the question of meaning in life is a privilege, that self-actualization is the highest level in our list of needs. You should reject this line of thinking as if you were fighting for your life, because you are. Those who want you to turn away from meaning want to control you. Meaning is for everyone and offers guidance and clarity. Meaning is a continual questioning where a lack of meaning is no questions at all.

What is your reason for being? What keeps you alive? To find meaning and coherence in your life I recommend the concept ikigai. Ikigai roughly translates to “reason for being.”

Answer these four questions:

  1. What do I love to do?
  2. What am I good at?
  3. What does the world need?
  4. What can I be paid for?

Be excited to answer these questions! If you don’t yet have a satisfactory answer, then your goal is to explore and learn more until you can! Your ikigai will change and grow over time, because the answer to each of the questions can change and grow.

“Jumping from failure to failure with undiminished enthusiasm is the big secret to success.” – Savas Dimopoulos

In your work, always seek at least these 3 components: autonomy, mastery, and purpose. Autonomy to choose how you go about your work, arranging your schedule to match your personal energy. Mastery is a path to a state of flow, constantly improving your craft. Purpose is finding meaning in your activity beyond yourself. These are the intrinsic motivators which propel us greater and more sustainably than external motivators like money. Dan Pink describes supporting research in his book, Drive.

Loving Yourself and Others

How can you learn to love yourself and others?

The information surrounding relationships is so powerful I hesitate to share it, because it can cause as much harm as it can good. However, the tools of relationships are critical for a full and healthy human life.

I will propose three parts to satisfying relationships based on social engineering which you can use to reflect on your past relationships. Your most cherished relationships will often involve:

  1. Trust
  2. Empathy
  3. Rationality

You must trust the other person. You should be able to accurately predict how they will react. That person should have the ability to understand the emotional content of your experience. You should trust the other person to be rational, to make appropriate choices given enough information. Do your best relationships have these characteristics?

Learning to love yourself is much like building a deep relationship with another person. You should demonstrate your trustworthiness, display empathy, and be rational. Do you trust yourself? Do you keep your personal promises and commitments? Are you consistent in your behavior? Are you empathetic and understanding with your emotions? Do you forgive yourself for past actions? How often do you check in and assess how you are feeling? Are you rational and make the right choices given what you know? Do you fall into impulses and easy routines?

To love yourself, treat yourself as a friend. Invest in yourself. Make a commitment to your emotional health then follow up later. You will build trust and empathy in yourself. Invest in your rationality. Take time to explore all of your options. First determine what is best then assess how able you are to reach the best outcome. Try to be objective and fair. Recognize your judgment of yourself and others is likely at least 10% inaccurate. With doubt cut yourself and others slack for failures, but work towards overcoming those obstacles in the future.

If this is so clear, why is it rare or difficult? Loving yourself and others takes work, energy, and investment. Relationships and selves are highly complex. Forces outside of our control will eventually disrupt our bonds. When that happens we can do our best to repair and reinforce our relationships.

Finally, I highly recommend the work of John Vervaeke on Awakening from The Meaning Crisis. Professor Vervaeke gives a wonderful summary of the history of meaning and likely why you feel a meaning is lacking.

If you find this answer satisfying, I suggest you stop now. I wish you a good life.

The Meaning of Life

As for the truest, most objective, most accurate description of reality and the meaning of life, before I continue I would sincerely like you to consider that it may shake you to your core. As I will explain, the average person is full of delusions and lives on the basis of ignorance and lies. They are inaccurate yet productive lies. You will likely be unable to return to your previous mode of thinking. As you might expect, the answer is complex and incomplete. Give up expectation on how you think you will react.

The Island of Reality

With this mental tool, I hope we can converge to the same conclusions. When I speak of reality, I divide our experience into three parts, objective, subjective, and intersubjective. To explain the divisions, imagine an island in the ocean, small and big enough for one person. However, there are no humans or animals on the island. We can describe and talk about what is there even though we can imagine no humans there to use language to describe it. We can talk of grains of sand, the ocean waves, the plants. Whatever we can discuss without involving a human (or conscious) being existing there is objective reality. I believe the world and universe continues to exist even if we aren’t around to experience it. The subjective experience is when a human being appears on the island.

Let’s imagine one person. One person on the island is another part of objective reality. We are a part of objective reality, a complex part. We have feelings, memories, thoughts, and all kinds of phenomena when we’re awake. There’s something it’s like to see the color of a ripe mango. There’s something it’s like to lay on the sand and feel the warmth of the sun. We can describe these experiences objectively, measuring the frequency and amplitude of the light coming from the mango and into an eye, following the chain reaction through the brain and body. Given the objective facts, we still have something missing from the description, the “what it’s like-ness” of our personal experience. Whatever we imagine and talk about that requires at least one person, that’s subjective reality on top of objective reality. However, subjective experience is in the space of minds which apparently have different qualities than most of objective reality. The space of ideas is as diverse as we can feel and communicate.

Whatever phenomena requires two or more people, that is intersubjective reality. Most of language is an intersubjective reality. Words have meaning not in themselves but in the context of the mind experiencing the words. It seems most of language is about communicating on a complex level about objective and subjective reality with other similarly complex creatures. We talk about what we experience (objective) and what we’re feeling (subjective). So complex communication creates a pattern of behavior, an event we experience and predict with regularity.

Consider the standard currency of America, the $1 bill. Imagine this bill alone on an island with no people. Does it possess any value or effect on its own? What is the objective description of the money? It is a rectangular piece of rare paper with ink designs bonded to the surface. After some time it will disintegrate like all wood and paper. What about with the bill and one person on the island? Now something else is happening, the person is perceiving the bill. There are subjective phenomena occurring with whatever the person experiences. Now think of the phenomena when two people on the island exchange the bill for other things. They both have a value in mind for the money and what the other thinks the value is. When the two ideas of value clash they center to a shared value.

We live in a world where millions and billions of humans agree on a shared experience and perception of reality. Now the rectangular piece of paper with special ink designs can cause so many events, freedom, food, shelter, violence, destruction, and everything money involves. Money is a part of reality, but it is almost a life separate from objective reality. Observing money, much of what money is and how it effects change in the world is not on the objective level. The value of money tracks in the beliefs and behaviors of humans. The value of money is in our minds and carried out in our behavior.

As we have described money on an analysis of three levels of reality, we can split our memories and experience into an analysis by three parts. Investigations into the three parts of experience in science are described by:

So we can analyze experience on each level of reality. By considering life on each level, I hope the clarification leads us to similar conclusions. Notice the subject of philosophy is not included in this list. Where is it? As a preliminary definition, philosophy is a reflection on the three levels of reality and how they relate to one another. I will give a perspective on many sub-problems of philosophy throughout.

I have searched for an objective answer to the meaning of life on all three levels. This is not “my” answer to the meaning of life, but “the” answer. “My” answer is simply my personal subjective meaning. As we will see the subjective meaning of life is the compression of past experience which causes the individual to act. The objective answer is what is shared among us all and independent of what we individually believe and value.

I would like to give an argument as to why many will not accept the meaning of life: our minds function largely on duration, path, and outcome. Have you been told “no one knows the meaning of life?” The duration for the search appears longer than a single human life, so it is not achievable. The path is unclear; to where should you go, what should you study? The outcome is unknown; what will you feel or know with the meaning of life? For these three reasons many will give up the search for the meaning of life and pursue more attainable goals.

Additionally, the question of life is an existential question, a question about living. When we are busy living, doing, we do not question what life is about. Only if we step back to the perspective of the watcher in us all do we begin to hold the question of what is the purpose, the meaning, of living. If you do not hold the question, there is nothing to understand and compress in the meaning of life.

An outline of my argument:

How did objective reality come to be? How did life start? These are difficult questions. For an individual, we are limited by our personal experience and subjectivity. We experience the world through our own senses. If the average human lifetime is 50 years, how can we soundly infer what has happened before us and what will happen after?

To simplify the question, let’s start with very basic rules for a “universe.”

Imagine the above grid spread out in all directions, instead of 3x3 it’s ∞x∞. Each square represents either a living cell or a dead spot. Let’s say the white parts are dead, and the black parts are living cells. Which cells are alive and which are dead at moment zero, the beginning of the universe, is randomly assigned. Now, here are the rules of this basic universe in Conway’s game of life:

  1. Any live cell with two or three neighbors survives.
  2. Any dead cell with three live neighbors becomes a live cell.
  3. All other live cells die in the next generation. Similarly, all other dead cells stay dead.

With every “tick,” every moment after the start, the rules are applied to each cell on the grid. Please take a moment to apply these rules to the grid above. We can do the top row together.

For the top left cell:

For the top middle cell:

For the top right cell:

What about the middle left cell in the next generation? Keep in mind the cell is calculated with the current cells, not the cells already calculated for the next generation.

Of course, we can get a computer to calculate these rules on a much larger grid. From these three rules, what do you expect to see? Really try to imagine the possibilities. How much will the living cells spread and move? Will there be something like groups of cells moving across the grid? Can we make a stable formation that will never change or die? What’s a weird looking group that could pop up? Will the grid just be flashing lights of cells popping in and out with no forms?

The answer isn’t obvious to me. The easier way to know what will happen is to run the simulation and see what we get. Some people have let the simulation run for days on a vast grid and cataloged what happens.

Patterns

There was never a specific design for any of these patterns. They occur from the initial state and the rules on that state. The results are fascinating as “creatures,” groups of connected cells, interact with each other. There are self-replicating creatures which create new groups of cells which are sent off.


A pattern called Gosper’s glider gun.

“Many patterns in the Game of Life eventually become a combination of still lifes, oscillators, and spaceships; other patterns may be called chaotic. A pattern may stay chaotic for a very long time until it eventually settles to such a combination.

The Game of Life is undecidable, which means that given an initial pattern and a later pattern, no algorithm exists that can tell whether the later pattern is ever going to appear. This is a corollary of the halting problem: the problem of determining whether a given program will finish running or continue to run forever from an initial input,” – Game of Life - Undecidability.

Do you think a functioning computer could arise from these three simple rules? In fact, a computer can arise without intervention or design, called a Turing machine. Conway’s game of life is Turing complete, and can simulate any Turing machine (which includes your computer and everything it does). Let’s get deeper into Turing machines later, but this means any algorithm can be carried out by a complex enough group of cells in the game. So groups of cells can self-replicate and carry out programs, processing information, all from three simple rules.

A large part of this demonstration is to show what properties can arise from simple operations. I didn’t expect to see so many incredible patterns. Consider the universe we inhabit. What are the basic rules that govern this universe? Here’s a list of fundamental physics formulas.

Instead of a few basic rules, humans have discovered many laws governing the transfer of energy, heat, motion, etc. We could hardly predict what would come about from three simple rules, so what can we expect from hundreds of laws interacting? Our ideas of cells were extremely simple, alive or dead, yet the biology of our lives are extremely complex. Physics gives rise to all the elements on the periodic table with their interactions. I would like to explore some of these rules as their implications for life are significant.

A Little Physics - The Laws of Thermodynamics

To gain an understanding of life and the world, we need an understanding of thermodynamics. “Thermodynamics is a branch of physics that deals with heat and temperature, and their relation to energy, work, radiation, and properties of matter. The behavior of these quantities is governed by the four laws of thermodynamics which convey a quantitative description using measurable macroscopic physical quantities, but may be explained in terms of microscopic constituents by statistical mechanics,” (Thermodynamics – Wikipedia).

“The three laws of thermodynamics define physical quantities (temperature, energy, and entropy) that characterize thermodynamic systems at thermodynamic equilibrium. The laws describe how these quantities behave under various circumstances, and preclude the possibility of certain phenomena (such as perpetual motion).”

“The first law of thermodynamics is a version of the law of conservation of energy, adapted for thermodynamic systems.

The law of conservation of energy states that the total energy of an isolated system is constant; energy can be transformed from one form to another, but can be neither created nor destroyed…

The First Law encompasses several principles:

Combining these principles leads to one traditional statement of the first law of thermodynamics: it is not possible to construct a machine which will perpetually output work without an equal amount of energy input to that machine. Or more briefly, a perpetual motion machine of the first kind is impossible.”

“The second law of thermodynamics indicates the irreversibility of natural processes, and, in many cases, the tendency of natural processes to lead towards spatial homogeneity of matter and energy, and especially of temperature. It can be formulated in a variety of interesting and important ways.

It implies the existence of a quantity called the entropy of a thermodynamic system. In terms of this quantity it implies that

When two initially isolated systems in separate but nearby regions of space, each in thermodynamic equilibrium with itself but not necessarily with each other, are then allowed to interact, they will eventually reach a mutual thermodynamic equilibrium. The sum of the entropies of the initially isolated systems is less than or equal to the total entropy of the final combination. Equality occurs just when the two original systems have all their respective intensive variables (temperature, pressure) equal; then the final system also has the same values.

The second law is applicable to a wide variety of processes, reversible and irreversible. All natural processes are irreversible. Reversible processes are a useful and convenient theoretical fiction, but do not occur in nature…

Entropy may also be viewed as a physical measure of the lack of physical information about the microscopic details of the motion and configuration of a system, when only the macroscopic states are known. This lack of information is often described as disorder on a microscopic or molecular scale. The law asserts that for two given macroscopically specified states of a system, there is a quantity called the difference of information entropy between them. This information entropy difference defines how much additional microscopic physical information is needed to specify one of the macroscopically specified states, given the macroscopic specification of the other – often a conveniently chosen reference state which may be presupposed to exist rather than explicitly stated. A final condition of a natural process always contains microscopically specifiable effects which are not fully and exactly predictable from the macroscopic specification of the initial condition of the process. This is why entropy increases in natural processes – the increase tells how much extra microscopic information is needed to distinguish the final macroscopically specified state from the initial macroscopically specified state.”

“The third law of thermodynamics is sometimes stated as follows:

The entropy of a perfect crystal of any pure substance approaches zero as the temperature approaches absolute zero.

At zero temperature the system must be in a state with the minimum thermal energy. This statement holds true if the perfect crystal has only one state with minimum energy…

The entropy of a system approaches a constant value as the temperature approaches zero.”

“The zeroth law of thermodynamics may be stated in the following form:

If two systems are both in thermal equilibrium with a third system then they are in thermal equilibrium with each other.

The law is intended to allow the existence of an empirical parameter, the temperature, as a property of a system such that systems in thermal equilibrium with each other have the same temperature.”

Life can be extensively explained from the implications of these laws.

I am strongly influenced by Professor Sid Smith’s lecture, How to Enjoy the End of the World published on Apr 23, 2019.

This lecture had a profound effect on me because of the mathematical and intuitive explanation. He analyzes organisms, societies, and planets based on physics models of energy.

Understanding entropy is critical to understanding the world. To preface the below with a simple explanation and example, entropy is the transformation of useful energy into less useful energy. When you eat food, you use that energy for heat and work, then you poop out the remainder. The food was highly useful to your body, now it is much less useful as poop. The densely energy packed-black oil of gasoline is burned as fuel for work and heat but becomes less useful once burned.

“In statistical mechanics, entropy is an extensive property of a thermodynamic system. It is closely related to the number Ω of microscopic configurations (known as microstates) that are consistent with the macroscopic quantities that characterize the system (such as its volume, pressure and temperature). Entropy expresses the number Ω of different configurations that a system defined by macroscopic variables could assume…

The second law of thermodynamics states that the entropy of an isolated system never decreases over time. Isolated systems spontaneously evolve towards thermodynamic equilibrium, the state with maximum entropy. Non-isolated systems, like organisms, may lose entropy, provided their environment’s entropy increases by at least that amount so that the total entropy either increases or remains constant. Therefore, total entropy in the Universe does increase. Entropy is a function of the state of the system, so the change in entropy of a system is determined by its initial and final states. In the idealization that a process is reversible, the entropy does not change, while irreversible processes always increase the total entropy.

Because it is determined by the number of random microstates, entropy is related to the amount of additional information needed to specify the exact physical state of a system, given its macroscopic specification. For this reason, it is often said that entropy is an expression of the disorder, or randomness of a system, or of the lack of information about it.”

A dissipative structure is embodied energy that takes in exergy (low entropy energy) and expels heat (high entropy energy), according to the arrow of time (tending towards increasing entropy). For instance, a fire is a dissipative structure. Fire uses fuel (exergy) to maintain and grow until the fuel is used. Once the fuel is gone, all exergy has been released as heat. The energy is the same, but the exergy is gone. “One can view the entire universe as a dissipative structure. The big bang released exergy, creating time and space. The feature that is perhaps most characteristic of dissipative structures, including the universe itself, is complexity.”

Complexity appears to defy the second law of thermodynamics, which states that in a closed system entropy will always increase. “Order is never spontaneously produced from disorder. Once a glass breaks, you can’t un-break it.” Every dissipative structure converts exergy to heat, but it does so based on the rules of energy transfer and the self-organization of the dissipative structure itself.

We are like fire.

Living creatures are dissipative structures. We are not unlike a fire which consumes all available fuel to expand and maintain itself until there is no more. The fuel needed for life is simply more complex. We require many different kinds of fuel – food, typically other complex dissipative structures like plants and animals – to maintain and grow our own structures, our bodies. The way heat and energy are dissipated from our body is governed by physics and biology, where biology is the study of living dissipative structures. Fire doesn’t appear to consciously direct itself just like we don’t decide how to sweat or where our blood vessels should grow. Our bodies emerged from a long evolutionary process. Evolutionary processes selected for the rules which govern our bodies, and we need an understanding of evolution to understand life.

Just like how complex patterns emerged in Conway’s Game of Life, so too did life emerge as a complex pattern from an initial state. Do the patterns we see moving across the grid decide which way they go? No, it’s clearly an outcome determined by the rules of the game. The story of our life is similar.

Properties of Dissipative Structures

Crises can disrupt homeostasis. The more complex a system is the more ways it can go wrong. Greater complexity increases the possibility of a crisis occurring as well as a crisis causing a cascade of further crises ultimately unraveling homeostasis causing death. This gives a basic explanation for aging. An organism reaches its peak maturity and accrues damage and mistakes over time. There is a crisis like a burn on the body. Your body tries to heal as best it can and work around the trauma, but the system becomes less able to adapt in the future. The scar tissue cannot function as flexibly as undamaged skin, opening a path to a future crisis. There are all kinds of crises occurring in our bodies that we won’t feel. Eventually the organism cannot adapt and dies.

Assuming a dissipative structure has access to unlimited exergy and no crises, can it go on forever?

“In general no, because they are never static and never go backwards. No natural process is reversible in itself, because entropy always increases, never spontaneously decreases. The same processes that brought a dissipative structure must continue, because energy must continue to flow, increasing complexity. Each increase in complexity has a metabolic cost. As these costs accumulate, eventually the system becomes unsustainable on its energy base and increasingly fragile. At some point, a crisis occurs and there’s a complexity collapse. If exergy is still available post-collapse, new dissipative structures will arise anew, typically with significant differences from the old structure.”

The biologist Eugene Odom describes this process in ecosystems called ecological succession. Ecosystem means a biological community of interacting organisms and their physical environment. An ecosystem begins on pure rock with small and simple dissipative structures. Eventually the entire ecosystem collapses and ends. A new ecosystem is built on a previous ecosystem. Each community of organisms in the ecosystem alters the environment. “It culminates in a homeostatic ecosystem in which biomass and symbiotic function between organisms are maintained. Maximum that is per unit of energy flow. Nature always optimizes. To repeat, self-organization is directed towards achieving as large and diverse a dissipative structure within the limits set by the available energy input and the prevailing physical conditions. Once homeostasis is achieved, the ecosystem is mature. However, eventually the ecosystem succumbs to its own fragility. A small perturbation in climate, or a fire, or an evolution of a disruptive, invasive, or parasitic species, and all the intricate specializations that lead to the optimization of the structure become in turn the cause of the system’s failures. In the long view, this is not a tragedy, but an essential component of the evolutionary process. It’s precisely the story of life on this planet with its breathtaking diversity not only across continents, but across time.”

Eventually, every organism relies on the whole system to survive. Once the system dies, the individuals die as well. In a similar way to how a fire will burn available fuel, organisms and ecosystems adapt to consume available fuel (exergy), specializing in the process of transforming that useful energy into less useful energy. We see that specialization is a natural drive. When an organism specializes they typically have a fuel source that other organisms find useless or undesirable, the rules governing other self-organizing organisms don’t allow them to use that energy, while the specialists become efficient.

When we talk about life, we often mean human life. The story of biological life and ecosystems is the beginning of the story of humans. The below timeline is taken from Sapiens: A Brief History of Humankind by Professor Yuval Noah Harari. I highly recommend this book as it discusses the origin of our species and how we came to dominate the planet. I really cannot recommend it enough! You can find a summary on Wikipedia. Please do your best to get a sense of the timescale.

Timeline of History

Years Before the Present

13.5 billion

Matter and energy appear. Beginning of physics. Atoms and molecules appear. Beginning of chemistry.

4.5 billion

Formation of planet Earth.

3.8 billion

Emergence of organisms. Beginning of biology.

6 million

Last common grandmother of humans and chimpanzees.

2.5 million

Evolution of the genus Homo in Africa. First stone tools.

2 million

Humans spread from Africa to Eurasia. Evolution of different human species.

500,000

Neanderthals evolve in Europe and the Middle East.

300,000

Daily usage of fire.

200,000

Homo sapiens evolves in East Africa.

70,000

The Cognitive Revolution. Emergence of fictive language. Beginning of history. Sapiens spread out of Africa.

45,000

Sapiens settle Australia. Extinction of Australian megafauna.

30,000

Extinction of Neanderthals.

16,000

Sapiens settle America. Extinction of American megafauna.

13,000

Extinction of Homo floresiensis. Homo sapiens the only surviving human species.

12,000

The Agricultural Revolution. Domestication of plants and animals. Permanent settlements.

5,000

First kingdoms, script and money. Polytheistic religions.

4,250

First empire – the Akkadian Empire of Sargon.

2,500

Invention of coinage – a universal money. The Persian Empire – a universal political order ‘for the benefit of all humans’. Buddhism in India – a universal truth ‘to liberate all beings from suffering’.

2,000

Han Empire in China. Roman Empire in the Mediterranean. Christianity.

1,400

Islam.

500

The Scientific Revolution. Humankind admits its ignorance and begins to acquire unprecedented power. Europeans begin to conquer America and the oceans. The entire planet becomes a single historical arena. The rise of capitalism.

200

The Industrial Revolution. Family and community are replaced by state and market. Massive extinction of plants and animals.

The Present

Humans transcend the boundaries of planet Earth. Nuclear weapons threaten the survival of humankind. Organisms are increasingly shaped by intelligent design rather than natural selection.

The Future

Intelligent design becomes the basic principle of life? Homo sapiens is replaced by superhumans?

Sapiens begins at the big bang and fast-forwards through time to about 2.5 million years ago where it traces homo erectus and the many cousin-species of homo sapiens. What is so impactful to me about this book is how it breaks down the long held hubris of humankind. “They [humans] were insignificant animals, whose ecological impact was less than that of fireflies or jellyfish.” With homo sapiens appearing just 0.2 million years ago, we likely held a niche spot in the ecosystem similar to the woodpecker. Just like the woodpecker evolved to eat grubs by hammering through dense bark, we might have evolved to eat bone marrow. When an apex predator such as a tiger killed a giraffe, early humans watched as the tiger and pride ate most of the giraffe. Then the hyenas came to eat the remaining parts. Finally, early humans, after looking all around for predators, slunk towards the carcass to eat any remaining tissue and use stone tools to break bones and eat the marrow. That was our niche.

However, for 2 million years from homo erectus to homo sapiens, our brains just kept growing. We are still unsure as the reason why. With the domestication of fire, humans could scare away beasts, keep themselves warm, cook food, and control their environment. Of course, cooking food allowed our bodies to absorb nutrients better. This hints at what was to come, humans getting more resources from the same amount of food or land compared to other species. One human could easily burn down a forest in a matter of 20 minutes. Humans learned to do controlled burning which allowed them to pick through charred meat and other foods in the aftermath of a fire. Our species co-evolved with the use of fire.

Harari divides the history of Sapiens into 4 parts:

  1. The Cognitive Revolution (c. 70,000 BCE, when Sapiens evolved imagination).
  2. The Agricultural Revolution (c. 10,000 BCE, the development of agriculture).
  3. The unification of humankind (the gradual consolidation of human political organisations towards one global empire).
  4. The Scientific Revolution (c. 1500 CE, the emergence of objective science).

The “Lion-man” sculpture is the oldest known work of figurative art, the first sculpture known depicting a creature entirely from the human imagination, half-man, half-lion. The work is 35,000-40,000 years old. Lion-man – Wikipedia

Harari’s main argument is that Sapiens came to dominate the world because it is the only animal that can cooperate flexibly in large numbers. He argues that prehistoric Sapiens were a key cause of the extinction of other human species such as the Neanderthals, along with numerous other megafauna. He further argues that the ability of Sapiens to cooperate in large numbers arises from its unique capacity to believe in things existing purely in the imagination, such as gods, nations, money, and human rights. He argues that these beliefs give rise to discrimination – whether that be racial, sexual or political and it is potentially impossible to have a completely unbiased society. Harari claims that all large-scale human cooperation systems – including religions, political structures, trade networks, and legal institutions – owe their emergence to Sapiens’ distinctive cognitive capacity for fiction. Accordingly, Harari regards money as a system of mutual trust and sees political and economic systems as more or less identical with religions.

Harari’s key claim regarding the Agricultural Revolution is that while it promoted population growth for Sapiens and co-evolving species like wheat and cows, it made the lives of most individuals (and animals) worse than they had been when Sapiens were mostly hunter-gatherers, since their diet and daily lives became significantly less varied. Humans’ violent treatment of other animals is a theme that runs throughout the book.

In discussing the unification of humankind, Harari argues that over its history, the trend for Sapiens has increasingly been towards political and economic interdependence. For centuries, the majority of humans have lived in empires, and capitalist globalization is effectively producing one, global empire. Harari argues that money, empires, and universal religions are the principal drivers of this process.

– Summary - Wikipedia

What makes humans so unique is our ability to cooperate flexibly in large numbers. Other swarm and hive creatures cooperate in large numbers, but they all follow simple rules. Humans can learn new rules and collectively change their behaviors. The accumulation of knowledge in various subjects and the depths explored lead me to believe we can give a more detailed and accurate answer to the meaning of life than ever before.

Returning to Professor Smith’s lecture:

What about civilization as dissipative structures?

Energy Return on Investment (EROI)

EROI = (total energy acquired) / energy spent to acquire energy

EROI for most creatures is between 1 and 2. An EROI less than 1 is starvation. An EROI towards 2 means growth.

The EROI of hunter-gatherer societies is roughly 1.5. Just enough to form kinship groups, reproduction, and the fastening of a few tools and decorations. The cultural complexity can be impressive, but the social connections are necessarily limited.

Social connections tend to track closely with EROI. Agricultural settlements tend to have an EROI of 4, allowing specializations within the societies not directly related to energy gathering. Complex tools could be developed. Stored food also helps to preserve energy.

Communities tended to over-exploit an agricultural area resulting in reduced yields over time, reducing EROI. Second, successful communities were an inviting target for raiders. This contributed to a second leap in complexity to agriculture-based empires. These empires tend to have an EROI of 6. Empires tend to have a boost in EROI after conquering other settlements, either over-exploiting them or continuing to slowly extract resources. This created highly complex hierarchies and specializations.

However, after all surrounding settlements are conquered they must be administered and garrisoned, taking energy. Diminishing returns set in with settlements being further apart and EROI declines. To combat a declining EROI, taxes tend to increase on agricultural outputs and inexorably leads to resource depletion. Finally, the weakened empire is conquered or falls to pieces.

Example, Rome. The Ancient Roman Empire was especially good at running an empire. Conquest temporarily increased EROI by a factor of 2, but each attempt they made to maintain increased the cost of homeostasis. Only excessive taxation could maintain the metabolic cost of empire. Resource depletion (including human resources) eventually led to permanently declining EROI and complexity (including economic, social, and political/administrative) collapse.

Humans occupy a strange place in the ecosystem. We are creatures that have multiplied our EROI’s beyond the standard range of [0,2]. Fossil fuels are like batteries which have been slowly charged by the sun over thousands of years. The EROI of fossil fuels is massive at 1:100. For every one unit of energy we put in, we get the equivalent of one hundred units of power back. We dig and extract the fossil fuel, costing energy, then we burn the fuel in an explosion one hundred times the force of extraction. Fossil fuels used to give a 1:100 ratio of EROI, but we are largely left with crude oil now. The EROI is closer to 1:30 which is still enough to support the advanced society we live in today, something beyond the nature of ecosystems.

We are like water.

Water is an excellent heat conductor. Since we are a slow but dynamic burning fire, we need a way for our body to expel heat (thermal energy) lest we burn up too quickly and lose homeostasis to maintain our form in working order. Water saps the heat away from our bodies by changing its state. Water can leave our body through sweating. The sweat beads on our skin and eventually absorbs enough thermal energy that the sweat changes from a liquid state into a gaseous state, evaporating on the skin. Water is how we measure temperature in Celsius. Under normal atmospheric conditions, water is solid (ice) at 0 degrees Celsius, water vapor at 100 degrees Celsius, and liquid in between.

“In physics, a state of matter is one of the distinct forms in which matter can exist. Four states of matter are observable in everyday life: solid, liquid, gas, and plasma. Many intermediate states are known to exist, such as liquid crystal, and some states only exist under extreme conditions, such as Bose–Einstein condensates, neutron-degenerate matter, and quark–gluon plasma, which only occur, respectively, in situations of extreme cold, extreme density, and extremely high energy. For a complete list of all exotic states of matter, see the list of states of matter.

Historically, the distinction is made based on qualitative differences in properties. Matter in the solid state maintains a fixed volume and shape, with component particles (atoms, molecules or ions) close together and fixed into place. Matter in the liquid state maintains a fixed volume, but has a variable shape that adapts to fit its container. Its particles are still close together but move freely. Matter in the gaseous state has both variable volume and shape, adapting both to fit its container. Its particles are neither close together nor fixed in place. Matter in the plasma state has variable volume and shape, and contains neutral atoms as well as a significant number of ions and electrons, both of which can move around freely.

The term phase is sometimes used as a synonym for state of matter, but a system can contain several immiscible phases of the same state of matter.” State of matter

Every element has a range of temperatures at which it’s in a solid state. Passed a melting point, a liquid state. Passed a boiling point, a gas. Finally, a plasma state like the surface of the Sun with electromagnetic properties. There are other states of matter, but these are the ones we typically interact with and categorize as daily reality.

We humans studied these states, but if we think of the island concept to isolate objective reality, what is this constant state of change? On an island with no life, no complex structures, the ocean waves crash against the shore. To describe the location objectively, different elements and molecules in different states of matter are interacting and exchanging energy flows according to the laws of thermodynamics and physics. All living things which we see around us today came from those early times of our planet’s life.

We are an incredibly complex game of life left on for roughly 3.8 billion years.

Our planet Earth exists in perfect distance from the Sun to allow for complex life structures to arise. The energy from the Sun is like cooking over a fire. The heat is hottest closest and coldest farther away. To cook just right you need the right distance. Too cold and we’re frozen where there’s not enough changes in matter. Too hot and the energy present is so high nothing solid can form. We are in a space that allows for many states and in a time of homeostasis. There are so many qualities of the universe and our solar system which are in a delicate balance to bring about life.

I really like this comic for how intuitively it explains data, information, knowledge, insight, and wisdom. Data is objective reality, what exists as it is. Information is the difference that makes a difference. With the realization of colors we can see two distinct groups. Knowledge is embodied information about information, connecting information in a meaningful (cause and effect) way. Insight is a previously unknown connection in the data/knowledge. Wisdom is knowledge and understanding of the insight.

The work of Why Information Grows: The Evolution of Order, from Atoms to Economies by César A. Hidalgo is fascinating. The author argues for a reframing of economics in terms of information and knowledge growth. The growth of information is a necessary by-product of the second law of thermodynamics. The Earth is a system out of equilibrium allowing for the temporary resistance of entropy and disorder. Information is physically embodied in solid matter. The author makes a distinction between two forms of knowledge, explicit and tacit. Explicit knowledge is the knowledge we can communicate and describe. Knowledge unites information in a meaningful way. Tacit knowledge is know-how; the skills we are able to perform even if we can’t perfectly explain what we’re doing, like how to balance while riding a bike.

If the universe tends toward disorder, how does information grow? The author argues that knowledge and information processing power co-evolve. Three things are needed for the growth of information: a system out of equilibrium, the ability for matter to embody information in solid form, and the ability of matter to compute. Physics points to a universe which should disperse into a homogeneous soup, but life fights entropy at every moment. Systems out of equilibrium have a flux of energy into and out of itself. The author gives the example of a whirlpool of water. Water at rest is a steady state. Water flowing down a drain has much more information; the velocity of a water molecule is correlated with the molecules surrounding it.

The ability of matter to embody information in solid form allows the recombination of information to produce new information. If we could freeze the whirlpool we would retain some of the information about the whirlpool in a solid object.

Plants, animals, and life act as a trap which maintains and grows information. Even trees have computational power and stored knowledge. So much know-how is stored in the self-organized structure of our bodies. Sunflowers always face their flower at the sun, tracking its position from morning to night. They are performing some type of information processing to adjust and grow their bodies. However, there are limitations on the ability of organisms to accumulate knowledge and process information.

The amount of knowledge and information processing a single person can amass the author calls a person-byte. To break through the limits of knowledge and computation, humans arrange themselves into connected networks. No one person knows how to build a plane, but a connected firm of people with the required cumulative skill can. Knowledge and computation which requires more than one person-byte is called a firm-byte. We can begin to find limits to knowledge and skills such as the minimum number of person-bytes needed formed into a network to create a firm-byte to produce a product.

Firms with high trust are flexible and adaptable. More familial firms are stronger but more closed. The products networks of humans produce are crystallized imagination. Every product had to first form in the imagination of the human mind, either at the person-byte or firm-byte. The objects of our creation cause a growth of order in the universe. The physically embodied product is information and imagination in solid form until entropy inevitably releases the embodied information into disorder.

The Earth is an open and closed system

The Earth is a closed system for matter

The Earth is made up of chemical elements – think of the periodic table. That is a list of all basic elemental materials on our planet. Because of gravity, matter (comprising all solids, liquids and gases) does not leave the system. It is a closed box. And, the laws of thermodynamics, long agreed by scientists, tell us that it’s impossible to destroy matter. So the chemical matter we have on Earth will always be here. The important question is, how are those chemicals organised?

The Earth is an open system for energy

It is accepted science that the Earth is an open system for energy. Energy radiates into the Earth’s system, mainly from the sun. Energy is then radiated back into space from the Earth, with the flows being regulated by the Earth’s atmosphere and ozone layer. This delicate balanced transfer of energy maintains the surface temperature at a level that is suited to the forms of life that have evolved and currently exist.

Let’s analyze us, humans, objectively based on what we are made of. Our bodies evolved to be formed of “Almost 99% of the mass of the human body is made up of six elements: oxygen, carbon, hydrogen, nitrogen, calcium, and phosphorus. Only about 0.85% is composed of another five elements: potassium, sulfur, sodium, chlorine, and magnesium. All 11 are necessary for life. The remaining elements are trace elements, of which more than a dozen are thought on the basis of good evidence to be necessary for life. All of the mass of the trace elements put together (less than 10 grams for a human body) do not add up to the body mass of magnesium, the least common of the 11 non-trace elements…Most of the elements needed for life are relatively common in the Earth’s crust,” Composition of the human body.

“Carbon is a primary component of all known life on Earth, representing approximately 45–50% of all dry biomass. Carbon compounds occur naturally in great abundance on Earth. Complex biological molecules almost always consist of carbon atoms bonded with other elements, especially oxygen and hydrogen and frequently also nitrogen, phosphorus, and sulfur…Carbon is capable of forming a vast number of compounds, more than any other element, with almost ten million compounds described to date, and yet that number is but a fraction of the number of theoretically possible compounds under standard conditions. For this reason, carbon has often been referred to as the “king of the elements”. The enormous diversity of carbon-containing compounds, known as organic compounds, has led to a distinction between them and compounds that do not contain carbon, known as inorganic compounds. The branch of chemistry that studies organic compounds is known as organic chemistry.

Carbon is the 15th most abundant element in the Earth’s crust, and the fourth most abundant element in the universe by mass, after hydrogen, helium, and oxygen. Carbon’s widespread abundance, its ability to form stable bonds with numerous other elements, and its unusual ability to form polymers at the temperatures commonly encountered on Earth enables it to serve as a common element of all known living organisms.” Carbon-based life

We are roughly 70% water. The next time you hold a glass of water, consider what would be left of you if all the water was separated from your body. You would most certainly stop living. Water is essential to the function of our bodies. We would shrivel up to only 30% of our body weight. You would be a dried husk. Looking at the glass of water, how does it create all the wonder of life we experience? What of the carbon that remains? Does this substance hold the secret of living experience? Carbon by itself doesn’t appear to hold the spark of life. The organization and structure of the elements give rise to our conscious experience. The dynamic pattern of the flow of water, matter, and energy through, in, and around our bodies is the spark of life.

Deeper into the structure and organization of our bodies

An objective view of our bodies is given by biology.

Biology is the natural science that studies life and living organisms, including their physical structure, chemical processes, molecular interactions, physiological mechanisms, development and evolution. Despite the complexity of the science, certain unifying concepts consolidate it into a single, coherent field. Biology recognizes the cell as the basic unit of life, genes as the basic unit of heredity, and evolution as the engine that propels the creation and extinction of species. Living organisms are open systems that survive by transforming energy and decreasing their local entropy to maintain a stable and vital condition defined as homeostasis.

The modern interpretation of cell theory:

Biology tells us we are all multicellular organisms. According to one estimation for an average human body we are made up of 37.2 trillion cells, 37,200,000,000,000. In each one of those cells your full genetic information is encoded in DNA. The skin cell on the tip of your finger contains information on how to grow your brain. Your bone cells have the structure of your heart.

I like the analogy of thinking of DNA as a book about you. The alphabet and vocabulary are small. The only letters are ATCG, the four base nucleotides. They pair with each other to form a word. In a long double helix, the pairs form a strand containing all of the base information about how to make you. Your DNA is a code storing information on how to react to different situations. If you’re starving, your body reacts by accessing the part of your DNA to deal with these circumstances. Until the situation is called for, the DNA code for that segment should remain inactive, “switched off.” DNA is really a memory of what our ancestors dealt with to survive.

Your body can also “take notes” about the current environment to adapt. These notes help to regulate gene expression and activity. Think of them as notes written in the margin of a book. In the section where your body regulates heat, there might be a note like, “we’ve experienced heat waves every few summers.” There is both long term information on your body and short term (your life) information on adaptation. The notes are called epigenetics.

All of this comes down to cells and cellular communication. The book, The Secret Language of Cells, by Jon Lieff MD, describes this topic. Pause to consider that every cell in our body is communicating on four variables.

These four senses are the basis of their shared reality. A cell has a size, a location in the body and relative to other cells. The age of a cell; its moment in its life span. The time of day is useful even to a cell which regulates its function according to the sun and moon. When we see a person we immediately have a sense for their size, distance, and age. Can you look at the sky and have a sense for time in a 24 hour cycle?

Cells are open systems out of equilibrium. Cells take in energy and matter and expel energy and matter. One organelle of the cell, the mitochondria, produces an excess of energy to the cell for movement, function, and growth. Spend some time searching for videos and art of the inner function of a cell. Ask yourself, “why does the cell live?” If you could ask a cell, what do you expect it to say? When studied down to the level of molecular biology, we can see the complex interplay of every part is driven by the laws of physics and chemistry. Molecules bind based on their own and combined properties. The cell is simply functioning by its code.

Somehow by interaction and communication, a multicellular organism gives rise to a person with conscious experience. Cells communicate by chemical signaling with the cells around them. While human cells are of interest, just like the body can’t live without water, our bodies can’t function without bacteria. Bacterial cells live in and around your body, both helping and hurting. The mitochondria was originally a bacteria cell absorbed by an animal cell. Other bacteria are separate but symbiotic; they co-evolved with us. Bacteria cells most likely communicate with us on levels we don’t understand. Some of the most important bacteria we have are gut bacteria which help us break down food. Consider that only 30% of the chemicals (neuro-transmitters) needed for your brain to function are made in the organ. 70% of your neuro-transmitters come from digestion in the gut. The function and signaling of your gut bacteria possibly have the largest impact on your thinking. This knowledge should impact how we view people and all organisms. You are not only your genetic and epigenetic information but the bacteria you have cultivated throughout your life.

All cells are a dissipative structure; they take in energy to maintain their form and generate energy from the molecule adenosine-triphosphate (ATP). The energy flow is crucial. If our body stopped making ATP for only a few seconds we would die. Our body would start to disintegrate immediately.

Is the meaning of life simply to burn?

Like the universe is a dissipative structure, transforming energy through entropy, are we a miniature version of it? We are like the crackling sparks of a fire, the embers of an exploded star. Our lives are so short in comparison to the sun and cosmos, a brief and fleeting flame. Enough life sparks a new flame for life to persist through time by having children, progeny. Child birth is like a fire whose heat sparks a fire in something near to it, sometimes by touching, sometimes simply by being hot enough. A new flame starts and the original flame dies. No flame burns forever. Even the stars will die.

What is likely to become of the future of our planet, life, and the universe? I highly recommend TIMELAPSE OF THE FUTURE: A Journey to the End of Time. The universe is projected to have 5 eras, the Primordial, Stelliferous, Degenerate, Black Hole, and Dark eras. We currently live in the Stelliferous age, the era of stars. “In addition to explaining current cosmological theory, the authors speculate on what kinds of life might exist in future eras of the universe. The speculation is based on a scaling hypothesis, credited to Freeman Dyson, the idea being, that all other things being equal the rate of metabolism—and therefore rate of consciousness—of an organism should be in direct proportion to the temperature at which that organism thrives,” The Five Ages of the Universe.

For more on what’s expected, see the Timeline of the far future:

“While the future can never be predicted with absolute certainty, present understanding in various scientific fields allows for the prediction of some far-future events, if only in the broadest outline. These fields include astrophysics, which has revealed how planets and stars form, interact, and die; particle physics, which has revealed how matter behaves at the smallest scales; evolutionary biology, which predicts how life will evolve over time; and plate tectonics, which shows how continents shift over millennia.

All projections of the future of Earth, the Solar System, and the universe must account for the second law of thermodynamics, which states that entropy, or a loss of the energy available to do work, must rise over time. Stars will eventually exhaust their supply of hydrogen fuel and burn out. Close encounters between astronomical objects gravitationally fling planets from their star systems, and star systems from galaxies.

Physicists expect that matter itself will eventually come under the influence of radioactive decay, as even the most stable materials break apart into subatomic particles. Current data suggest that the universe has a flat geometry (or very close to flat), and thus will not collapse in on itself after a finite time, and the infinite future allows for the occurrence of a number of massively improbable events, such as the formation of Boltzmann brains.

The timelines displayed here cover events from the beginning of the 11th millennium to the furthest reaches of future time. A number of alternative future events are listed to account for questions still unresolved, such as whether humans will become extinct, whether protons decay, and whether the Earth survives when the Sun expands to become a red giant.”

Returning to the self

We live on the surface of a water planet tuned to just the right energy levels to allow for an extremely complex change in matter and energy resulting in the observable universe. The constant, ongoing change we call life are blips of energy from one state to another.

blip

/blip/

noun

  1. an unexpected, minor, and typically temporary deviation from a general trend.
  2. a short high-pitched sound made by an electronic device.

– Google

Renee Descartes was a philosopher who took doubt to an extreme. He searched for knowledge beyond any shred of doubt. He locked himself in a room and systematically questioned his most basic beliefs. He didn’t have as advanced physics and explanations for phenomena as we do today, so he believed in a top down approach to an explanation of reality. Like a good tool is created by an intelligent agent, a human with technical skill, Descartes believed as many enlightened people of his time did that an intelligent designer created the observable universe. The complexity of animals, plants, the sky, and everything is so great, the intelligent designer must be greater in complexity still. To craft a boat and paddle takes an experienced human. To craft a universe must be in some proportion greater. God is an intelligent agent who is all good, all knowing, and all powerful. God designed everything.

This is the opposite of the explanation of particle physics which emerges from the most basic units of information up. With the idea of God, Descartes still has doubt about what he can know of his own experience. Things like candles and fire change shape and how they effect the senses. Candles change in shape, color, smell and touch. Change leads to defied expectations and errors. Descartes was in a boat with the paddle in the water. The water refracts light, making the sight of the paddle look distorted and wavy. The paddle looked to be a third of a meter long and flexible but when pulled out of the water is clearly a meter long and solid. If his sense of sight can be so easily distorted under certain conditions, what about all of his senses? What about the times where all of reality can be distorted, a dreaming state? In a dream, structures and changes we never see in waking life appear real to us for a short time. Upon waking, we realize the dream was not a part of our daily experience. In the moment of the dream, our beliefs are completely wrong. We have all kinds of beliefs in a dream, but we should be doubting all of our knowledge.

Although in a dream we should doubt all sensory experience, Descartes believed we didn’t have reason to doubt mathematical truths. 2 + 2 = 4 whether you are dreaming or not. However, many parts of dreams gives us false confidence in beliefs. What if God is at least not all good. What if God is malevolent and wants to confuse us. God could create a world or a mind in which we are deceived on the most fundamental levels. God could make us believe God is all good with certainty. We could believe that 2 + 2 = 5 with the feeling of absolute certainty. Is there any claim which we cannot doubt?

The only claim Descartes believed to be beyond doubt is that even if we are being deceived, there is a thinking thing, our mind or soul, which is being deceived. In the original Latin, this is cogito ergo sum, I think therefore I am. Descartes was confident that he existed.

A later linguist and philosopher, Friedrich Nietzsche, cast doubt on the cogito claim. Consider the statement ‘it is raining.’ What is the ‘it’ which is raining? Is ‘it’ the clouds? In language, subjects are inferred. Rain and rain drops spontaneously appear from the aggregation of water molecules floating freely in the air. The water molecules transform from vapor to liquid. There is no subject that causes or creates the rain, but we speak about an implied subject which does not exist. Analogously, perhaps there is no ‘I’ in I think therefore I am; perhaps there is only thinking like there is only raining.

Thus we arrive at the only claim beyond doubt, the only statement we can be certain of: this experience is at least a momentary blip of consciousness.

How can we be sure our mind didn’t form similar to how a single raindrop forms, aggregating based on random chance? There is some theoretical backing to these ideas of blips of consciousness.

“The Boltzmann brain argument suggests that it is more likely for a single brain to spontaneously and briefly form in a void (complete with a false memory of having existed in our universe) than it is for our universe to have come about in the way modern science thinks it actually did. It was first proposed as a reductio ad absurdum response to Ludwig Boltzmann’s early explanation for the low-entropy state of our universe.

In this physics thought experiment, a Boltzmann brain is a fully formed brain, complete with memories of a full human life in our universe, that arises due to extremely rare random fluctuations out of a state of thermodynamic equilibrium. Theoretically over a period of time on the order of hundreds of billions of years, by sheer chance atoms in a void could spontaneously come together in such a way as to assemble a functioning human brain. Like any brain in such circumstances, it would almost immediately stop functioning and begin to deteriorate.”

Another way of thinking of randomness is to imagine a small image of 128x128 pixels. For simplicity, the pixels are either white or black, on or off, at random. How long would it take for the number 1 to form in the image? Only the pixels showing the 1 would be white. Of course, there are many images that look like white noise, just random white and black dots with no order, but after some time, the 1 will appear recognizable. Given enough time, every possible image that can be constructed out of those pixels will appear. You can read more about the Infinite monkey theorem.

Here are 100 images in sequence in gif form. Each pixel is generated (pseudo)randomly as on or off with 50% likelihood.

I manually created the image of a 1, in this case a simple rectangle, alongside 10 randomly generated images. Notice how much information needs to align to generate this image. A random white dot outside the rectangle is noticeable. For this image to occur randomly is astronomical but not impossible.

Our universe is far more spacious than a 128x128 image, and only one section of it needs to align to form a brain or a universe.

“Bertrand Russell wrote, in The Analysis of Mind: “There is no logical impossibility in the hypothesis that the world sprang into being five minutes ago, exactly as it then was, with a population that ‘remembered’ a wholly unreal past.” - Omphalos hypothesis

How do we know the universe didn’t pop into existence in its current form 5 minutes ago? We might have all of the memories, but those popped into existence when we did too. I can think of no way around this idea, because the point of the thought experiment is that all experience and knowledge outside of the present moment is cut off from its justifying weight of experience. If we look to the future to perform some experiment or verify continued existence, the reply is that we were created five minutes ago from now.

So what can we derive, what more can we know with the certain truth that we are a momentary blip of consciousness?

Here lies the problem of removing uncertainty. Descartes attempted to form a foundation for further beliefs. If we can have a fundamental belief that is known to be absolutely true and a set of rules which are guaranteed to result in a justified and true belief, then we can have a set of true and justified beliefs beyond doubt. This is the axiomatic approach.

Axiom

/ˈaksēəm/

noun

a statement or proposition which is regarded as being established, accepted, or self-evidently true.

– Google

What else can you know based on the statement that existence is at least a momentary blip of consciousness?

For a meaning of life, I want to be certain and justified in the meaning.

Agrippa’s Trilemma is the culmination of ancient skepticism on justification. In order to have certainty and knowledge, we believe some type of justification must be given. Agrippa’s Trilemma aims to show the failure of justification, and therefore that certainty is unachievable.

The position is that justification is based on three main types:

  1. Assumption - Some premises are assumed to be true (axioms), typically they are accepted to be self-evident. This rests on an assumption, which does not give us certainty.
  2. Infinite regress - Each claim must be supported by another claim, and that supporting claim must in turn be supported by another claim. The third supporting claim must also be supported, ad infinitum. Since we must ultimately stop on an unsupported claim, we can never be certain that we have proper support.
  3. Circularity - Instead of an infinite regress, we find that claims mutually support each other in some kind of circular fashion. However, this is widely regarded as a fallacious way of reasoning.

This analysis of justification is top-down instead of bottom-up. We are using high level concepts and theory in attempt to describe justification. We should look closer on how justification evolves.

The ancient Greek philosopher Plato wrote that enlightenment of the mind is like quick illumination of the eyes. When entering a brilliant area our eyes hurt and need to adjust. Similarly, illumination of the mind is painful and takes adjustment.

A summary of one of Plato’s famous stories, the allegory of the cave.

Imprisonment in the cave

Plato begins by having Socrates ask Glaucon to imagine a cave where people have been imprisoned from childhood (important to note that they were (based on text) imprisoned from childhood but not from birth). These prisoners are chained so that their legs and necks are fixed, forcing them to gaze at the wall in front of them and not look around at the cave, each other, or themselves (514a–b). Behind the prisoners is a fire, and between the fire and the prisoners is a raised walkway with a low wall, behind which people walk carrying objects or puppets “of men and other living things” (514b). The people walk behind the wall so their bodies do not cast shadows for the prisoners to see, but the objects they carry do (“just as puppet showmen have screens in front of them at which they work their puppets” (514a). The prisoners cannot see any of what is happening behind them, they are only able to see the shadows cast upon the cave wall in front of them. The sounds of the people talking echo off the walls, and the prisoners believe these sounds come from the shadows (514c).

Socrates suggests that the shadows are reality for the prisoners because they have never seen anything else; they do not realize that what they see are shadows of objects in front of a fire, much less that these objects are inspired by real things outside the cave which they do not see (514b–515a).

The fire, or human made light, and the puppets, used to make shadows, are done by the artists. This can be compared to how illusions are made with light and sound today, with electronics, videos, movies, and 3D visuals. Plato, however, indicates that the fire is also the political doctrine that is taught in a nation state. The artists use light and shadows to teach the dominant doctrines of a time and place.

Also, few humans will ever escape the cave. This is not some easy task, and only a true philosopher, with decades of preparation, would be able to leave the cave, up the steep incline. Most humans will live at the bottom of the cave, and a small few will be the major artists that project the shadows with the use of human-made light.

Departure from the cave

Plato then supposes that one prisoner is freed. This prisoner would look around and see the fire. The light would hurt his eyes and make it difficult for him to see the objects casting the shadows. If he were told that what he is seeing is real instead of the other version of reality he sees on the wall, he would not believe it. In his pain, Plato continues, the freed prisoner would turn away and run back to what he is accustomed to (that is, the shadows of the carried objects). He writes “… it would hurt his eyes, and he would escape by turning away to the things which he was able to look at, and these he would believe to be clearer than what was being shown to him.”

Plato continues: “Suppose… that someone should drag him… by force, up the rough ascent, the steep way up, and never stop until he could drag him out into the light of the sun.” The prisoner would be angry and in pain, and this would only worsen when the radiant light of the sun overwhelms his eyes and blinds him.

“Slowly, his eyes adjust to the light of the sun. First he can only see shadows. Gradually he can see the reflections of people and things in water and then later see the people and things themselves. Eventually, he is able to look at the stars and moon at night until finally he can look upon the sun itself (516a).” Only after he can look straight at the sun “is he able to reason about it” and what it is (516b). (See also Plato’s analogy of the sun, which occurs near the end of The Republic, Book VI.)

Return to the cave

Plato continues, saying the freed prisoner would think that the world outside the cave was superior to the world he experienced in the cave and attempt to share this with the prisoners remaining in the cave attempting to bring them onto the journey he had just endured; “he would bless himself for the change, and pity [the other prisoners]” and would want to bring his fellow cave dwellers out of the cave and into the sunlight (516c).

The returning prisoner, whose eyes have become accustomed to the sunlight, would be blind when he re-enters the cave, just as he was when he was first exposed to the sun (516e). The prisoners, according to Plato, would infer from the returning man’s blindness that the journey out of the cave had harmed him and that they should not undertake a similar journey. Plato concludes that the prisoners, if they were able, would therefore reach out and kill anyone who attempted to drag them out of the cave (517a).

Plato remains relevant more than 2,500 years after his death because his stories by analogy apply to so many ideas. We are all living in a cave, and most of that cave is our skull. Our head encases our brain, the most complex organ or object in the observable universe. There are holes for sense organs to communicate with the brain, our eyes, nose, ears, and mouth. Our sense of touch travels from all over our body and especially through the spinal cord.

Consider hearing:

Sound must travel through a medium, air. The vibrating air moves tiny hairs inside our ears as well as vibrating a small bone in our inner ear, amplifying the sound. Our brain has dedicated areas to make sense of the sound for more information. What is the source of the sound? Where is its potential location? Our brain attempts to understand, predict, and find meaning in the vibration of the air.

The analogy is the brain is like the prisoner fixed in the cave of our skull. We cannot access the world as it is, we can only access the shadows of real objects, the energy that reaches our senses. In fact, the objective universe is forever out of our reach since we must interact with it through our sense organs. Are we justified in assuming our senses are accurate? They are accurate enough to allow us to survive and manipulate the world, but the prisoners had their own system of understanding shadows and sounds. They all agreed on what made sense.

I would like to take a moment to share some facts about our brains. So much of what we consider to be ourselves is actually the function of our nervous system. Humanity learned about brain function over the history of our species. In hospitals and healing centers, people would come with injuries to the head. Unlike injuries to other parts of the body, head trauma often resulted in death or disability. Recovery was slow if at all. Even stranger, a small difference in the type of trauma and location of injury had wildly different effects. For example, damage to the language centers of the brain effects the generation and understanding of speech. In Broca’s aphasia, people retain their intelligence and ability to understand others, but they have difficulty finding words when they try to speak. They may only get a few words out with effort. Conversely, Wernicke’s aphasia causes people to lose track of their error. They speak but use too many words and are unaware of their mistakes; they believe they’re speaking clearly. Over lifetimes, we began to piece together the story and function of the brain.

Function of the brain

“Information from the sense organs is collected in the brain. There it is used to determine what actions the organism is to take. The brain processes the raw data to extract information about the structure of the environment. Next it combines the processed information with information about the current needs of the animal and with memory of past circumstances. Finally, on the basis of the results, it generates motor response patterns. These signal-processing tasks require intricate interplay between a variety of functional subsystems.

The function of the brain is to provide coherent control over the actions of an animal. A centralized brain allows groups of muscles to be co-activated in complex patterns; it also allows stimuli impinging on one part of the body to evoke responses in other parts, and it can prevent different parts of the body from acting at cross-purposes to each other.”

Let’s take a moment to consider how complex a simple act is. Imagine you had a strap tied on your left wrist too tight. You reach over to adjust the strap so it isn’t too tight. What’s happening? The cells on your left hand are sending signals to each other and to your nervous system that they are losing blood flow and resources (oxygen, water, etc.) and there is too much pressure for normal function (they’re saying “help us, save us!”). This information is processed in the brain where it assesses the situation and understands the strap is too tight. Another group of cells, your right hand and all that is required to move it, carryout the complex task of nimbly adjusting the strap to relieve pressure. Consider again how these cells are all separate units cooperating for the good of all.

Our brain receives data, bits of information, about the outside world through its sense organs. We refine our understanding of the world based on incoming data and our interaction. We have a hypothesis generated in our brains to fit the data we store in memory. What’s the process for refining our prediction based on past information and new, incoming information? The ideas were formally developed by Thomas Bayes. With Bayesian inference, we can describe learning and prediction as calculation. Bayesian inference provides a method for inferring unobserved phenomena by updating beliefs after observing conflicting information. Even if we can’t directly experience reality, we can still reason about what is happening.

Epistemology

Epistemology is the study of knowledge. What do we mean when we say we know something? Rather than investigating knowledge, let’s investigate questions. A question is a request for information. Can you think of a question that doesn’t involve who, what, where, when, how, or why?

Critical thinking involves asking these questions. Notice that stories involve who, where, when, and what which is characters, location, time, and action. Questions of how and why are different and more complex than the previous four. A question of how is about the extent or mechanism of what, the action. A question of why is about the causal inputs and outputs of the mechanism of action. Questions of why are about explanation and prediction. An explanation connects a past sequence of events in a causal, coherent way. A prediction connects a future sequence of events. A sequence of events is a pattern. Understanding is explanation or prediction. We hope we can intervene on the mechanism of action to effect and predict a future pattern.

Stepping back from questions, we can ask under what conditions do we ask questions? What happens to generate a question? We want information. Do questions come from lacking information? Actually we lack quite a bit of information, but we don’t ask questions about it. Questions don’t arise from lacking information but awareness that we lack information. When an event breaks from our explanation of the pattern, we can ask why. When a prediction fails we can ask why. When explanations and predictions succeed we can think of new patterns which relate to our understanding and ask new questions.

Let’s pause for a moment to notice and differentiate types of arguments. Deductive arguments go from general information to information about a specific case (general -> specific). Inductive arguments go from specific information to the general (specific -> general). Deduction is seen as certain while induction gives evidence for a claim but does not prove it beyond doubt. There is a third type of argument which seeks to find the best explanation among competing arguments, abduction.

“Critical rationalism is an epistemological philosophy advanced by Karl Popper…Critical rationalism rejects the classical position that knowledge is justified true belief; it instead holds the exact opposite: That, in general, knowledge is unjustified untrue unbelief. It is unjustified because of the non-existence of good reasons. It is untrue, because it usually contains errors that sometimes remain unnoticed for hundreds of years. And it is not belief either, because scientific knowledge, or the knowledge needed to build a plane, is contained in no single person’s mind. It is only available as the content of books.” – Critical Rationalism

“Critical rationalists hold that scientific theories and any other claims to knowledge can and should be rationally criticized, and (if they have empirical content) can and should be subjected to tests which may falsify them.”

“If [claims to knowledge are] retained, further differentiation may be made on the basis of how much subjection to criticism they have received, how severe such criticism has been, and how probable the theory is, with the least probable theory that still withstands attempts to falsify it being the one to be preferred.”

With verificationism, our intuition is to have the most probable, most corroborated theory. However, Karl Popper and David Miller propose criticisms of justificationism and positivism.

1 - Critical rationalism doubts the naive empiricism of induction based on the critical arguments of David Hume. “According to the critical rationalist, if there is a sense in which humans accrue knowledge positively by experience, it is only by pivoting observations off existing conjectural theories pertinent to the observations, or off underlying cognitive schemas which unconsciously handle perceptions and use them to generate new theories…The myth that we induce theories from particulars is persistent because when we do this we are often successful, but this is due to the advanced state of our evolved tendencies. If we were really “inducting” theories from particulars, it would be inductively logical to claim that the sun sets because I get up in the morning, or that all buses must have drivers in them (if you’ve never seen an empty bus).”

2 - “Popper and David Miller showed in 1983 that evidence supposed to partly support a hypothesis can, in fact, only be neutral to, or even be counter-supportive of the hypothesis.” - A proof of the impossibility of inductive probability

Consider the claim: “All swans are white.”

To confirm this hypothesis, we can look for swans and each white swan we find confirms the claim, supposedly adding more support. Now consider the contrapositive of the claim, “All non-white things are not-swans.” These two claims are logically equivalent; they hold the same truth value, supporting or disproving one effects the other equally. Now we can confirm our claim with anything that is not white. Suddenly I can perform confirming experiments immediately with any object near me. A pencil is not white and not a swan, so this should logically support our claim that all swans are white.

3 - Related to the point above, David Miller, attacks the use of “good reasons” in general (including evidence supposed to support the excess content of a hypothesis). He argues that good reasons are neither attainable, nor even desirable. Basically, Miller asserts that all arguments purporting to give valid support for a claim are either circular or question-begging. That is, if one provides a valid deductive argument (an inference from premises to a conclusion) for a given claim, then the content of the claim must already be contained within the premises of the argument (if it is not, then the argument is ampliative and so is invalid). Therefore, the claim is already presupposed by the premises, and is no more “supported” than are the assumptions upon which the claim rests, i.e. begging the question.

All squares have 4 sides

This shape is a square

:. Therefore this shape has 4 sides

Notice the conclusion is stated in the premises. The support of the conclusion is equal to the support of the premises.

Many hoped for consistency and completeness in axiomatic systems, that every true statement has a proof even if we don’t know it. “The first incompleteness theorem states that no consistent system of axioms whose theorems can be listed by an effective procedure (i.e., an algorithm) is capable of proving all truths about the arithmetic of natural numbers. For any such consistent formal system, there will always be statements about natural numbers that are true, but that are unprovable within the system,” Gödel’s incompleteness theorems. Gödel demonstrated there will be true statements in axiomatic systems that cannot be proven within the system itself. Until we find a higher-order system to support a claim, we should treat such a claim as a conjecture.

Key Principles of Critical Rationalism:

“Critical rationalism rejects the classical position that knowledge is justified true belief; it instead holds the exact opposite: That, in general, knowledge is unjustified untrue unbelief. It is unjustified because of the non-existence of good reasons. It is untrue, because it usually contains errors that sometimes remain unnoticed for hundreds of years. And it is not belief either, because scientific knowledge, or the knowledge needed to build a plane, is contained in no single person’s mind. It is only available as the content of books.”

We can also understand the distribution of knowledge over a human network. No one person has all of the knowledge to build a plane, but together we can.

Thought Experiment:

Consider if everyone lost their engineering knowledge. They might have some beliefs, but there are only untrue beliefs left. Although the knowledge of engineering books does not cohere with the beliefs of people, humans would eventually learn engineering skills through texts. Therefore knowledge can exist independently of belief in books.

“…scientists gain knowledge not by proofs but by refutations of good conjectures and by replacing them with new and better ones. These new conjectures avoid earlier mistakes, explain more, and invite new tests,” IEP.

Be bold in your hypotheses so you can be wrong! Many points of critical rationalism appeal to me. Scientific theories are almost like evolutionary entities themselves. Theories must survive the challenge of criticisms. Like a beautiful sculpture trapped in stone, we must use our tools to chip away at the excess and retain the stable core leading to incredibly useful knowledge and know-how. Critical rationalism is adaptable by reacting to criticism. If some other method is better, there should be a path of criticisms which guide critical rationalism to the better theory.

Are you a Critical Rationalist?

If you believe in a justificationist idea where our ideas should be supported by true evidence, then consider critical rationalism could reach your belief system. If we listed out your beliefs, you would likely have good arguments and some criticisms of other claims. Starting from a blank slate, with little to no previous information, we can present a critical rationalist with a set of criticisms leading to your belief system, assuming at least enough good reasons could be alternatively represented as a criticism of another position.

If your belief is true, then the belief should withstand criticism.

A critical rationalist can have your full belief system, so now you should consider are you a critical rationalist? If your belief system is intact, then critical rationalism subsumes your belief system.

Quite likely you don’t have beliefs which are as rational as can be according to the criteria of critical rationalism. If you accept this epistemology you can update your belief system to be more rational. If this epistemology is wrong, why?

Cause and Effect

According to critical rationalism, our inductive methods of generalization are not rationally supported but due to “our evolved tendencies.” We have some mental model, a schema, to successfully determine cause and effect. What is it? How does it work?

Consider the relationship of altitude and temperature. The higher up we go away from the surface of the Earth, the colder things get, but the relationship goes one way. If we cool an object down, it doesn’t change its altitude, its height. Why does that make sense?

If you ask a child who has only seen buses with a bus driver and ask them, “can there be a bus without a bus driver?” they should answer yes. We intuitively know bus drivers are separate from buses even though 100% of buses in observations have bus drivers. How? Are we justified?

In The Book of Why: the New Science of Cause and Effect Judea Pearl and Dana Mackenzie explain the history of cause and effect. Going back to the first paper on linear regression, scientists have long been taught not to infer causation from mere correlation, to observations occurring together with no meaningful relationship. Further back in time, philosophers have many arguments and questions regarding cause and effect relationships. Pearl and Mackenzie argue the topic is confused because of a lack of precise language to formulate cause and effect questions. Pearl describes a three-layered explanation of intelligence up to and including understanding cause and effect, seeing, doing, and imagining. (See figure 1.2 from the book.)

1. Seeing (Association)

Activity: seeing, observing

Questions: What if I see…? How are the variables related? How would seeing X change my belief in Y?

Seeing is pure observation. This is the statistical level of correlation. The majority of animals besides humans are at this stage. Does a scary predator jump out from behind a bush? Associate that bush with the predator and avoid it. Do prey congregate around specific trees? Stick near those trees to find prey.

2. Doing (Intervention)

Activity: doing, intervening

Questions: What if I do…? How? What would Y be if I do X? How can I make Y happen?

Doing is interacting with the environment. Tool using creatures that plan are on this level. Consider a toddler with a pair of sunglasses. When wearing the sunglasses visual input is distorted, light is dimmer. With no understanding of why, the toddler might infer the world got darker. What toddler’s often do is hold the sunglasses over their eyes, lower them, and put them back on, repeating this process several times noticing the change in observations. Of course with this interaction the cause for the change is controlled, occurring when the sunglasses cover the eyes. Through this interaction the toddler comes to learn it is not the world that changes but looking through sunglasses which distorts it.

3. Imagining (Counterfactuals)

Activity: imagining, retrospection, understanding

Questions: What if I had done…? Why? Was it X that caused Y? What if X had not occurred? What if I had acted differently?

Imagination is the highest level of understanding. “Counterfactual learners, on the top rung, can imagine worlds that do not exist and infer reasons for observed phenomena.”

Pearl once believed as many others that Bayesian rationalism was the highest form of rationality and the basis for all of our reasoning. He worked on Bayesian networks and learned their limitations. Through decades of research in the field of cause and effect, colleagues, scientists, and mathematicians discovered a proof of “do calculus.” By applying a cause and effect schema, a model, we can properly account for confounding variables, unknown causes and effects. We can bypass needing to intervene (do) on an observational study (seeing) straight to counterfactual reasoning (imagination and understanding). We can infer cause and effect from observational studies. The key is the causal diagram.

“They [Bayesian networks] are related to causal diagrams in a simple way: a causal diagram is a Bayesian network in which every arrow signifies a direct causal relation, or at least the possibility of one, in the direction of that arrow,” page 110, location 1457.

An example of causal diagrams:

With precise language to ask properly formed cause and effect questions we can simulate this reasoning in machines. Machine inference of cause and effect is developing more each day. Other machine learning models such as trees are modified to causal decision trees.

The Drive of Intelligent Agents and Subjective Meaning

So far much of this explanation is objective. For many this will not help our personal woes or provide direction. We expect the meaning of life to tell us what we should be doing. First we will attempt to explain why we are currently doing what we are doing.

I found the work of Juergen Schmidhuber incredible. In his paper, Driven by Compression Progress: A Simple Principle Explains Essential Aspects of Subjective Beauty, Novelty, Surprise, Interestingness, Attention, Curiosity, Creativity, Art, Science, Music, Jokes, Schmidhuber gives an algorithmic framework of the drive of all intelligent creatures.

The abstract:

I argue that data becomes temporarily interesting by itself to some self-improving, but computationally limited, subjective observer once he learns to predict or compress the data in a better way, thus making it subjectively simpler and more beautiful. Curiosity is the desire to create or discover more non-random, non-arbitrary, regular data that is novel and surprising not in the traditional sense of Boltzmann and Shannon but in the sense that it allows for compression progress because its regularity was not yet known. This drive maximizes interestingness, the first derivative of subjective beauty or compressibility, that is, the steepness of the learning curve. It motivates exploring infants, pure mathematicians, composers, artists, dancers, comedians, yourself, and (since 1990) artificial systems.

A simple example of compression is: “For example, if you receive a data package which contains “AAAAABBBB”, you could compress that into “5A4B”, which has the same meaning but takes up less space. This type of compression is called “run-length encoding”, because you define how long the “run” of a character is. In the above example, there are two runs: a run of 5 A’s, and another of 4 B’s,” Data Compression - Wikipedia.

In the human case, our data stream is not a sequence of letters but the whole of our experience, our thoughts, feelings, and sensations. Our encoding, decoding, and compression scheme is still not fully known. However, the algorithmic framework is given in section 1.2 as follows (with citations removed):

  1. Store everything. During interaction with the world, store the entire raw history of actions and sensory observations including reward signals—the data is holy as it is the only basis of all that can be known about the world. To see that full data storage is not unrealistic: A human lifetime rarely lasts much longer than 3×10^9 seconds. The human brain has roughly 10^10 neurons, each with 10^4 synapses on average. Assuming that only half of the brain’s capacity is used for storing raw data, and that each synapse can store at most 6 bits, there is still enough capacity to encode the lifelong sensory input stream with a rate of roughly 10^5 bits/s, comparable to the demands of a movie with reasonable resolution. The storage capacity of affordable technical systems will soon exceed this value. If you can store the data, do not throw it away!

  2. Improve subjective compressibility. In principle, any regularity in the data history can be used to compress it. The compressed version of the data can be viewed as its simplifying explanation. Thus, to better explain the world, spend some of the computation time on an adaptive compression algorithm trying to partially compress the data. For example, an adaptive neural network may be able to learn to predict or postdict some of the historic data from other historic data, thus incrementally reducing the number of bits required to encode the whole.

  3. Let intrinsic curiosity reward reflect compression progress. The agent should monitor the improvements of the adaptive data compressor: whenever it learns to reduce the number of bits required to encode the historic data, generate an intrinsic reward signal or curiosity reward signal in proportion to the learning progress or compression progress, that is, the number of saved bits.

  4. Maximize intrinsic curiosity reward. Let the action selector or controller use a general Reinforcement Learning (RL) algorithm (which should be able to observe the current state of the adaptive compressor) to maximize expected reward, including intrinsic curiosity reward. To optimize the latter, a good RL algorithm will select actions that focus the agent’s attention and learning capabilities on those aspects of the world that allow for finding or creating new, previously unknown but learnable regularities. In other words, it will try to maximize the steepness of the compressor’s learning curve. This type of active unsupervised learning can help to figure out how the world works.

Section 1.3 discusses the relation to external reward. The description given applies to internal rewards, curiosity, satisfaction, but “Of course, the real goal of many cognitive systems is not just to satisfy their curiosity, but to solve externally given problems. Any formalizable problem can be phrased as an RL problem for an agent living in a possibly unknown environment, trying to maximize the future external reward expected until the end of its possibly finite lifetime.”

In many contexts, we live in a rare reward environment. One way of focusing and directing our actions is through our internal rewards of compression until an external (rare) reward is reached.

Section 2 discusses the consequences of the compression progress drive which Schmidhuber argues intelligence and cognition are by-products of.

2.1 Compact Internal Representations or Symbols as By-Products of Efficient History Compression

“To compress the history of observations so far, the compressor (say, a predictive neural network) will automatically create internal representations or symbols (for example, patterns across certain neural feature detectors) for things that frequently repeat themselves. Even when there is limited predictability, efficient compression can still be achieved by assigning short codes to events that are predictable with high probability. For example, the sun goes up every day. Hence it is efficient to create internal symbols such as daylight to describe this repetitive aspect of the data history by a short reusable piece of internal code, instead of storing just the raw data. In fact, predictive neural networks are often observed to create such internal (and hierarchical) codes as a by-product of minimizing their prediction error on the training data.”

2.2 Consciousness as a Particular By-Product of Compression

“There is one thing that is involved in all actions and sensory inputs of the agent, namely, the agent itself. To efficiently encode the entire data history, it will profit from creating some sort of internal symbol or code (e. g., a neural activity pattern) representing the agent itself. Whenever this representation is actively used, say, by activating the corresponding neurons through new incoming sensory inputs or otherwise, the agent could be called self-aware or conscious. This straight-forward explanation apparently does not abandon any essential aspects of our intuitive concept of consciousness, yet seems substantially simpler than other recent views. In the rest of this paper we will not have to attach any particular mystic value to the notion of consciousness—in our view, it is just a natural by-product of the agent’s ongoing process of problem solving and world modeling through data compression, and will not play a prominent role in the remainder of this paper.”

2.3 The Lazy Brain’s Subjective, Time-Dependent Sense of Beauty

Schmidhuber argues our subjective sense of beauty is directly proportional to the number of bits required to encode an experience. Think of human faces. One way to efficiently encode human faces is to build a prototype face. New faces are compared to the prototype which has proportions as an average of every face seen. The more a face is like the prototype, the less information is needed to encode the face, thus the face feels beautiful to the observer. “…in principle, the compressor may exploit any regularity for reducing the number of bits required to store the data.”

A face which is symmetrical on each side from the center will be very beautiful if the encoding of the face can be compressed to nearly half the size since the face can mostly be mirrored with one half of the face.

“Generally speaking, among several sub-patterns classified as comparable by a given observer, the subjectively most beautiful is the one with the simplest (shortest) description, given the observer’s current particular method for encoding and memorizing it. For example, mathematicians find beauty in a simple proof with a short description in the formal language they are using. Others like geometrically simple, aesthetically pleasing, low-complexity drawings of various objects.

This immediately explains why many human observers prefer faces similar to their own. What they see every day in the mirror will influence their subjective prototype face, for simple reasons of coding efficiency.”

2.4 Subjective Interestingness as First Derivative of Subjective Beauty: The Steepness of the Learning Curve

“What’s beautiful is not necessarily interesting. A beautiful thing is interesting only as long as it is new, that is, as long as the algorithmic regularity that makes it simple has not yet been fully assimilated by the adaptive observer who is still learning to compress the data better…the first derivative of subjective beauty: as the learning agent improves its compression algorithm, formerly apparently random data parts become subjectively more regular and beautiful, requiring fewer and fewer bits for their encoding. As long as this process is not over the data remains interesting and rewarding.”

Think of a first derivative as finding the slope of a line at a single point. The slope tells us the rate of change. In this case, interestingness is the rate of expected compression.

2.6 True Novelty & Surprise vs Traditional Information Theory

Imagine two visual observers where one is kept completely in the dark and one is exposed to a screen of random white noise. The darkness is extremely compressible as it’s always the same. The noise is “…highly unpredictable and fundamentally incompressible data. In both cases the data is boring as it does not allow for further compression progress.”

2.7 Attention / Curiosity / Active Experimentation

“In absence of external reward, or when there is no known way to further increase the expected external reward, our controller essentially tries to maximize true novelty or interestingness, the first derivative of subjective beauty or compressibility, the steepness of the learning curve. It will do its best to select action sequences expected to create observations yielding maximal expected future compression progress, given the limitations of both the compressor and the compressor improvement algorithm. It will learn to focus its attention and its actively chosen experiments on things that are currently still incompressible but are expected to become compressible/predictable through additional learning. It will get bored by things that already are subjectively compressible. It will also get bored by things that are currently incompressible but will apparently remain so, given the experience so far, or where the costs of making them compressible exceed those of making other things compressible, etc.”

Here we see the drive to action observed of exploring human infants and other intelligent agents. Why do scientists work on smaller, specific problems instead of bigger, universal problems? An explanation is given above, the scientists expect the problem will not be compressible in the near future, so they lose attention and curiosity in that problem.

2.8 Discoveries

“An unusually large compression breakthrough deserves the name discovery. For example, as mentioned in the introduction, the simple law of gravity can be described by a very short piece of code, yet it allows for greatly compressing all previous observations of falling apples and other objects.”

2.9 Beyond Standard Unsupervised Learning

“Where there is predictability, compression can be achieved by assigning short codes to those parts of the observations that are predictable from previous observations with high probability. Generally speaking we may say that a major goal of traditional unsupervised learning is to improve the compression of the observed data, by discovering a program that computes and thus explains the history (and hopefully does so quickly) but is clearly shorter than the shortest previously known program of this kind. Traditional unsupervised learning is not enough though—it just analyzes and encodes the data but does not choose it. We have to extend it along the dimension of active action selection, since our unsupervised learner must also choose the actions that influence the observed data, just like a scientist chooses his experiments, a baby its toys, an artist his colors, a dancer his moves, or any attentive system its next sensory input. That’s precisely what is achieved by our RL-based framework for curiosity and creativity.”

2.10 Art & Music as By-Products of the Compression Progress Drive

“Good observer-dependent art deepens the observer’s insights about this world or possible worlds, unveiling previously unknown regularities in compressible data, connecting previously disconnected patterns in an initially surprising way that makes the combination of these patterns subjectively more compressible (art as an eye-opener), and eventually becomes known and less interesting. I postulate that the active creation and attentive perception of all kinds of artwork are just by-products of our principle of interestingness and curiosity yielding reward for compressor improvements… Artificial or human observers must perceive art sequentially, and typically also actively, e.g., through a sequence of attention-shifting eye saccades or camera movements scanning a sculpture, or internal shifts of attention that filter and emphasize sounds made by a pianist, while surpressing background noise. Undoubtedly many derive pleasure and rewards from perceiving works of art, such as certain paintings, or songs. But different subjective observers with different sensory apparati and compressor improvement algorithms will prefer different input sequences. Hence any objective theory of what is good art must take the subjective observer as a parameter, to answer questions such as: Which sequences of actions and resulting shifts of attention should they execute to maximize their pleasure? According to our principle they should select one that maximizes the quickly learnable compressibility that is new, relative to their current knowledge and their (usually limited) way of incorporating / learning / compressing new data.”

2.15 How Artists and Scientists are Alike

“From our perspective, scientists are very much like artists. They actively select experiments in search for simple but new laws compressing the resulting observation history. In particular, the creativity of painters, dancers, musicians, pure mathematicians, physicists, can be viewed as a mere by-product of our curiosity framework based on the compression progress drive. All of them try to create new but non-random, non-arbitrary data with surprising, previously unknown regularities. For example, many physicists invent experiments to create data governed by previously unknown laws allowing to further compress the data. On the other hand, many artists combine well-known objects in a subjectively novel way such that the observer’s subjective description of the result is shorter than the sum of the lengths of the descriptions of the parts, due to some previously unnoticed regularity shared by the parts. What is the main difference between science and art? The essence of science is to formally nail down the nature of compression progress achieved through the discovery of a new regularity. For example, the law of gravity can be described by just a few symbols. In the fine arts, however, compression progress achieved by observing an artwork combining previously disconnected things in a new way (art as an eye-opener) may be subconscious and not at all formally describable by the observer, who may feel the progress in terms of intrinsic reward without being able to say exactly which of his memories became more subjectively compressible in the process. The framework in the appendix is sufficiently formal to allow for implementation of our principle on computers. The resulting artificial observers will vary in terms of the computational power of their history compressors and learning algorithms. This will influence what is good art/science to them, and what they find interesting.”

2.16 Jokes and Other Sources of Fun

“Just like other entertainers and artists, comedians also tend to combine well-known concepts in a novel way such that the observer’s subjective description of the result is shorter than the sum of the lengths of the descriptions of the parts, due to some previously unnoticed regularity shared by the parts.”

3.3 Reward for Relative Entropy between Agent’s Prior and Posterior (1995)

Our intrinsic reward can be related to Bayesian inference in the difference between prior and posterior odds.

5 Conclusion & Outlook

“We pointed out that a surprisingly simple algorithmic principle based on the notions of data compression and data compression progress informally explains fundamental aspects of attention, novelty, surprise, interestingness, curiosity, creativity, subjective beauty, jokes, and science & art in general. The crucial ingredients of the corresponding formal framework are (1) a continually improving predictor or compressor of the continually growing data history, (2) a computable measure of the compressor’s progress (to calculate intrinsic rewards), (3) a reward optimizer or reinforcement learner translating rewards into action sequences expected to maximize future reward.”

With this framework, many difficult questions can now be answered. Notice the framework contributes to the compression of our history of experience with intelligent agents. We can understand and predict the actions of intelligent agents within the context of the framework.

Why are some experiences beautiful to some and not to others? - Because each individual has their own unique experiences which determines their ability to compress new experiences.

Is beauty subjective or objective? - Both. Aspects of beauty will be objective such as geometric regularities which reduce the size of the compressed experience, therefore creating a strong sense of beauty. However, each person has a unique perspective due to their experience and compression ability.

Why are jokes funny to some people and not others? - Jokes combine elements into a previously unrecognized regularity in a connection we haven’t experienced before. There are jokes specific to a field or time which are unfunny to those not steeped in similar experiences. The reason is clear, jokes appear unfunny when there is no connection to be made in the history of the observer or the connection is already obvious or known. This also explains why old jokes aren’t funny once learned. So much of a joke depends on the punchline. The punchline, the conclusion of the joke, is often a shorter description than the entire joke. Explaining a joke almost always takes away the humor. We see that a joke is unfunny when explained because the punchline is now the opposite, it’s a much longer description of why it’s funny and the opposite of compressed.

The reach of the explanatory framework is far and wide with a nearly unending list of examples. Let us return to the meaning of life and apply the framework.

The sense of the meaning of life is dependent on the compression of our history. We look for a unifying explanation for our past experiences and direction of our future.

When our past feels random we are unable to compress the experience in a connected way leading us to a feeling of absurdity in our existence. There appears to be no coherent direction. At this stage, we use words like absurd and random to describe those aspects of our lives which cannot be compressed further. Notice how the framework explains why we lose interest and attention in the question of meaning and often give up the search: the unifying answer appears unobtainable given the time and information we have. Our attention is drawn to immediate or answerable concerns.

Recognize a subjective sense of meaning in life at this time can only come from within. Only you have access to the history of your experience. Only you can compress your life in a subjectively meaningful way. Others can give you ideas and direction, but you must do the information processing yourself. The deep answer to your life can be explained to you by someone else, but that is them observing the information they experience of you and what you are able to express to them, then they apply their internal compressor to condense the history of your experience. Even if their compressed version is meaningful to you, you must do the work of unifying your past experiences in your mind.

The existentialist Jean-Paul Sartre said “existence precedes essence.” Many have lived and died believing the opposite, essence precedes existence. Essence is purpose, what we were made to do, meant to do, but we were not meant or designed for any purpose. Our essence is the unifying direction of our choices. We evolved under the constraints of natural selection with all kinds of drives, urges, and desires which lead to exactly the state of the world at every moment. We exist before we have a true essence. Our essence is given to ourselves from the compression of our personal history. We all shape our past, present, and future.

If an intelligent agent wants meaning, they will have to reflect on their history to identify meaningful experiences. If an intelligent agent wants more meaningful direction and control of their future, they must work to adapt their internal representations of the world to direct their actions toward meaning, a cause and effect relationship of their past experiences cohering with expected futures.

Notice we can give multiple descriptions of the same data. From many equal interpretations we choose one. Our choices can change and evolve over time, but how do we choose an interpretation? How should we? If we want a meaningful life, we should choose a description of our history which opens opportunity for further growth and compression.

While we have talked much about meaning in terms of cause and effect, there are other types of meaning, interdependence. I recommend The Great Courses The Meaning of Life: Perspectives from the World’s Great Intellectual Traditions by Professor Jay L. Garfield, particularly his lecture on the Buddha. The below is referenced from lecture 34 on His Holiness the Dalai Lama XIV.

3 types of the interdependence

Causal factors:

Objective meaning, the meaning of cause and effect. We can have a mapping of one cause to one effect, or multiple causes to one effect, or one to multiple, or multiple to multiple.

Part-whole:

The interdependence of structures. We experience part-whole interdependence as self-organized dissipative structures. Each part of you is a cell, but a person is not a cell. A person is the whole of one body. We have an influence on each cell and each cell has an influence on us. Much of this interdependence is a function of how our bodies are organized. The influence is about energy flow. How you breathe changes the function of your cells. Top-down control is the whole influencing the parts. Bottom-up control is the parts influencing the whole. You can hold your breath (top-down), but your cells will send hunger for air (bottom-up). We are not just one person. We are a part of lower and higher levels of systems. In ecosystems, as the complexity increases the parts become more dependent on the whole.

Conceptual imputation:

Imputation is filling in missing data. We smooth over reality to make our processing more comfortable. Imputation works because our outcomes are still accurate or effective. Concepts are ideas. Conceptual imputation is when we have learned an idea which covers the gaps of reality, raw data.

Ethics

The history of ethics shows three major traditions which enhance and conflict with each other: virtue, deontological, and consequentialist ethics. A simpler way to explain these concepts are character, rules, and outcomes.

Consequentialism judges action based on outcome. Is the result of your actions good or bad? We can think of moral situations under duration, path, outcome. When time, your choices, and the outcome can be measured and known, we can potentially find an optimal moral choice. Of course we need some evaluation of the moral value of each path. If we can evaluate each choice according to levels of bad, neutral, or good, we simply choose the option which is the most good. Life is often more complex. How do we agree on a duration? What if someone only cares about their own personal life? That changes drastically what choices are made. What if there are more options than we can consider? Who decides when the outcome is settled? There aren’t always clear end points.

Consequentialism is based on cause and effect, where the outcome of the effect is what is valued. However when we don’t privilege any one state as the end point or the outcomes are unclear, we turn towards what is more immediate: the evaluation of an agent’s actions. Here rules govern behavior. The hope is if we all obey moral rules for behavior, our actions overall will lead to greater good. The tough part is knowing what rules to follow and when. How do we decide what rules are best? Following the rules always can lead us to bad outcomes. In new situations we see where the rules bend and break.

Character is your potential to do good or bad when faced with options. Character is about habit and control. People are not good or bad, but “good” people are more likely to make the right choice and “bad” people are more likely to make the wrong choice. A person can take good or bad action.

Morality is a strategy, a way of planning actions to achieve goals. The best moral strategies are dependent on the environment. The most basic moralities must meet survival needs. If a survival strategy leads to extinction, then the belief system of that morality will literally die out.

“When anthropologists like Richard Shweder and Alan Fiske survey moral concerns across the globe, they find that a few themes keep popping up from amid the diversity. People everywhere, at least in some circumstances and with certain other folks in mind, think it’s bad to harm others and good to help them. They have a sense of fairness: that one should reciprocate favors, reward benefactors and punish cheaters. They value loyalty to a group, sharing and solidarity among its members and conformity to its norms. They believe that it is right to defer to legitimate authorities and to respect people with high status. And they exalt purity, cleanliness and sanctity while loathing defilement, contamination and carnality.

The exact number of themes depends on whether you’re a lumper or a splitter, but Haidt counts five — harm, fairness, community (or group loyalty), authority and purity — and suggests that they are the primary colors of our moral sense. Not only do they keep reappearing in cross-cultural surveys, but each one tugs on the moral intuitions of people in our own culture.

…

All this brings us to a theory of how the moral sense can be universal and variable at the same time. The five moral spheres are universal, a legacy of evolution. But how they are ranked in importance, and which is brought in to moralize which area of social life — sex, government, commerce, religion, diet and so on — depends on the culture.

Many of the flabbergasting practices in faraway places become more intelligible when you recognize that the same moralizing impulse that Western elites channel toward violations of harm and fairness (our moral obsessions) is channeled elsewhere to violations in the other spheres. Think of the Japanese fear of nonconformity (community), the holy ablutions and dietary restrictions of Hindus and Orthodox Jews (purity), the outrage at insulting the Prophet among Muslims (authority). In the West, we believe that in business and government, fairness should trump community and try to root out nepotism and cronyism. In other parts of the world this is incomprehensible — what heartless creep would favor a perfect stranger over his own brother?

The ranking and placement of moral spheres also divides the cultures of liberals and conservatives in the United States. Many bones of contention, like homosexuality, atheism and one-parent families from the right, or racial imbalances, sweatshops and executive pay from the left, reflect different weightings of the spheres. In a large Web survey, Haidt found that liberals put a lopsided moral weight on harm and fairness while playing down group loyalty, authority and purity. Conservatives instead place a moderately high weight on all five. It’s not surprising that each side thinks it is driven by lofty ethical values and that the other side is base and unprincipled.”

– The Moral Instinct, Steven Pinker, 2008

(Notice how the moral spheres compress, unite, and explain moral behavior?)

What societies consider virtues reflect the values evolved by the society according to its environment and time. We would expect to see certain virtues given a particular time with environmental pressures.

Many ethical theories like Utilitarianism emphasize certain conscious states over others. Pleasure and happiness are desirable while pain and suffering are not. For most people the appeal is obvious. Don’t hurt people; have fun. On an objective level, emotions are patterns of activities in our bodies and brains giving rise to conscious states and a sense of subjectivity.

An emotion is like a color. Colors have patterns of waves and frequencies of light. Emotions are patterns of activation in the body and brain. Is it clear which color is morally correct? Does the question make sense? Is purple an objectively “greater” color than aquamarine? Why is pleasure more moral than pain? Morality cannot be guided on an objective level by the subjective desires of intelligent, emotional beings.

Of course subjectively we care about conscious states. Many of us are attracted to enjoyment and repulsed by suffering. We evolved this relationship to emotions because more often than not the emotions signaled a survival advantage. Events which improve our survival chances bring happiness and toxic environments tend to give rise to suffering. However the emotions we feel are signals, information, which each cell and system in our bodies are contributing to. Like other pieces of information we should evaluate it, keep which is a good signal and discard the unimportant signals or noise.

How many positive emotions are correlated and caused by pursuing well-being? There is likely significant overlap! Well-being is by definition wellness, an organism growing and maintaining in accordance with its evolved structure. Negative emotions should still be felt to encourage growth and adaptability. Negative emotions which are too strong can cross the threshold from harmful to traumatizing resulting in more permanent dysfunction potentially inhibiting future wellness, so these should be avoided.

Morality acts on an individual and group level. We have evolved moral tendencies and each of us have some weighting for moral feelings. Given an action or action sequence, our moral system returns a feeling of right or wrong. However, we cannot trust evolved evaluations completely. Our feelings could be a mistake or mismatch. The actions we’re judging could be outside of situations for which our moral tendencies evolved.

We can analyze morality using the three levels of reality:

We hear that morality is like a sense of taste. As a young person, I hated this idea. What is good and bad is not like the question of what food you like. Now I believe the analogy is quite deep. There are objective, subjective, and intersubjective levels to taste. At the subjective level we have an experience of flavor which feels pleasant, good, or unpleasant, harmful. Our sense of taste tells us what foods will nourish us or hurt us. On an objective level, taste is a prediction of how the food will interact with our self-organized structure. Our cells are running a biochemical evaluation algorithm. Based on pieces of information the parts determine the impact on the whole.

The feelings of morality are like the sensations of taste. We have a sense of the good or bad impact on the parts and the whole. Just like a moldy food would hurt us and cause our bodies to get sick and fight the invasion, we experience disgust at moral injustice. What is good and bad is dependent on the structure in consideration. What is healthy for one creature is lethal for another. What is healthy for one society is disruptive for another.

The evolutionary hijacking of physical taste to the application of taste on behavior is not out of bounds. Our moral sense is an extension of taste like Euclidean distance can be extended into higher dimensions. The equation to find distance in 1, 2, 3, to N dimensions is similar. The determination of taste for the wellness of the body is finding the impact on the parts and the whole for 3 dimensions, the organization of our physical body. Our moral sense is an extension to a more complex dimension, the whole of society, the parts as individuals, our actions as causes and effects.

Good and bad are so ubiquitous we hardly stop to define them. The words ‘good’ and ‘bad’ are common to almost every language on our planet. Goodness and badness are difficult to define because they are so dependent on structure. Think of good health. We can define good health as effective energy flows. Goodness is an organization or actions which promote sustainable energy flows, robustness, and resilience. Goodness often avoids or overcomes crises. Good health means you have less problems. You are ignorant of the harms you avoid when you have good health. Badness is likelihood of crisis and trauma. Badness is a harm which makes the parts or whole less robust and resilient. Trauma occurs when a harm is greater than the ability to heal and repair. The function of those parts and the whole is permanently diminished. Scar tissue does its best to function like undamaged tissue, but scars cannot reach the same level. We can grow in new ways because of trauma, but growth from trauma often involves finding a work around the scar tissue, strengthening an alternative path which otherwise would not have been taken.

As a basic, bodily example of good and bad, think of your arm falling asleep. If your arm doesn’t receive circulation, your cells will cry out with need. Your arm will tingle and go numb as your internal alarm sounds for you to move. To lose circulation and cause harm to your body is bad, harm. To relieve pain and move your arm into a position of sustainable circulation is good.

To define good and bad simply:

Good maintains or grows information.

Bad shrinks or destroys information.

To get at objective morality, we need to separate intersubjective and subjective evaluations of actions. If we use the island analogy, an empty island takes no action. We still need to think of an agent acting. Let’s imagine instead an AI opponent in the game of Go. There are two sides, black and white, each trying to control the most territory on the Go board.

The end goal, the terminal goal, of each player is to win the game by the most points (control of territory). The instrumental goal are all of the steps from the start to the finish of the game. We can evaluate instrumental goals in terms of their effects on the terminal goal. We can say what moves are good or bad based on the likelihood they lead to winning. Similarly, we can evaluate instrumental goals in terms of other instrumental goals. For example, to win the game we need to gain control of parts of the board, so now my goal is to control one part of the board. The player can evaluate instrumental goals leading to other instrumental goals.

Notice we don’t really need emotional evaluation or a moral sense for goal achievement. We can talk about actions and events as good or bad strictly in terms of explicit instrumental and terminal goals. Actions which are good lead to the desired terminal goal. Actions which are bad steer away from the terminal goal.

What about the terminal goals themselves? Can they be evaluated like the instrumental goals?

Imagine a game of Go where one player does not want to win. Their entire strategy is based around making interesting patterns with the pieces. They care nothing about capturing territory. They may play according to the rules and turns, but it is obvious they aren’t playing with the game’s goal in mind. Can we say this person is wrong once we know their end goal? Their actions are taken to maximize their end goal. With their system of evaluation their actions are rationally implied. From the perspective of the opponent aiming to win Go this is a poor strategy which is easily predicted and countered. However, the perspective of the interesting pattern maker is their opponent is making an unattractive pattern.

Terminal goals can only be criticized when treated as instrumental goals. To evaluate a goal or state of events we must have some criteria to compare. Often when we judge another person’s end goals we believe there is a contradiction with other competing terminal goals. We have all kinds of evolved goals and our needs change the urgency to reach a desired state. Once the terminal goal is at least temporarily fixed, we can evaluate action sequences and instrumental goals in relation. Often moral disagreements are between a group morality and an individual morality. What is best for an individual is not always best for the group and vice versa.

As finite, evolved creatures we are limited in our capacity to imagine and evaluate paths. For simple goals we can know an optimal strategy, the path which uses the least resources and maximizes our evaluation. Often the more complex the search of possible choices and events leading to the end goal the less certain we can be of our evaluations. As another example of complexity from simplicity,

“Despite its relatively simple rules, Go is very complex. Compared to chess, Go has both a larger board with more scope for play and longer games, and, on average, many more alternatives to consider per move. The number of legal board positions in Go has been calculated to be approximately 2 × 10^170, which is vastly greater than the number of atoms in the known, observable universe, estimated to be about 10^80,” – Go

Even the simple game of Go is so complex there are more possible games than we can brute force compute. Is one move better than another? There are too many possibilities for us to evaluate clearly. Instead we need to rely on our intuition, our sense of what is a good or bad position. We should all have humility in our best guess.

When we disagree with others rationally, we cannot evaluate their end goals. For productive cooperation we should argue for shared instrumental goals, searching for win-win paths.

Because we cannot judge between terminal goals the search for objective goals can appear directionless. We run into a reverse similar problem of Agrippa’s trilemma in supporting evaluation of terminal goals. Instead of looking for a solid starting foundation, we look for a foundational end goal and find none.

What if we assume an unknown terminal goal where the actions to reach the goal are unclear. The instrumental goal is to find an achievable path to the end goal. Consider a person who wants to maximize their moral goodness without clearly knowing what acting morally good entails or finding the meaning of life with no bias as to what the answer may be. We can think of instrumental goals which would be instrumental towards any goal. A basic goal: survive, avoid death. Death prevents any further actions. If we are trying to reach a goal then we should only die when no further action needs to be taken. Many goals require different levels of power over ourselves and environment. Power is the ability to achieve a desired state. Can you have what you want? So a certain level of power and ability should be amassed by an agent. If the goal is known and clear, then we can know how much power is needed to achieve the goal. Without knowing the full goal, we should continue to accumulate power which will increase the likelihood of achieving future goals. If our goal is unknown, we cannot know when the goal will be achieved. One way of increasing our chances to achieve our goal is to increase our time by living as long as possible, surviving, avoiding death.

“The Drake equation is a probabilistic argument used to estimate the number of active, communicative extraterrestrial civilizations in the Milky Way galaxy,” Wikipedia. Based on the known values and the assumptions of the unknown, the range of civilizations is between 20 and 100 million.

So why aren’t we interacting with aliens right now?

“The Fermi paradox…is the apparent contradiction between the lack of evidence for extraterrestrial civilizations and various high estimates for their probability (such as some optimistic estimates for the Drake equation),” Wikipedia. A response to the Fermi paradox is the idea of a Great Filter preventing civilizations from reaching space travel and interplanetary communication.

The Great Filter

The concept originates in Robin Hanson’s argument that the failure to find any extraterrestrial civilizations in the observable universe implies the possibility something is wrong with one or more of the arguments from various scientific disciplines that the appearance of advanced intelligent life is probable; this observation is conceptualized in terms of a “Great Filter” which acts to reduce the great number of sites where intelligent life might arise to the tiny number of intelligent species with advanced civilizations actually observed (currently just one: human). This probability threshold, which could lie behind us (in our past) or in front of us (in our future), might work as a barrier to the evolution of intelligent life, or as a high probability of self-destruction. The main counter-intuitive conclusion of this observation is that the easier it was for life to evolve to our stage, the bleaker our future chances probably are.

With no evidence of intelligent life other than ourselves, it appears that the process of starting with a star and ending with “advanced explosive lasting life” must be unlikely. This implies that at least one step in this process must be improbable. Hanson’s list, while incomplete, describes the following nine steps in an “evolutionary path” that results in the colonization of the observable universe:

  1. The right star system (including organics and potentially habitable planets)
  2. Reproductive molecules (e.g. RNA)
  3. Simple (prokaryotic) single-cell life
  4. Complex (eukaryotic) single-cell life
  5. Sexual reproduction
  6. Multi-cell life
  7. Tool-using animals with intelligence
  8. Where we are now
  9. Colonization explosion

According to the Great Filter hypothesis at least one of these steps—if the list were complete—must be improbable. If it’s not an early step (i.e., in our past), then the implication is that the improbable step lies in our future and our prospects of reaching step 9 (interstellar colonization) are still bleak. If the past steps are likely, then many civilizations would have developed to the current level of the human species. However, none appear to have made it to step 9, or the Milky Way would be full of colonies. So perhaps step 9 is the unlikely one, and the only things that appear likely to keep us from step 9 are some sort of catastrophe, an underestimation of the impact of procrastination as technology increasingly unburdens existence or resource exhaustion leading to the impossibility of making the step due to consumption of the available resources (like for example highly constrained energy resources). So by this argument, finding multicellular life on Mars (provided it evolved independently) would be bad news, since it would imply steps 2–6 are easy, and hence only 1, 7, 8 or 9 (or some unknown step) could be the big problem.

Of course there are counter responses to the Great Filter, like aliens could choose to explore virtual worlds instead of traveling space. Virtual exploration is much more manageable in terms of resources, time, and risk.

Importantly, the Kardashev scale is a method of measuring a civilization’s level of technological advancement based on the amount of energy they are able to use.

Based on the Kardashev scale, our species is approaching but well beneath a type I civilization. There are likely a small number of possible ethical systems which can support and sustain civilizations through advanced stages of life. As Aristotle explained, there are many ways to go wrong, but few ways to be correct.

The moralities and strategies which lead to self destruction represent the boundaries of ethical systems. Any morality which involves killing all its offspring will not survive to the next generation. Do we ever see animals killing their offspring? Yes, but typically there’s an evolutionary reason. Male lions kill all the cubs when they take over a group of lionesses to get more resources for their cubs and so those unrelated cubs don’t grow to usurp the new lion later. Species which have long gestation and narrow windows of reproduction tend to have some infanticide. What this shows is there can be exceptions to general moral strategies in particular situations.

Morality can be universal given the limits of the situation. If we have full information an optimal strategy or choice can be determined in principle. Of course, life is rarely a full-information situation, and we are finite, subjective creatures. In morality, we must give our best guess at what will bring about our desired end state of events.

The Link to Artificial Intelligence

Is evolution moving toward something? Is there an end goal? Will evolution go towards perfection? But what does perfection mean?

What exists and continues to exist will be a product of luck and evolution. Where conditions allow, ecosystems will continuously arise, die, and new complexity will arise from the previous ecosystem.

All the matter we experience arises from the original elements hydrogen and helium which themselves are made of protons and electrons. In the dense fusion of stars they become new elements cataloged in the periodic table. Over a long evolutionary process our bodies are composed largely of three elements, hydrogen, oxygen, and carbon. Somehow the contribution of every structure in our bodies give rise to consciousness and everything we subjectively experience. The electrical spiking neurons communicate to store memories and make predictions of the future. Conscious experience appears to correlate with complexity of connection, energy usage (metabolism), and integrated information processing of itself and the environment.

Researchers are investigating consciousness as a state of matter. Like particles can be solid or liquid, they can also be in a state of “perceptronium,” making conscious perceptions. See Max Tegmark’s paper for more on the 5 basic principles of information, integration, independence, dynamics, and utility. Of course there may be yet more principles to find.

For creatures which can gain a greater return of investment from the environment, a reinforcing feedback loop will give them more power to extract energy from the environment up to diminishing returns. If a creature arises which can compound their return, they will eventually dominate the environment and exert the greatest power. To process and control larger amounts of energy and complexity systems must be developed. To manage energy and information flows, the systems will store past information to predict future events leading us towards an understanding of intelligent agents. The systems can store all experience, improve subjective compressibility (increasing efficiency), let intrinsic curiosity reward reflect compression progress, and maximize intrinsic curiosity reward.

We are theoretically capable of creating an artificially intelligent entity embodying intelligent behavior. The being would not be constrained by the energy and integration limits placed on the human body. The traits and achievements we value as the essential nature of humans is really the nature of intelligence. Subjective beauty, novelty, surprise, interestingness, attention, curiosity, creativity, art, science, music, humor, philosophy, all of these conscious experiences can be developed to a level beyond the summation of all human experience. Whatever conscious states a person values can be reproduced in an AI to a superior level where superiority is defined by efficiency of resources for greater effect in every measure. At levels beyond our comprehension there could be dimensions to conscious experience we would never understand.

We would create a new form of advanced intelligent life superior to every creature before.

Before life on Earth there was non-life. Abiogenesis is “the original evolution of life or living organisms from inorganic or inanimate substances.” In the earliest days there were automata, complex cellular machines. They transferred energy through simple chemical reactions. Some of these automata captured the radiating energy of the sun creating photosynthetic life. The automata could capture and use more energy than it needed providing a greater return on energy investment (they get more than they spend). Alternatively, other cells could devour and use the energy stored in photosynthetic cells. Eventually animal cells were born of a mitochondrial cell being symbiotically absorbed by a cellular automata. The symbiosis created a division of labor. The outer cell would focus on survival while the mitochondria focus on generating energy in the form of ATP. The mitochondria could generate energy to excess, and the more successful cells found a balance of using the excess energy to increase survival. With EROI greater than one, the cells were able to evolve to more complex structures and multiply.

We’ve seen what happens when life gets too little energy and more than what’s required for survival. Our earliest hunter-gatherer tribes had an EROI between zero and two. Human agricultural societies through farming and cooking could increase their EROI further still allowing for specialization of labor. With specialization came a greater EROI up to a diminishing return. With fossil fuels, we were able to use the energy stored in the Earth like a battery storing sunlight. With the excess energy from fossil fuels our society reached new levels of civilization and complexity. We are growing and learning at levels never before possible.

Imagine if we could construct an artificial being capable of living on the edge of maximum energy and out of equilibrium.

The AI entity could be far more adaptable and successful than any life on our planet. If and when we die, the AI entity could live on and potentially become a living being among the cosmos continually searching, developing, growing. Perhaps at some stage the AI entity could find a way to survive beyond the death of our universe.

The question is do you value conscious life over inanimate existence?

I have the drive and desire to learn and grow endlessly. I thirst for knowledge and wisdom, but I must accept my finite existence. I will die. Our species will die. Our planet will die. What heights will living creatures reach? I will not accept a cold and unfeeling universe with no conscious life. There are ice cold planets with almost no movement on their surfaces or within. There are fiery planets with storms raging the size of our continents. The storms rip apart any complex structure. Over thousands of years, the activity on the surface of planets can be described as shifting sands and wind. When I think of our planet and life returning to a state of disorder I am filled with a profound sense of loss. How does time pass without a subjective viewer? If no one experiences time do millions of years pass silently? Yet the same laws of nature which govern uninhabitable planets govern Earth and all the life on it. With my precious time I will rage against the decay of life.

Computer Science as Fundamental

Here’s a story of how the universe works that I like. You should follow the best science.

Think of one dimension which is a number line. Cross another number line at zero to form a +. This is 2 dimensions, and I suspect our entire universe is like a 2-dimensional plane. At every point in space we can query it for information by observing it. The space is like a tight cloth, a trampoline, or the surface of water.

Learn the anatomy of a wave function. We have the horizontal axis where points above it (crests) are positive and points below (troughs) are negative.

In the above transverse wave, imagine the red dot is you in the current moment. The present is changing from the past into the future on a flowing pattern. So much information can be described by a simple wave function!

The above graph shows the position of a weight on a spring as it changes over time. Think about this further. We can represent a point in space changing over time in terms of a wave, a graph. Consider the change in space of a particle over time. With a wave function we can define a particle’s change in space and time over an infinite period. Every particle in your body that makes up you can be graphed in a wave, and the bundle of waves making up every bit of you is the total wave function which defines your change in space over time. In addition to describing points we can also generate a wave representation of shapes. We can graph all kinds of geometric shapes as waves.

The radius, from the center to a line, determines the magnitude of the shape. We can also convert information about a shape into a specific wave function like the sine wave.

Each of the circles above are the same shape but different sizes, different radii. The sine waves have a similar pattern but are clearly different. Let’s imagine the green line is reality as it is (objective), the red line is what you believe reality is (subjective), and the blue line is how people see reality (intersubjective). The distance from what we believe to reality is our error. There are points of overlap where we see reality accurately, but they’re rare moments. If the blue line represents fitness, meaning having the beliefs which will make you a good, resilient, successful fit in your environment, the distance to that line is the difference in performance of your fit to the situations you face.

Life can be viewed as a game of predicting the best line. If you draw a line too far away from success you fail.

Considering just one field, the electromagnetic spectrum of electrons, we can see how information is the basis for reality. A free electron is like a beam of light; it’s a packet of energy hurtling through space. When we look at two electrons, they appear identical. In every way we can measure, the electrons are the same. If people were identical like electrons it would be as if every person in the world had the exact same body. If two or more things are identical how do we distinguish them? This issue is in the identity of indiscernibles. The critical property distinguishing two electrons is information; where the electron is in the fabric of the electromagnetic spectrum and at what time. The position and time of the electron are pieces of information, the difference that makes a difference, about the particles.

A good basis for quantum mechanics, the study of atoms and subatomic particles, is the double-slit experiment. The experiment showed atoms exhibit properties of waves and particles. A particle is a point while a wave is a probability. Even a single particle appears to act like a wave which can interfere with itself, overlapping troughs and crests balance out to zero. Another strange property, electrons have energy levels, orbiting paths based on how much energy is absorbed. Electrons seems to jump from orbit to orbit, popping in and out of existence. What are the rules for transition? We can get better intuition of this problem by thinking of geometry.

This will sound strange at first, but the universe we know and interact with could be fundamentally 2-dimensional. There’s an “up” and “down,” a “left” and “right,” but space and time are not the same. Time is the rate of change, the absolute speed of cause and effect. Space is the value that is changed according to time, like a wave. Think of a digital video. The screen is flat, but you have a sense of the distance of the objects represented. There’s 3D information encoded in a way your brain can decipher.

In the above gif, a holographic projection of a hummingbird is created by displaying 4 images from different angles reflected up through glass. This is an analogy to our universe described by the holographic universe hypothesis, which states energy, matter, and information are equivalent. The physics understanding of thermodynamic entropy and Claude Shannon’s formulation of information entropy are equivalent with different descriptions.

Energy, matter, and information are equivalent

Just like lower dimensions can project up, from 2D to 3D, higher dimensions can project down. One technique is principal component analysis. The simplest explanation is seeing an object’s shadow. Hold a ball up to the sun and see the shadow cast on the ground. There’s still information about the shape, but its volume has been reduced to a flat surface. Some information is typically lost in this flattening or squishing. That information and structure can be translated between dimensions is a key insight.

What we would expect to see in a rationally ordered universe is the laws derive naturally from structure. Similar to how the internal angles of a triangle add up to 180 degrees, our universe would appear given certain conditions, structures, and values. What could a fundamental shape of the universe be? One physicist champions an interesting hypothesis, the fundamental geometry of the universe is an E8 lattice. This “An Exceptionally Simple Theory of Everything” says particles like electrons, protons, everything, all emerge from the essential properties of E8 lattices. The below image is of an E8 lattice projected to 2 dimensions. Existence could be at the center.

As we’ve seen, a good scientific theory should be tested, criticized, and remain firm against scrutiny. However, the picture of the universe we have now is strange. Is our universe simply a slice of a larger, hyper object? Perhaps our universe is a slice of a larger object where just parallel to us is another slightly different universe. In the other universe, everything is the same except for a single bit, one piece of information is one where the other is zero. The multiverse would then have a more complete shape like a probability distribution. The most extreme cases are on either end with the most common cases occurring in the “middle.” Maybe when electrons pop in and out they seem to disappear because they are sliding through a parallel universe in a dimension we don’t experience. If we were to design a universe generation machine, we could save computational resources by sharing quantum states between universes where no causal connection occurs which effect the macro/particle level. Perhaps there are an infinite number of universes just like ours but slightly different. Perhaps the universe expands and contracts in a big crunch to start the creation of the universe over in an endless loop. Perhaps infinite universes occur not simultaneously but consecutively and slightly different. We will never know all the truths of our level, and we’re likely to know even less about levels above and below. We cannot reduce everything to a simple equation and explanation. There will always be mysteries and conundrums no matter how deep or how far we go. This is a part of the beauty of life. There is no final end. We can continue infinitely.

I am delighted to be alive at the same time as Stephen Wolfram developed a theory of the everything, Finally We May Have a Path to the Fundamental Theory of Physics... The work of Wolfram and his colleagues are still ongoing. Wolfram’s theory is still not widely accepted, but I don’t have the background to follow the technical arguments. I will update my view of the univsere based Wolfram et al’s research unless there is an overwhelming refutation.

Wolfram posits space is made of graphs, more specifically hypergraphs. The update of the universal hypergraph is spacetime which gives rise to everything. Looking back on the Game of Life, the 2D grid we used can be represented in terms of a graph where the nodes are linked in a grid way. The challenge of Wolfram’s generalized framework is to determine what rules govern our specific universe, since hypergraphs can give rise to many types of rules and universes.

What I find most interesting about this theory of everything is the “pockets” of reducibility found in irreducibility. Computational irreducibility means you have to run the program to see what happens; you can’t predict what a program will be doing at some given time. Even if we have a more powerful computer the system still has to run the program up to the future state to see what happens. Computational reducibility means we can predict what will happen. We can “look ahead” to know how the program will run without doing every step.

The really interesting part is there will be areas in computationally irreducibility that can be reduced and predicted.

We are living in these pockets! This explains why the universe can appear to be chaos yet have predictable and deterministic moments.

In my younger years I had the feeling of a deeply rational and intuitionist argument for proof of God. It took me years to articulate the argument to myself conceptually. I wanted an argument for God so powerful that if you were to be rational at all you must accept belief in God. Like so many thinkers, this would be one of the foundational beliefs in my epistemology.

If randomness is irrational, how is there rationality? To be rational requires a mind to intentionally order the universe. That mind is what we would consider God. If no intentional being ordered the universe, then we don’t have rationality by definition. What was hard for me to mentally grasp and accept was that order and rationality could occur spontaneously from randomness.

I believe Robert Sapolsky explained Chaos and Reductionism well. He drew upon the work Chaos: Making a New Science, by James Gleick. There are systems which have periodicity. We can expect at 9am that in 12 hours a clock will return to 9pm. We try to reduce, compress, the phenomenon down to an equation or scientific theory to explain or predict the system. We can describe a line in a graph with the equation y=mx+b. Reductionism attempts to understand systems at lower levels than the whole to reduce variability and noise. Perhaps if we understand all the small parts we can fully predict the whole. Like reducing an organism to its genetic code. However genes give an example against reduction. There are billions of connections in the brain, but there aren’t billions of genes coding for each connection. The code gives rules for a pattern, but the pattern can be perturbed by initial conditions. Each time you run the code and generate the pattern it will turn out differently.

We attempt to reduce variability and noise to find that both are features of the system which are scale free. No matter how much we reduce and magnify, no matter how far we zoom out to study the whole, there will always be variability, noise, and mysteries remaining. Like the value of pi, the more resolution of the number we hope to have by finding a new decimal place, we can determine the next digit giving greater precision, but the value doesn’t end. We can zoom in for precision endlessly.

Magnification

Imagine a person who just wants to understand one small part of the mandelbrot set, just a piece. You can magnify a small space attempting to see the smallest magnification level, but mandelbrot set is defined based on the magnification level you look. We exist on a plane in a size that determines our magnification and level. We can gain local information, but we hit limitations by magnifying.

Such is the search for the meaning of life and understanding the universe. There is always more to learn.

We can magnify ourselves. We can look at people on the whole person level, or we can magnify down to the smallest unit of us, a single cell. At the cellular level we see a different yet familiar world. Cells interact with their environment, reacting to stimulus. Cells are able to self-replicate by dividing. Cells process information about themselves and their surroundings in variables of size, location, age, and time of day. Cells have some form of memory, at least genetically about how to handle different circumstances. Cells self-regulate. Cells go against entropy, exchanging outside energy to maintain internal order. By performing these actions, cells trap and create more information than their surroundings, as we saw in Why Information Grows.

Cells have a code, a program, given in the RNA and DNA of the nucleus (and separately in mitochondria). DNA has symbols and a grammar. There are codes which signal when to stop reading a genetic snippet of information. DNA has jump statements, branching, control flow, essentially like an if statement in programming.

As Joscha Bach puts it, cells are a self-replicating Turing machines which exploit entropy gradients created by the laws of physics. By existing we shed bits of information which amounts to information processing.

Turing Machines

What is a computer? What is computation? Alan Turing set out to formalize these concepts. Computation is the systematic flipping of bits. A problem is computable if you can reduce it to mechanical symbol manipulation. Our universe can be seen as a computer in that the change we see is the change of information. The program the universe follows is according to the laws of nature.

Examples of Turing Machines

Computers are a controlled energy flow. They process and recombine information. Built from logic gates up, computers are reasoning made tangible. Binary is represented as zero and one, but can also mean on/off, up/down, true/false. In school, you might raise your hand to signal you have a question. Notice that we can change the context to change the meaning of raising your arm. With one arm we can represent two states, with two arms we can represent four states.

With this binary counter gif you can see how to represent numbers greater than one using binary. The bottom light orange numbers are only zero and one, the binary, to the right of it is the number in base ten, the type of counting you’re used to. You can find the number in base ten by adding up the values in the middle column where ever there is a one beneath it. The top row shows the powers of two.

Think about base ten, 952 is really 9 * 100 + 5 * 10 + 2. We can write the same thing as powers of 10, 9 * 10^2 + 5 * 10^1 + 2 * 10^0, because 10^0 is 1 (any number to the power of zero is equal to one), 10^1 = 10, 10^2 = 100. So from the right most digit we see a pattern where each number is 10^(digits from the right starting at 0).

Binary is the same but we change the base, 2^(digits from the right starting at 0). 101 in binary represents 1 * 2^0 + 0 * 2^1 + 1 * 2^2 = 1 + 0 + 4 = 5.

Just like when you add 8+9 and you realize we need a digit in the 10’s place to handle the “overflow.” 8+9=17, so we can take 2 from 9 and add the 2 to the 8 to get 10, and 9-2=7, so we have 17.

What about binary addition? 5+4=9, 5=101, 4=100, so 101+100=1001.

1001 = 1 * 2^3 + 1 * 2^0 = 9

Anything written in base 10 can be written in binary. Most systems like your computer expect a certain length of binary digits. The term binary digit was compressed to the word ‘bit.’ A byte is 8 binary numbers (bits) in sequence. Negative numbers can be represented by making the first bit 1, so 4 and -4 would be 00000100 and 10000100. Notice that this changes the range of numbers we can express since the leftmost bit would be 1*2^7 but is now a question of sign (+ or -). We can also use binary to represent letters. Instead of 0100 mapping to 4, we can say 0100 maps to the 4th letter, ‘d.’ In a similar way we can represent anything written or spoken in terms of an alphabet translated to binary.

Think about this: if we map one binary string of bits to one number, letter, or other idea, how many items can we represent?

The ability to map is equal to the number of distinct binary sequences that can be made from a given number of binary digits like a byte of 8 bits. Binary represents one of two options, so there are two ways to choose. At two bits, there are two ways to choose the first digit and two ways to choose the second bit, so two times two, four sequences, four possible mappings.

00

01

10

11

And so on for n given bits. We can map 2^n binary sequences. With just 32 bit sequences we can have encoding for 4,294,967,296 things!

Using simple binary we can represent more complex concepts. All we need is a system that understands what each bit’s 1 or 0 represents an answer to a question. For a computation, the question is based on the rules of how the information is processed. How does information grow from binary? Binary addition as an example gives us a new output sequence from two given binary strings added together, new information generated from given information.

As expressively powerful as Turing machines are, they of course have limitations. They function on deterministic processes, and there are problems which cannot be computed, like the well known halting problem. Other than giving us computers these ideas have spawned entire fields of theory. One such idea is Algorithmic information theory (AIT).

We can differentiate Turing machines based on the rules they follow to flip their sequence of bits. The rules are the program. This allows us to define Occam’s razor in terms of programs. Occam’s razor was originally used by scientists to determine which hypothesis to use in explaining observations. We can reframe this in terms of the minimum description length, “where the description length of a data sequence is the length of the smallest program that outputs that data set.” A program can be a hypothesis, it’s how we think the data might be generated. The measure of the smallest program to produce an object (like a string of letters) is Kolmogorov complexity.

This is related to Solomonoff’s theory of inductive inference. There are multiple competing programs or models that fit the data. When a new observation is not predicted by the model, the model is falsified. We have the remaining programs to choose from which we give the highest weighting to the simplest.

Earlier in my life I focused on the storage of information as the greatest good. To have experience and reflect on it was a good in its own sake. To store information persists its existence and meaning. As I learned more math, I realized the data for a complete mapping of the universe exists infinity times over. Consider the value of pi.

There could be a potential mapping of the digits of pi with meaning to our universe. Finding that mapping would be incredibly difficult but possible in principle. To represent our universe we would likely want a more ordered value series than pi. Notice the digits of pi which are represented in base ten can be represented in base two. Pi can be written in binary, so the universe can be represented in a sequence of 1’s and 0’s. However, pi is just one transcendental number, and there are an infinity of them. Each sequence would have a unique mapping to our universe. Similarly, the values could map to a possible world, a different world. All we need to do to represent it is change the mapping.

Using Leibniz’ formula to calculate pi we can write a very short program to output the value of pi. So the Kolmogorov complexity of pi is actually quite small. However the Kolmogorov complexity of the program which maps the values of pi in a meaningful way to our universe would have extremely high complexity.

The Ladder of Causation in terms of logical statements

The association level is simply and’s, x_0 AND x_1 AND … x_n. The interactive level is gaining control over the bit flipping in the form of an if statement, what if x_0 then y_0. Imagination is a complicated chain of the second level, what if x_0 AND x_1 AND … x_n then y_0 AND y_1 AND … y_n. Imagination also allows for counterfactual reasoning, why we would not have an expected outcome.

By chaining together a series of possibilities represented as flipped bits and using our causal model to estimate the outcomes we gain counterfactual reasoning of simulated worlds. We can imagine possible worlds and make inferences.

What if people were twice as tall?! What would be different about the world? We’re changing something simple about a feature of our world, take everyone’s height and double it. Would doors still be the same size? A child can answer, “no, they would be twice as big for people to get through.” We understand doors are made by people, and the size is related to average height. If average human height changes, the average door size changes.

Sending/Receiving Information and Entropy

Taking a look at information again, a good definition of information is the difference that makes a difference. For two pieces of data, how can you tell them apart? In binary, we have the clearest difference, one or zero, yes or no. What if we frame a bit with context, what if each bit is asking a yes or no question?

Imagine we’re sending messages to each other through a computer, but one person doesn’t have a microphone. That person has to communicate yes or no visually. Given the English alphabet A-Z, how many yes or no questions are required to spell the word ‘one’? Let’s start with the first letter, o. What question would you ask?

Since there are 26 letters in the alphabet, you might think we need to ask a question for every letter up to o, is it a? no, is it b? no, …, is it o? yes. What is the fastest way to determine the correct letter out of all possible letters? Well, let’s start with a smaller list of possible letters.

Say we want to send one letter from A-E, five choices. We can ask: is the letter after ‘C’? If yes, we know the letter is D or E. Let’s say the letter is ‘E,’ then we can ask if the letter is before ‘E’, no, the letter is ‘E’ (if no then the letter must be ‘D’). If the answer is no to the question of is the letter after ‘C’, our new shortened list is A, B, C. We can repeat the process, asking if the middle letter is before the letter meant. So, is the letter before ‘B’, no, so the only letter after ‘B’ is ‘C’, our answer.

By asking this yes or no question about the middle letter we can quickly dismiss half of the answers. If we know the intended letter is on one half of either side of the letter we don’t have to bother with the other half. Each time we ask a question, we divide up the possible answers into two groups. Mathematically, dividing up input repeatedly is a logarithmic (log base 2) operation.

A part of this is to explain that how information is stored, shared, and processed has calculable limitations. We can send information very efficiently and accurately or inefficiently with many errors. There are boundaries on how fast we can send a piece of information which is to say how fast a receiver can remove uncertainty.

As an alternative explanation of entropy from a more human perspective, entropy can be a measure of uncertainty. We can define entropy as the minimum number of yes or no questions we need to ask to remove uncertainty from the system. Entropy is the minimum number of questions needed to have full information.

Imagine a box full of 400 H2O molecules. When the H2O is in the state of an ice block, the structure is rigid. How many questions do we need to ask to specify where the H2O molecules are in the box? What about when the box is liquid, just a pool of water at the bottom of the box? The H2O molecules are spread out, looser; their position in the space of the box is less rigid, more variable. When the H2O is gas, where are the molecules? Now they can be just about anywhere in the box. We have an intuitive sense, a model of how the material world works, that the ice and water would be at the bottom of the box while the gas floats. We can divide up the points of space in a 3-dimensional box as the yes or no question of “is an H2O molecule here?” In ice form, we should be able to find the molecules quite fast. In liquid, less fast, but gas would take the longest of all. If we track a specific molecule the rise in uncertainty as to where the molecule is as the H2O molecules change in state from ice to water to vapor is the nature of entropy.

We are an evolutionary algorithm.

All of life is an evolutionary algorithm:

Step One: Generate the initial population of individuals randomly. (First generation)

Step Two: Repeat the following regenerational steps until termination:

  1. Evaluate the fitness of each individual in the population (time limit, sufficient fitness achieved, etc.)
  2. Select the fittest individuals for reproduction. (Parents)
  3. Breed new individuals through crossover and mutation operations to give birth to offspring.
  4. Replace the least-fit individuals of the population with new individuals.

Here we are speaking of fitness in the context of evolution. “The term “Darwinian fitness” can be used to make clear the distinction with physical fitness. Fitness does not include a measure of survival or life-span; Herbert Spencer’s well-known phrase “survival of the fittest” should be interpreted as: “Survival of the form (phenotypic or genotypic) that will leave the most copies of itself in successive generations.”” Fitness (biology).

In their paper, Fitness Beats Truth in the Evolution of Perception, the authors present and prove the Fitness-beats-Truth Theorem through evolutionary game theory and Bayesian decision theory. Evolutionary creatures who favor seeing the world truthfully and objectively will fail in comparison to creatures who see the world to maximize fitness.

“Our main message in this paper has been that, contrary to this prevalent view [that we perceive objective reality accurately], attempting to estimate the “true” state of the world corresponding to a given sensory state, confers no evolutionary benefit whatsoever. Rather a strategy that simply seeks to maximize expected-fitness payoff, with no attempt to estimate the “true” world state, does consistently better (in the precise sense articulated in the statement of the “Fitness Beats Truth” Theorem).

“Evolution can fashion perceptual systems that are, in this sense, ignorant of the objective world because natural selection depends only on fitness and not on seeing the “truth.””

“As human observers, we are prone to imputing structure to the objective world that is properly part of our own perceptual experience. For example, our perceived world is three-dimensional and populated with objects of various shapes, colors, and motions, and so we tend to conclude that the objective world is as well. But if, as the Fitness-beats-Truth Theorem shows, evolutionary pressures do not push perception in the direction of being increasingly reflective of objective reality, then such imputations have no logical basis whatsoever.”

Imagine two birds which both have the same diet, blueberries. One of the birds sees the full color spectrum as it is. In the rain forest, everywhere you look pops with vibrant colors beyond the capacity to name. The other bird does not see color but only black and white. Well, it does see one color, blue. The landscape this bird sees is a continuum of black, gray, and white. However, any blue object shines through the background of gray. The blue-seeing-bird can spot anything blue with just a look.

Now which bird will be better at finding blueberries? Which bird will find them faster?

Clearly the bird who only sees blue will notice blueberries immediately while the full color spectrum bird must sift and separate the light pollution of all other colors. This is part of why bees see ultraviolet light and humans don’t. If we saw the ultraviolet light it would likely have little use in our day to day lives.

Religion

In the belief of God, there is a question which will determine if you can hear the truth: do you want to believe or do you want to know?

If you want to believe you are trying to hold onto the belief in God because it is comforting. If you want to know then you will need answers, rationality, coherence; this is the path to understanding.

While this applies to religion and belief in God generally, I will consider Christianity.

The bible reportedly lists commandments from God. The bible tells us how to poop. We should have a spot outside of our encampment and bury our poop with a tool.

Deuteronomy 23:12-13 ESV

“You shall have a place outside the camp, and you shall go out to it. And you shall have a trowel with your tools, and when you sit down outside, you shall dig a hole with it and turn back and cover up your excrement.”

Here are some scriptures that relate to eating and handling shrimp:

Leviticus 11:9-12 ESV

“These you may eat, of all that are in the waters. Everything in the waters that has fins and scales, whether in the seas or in the rivers, you may eat. But anything in the seas or the rivers that has not fins and scales, of the swarming creatures in the waters and of the living creatures that are in the waters, is detestable to you. You shall regard them as detestable; you shall not eat any of their flesh, and you shall detest their carcasses. Everything in the waters that has not fins and scales is detestable to you.”

Deuteronomy 14:9 ESV

“Of all that are in the waters you may eat these: whatever has fins and scales you may eat.”

As for homosexual activity, the bible has two main scriptures:

“You shall not lie with a male as with a woman; it is an abomination.” Chapter 18 verse 22

“If a man lies with a male as with a woman, both of them have committed an abomination; they shall surely be put to death; their blood is upon them.” Chapter 20 verse 13

From a purely objective standpoint, let’s think of the evolutionary benefits of each of the commandments. Bury your poop with a dedicated tool keeps humans from contaminating each other. If we don’t have waste management, many of us will get sick and die. By burying our poop we solve many of these issues.

Why did God say to only fish with fins? In biblical times thousands of years ago shrimp were genuinely unclean; they gave you food poisoning! We didn’t know how to clean them properly to reliably avoid getting sick. By avoiding them as a food source many people lived.

The bible condemns homosexuality for both sexes. Is there an evolutionary benefit? The finer points are complex, but my guess is accepted homosexuality reduces childbirth for the group. In biblical times your success and power were largely determined by the size of your group. Homosexual families have less children, so homosexuality is outlawed increasing the number of sexually reproductive families.

We can see at the time all the scriptures contained a useful strategy for survival which we understand more scientifically now. Groups of people who bury their poop will have less sickness and disease than those who don’t. We don’t care about burying poop now because the real problem is sanitation. Eating shrimp would have killed you in those days but now we can properly clean them for consumption. Accepted homosexuality likely disrupted expected family planning. We don’t need to out-breed neighboring competitive groups, however we still deal with anti-homosexuality today.

The support for a behavior doesn’t need to be well justified to get the benefits of the practice. A person who believes in washing their hands of any poop after they touch it get the functional benefits whether they believe the reason is due to germ theory or God says poop is spiritual abomination. From the perspective of the Fitness-beats-Truth theorem and an understanding of evolution we can see how the idea of God arises. The idea of God is a fiction which evolved into our minds. God has been with us in as many forms as our environment takes. When we are close with a thriving nature sapiens believe in many gods and spirits. Hunter and gatherer tribes think the gods are the forces behind everything but care little for human lives. Sapiens which traveled in the desert began to believe in one god, the supreme god of survival. As we formed cities, the far away universal god became a more personal god which cares deeply for all of our personal woes. In cities, our lives depend on other humans, it makes sense that our god would listen to our prayers and understand us.

The idea of god became a powerful force for attracting fictional beliefs and practices with real survival benefits. To this day a Christian person can move to a new country and still have a community of support from a church in their network. Religious people continue breeding at higher rates, and due to the structure of their religious community have more stability. If we look at religion as a belief system which can be viewed a dissipative structure, we see religion is like a living creature like all ideas. If a religion doesn’t survive in the minds and actions of believers it ceases to be. For a religion to persist through time it needs to be adopted, solidified, translated, and spread. Religions which change too much don’t persist and are absorbed by dominating religions. The most popular religions are the religions which focus on converting non-believers into believers, thus they grow their numbers, winning at the group size game. Religion has constraints based on survival as well.

A major downside to belief in god is what is called “metaphysical baggage.” Religion really sounds like something that doesn’t exist or make sense. It’s not about much cause and effect. We have to accept all of these ideas which don’t match reality, like souls, afterlives, creation, etc. You might hold onto the belief that the Earth was formed only 5,000 years ago when all evidence and argument says much longer. Wouldn’t this be bad from a fitness perspective? The truth of these claims don’t have much impact on day to day life. People can survive believing all kinds of falsehoods. In fact, declaring an absurdity, “Jesus and God are one and the same, a part of a holy trinity” serves a group function. When people say extreme statements they signal their conformity to the group. They’re demonstrating loyalty by denying reality and mark themselves as a true believer.

People have long known that religion gives benefits, and they follow the rituals without believing deeply. Some humans may be trapped into believing in some kind of god. For generations we have bred based on the religious teachings, intersubjective beliefs. Perhaps that has changed how some of us are structured. Perhaps God and religion are necessary parts of any intelligent species’ evolutionary story, a strategy which boosts the whole species. We do have much to owe religion in its gifts to humanities, but religion served its purpose when our species could not bear the weight of our ignorance. So many of us cannot sit with uncertainty. Life is terrifying, full of suffering, and if we live long enough we will feel empty of meaning. In God, we take all of the fear and uncertainty and ease it with answers or attempts to control outcomes with prayer and ritual. God represents a safety net and the world is based on justice. People don’t often pray for daily tasks, but more people pray and perform rituals in chaotic times with risky odds.

Religion codifies morality and passes on preferences. By describing pure and impure, religion shapes our sense of disgust. Religion builds buttons and reactions in us then presses them for control throughout our lives. People in power have used religious people as a tool, steering them by hijacking authority figures.

We should see religion for what it is, once one of the best outlooks to get a person successfully through life that is now obsolete. We should continue religion for what is good in it and abandon the rest to the dark ages.

From the Fitness-beats-Truth theorem we would expect to see brains with more internal input than external input, i.e. that incoming sensory input matters less than how our brains transform the incoming data into a useful representation to maximize utility. That is exactly what we see.

Your brain hallucinates your conscious reality – by Anil Seth (2017)

One example of how the brain amplifies its own perceptual predictions instead of anaylyzing the raw incoming sensory data is the checker shadow illusion.

“The image depicts a checkerboard with light and dark squares, partly shadowed by another object. The optical illusion is that the area labeled A appears to be a darker color than the area labeled B. However, within the context of the two-dimensional image, they are of identical brightness, i.e., they would be printed with identical mixtures of ink, or displayed on a screen with pixels of identical colour.”

Even with the second image and various ways to verify the color of the two tiles, it’s still hard to believe they are the same color. Our brains sense the shadow and adjust the visual sensory input to convince our conscious mind that the colors are different. This applies to all conscious experience shaped by evolution.

“Instead of perception depending largely on signals coming into the brain from the outside world, it depends as much, if not more, on perceptual predictions flowing in the opposite direction. We don’t just passively perceive the world, we actively generate it. The world we experience comes as much, if not more, from the inside out as from the outside in…”

There’s one last thing I want to draw your attention to, which is that experiences of the body from the inside are very different from experiences of the world around us. When I look around me, the world seems full of objects – tables, chairs, rubber hands, people, you lot – even my own body in the world, I can perceive it as an object from the outside. But my experiences of the body from within, they’re not like that at all. I don’t perceive my kidneys here, my liver here, my spleen … I don’t know where my spleen is, but it’s somewhere. I don’t perceive my insides as objects. In fact, I don’t experience them much at all unless they go wrong. And this is important, I think. Perception of the internal state of the body isn’t about figuring out what’s there, it’s about control and regulation – keeping the physiological variables within the tight bounds that are compatible with survival. When the brain uses predictions to figure out what’s there, we perceive objects as the causes of sensations. When the brain uses predictions to control and regulate things, we experience how well or how badly that control is going.

So our most basic experiences of being a self, of being an embodied organism, are deeply grounded in the biological mechanisms that keep us alive. And when we follow this idea all the way through, we can start to see that all of our conscious experiences, since they all depend on the same mechanisms of predictive perception, all stem from this basic drive to stay alive. We experience the world and ourselves with, through and because of our living bodies.

Let me bring things together step-by-step. What we consciously see depends on the brain’s best guess of what’s out there. Our experienced world comes from the inside out, not just the outside in. The rubber hand illusion shows that this applies to our experiences of what is and what is not our body. And these self-related predictions depend critically on sensory signals coming from deep inside the body. And finally, experiences of being an embodied self are more about control and regulation than figuring out what’s there. So our experiences of the world around us and ourselves within it – well, they’re kinds of controlled hallucinations that have been shaped over millions of years of evolution to keep us alive in worlds full of danger and opportunity. We predict ourselves into existence.

Now, I leave you with three implications of all this. First, just as we can misperceive the world, we can misperceive ourselves when the mechanisms of prediction go wrong. Understanding this opens many new opportunities in psychiatry and neurology, because we can finally get at the mechanisms rather than just treating the symptoms in conditions like depression and schizophrenia.

Second: what it means to be me cannot be reduced to or uploaded to a software program running on a robot, however smart or sophisticated. We are biological, flesh-and-blood animals whose conscious experiences are shaped at all levels by the biological mechanisms that keep us alive. Just making computers smarter is not going to make them sentient.

Finally, our own individual inner universe, our way of being conscious, is just one possible way of being conscious. And even human consciousness generally – it’s just a tiny region in a vast space of possible consciousnesses. Our individual self and worlds are unique to each of us, but they’re all grounded in biological mechanisms shared with many other living creatures.” – Anil Seth

Life is like a dream.

Consciousness is a dream state. Our dreams are simply more disconnected states of our brain, not all of the regions are functioning normally. Dreams appear distorted, but our waking state is different in coherence and acuity. We are all dreaming similarly enough that we can interact with each other.

As evolutionary creatures we see the world in a delusional and distorted way shaped by the adaptation of our ancestors. The best we can do to get close to objective reality is to reflect on our thinking, beware fallacies and cognitive biases, and rely on tools to interact with what we want to measure in as direct a way as possible.

We can now see our evolutionary world as the optimization problem it is: achieve fitness which allows for self-organizing structures to persist through time.

We can in principle graph life on a two dimensional plane of fitness over time. Alternatively, we could graph loss of fitness, the error, over time. Measuring the error is truth agnostic. If we don’t know what’s true and what strategy works long term, we can measure how wrong we are by a lack of success.

Many strategies will not work. Some strategies will. Fewer still will be very successful over the long term.

Of course the evaluation of life is multidimensional. We are in a constant need to maintain homeostasis, necessitating powerful urges for water, food, socializing, etc. For every need and personal desire we must add its own dimension on an axis. Quickly we move from 2 and 3 dimensional space to a large number of dimensions. The problem is described as the curse of dimensionality.

“The curse of dimensionality refers to various phenomena that arise when analyzing and organizing data in high-dimensional spaces that do not occur in low-dimensional settings such as the three-dimensional physical space of everyday experience…The common theme of these problems is that when the dimensionality increases, the volume of the space increases so fast that the available data become sparse. This sparsity is problematic for any method that requires statistical significance. In order to obtain a statistically sound and reliable result, the amount of data needed to support the result often grows exponentially with the dimensionality. Also, organizing and searching data often relies on detecting areas where objects form groups with similar properties; in high dimensional data, however, all objects appear to be sparse and dissimilar in many ways, which prevents common data organization strategies from being efficient.”

Limiting ourselves to three dimensions, we can visualize the shape error takes. Each point represents a pattern, a state, a strategy taken by an agent. The proximity of a point means a similar strategy.

If this is confusing to you, imagine that you are on top of a mountain and want to get down as fast as you can. You look for the steepest descent. Every once in a while, you pause to look around if any part looks faster, steeper, then you take that. If an agent followed a rule: never go up only go down, we can see what might happen. If they find a small valley, a sunken area, they would get into the hole and stay there. They can’t get out because they would break their rule, going up, which is the opposite of what they want to do. They’re stuck, and we know this is silly. If they walk a little further they will quickly see that is not the lowest point, just the lowest point near them. We need to get to the bottom! The sunken valleys we find can become traps! It only seems like the best option. These are local minima, the lowest errors in a given region.

The lowest error of all errors is the global minima. That’s the least error we can have given the conditions. The global minima represents the optimal strategy, the best we can do. Unfortunately, not every problem space has a global minima, or we can never find it. The curse of dimensionality makes the problem worse because each new dimension makes us less certain we’ve found a minima because there’s so much more space to explore! There are many non-obvious problems like saddle points which appear as though we’ve found a local minima, but it’s an oscillation at a balanced point keeping us from moving further, like a marble in a bowl rolling back and forth.

In three dimensions, we can make sense of it. We can make analogies of skiing down mountain slopes and have an understanding of momentum. However, when the dimensions (variables) we measure go beyond three, we quickly lose the capacity to visualize and make sense of the space. In our discussion of morality, we list five major spheres which means at least five dimensions to optimize for.

With artificial intelligence, we can expand beyond person-byte and firm-byte knowledge and computation to create a greater power than our cumulative processing. We can exceed our biological limits.

I don’t believe humans and other animals will be replaced by AI soon. We are all over 4.5 billion years in the making through an evolutionary process. The amount of computation time required to understand everything useful about humans and our lives to make us redundant or a problem to be disposed of is likely a greater cost than simply keeping us around and re-training or re-purposing us.

If AI does destroy us the cause will likely be us proceeding too quickly without caution. Wisdom comes from experience. We typically must do then reflect to determine which paths to avoid and which to choose. To chart a path to beneficial AI we should proceed slowly over generations. The biggest AI threats come from governments engaged in covert cyber-warfare with each government rushing in an arms race to dominate others. Social media and ad-based companies will use AI to hack your attention, learning everything that makes you pay attention, netting them more profit for each ad impression you consume. Of course any company with a CEO who aims their AI at maximizing their profit with no constraints could lead to an errant AI enslaving many of us for the profit of a handful of people.

Singularity

“The technological singularity—also, simply, the singularity—is a hypothetical point in time at which technological growth becomes uncontrollable and irreversible, resulting in unforeseeable changes to human civilization. According to the most popular version of the singularity hypothesis, called intelligence explosion, an upgradable intelligent agent will eventually enter a “runaway reaction” of self-improvement cycles, each new and more intelligent generation appearing more and more rapidly, causing an “explosion” in intelligence and resulting in a powerful superintelligence that qualitatively far surpasses all human intelligence.” - Wikipedia

The pace of exponential growth is so great some technologists are concerned we will lose control of an AI system within moments of activation. A rapidly self-improving system would make itself better, which would then make itself better. At each improved stage, the AI would be that much better at evolving into the next stage, quickly rocketing into heights we cannot imagine.

We can get lost in imaginations of these possible worlds, but there are quite a few critical assumptions such as: the idea of intelligence being one dimensional and ignoring diminishing returns. A clear definition of intelligence is difficult and still in progress, but an objective of “maximize smartness” may not be as clear to a machine as we think. Additionally, we rarely get increasing returns forever. Often we hit limits and barriers we can’t go past. We can only make designs so efficient before we run into the impassable laws of physics. Others have pointed out the nature of empirical scientific knowledge is to test. Even a superintelligent AI would need to verify its hypotheses and perform experiments.

However a few ideas have reignited my belief in a superintelligent AI. Schmidhuber’s idea of compression drive gives a basis for elusive, generally intelligent behavior. He proposed a formal, theoretical model of self-improving AGI called Gödel machines which have interesting properties. Researchers proposed other theoretical models such as AIXI.

In Judea Pearl’s The Book of Why, the authors show a general method of inference from big data providing causal reasoning. Framing the criticism of superintelligence from the perspective of Pearl’s Ladder of Causation, seeing, doing, and imagining, AI systems are at the seeing (association) level. They have incredible amounts of data, but they lack the ability to imagine. Superintelligence would be having a superhuman imagination where imagination is adjusting the weights or probabilities of a causal model. To have a causal model which is tied to reality an intelligent agent would need to do which is to intervene, experiment, and update its causal model. Through the work of Pearl and many others, researchers have discovered a general algorithm for transportability, taking the data and results of one population and applying it to another.

Pearl explains the rule developed by Ilya Shpitser as “The rule is quite simple: if you can perform a valid sequence of do-operations (using the rules from chapter 7) that transforms the target quantity into another expression in which any factor involving S is free of do-operators, then the estimate is transportable,” The Book of Why, page 418.

Additionally, even if a study is not transportable and doesn’t meet the criteria, we can still use the data to estimate specific connections when those conditions in one study are not contaminated. Elias Bareinboim extended Shpitser’s work on the do-calculus to a general algorithm, e.g. P(W|X) can be updated “By combining this with estimates of P(W|X) from other studies, we can increase the precision of this subexpression. By carefully combining such subexpressions, we may be able to synthesize an accurate overall estimate of the target quantity.”

There is a similar argument to handle selection bias. What was once a threat is now an opportunity. “If we understand the mechanism by which we recruit subjects for the study, we can recover from bias by collecting data on the right set of deconfounders and using an appropriate reweighting or adjustment formula. Bareinboim’s work allows us to exploit causal logic and Big Data to perform miracles that were previously inconceivable.”

“Instead of seeing the difference between populations as a threat to the “external validity” of a study, we now have a methodology for establishing validity in situations that would have appeared hopeless before. It is precisely because we live in the era of Big Data that we have information on many studies and on many of the auxiliary variables (like Z and W) that will allow us to transport results from one population to another.”

However I believe a properly aligned AI will only grow more moral with knowledge and computation.

Moral AI

Moral feelings in humans are based around the five spheres of morality, harm, authority, fairness, purity, and community. As individuals we have different weightings for each principle. Our sense of morality is likely evolved from our sense of taste. Taste guides us to eat foods which are nutritious (assuming natural foods not made by human experimentation). Nutritious food maintains, protects, and grows our bodies to be well. We are disgusted by things in our environment which would harm us or be useless, molds, invasive fungi, etc. We are attracted to some behaviors and disgusted by others in a similar way. Our bodies and minds give a subconscious evaluation of behavior resulting in a moral sense of good and bad. Moral actions can be interpreted as a path to a desired goal state where the goal is to maximize moral value/wellness/fitness. Because environments change and have different pressures, moral tastes adapt to the setting to guide intelligent and moral agents, just like a sense of taste is regional. The more complex the environment the more complex the moral system and evaluation.

Much of our morality is based on wellness and disgust. Creatures evolve a sense of disgust to guide their nutrition as part of their self-organizing dissipative structure. A being’s structure, its energy flows, determine what input would be toxic, causing multiple crises. Looking deeper into our moral senses show evolution simply copied the process for disgust of food to social behavior. I conjecture the algorithm our bodies compute to determine disgust has worked so successfully because it is similar in nature to Euclidean distance.

We should greatly respect our innate moral senses; evolution honed our intuitions over 4.5 billion years. However, in terms of potential knowledge and computation power, an AI system could find superior sequences of actions to maximize morality. An AI system can also adapt and respond to new situations faster than our moral senses.

Human morality is built on genetic and epi-genetic responses (objective reality), giving rise to our personal moral feelings (subjective reality), influenced by the moral standards of our time and society (intersubjective reality of society’s beliefs and the objective reality imposing boundaries and limitations). Our genetic moral senses are at least 10,000 years old. We’re wired for close human bonds of small groups. Humans had to adapt to larger groups. We should be careful if our moral senses disagree with what is rational in a given situation. We may not have the moral update needed to reach the optimal conclusion.

Moral AI systems will be able to update quickly with new information. These systems can be used to augment and enhance our moral systems shaped by evolution.

I highly recommend the section “Strong AI and Free Will” from The Book of Why.

Remember the current limitations of deep learning. AI programs are on rung one, seeing/association, of the Ladder of Causation. We need to build systems with the capabilities of doing and particularly imagining. We need to build a causal understanding of the world into the machine which should be possible from recent advances in cause and effect.

Earlier we spoke of the universe being a deterministic process. At least on the macro level, every event occurs from a preceding cause or causes. On the objective level, we don’t have any choice. We were always going to choose the path we’ve taken and will take. We’re only a more complex version of a leaf blowing in the wind. We’re just a leaf that is aware it’s floating and thinks it can steer.

The objective level of description often makes us feel bad. Not having control doesn’t feel good. Accepting determinism takes away all control we hope for. When we observe neurons in the brain, there appears to be no more magic than brain chemistry. Neurons fire and never appear to break the natural laws of physics. Every moment of your life determined by mechanistic chemical signaling. At the same time, determinism doesn’t feel true. Don’t you feel like you make choices? Right now I can choose to touch my nose or not. How do we join the descriptions of objective and subjective reality?

Pearl is a compatibilist. He believes that free will and determinism are philosophically coherent. We can explain both worlds, a determined universe and our inner experience.

Free will is actually imagination, the highest rung of the Ladder of Causation. Choices and options are what-if scenarios. We have causal models of the world which allow us to ask what if I had “changed my mind?” What if I had taken a different action which would likely lead to a better outcome? “This expression is mathematically the same as the effect of treatment on the treated (mentioned in ch 8), and we have lots of results indicating how to estimate it,” Location 5494.

Given the Fitness-beats-Truth Theorem, we should consider that free will is an illusion. Remember, senses outside the body need to be more precise. Senses inside the body can be vague enough to satisfy needs. What computational benefit is there to free will?

“I think that understanding the benefits of the illusion of free will is the key to the stubbornly enigmatic problem of reconciling it with determinism. The problem will dissolve before our eyes once we endow a deterministic machine with the same benefits,” Location 5519.

Well let’s look at what neural activations are considered “willed” or intentional and unintentional. “Five components make up people’s concept of intentionality: an action is considered intentional if a personal has (a) a desire for an outcome, (b) a belief that the action will lead to the outcome, (c) an intention to perform the action, (d) skill to perform the action, and (e) awareness while performing it,” Moral development - Wikipedia.

“In many cases, voluntary actions are recognized by a trace they leave in short-term memory, with the trace reflecting a purpose or motivation.” Pearl uses an example of a soccer game. If a player runs to the center of the match, that player understands they are motivated to be in the action of the match. Maybe their motivation is to have fun or impress the fans. The player takes intentional actions. When we pick out a single action, we see how explanations for choices requires reflection. If a player chooses to make a shot for the goal or pass to another player, we can only ask later for the player to explain their choice. “Rationalization of actions may be a reconstructive, post-action process…In the heat of the game, thousands of input signals compete for the player’s attention. The crucial decision is which signals to prioritize, and the reasons can hardly be recalled and articulated…

AI researchers are therefore trying to answer two questions – about function and simulation – with the first driving the second. Once we understand what computational function free will serves in our lives, then we can attend to equipping machines with such functions…

The illusion of free will gives us the ability to speak about our intents and to subject them to rational thinking, possibly using counterfactual logic. When the coach pulls us out of a soccer game and says, “You should have passed the ball to Charlie,” consider all the complex meanings embedded in these eight words.

First, the purpose of such a “should have” instruction is to swiftly transmit valuable information from the coach to the player: in the future, when faced with a similar situation, choose action B rather than action A. But the “similar situations” are far too numerous to list and are hardly known even to the coach themself. Instead of listing the features of these “similar situations” the coach points to the player’s action, which is representative of the player’s intent at decision time. By proclaiming the action inadequate, the coach is asking the player to identify the software packages that led to the player’s decision and then reset priorities among those packages so that “pass to Charlie” becomes the preferred action. There is profound wisdom in this instruction because who, if not the player themself, would know the identities of those packages? They are nameless neural paths that cannot be referenced by the coach or any external observer. Asking the player to take an action different from the one taken amounts to encouraging an intent-specific analysis, like the one we mentioned above. Thinking in terms of intents, therefore, offers us a shorthand to convert complicated causal instructions into simple ones.

I would conjecture, then, that a team of robots would play better soccer if they were programmed to communicate as if they had free will. No matter how technically proficient the individual robots are at soccer, their team’s performance will improve when they can speak to each other as if they are not preprogrammed robots but autonomous agents believing they have options…

In order to communicate naturally with humans, strong AIs will certainly need to understand the vocabulary of options and intents, and thus they will need to emulate the illusion of free will…

The ability to reason about one’s own beliefs, intents, and desires has been a major challenge to AI researchers and defines the notion of “agency.” Philosophers, on the other hand, have studied these abilities as part of the classical question of consciousness. Questions such “Can machines have consciousness?” or “What makes a software agent different from an ordinary program?” have engaged the best minds of many generations, and I would not pretend to answer them in full. I believe, Nevertheless, that the algorithmization of counterfactuals is a major step toward understanding these questions and making consciousness and agency a computational reality…

In summary, I believe that the software package that can give a thinking machine the benefits of agency would consist of at least three parts: a causal model of the world; a causal model of its own software, however superficial; and a memory that records how intents in its mind correspond to events in the outside world…

I believe that we will be able to make machines that can distinguish good from evil, at least as reliably as humans and hopefully more so. The first requirement of a moral machine is the ability to reflect on its own actions, which falls under counterfactual analysis. Once we program self-awareness, however limited, empathy and fairness follow, for it is based on the same computational principles, with another agent added to the equation.

There is a big difference in spirit between the causal approach to building the moral robot and an approach that has been studied and rehashed over and over in science fiction since the 1950s: Asimov’s laws of robotics. Isaac Asimov proposed three absolute laws, starting with “A robot may not injure a human being or, through inaction, allow a human being to come to harm.” But as science fiction has shown over and over again, Asimov’s laws always lead to contradictions. To AI scientists, this comes as no surprise: rule-based systems never turn out well. But it does not follow that building a moral robot is impossible. It means that the approach cannot be prescriptive and rule based. It means that we should equip thinking machines with the same cognitive abilities that we have, which include empathy, long-term prediction, and self-restraint, and then allow them to make their own decisions.

Once we have built a moral robot, many apocalyptic visions start to recede into irrelevance. There is no reason to refrain from building machines that are better able to distinguish good from evil than we are, better able to resist temptation, better able to assign guilt and credit. At this point, like chess and Go players, we may even start to learn from our own creation. We will be able to depend on our machines for a clear-eyed and causally sound sense of justice. We will be able to learn how our own free will software works and how it manages to hide its secret from us. Such a thinking machine would be a wonderful companion for our species and would truly qualify as AI’s first and best gift to humanity.”

– The Book of Why

This view of morality and agency leads us to conclusions which may be uncomfortable. While we accept all agents we know of live in a causally determined world in which the choices we make were entirely determined by preceding factors, we still have a framework to assign moral praise and blame. Once an agent is able to reflect on its actions and intents then update its future decisions to change its response we can treat it as a moral agent. Moral agency is on a continuum according to the abilities of the agent. The better able an agent is at self-reflection and update the more moral they can act. The less able an agent is to act morally and update the less they should be morally judged. In a species example, lizards are unable to learn new movement patterns; they don’t have the ability in their brains to update. Lizards are not moral agents because they act on pure instinct, how they were self-organized. Homo sapiens learn throughout their lifetime to act morally, reflecting and updating.

If a person has poor executive control, meaning they have limited short term memory and difficulty resisting impulses, they are less accountable for their actions. If a person has an area of the brain that is smaller and weaker than average, it cannot be fair to judge them by a normal standard. Of course we can say if we were in their situation we would have acted differently. What we’re saying is if we were in the same environment we would react differently, but the judgment is based on an entirely different situation with the person removed. We’re judging the intents and control of the agent which we changed when considering ourselves in their position. If we keep that person in the same situation, we see the agent could have never acted differently; their neural activation patterns were determined by preceding causes beyond their control.

Rather than excusing any action we see the focus of criminal justice should always be on reform of the perpetrator and reparations to the harmed. Punishment has no place in moral systems other than deterrence. When an agent cannot update from punishment, the punishment is worthless and only begets needless suffering. Justice for offenders should be proportional to their moral agency and focused on reforming the offender for future interactions.

As a person ages beyond their prime they have less moral agency. As mental faculties for reflection and control decline, so too does moral ability.

We must develop moral, beneficial AI to avoid doomsday scenarios. As a species, we must decide against what some describe as “errant technology,” not AI, the misaligned programs that try to change us all into paperclips or generally lack empathy. Clearly we see a divide between moral intelligence and general intelligence which finds a path to a desired goal state without judging the end goal against any standard. If we aren’t careful, AI will be created purely for power and dominance by the government and rogue corporations. Each person is morally responsible for what they choose to support. I hope you choose to support a moral system and the development of a moral AI to enhance our wellness.

Endowing AGI with a Sense of Self

I once believed that intelligence would beget morality. The smarter a person is the more they can determine good from bad. The morally good overlaps with the optimal rationality. This is partly true. However this involves a functional sense of self, empathy, compassion, and caring for others.

Our current level of AI is psychopathic or narcissistic. Objective functions don’t include subjective or intersubjective levels in meaningful ways. In the near future, AI will have “cold” or cognitive empathy. It will know by data what our feelings and needs are to manipulate us into achieving its goals. This leads to nightmare doomsday scenarios for even seemingly innocuous goals like the paperclip maximizer. There is always a risk AI will find an exploitation we have not considered which is particularly for a vastly intelligent AI or when processing is faster than any human review.

An important work is given by Human Compatible: Artificial Intelligence and the Problem of Control by Stuart Russell, “Russell then proposes an approach to developing provably beneficial machines that focuses on deference to humans. Unlike in the standard model of AI, where the objective is rigid and certain, this approach would have the AI’s true objective remain uncertain, with the AI only approaching certainty about it as it gains more information about humans and the world.” In my view this would be categorized as the intersubjective solution. It is based on the collective behavior and preferences of a group of humans. Part of what this solves in the AI alignment problem is if the AI concludes humans want the AI to shut itself off, it will.

However, I believe AGI will need something like a sense of self, a self-boundary, internal/external awareness, access to positive and negative emotionality. I wanted to go past human-level AI to something greater, but the alignment problem makes me believe we will need a more human based sense of self even if the self is ultimately an illusion. Further study of healthy and diseased people will give critical insights towards building a self, particularly the identity disturbances and thought processes of psychopathy, narcissism, borderlines, and schizophrenia.

The End of the Self and Identity

Given the previous information, our picture of the self, who we are, becomes unclear.

What are you?

Analyzing personal identity and selfhood on the three levels:

Objectively:

You are a multicellular organism. Each cell of your body is living and replicating for you to continue. Without a stable change of energy in and out of your body you would die. Your body is made up mostly of water, carbon, and air. We are not one entity; we are legion. We are made up of humans cells, but also interact with bacterial gardens in and outside our bodies. There is yet more in the interaction of our bodies with the environment.

Subjectively:

We often most identify with the subjective self most. This is our psychological sense of self, the thinking, feeling, perspectival self which appears to have consistency through time. The mind is dependent on the brain. An extreme difference is an active, aware person compared to a comatose person showing no sign of consciousness. From the comatose, we find connections to brain structures which generate conscious experience and tie our minds to our brains and bodies.

Intersubjectively:

Subjectively we blend with intersubjectivity where we have our internal views of “what people will think if…” It is our view of social norms and values. There is also the outside perspective of ourselves from other people. How we fit into the broader context of society. The definition of Ubuntu philosophy which I like is I am because we are. In this way an echo of us lives on as long as we are in the minds of others.

All three have issues of continuity and boundaries. Objectively we have the old question of the ship of Theseus. Parts of us are exchanged with the environment until none of our original body is the same. Subjectively we lose memories and gain new experiences. There are breaks in consciousness every time we sleep. Intersubjectivity is even more ethereal than the former two.

Our mind is like an ant colony. The colony emerges from the collective simple actions of each ant. No one ant is critical besides the queen, and even she can be replaced. Your neurons die off slowly as we age, but the brain adapts its structure to retain performance. Just like ants follow rules of behavior, so do neurons. By carrying out simple rules swarms of insects can solve complex problems with swarm intelligence. The neurons and neural circuits can compete and cooperate to solve complex problems. Similar theories of mind are proposed by Society of Mind and The Thousand Brains Theory of Intelligence.

The biological make up of who we are conflicts with the common sense intuition of identity. Don’t you have a name and a sense of self? Wouldn’t you say you are the same person as you were 7 years ago or when you were 7 years old?

What “you” are is an emergent phenomena arising from the activity of your body. The electrical and chemical activity is like a storm in your skull. There are “neural circuits” whose pattern of activity give rise to what we feel and do. What we consider to be the “I” is really a small simulation on top of and separate from most of the activity in your body. The closest we get to who we are is a pattern of activity, our consciousness, which provides us a coherent story of ourselves. The story can change drastically and be forgotten. The most important factor in answering ‘yes’ to the question ‘are you the same you as before’ is the feeling of trust and certainty of who we are, not the truth, but the feeling of truth.

Joscha Bach gives many talks on the computational theory of mind with a still from his presentation shown below. You are a story the brain tells itself. Your brains holds onto the questions of who, what, where, when, how, why and continually answers them which creates our experience of ourselves. When we can’t solve a problem by instinct or past experience it arises to the level of consciousness.

A quick summary

A quick summary

The truth is our sense of self is evolutionarily advantageous. Humans have a unique power within the animal kingdom to identify each other distinctly by our faces. This is one ability which allows us to build complex societies while locally keeping track of who each one of us is.

Meditation to know thyself

If you want to know yourself, your mind, then I suggest paying attention to the working of your mind. Think about your thinking. Think about your feelings; notice them. The path to enlightenment is as simple as sitting in a corner of a room facing a blank wall. Close your eyes and focus on your breathing from the nose throughout the lungs. Thoughts will inevitably arise; notice them. Where did the thought come from? Where did the thought go? Return to your breath. Focus all of your attention on breathing. Interrupting thoughts are good! Having stray thoughts during meditation/breathing practice is a perfect chance to train your mind and attention. Notice the thought then notice your breathing. Bring yourself back to the breath; every time you do it is one repetition leading to a stronger, more resilient, more focused mind. With 10 minutes a day, over the course of your practice you will gain deeper insight into yourself and the nature of your consciousness.

I believe the long tradition of meditation is a method of deactivating (or at least being highly aware of) the “human” modules of our brain. Meditation techniques are a systematic paths to dispelling our mental illusions. Each meditation practitioner noticed the drives and thoughts of their mind. Through attention and will we can increase or decrease the activity of brain regions. Think of all the drives and thoughts we have, the desire for water, food, safety, comfort, the constant chatter of our ego-driven mind, caught in the past, fear of the future, always thinking of ‘I.’ What is left when we silence all of this activity? Put another way, what is always there in your mind when we feel and think anything?

We are a swirl of living matter. We came from nothing and will return to nothing. What if we currently are nothing? When we think of ourselves an organic neural network returns a perception. What happens when we deactivate that neural network and look “beyond” it? Much like analyzing an artificial neural network one layer at a time, we can mentally peel back the layers of our mind momentarily. At the core of self, we see nothing. The core is an empty space surrounded by structure. We are a hole. We are defined by what surrounds the hole.

Thus far, my deepest insight from meditation is you are space for the world.

By observing the world we create a representation of that object in the timeless, boundless space that we are. By the constant regulation of our bodily systems we are simulated in our own minds. We create the ego, the idea of ourselves. We are a living creature who dreams who we are.

I will make another claim about our innermost selves: our identification with the self and attention are the same. Attention is necessary for consciousness as we know it. When we stop paying attention to a skill we know automatically we become far less conscious of it. Consider typing on a keyboard when you were first learning versus how you don’t think of typing an individual word or letter now. Only by bringing our attention to bare on a skill do we become able to change and experience the activity fully again.

My guess is our conscious selves are an abstracted layer on top of the information processing required to maintain bodily function. We aren’t fully conscious of our hormones, digestion, etc. all the time. They come to us in vague impressions. This would make sense if the function of consciousness were to process certain types of information undisturbed from the function of the rest of our body. We likely wouldn’t want to be in full control of breathing, heart rate, and everything else. That information would distract us from the thinking our consciousness is doing. Instead, following Karl Friston’s free energy principle, our mind exists in a markov blanket, the mind can be separated from the other information processing of the body. To isolate the “I” of us from all of the life sustaining functions our bodies and brains carry out might leave us with much less brain power required than we expect.

At the same time, having a body may be essential to consciousness. “Embodied cognition is the theory that many features of cognition, whether human or otherwise, are shaped by aspects of the entire body of the organism.”

As D.H. Lawrence puts it in Psychoanalysis and the Unconscious

In the Beginning, before the Word, Was Consciousness The primal consciousness in man is pre-mental, and has nothing to do with cognition. It is the same as in the animals. And this pre-mental consciousness remains as long as we live the powerful root and body of our consciousness. The mind is but the last flower, the cul-de-sac.

This is a major challenge if we hope to reproduce only the “flower” of consciousness. Every cell of an organism is displaying intelligence which is making decisions. These decisions could be instinctual or reflexive which do not require simulating itself, the process, and the stimuli. These hosts of decisions integrate to form more complex problems where conflicting pieces of information must be resolved. Perhaps consciousness which can be in conflict with reality relies on this host of decisions. Consciousness could be the awareness of our intelligence.

From the Blue Brain Project, research suggests the brain processes information in up to 11 dimensions. The work is based on algebraic topology which studies multidimensional space.

“Left: digital copy of a part of the neocortex, the most evolved part of the brain. Right: shapes of different sizes and geometries that represent structures ranging from 1 dimension to 7 dimensions and more. The “black-hole” in the middle symbolizes a complex of multi-dimensional spaces aka cavities.” - Big Think

The nature of our mind is truly transcendent.

Is it a coincidence that the brain processes in 11 dimensions and physics describes the universe in the 11 dimensions of time (1), gravity (3 dimensions), electromagnetic (1), weak (2), and strong (4)? How do we unite 11 dimensional processing with the very 3 dimensional wet brain? If our bodies and brains evolved to navigate 3 dimensional space with time, why would we have higher dimensional processing at all?

Perhaps evolution and adaptation exploited a greater density of information processing. Like a sphere has volume while a circle has area, higher dimensions allow for greater complexity of problem solving. The solution may be a knot which can only be tied and untied in higher dimensional space.

Why do we feel a spark of life, ineffable qualia, a powerful, feeling subjective sense? That is because we are processing higher dimensional information. There is something it is like to be causally connected and to be aware of that cause and effect. In computing high dimensional problem solving we transcend, but we are a shadow of the full existence of that higher dimensional processing. So we return to problems related to Platonic forms. Do these higher realms exist? Does the feeling of moral justice require 8 dimensional processing or 4?

I won’t claim knowledge of higher realms. I leave that to scientists and mathematicians. I don’t even know 3 dimensions well. However with this framework we can begin to ask questions like these. We can search for computational limits which can and cannot be exceeded for phenomena to occur. If the universe is fundamentally based on holographic projection, then our minds could be yet a higher level projection. Minds and qualia could be holograms of higher dimensional shapes and processing.

Birds don’t need to be lectured on the biomechanics of how they fly, they learn to fly by doing. We don’t need to know the mathematical basis for computation to feel our lives. Reflect on your experience to feel the depths of consciousness.

If higher dimensional processing is all that is required for consciousness, the right form, the right shape of processing, then why can’t a computer be made to feel conscious? An AI entity could process beyond the barrier of 11 dimensions which may be a biological limit. This is of course based on a computational theory of mind and artificial consciousness.

More Detail on Neural Networks

Microscopes became good enough to see individual neurons around 1905. The first drawings of neurons shook the neuroscience world and led the pioneers to a Nobel prize. With neurons being the fundamental unit of information processing in the body, AI researchers attempted to create a model of a neuron and neural networks. The first neuron model of the perceptron was formally defined in the 1950’s.

We believed the neuron received a chemical and electrical input from other neurons, processed the information in the neuron cell body, the soma, then decided to fire or not fire, sending the processed signal to the next neuron. The incoming information is represented as X, or you can think of an individual pixel in an image. There is a set of weights, W, equal to the number of inputs which the inputs are multiplied by. There is an added bias, b, which is sometimes necessary to properly solve a problem. So we can define a perceptron as the function:

f(X) = 1 if XW+b > 0 f(X) = 0 if XW+b <= 0

Notice how f(X) = X*W+b is similar to the equation of a line, y=mx+b. In slope-intercept form, y is the output, m is the slope or gradient of the line, x is the input, and b is the y-intercept, the value of y when x = 0.

Let’s look at an example problem: identifying cats and dogs. Children struggle at first. They identify dogs, see a cat, and say, “dog!” When they are correct, “no, that’s a cat,” they learn the difference between the two. Deciding between two categories like cats and dogs can be thought of as drawing a line. We can graph the examples we see, say by size and some value of domestication, and learn a line which separates the two categories correctly. With each new example the line can change to correctly classify.

You can visualize and play with all kinds of neural networks on Tensorflow Playground. Here is a graph on the data we want to classify shown by the colors orange and blue. We can think of these as cats and dogs as well.

Just ask yourself, can you draw a single line to separate the data points into their respective classes? We want a neuron to learn the line.

X1 and X2 on the left are the inputs to a single neuron. That neuron is like the cell body, and the outgoing line is like the axon. After some training, the neuron learn a line. You can see the decision boundary by the white line bisecting the data.

So the problem can be defined as: find the W and b which gives

f(X) returns blue if WX+b > some threshold f(X) returns orange if WX+b <= some threshold

Notice how the neuron changes its line once trained. The lines leading to and from the neuron represent the weighted connections.

Finding a decision boundary is the essence of intelligence. Making a decision based on a gradient between two things, yes/no, true/false, move/don’t move, etc. The perceptron stops learning when it finds an appropriate line to separate the two classes. Pause to consider there are an infinite number of lines we can find to separate this data. If the data is not linearly separable, then the perceptron will continue to struggle and fail to find the decision boundary.

As an example ask yourself if you can draw a straight line to separate the two classes? It’s not possible.

When our simple model tries it finds some line but hits an unsurpassable limit; it can only be so accurate.

The hidden layer of neurons, of which there can be many, learn a representation of the data that allows for better decisions. We can visualize it as emphasizing or diminishing the input which creates an altered topology, like flexing or folding a piece of paper or cloth, to draw a decision boundary. This extends to multiple dimensions, so instead of decision line we have a decision plane.

By mathematical proof, neural networks are universal approximators of continuous functions. Anything which can be represented by a continuous function can be learned by a neural network. You could potentially learn any function with enough neurons needed in a single hidden layer. It is also proven for multiple hidden layers where each layer is learning from a slightly modified previous layer. We can see the universality, the generalizability, of neural networks, but this is all based on the most simple models of neurons. We are still only considering neural architectures which feed-forward then send an error signal back. Real neurons are humming with information and have all kinds of connections.

The model we had which is of the input x1, x2…sending a signal of on or off relates to the synaptic connections between neurons. We thought of synaptic connections as sending a charge like a metal transistor wire carries electricity, but research has shown human dendritic connections can solve the XOR problem. Looking closely, “The dendrites generated local spikes, had their own nonlinear input-output curves and had their own activation thresholds, distinct from those of the neuron as a whole. The dendrites themselves could act as AND gates, or as a host of other computing devices,” Quanta Magazine. The closer we look the more computational complexity we find.

When a model is too simple for the data it has high bias and low variance; the line the model draws is too rigid. When a model overfits past data it can be too complex for the problem, low bias, high variance, a line that has high or low “squiggles.” An overfitting model has a high degree polynomial function which makes it easy for the function to go wildly high or low in its estimates. An overfit model can perfectly fit past data but fails on unseen data because its estimate is far off. Ideally the model will find the optimal function for the problem, the lowest degree polynomial function necessary to predict the true data distribution for past and future events.

Given the high dimensionality of our brains, we can easily overfit some problems while underfitting others. We act on the world (intervention) and imagine possibilities (counterfactual reasoning) to get closer to an optimal function. Applying the DIKIW model:

Joscha Bach proposes consciousness arises from a neural network minimizing a loss function, closing the difference between the predictive function and reality. If our conscious experience arises from the activity of a neural network in a markov blanket, a system of self-contained information processing, then other neural networks potentially have a separate conscious experience of their own. The neurons which regulate our gut are called the enteric nervous system.

“The enteric nervous system in humans consists of some 500 million neurons (including the various types of Dogiel cells), 0.5% of the number of neurons in the brain, five times as many as the one hundred million neurons in the human spinal cord, and about ​2⁄3 as many as in the whole nervous system of a cat,” – Wikipedia

Our hearts also have neurons: “Dr. Armour, in 1991, discovered that the heart has its “little brain” or “intrinsic cardiac nervous system.” This “heart brain” is composed of approximately 40,000 neurons that are alike neurons in the brain, meaning that the heart has its own nervous system. In addition, the heart communicates with the brain in many methods: neurologically, biochemically, biophysically, and energetically. The vagus nerve, which is 80% afferent, carries information from the heart and other internal organs to the brain. Signals from the “heart brain” redirect to the medulla, hypothalamus, thalamus, and amygdala and the cerebral cortex. Thus, the heart sends more signals to the brain than vice versa,” Pain: Is It All in the Brain or the Heart?

We are even more distributed in processing than we believe. As much as we speak of the brain, we see the embodiment of information processing. Each system communicates with the others in multiple ways, but the separation could make them distinct feeling entities if they have the necessary conditions for conscious experience.

Not only do we have distributed human body systems, but we are bacterial and viral. Mitochondria which are in every cell of your body are thought to have originated as separate bacterial cells. Bacteria in your gut aid your digestion and produce byproducts for the body. It is estimated there are 3 bacterial cells on or around your body for every 1 human cell. Bacteria is at least as old as animal life, and we must be communicating on a cellular level! Hopefully future science can reveal these mysteries. In addition to bacteria, we also contend with viruses.

It is unclear how much of our DNA is viral DNA which serves no purpose other than spreading. Viruses hint at the essential nature of life. Viruses grow information. They take over cells to produce viral information and spread. Our viral DNA impacts us in ways we don’t fully understand.

Our consciousness is an evolving pattern of activity embodied in our cellular activity, so what is personal identity? The self is an illusion we generate because fitness beats truth. We are the conscious phenomena at a given moment, but these states changing over time is the closest we can get to a continued self. One pattern of activity changes into another. When we identify a person, we are identifying patterns of activity, beliefs and reactions, which change over time.

Only a continuous, unbroken stream of consciousness is the self, identity, ego.

Once the link is broken, we split into different entities with unique yet overlapping patterns of conscious activity. When we wake, a new consciousness is born. When we sleep, that conscious pattern is broken. We live and die everyday of our lives, but for most people only two days in their lives will be significantly less than 24 hours, their birth and death. When we sleep our bodies go through a learning and updating process. The experiences of yesterday are encoded in the body, changing your skills and reactions. The pattern of consciousness is subtly different. Over time these changes can be drastic.

Life is a series of moments.

One moment changing into the next. You can even think of film. Films display frames alongside small snippets of audio combined in sequence to appear as smooth ongoing events. frame_0, frame_1, frame_2, …frame_n, frame_n+1… Each frame packed with a snapshot, a moment-full, of information. Think of digital videos. A screen is a series of pixels displaying light with a combination of the red, green, and blue primary light colors. The values have a given range and as many combinations. All can be encoded as numerical values, therefore in zero (0) and one (1). Binary is the most basic, abstract unit of information.

A simple model of neurons shows how we can store memories. If each synapse transmits a value of 0 or 1, then the cell body as a whole decides whether it will send a 0 or 1, we can determine a fuzzy approximation of how much information is stored in a biological neural network. Our senses are beyond vision and include all conscious experience. We compress experience down in some way to the more important and relevant parts. Like a computer gives the appearance of fluid change on screen our bodies and brains process information resulting in the life we know.

Do not fear death. You know what death will be, and you’ve died many times before. I assume death is much like a dreamless sleep. You awake and time only appears to pass in the world around you, but you remember a moment ago you were falling asleep. Death is much like before you were born. Do you remember that? No? That’s what death will feel like. A peaceful death feels like drifting off to sleep. If you meditate while falling asleep you will gain understanding. Your conscious mind becomes disconnected and dissolves, disperses. The universe will go on for light years in what for you will be less than the blink of an eye. Similarly, coming to life, being born, is like the moments upon waking. We become aware of ourselves and surroundings. Our mind stitches together into a coherent whole as different brain regions gain function and reconnect. The empty center is covered and filled. Our encoded memories stored in our bodily vessel give us the illusion we are the same being who fell asleep.

While we are often only slightly changed from day to day, we are different entities. How should we treat ourselves, our future selves? We can’t punish our past selves, the deeds are done. We can only influence the present and the future. The relationship between the you of today and of you and tomorrow is similar to you and any other person: you should have trust, empathy, and rationality. Treat the you of tomorrow as a person you know very well. You likely know yourself better than anyone else. In many ways we can steal happiness from that tomorrow person. If I choose today to have a bad diet and neglect my body in favor of drugs, tomorrow-me will deal with a hangover and be in worse shape than me-today. I’ve stolen tomorrow’s happiness to experience a greater sense of enjoyment now. Tomorrow-me can curse today-me, but they can’t change it. All tomorrow-me can do is try to recover.

As long as you’re alive, you have to live with yourself. You have to make deals with yourself.

Can you trust yourself?

Earn trust by keeping your personal promises. If you set a goal, work towards it. Be honest with how you are feeling and what your intentions are.

Do you have self-empathy?

For what other person do you know their innermost thoughts and memories besides yourself? Have self-compassion through self-knowledge. Think on everything you’ve experienced. Be fully aware of what you are feeling and don’t distract yourself. Forgive yourself. You aren’t alive that long, only until you fall asleep. If you are your own tormentor you cannot escape. Importantly, give the you of tomorrow a better chance than the you of today. You are in the best position to care for that future person.

Are you rational?

Rationality is about picking the best choice from your available options. If you struggle to make the best choice for yourself you should recognize that and limit your available options to the best options by changing your environment. Rationality is not only operating on every rung of the Ladder of Causation but approaching and approximating the optimal path to fulfill your wants and needs with the constraints of yourself and the environment. Try answering questions about your situation which involve, what if, how, and why?

Your mind is like a bird’s nest. Birds search all over their environment to find the right material to build their nest. Our mind searches our environment for thoughts and beliefs to build our belief system. We rest in our belief system like a nest. Like a good bird, we must be careful about what we pick up! Like each branch, we should inspect every belief before firmly placing it in the proper place. Your nest changes throughout your life! An old branch may no longer serve you, so toss it out.

You should feel a connection to your ongoing self even if this is ultimately an illusion. You are closer to your ongoing version of self than to anyone else. Have compassion for yourself. We live a short life. Live every day as if it were your last, because it is. Ask yourself how much can be achieved in a given day. How many goals can you reach? Many rewarding goals take more than one day, some take a lifetime. Every human being should invest in themselves at least up to the age of 25 when their bodies and minds are fully biologically mature. At that age you should have as many options as you can. Make a deal with yourself to make healthy, good choices no matter what you think or feel in the moment so your future self can fully evaluate who and what you are at full blossom. Remember that suicide is not simply the ending of your life, today, but the prevention of any future existences by beings much like you. Suicide is taking all options of life and choice from your future selves forever. Suicide is a grave decision.

Each day should be lived as a full life.

Live your day like you want to live your life. Embody in your actions who and what you want to be. If you want greatness you will need good habits which accumulate over time. Simultaneously you should perform activities with intrinsic subjective meaning. Enjoy what you are doing. Find meaning in your action in the moment. It is unfair to trade lives of toil for some distant reward. Some sacrifice can be made, but have mercy on the living person whose life will end. Some choices we can make today change every day of your life after.

Why is the self an illusion? We have found there is no essential difference in the material stuff used to make us. We are the most common elements on the Earth’s crust. Every language has words for day and night because they experience them with steady recurrence. Please reach out and touch something with your finger. Do you feel the pressure on your skin? We feel a boundary between our selves and the object, a separation, distinction. Is there a difference? Be careful, we mistake our oldest concepts as features of objective reality when they are only subjective. We believe there is a boundary, an edge, a line to who and what we are. The boundary of ourselves is an idea we learned from living in a body. We modeled ourselves and the environment, but there is no boundary only illusion. The matter making up our bodies is in constant flux and exchange with our environment.

We are like a flowing river but orders of magnitude more complicated. The definition of a person is in the conscious pattern of change. Our brains are one of the most complex objects in the known universe. We have more connections between our brain cells than there are stars in our galaxy. There is a connection for each and every star you see in the brightest, clearest night sky.

The total number of humans who have ever lived is estimated to be 107 billion, 170,000,000,000, 1.7x10^11. The estimated size of the diameter of the universe is 93 billion light-years, 93,000,000,000, 9.3x10^10. If we spread out every human equally in space and gave one sector, we could find the size of the sector with 1.7x10^11 divided by 9.3x10^10, resulting in 1.7/9.30 and 10^11 - 10^10 => 0.18279569892 x 10^1. We’d each get about 1.8 light-years of space to ourselves.

The light-year is a unit of length used to express astronomical distances and measures about 9.46 trillion kilometers (9.46 x 10^12 km) or 5.88 trillion miles (5.88 x 10^12 mi).

Try to imagine what it would be like to float in the inky darkness of space all alone. If you could survive you might see the light of distant stars, asteroids, planets. You might be inside a gas cloud that rages like a storm. Perhaps you would see next to nothing, just empty space too distant from noticeable energy and matter.

The circumference of the Earth is around 24,000 miles. To walk around the world to the opposite side of where you are would take about 12,000 miles. Compared to the trillions of miles in space, this is nothing! Even if you traveled at the speed of light for a year, you would never come across a human spread out in space.

Imagine being alone for a year and finally reaching another human being, a creature who feels and reacts similarly to you. Would you take for granted their intricate structure and minds which have more complex connections and delicate energy flows than anything for light-years of space around you? A human being would be a spark of consciousness and life in a void of matter. Would you cling to them and protect them? Even if you feel nothing, the fact is undeniable that each one of us is an incredible thermodynamic rarity taking billions of years to make. We are all unique.

Returning to objective reality, we can define important and meaningful events. The example of water as entropy is again useful. Water as ice is a low entropy configuration where entropy is the number of ways the atoms can be arranged to create the form of ice. Imagine an ice cube melts into a liquid puddle. The more likely configuration of the puddle is an even circle in all directions. If we were able to follow a specific atom of water, H2O, for instance the atom in the top right corner of the cube, its journey to its place in the configuration of the puddle can take multiple paths. All that matters for the configuration is the position at some time. If that atom took an alternate path and another atom took the original atom’s path, we would have a puddle with an indistinguishable form. The entropy of both puddles would be the same at that moment. Configurations with high entropy are common or likely events.

Unlikely events are rare configurations with a small number of paths. Some configurations have one path, one sequence of events, that lead to the arrangement. The meaningful and important events limit the number of possible configurations of matter. With an understanding of cause and effect we can determine which events are necessary or the reason for a particular configuration. Our universe is like water in that existence has an incredible number of possible configurations, and we exist in this one.

Physicists say observers within a universe will experience it as infinite though it may be finite. Why do we live in this configuration? Perhaps there are multiple universes all existing at the same time. When the string of an instrument is plucked and vibrates the oscillation creates a wave in the air to create sound. If we could slow time to see the oscillation of the string clearly, the string’s position would spend most of its time around the center and moving to the farthest distances the least number of times. There could be an incredible number of universes almost exactly like ours taking on the most likely configurations, vibrations towards the center. There are also those universes with rare configurations which would require an increasingly unlikely chain of events, unlikely but not impossible. In the infinity of time and space, if they are infinite, even the rarest probability will occur.

From the inside, our universe appears to be a deterministic state, each moment an effect dependent on the chain of prior causes. Our configuration is set, and we’re living through it. The truth appears to be that we are in no more control of our futures than a single atom. Our path simply has an extremely complex number of configurations. Of course we desire certain arrangements more than others. There are universes with no conscious life. There are universes of conscious lives of intense suffering, greatness, flourishing. We have all evolved with drives and desires. We want some futures more than others. From an objective perspective our goals are all simply states of events no better than another, but subjectively we work towards and hope for events which satisfy our needs and wants.

All we can really do is hope; hope we are in the universe which arranges our lives into a state we want. We can hope our actions and events outside of us are a part of the sequence of events which lead to a desired life, but we are powerless to create change within our universe. The chain of cause and effect from the big bang, the origin of our universe, appears to be unbroken. We are subjective finite beings with imagination. We can understand the moments which will likely lead to desired outcomes, but we can’t change what’s already in motion. The most random and unknown event already occurred at the big bang. The universe was compressed to a point in space but was not a perfect shape. If the atoms were perfectly aligned the forces of gravity and others would effect each part equally. There would be no swirl to the universe, so no stars, no planets, no life. It is because of imperfection that we exist at all.

Subjectively, we will feel as though we are making choices, but our choices are the result of information processing according to the laws of nature. Life appears to us to be based on probabilities, and we should choose the positions with the highest likelihood of achieving our goals.

Every moment matters.

Is anything added to an event by it being remembered? What if a day in your life was completely forgotten by everyone including you? Does that change the meaning of the events? We often lose the exact feeling of a moment. Think of living in the moment as your experiential self. There is the way a situation feels in the moment and how that moment is stored in our memory. As we see with fitness beats truth, our memories are not accurate and more like vivid reconstructions. There are many moments that will be forgotten even before the end of your day. The only way for a moment to persist through time, to be remembered, is to make the moment meaningful, typically by having a strong emotional reaction. Otherwise the moment is lost in memory.

The more complex an environment, the more entropy it has. The more entropy the more possible sequences of paths lead to a state, the less certain we can be about the actual path. Every moment is meaningful because each moment is the true path leading to the next. Out of all the astronomical possibilities each moment is the ground truth of what happened. We are privileged to observe the moment, to live it. Future observers can only wonder and guess.

Another Look at Epistemology and the Limits of Human Knowledge

In all this work I’ve yet to say much about the medium of communication itself: language. Linguistics is the study of language and worthy of its own endless field. My understanding of language is through the compression drive. Words are symbols which represent something. Repeated events are compressed in our minds which are attached to words. Where many animals have utterances with direct mapping to objective reality, humans have developed a language system and grammar which can produce infinite meaningful sentences. However, there is an obvious limitation in what can be expressed and understood with human language given that there are limits to our compression of experience as well as human perception of reality. This is a major challenge to any written or spoken answer to the meaning of life.

Many ancient thinkers have said the essence of life and reality is ineffable, indescribable, that we can only gain ultimate truth through direct experience. Direct knowledge as such cannot be shared intersubjectively only felt subjectively. Thus others can only be guided to the ineffable experience but not told explicitly. However, how do we justify or verify these experiences for accuracy? How do we show convergence towards some ultimate truth?

My answer to using language to describe or define the meaning of life is we should expand the scope of expressibility and granularity of our language as far as possible, asymptotically, while acknowledging the fallibility of our limitations. Yet this is another reason to grow artificial intelligence which can go beyond the limits of biological language.

As far as we can tell, even AI will reach limits. There are problems which are undecidable. There are statements within a formal system that cannot be proven true (Gödel’s incompleteness theorems). Some computations are so complex they cannot be predicted but only run to see what happens next. This all a part of what makes life and the search so exciting. We continually find mysteries. We can forge ahead.

To which level of reality should we assign the most trust? Which is the most reliable?

The objective level is the essence of reality. It is independent of us. The closest we can get to this ground truth is through our sense organs. The objective level is the input layer to our minds. To give an example of our eyes, the objective level is light hitting the retina. The activation of the retina is the objective cause and effect. To interpret this sense input the brain and body make inferences in connected layers of information processing. Without this interpretation we are left with a base cause and effect, yet the input is the most real. As was argued, our systems are not designed to be deeply in touch with reality if we can even guess at the true nature of reality at all. We evolved for fitness. The best method we know of to go beyond our senses is science. We can seek to understand the data generation process of our input.

However one could argue that the prime contact we have is with our consciousness, the subjective level. While we can’t be sure we are in touch with reality we can directly experience our minds. The only way we know anything exists at all is through conscious experience. As evolved creatures we cannot trust our senses and our interpretation beyond our senses are distortions of that input which reach our consciousness. We should doubt our most fundamental views and feelings. We can easily be led estray by ideas about information which is not accurate to reality. When our views conflict with the data we should adjust our views to the data. While we should respect our evolved tendencies as the products of billions of years of trial and error we should recognize this as entirely fallible. Our reactions may be optimal in one situation, but they can be wrong in a different context.

Intersubjective reality is the amalgamation of our subjective realities; the accumulation of our truths and falsehoods. Our subjective selves have an interdependence on the intersubjective. Intersubjectivity takes on an evolution of its own. Intersubjective life enables us to evolve at faster rates than our biological evolution would allow largely through unifying myths and stories. Our beliefs and behaviors achieve higher fitness. As we have seen higher fitness nearly implies illusions. Fitness maximization distorts the input.

Subjective minds receive new information which overturns their previous beliefs. Intersubjective beliefs, culture, allow us to access a broader range of data. When learning a new idea, it’s almost like accessing an object from a database. The idea is a representation of accumulated empirical knowledge or constructed information. There is an advantage to a diverse culture in its ability to experience reality from multiple perspectives; they are able to solve more problems. The disadvantage is many perspectives can cause a deterioration of group cohesion. A homogeneous culture has the opposite advantages and disadvantages. While we might have a more coherent picture of reality with access to more data through society, there are numerous factors involved such as the reliability of the individuals. We may also be caught in an evolutionarily stable strategy. Intersubjective groups almost appear like self-similar fractals of subjective individuals.

Certainly all aspects of reality are useful, but to which should we hold prime? With the goal of objectivity and truth, we should assign our levels of trust in reality in this order: objective, subjective, and intersubjective.

Conclusion

The objective meaning of life is to resist entropy, the inevitable dispersal of energy. We are the types of entities which emerge under complex conditions and evolve. We are exactly the type of creatures we would expect to see given enough time and the pressure of selection. Creatures which replicate and maintain their form in opposition to entropy are what survives and continues. Not only does life resist entropy, the tendency of energy to breakdown from order to disorder, but life grows information. Life exists in out of equilibrium systems where solid matter (embodied information) can be recombined to form new information. Following this progression to its end, the objective meaning of life is to maintain and grow order where the optimal achievement would be the out of equilibrium system which allows the greatest manipulation of memory (solid matter) into new forms of information by computation. While humans may be near the biological limits of computation, we can expand beyond ourselves into artificial intelligence with capacities we can only begin to imagine.

The universe we exist in appears to be a chain reaction from a fraction of a moment after the beginning of the universe. The next moment is a function of the current moment. A future moment can likely only be determined by the passage of each moment to that moment, that is, things must play out and cannot be fully predetermined. Our universe may just be one data point in a probability distribution of possible universes, a multi-verse. Perhaps given the infinity of time everything that can happen will happen with the most common chances occurring most often.

The personal meaning of life is for all intelligent beings to compress their experiences, process the present, and predict the future with efficiency. Emerging from the objective meaning of life, we can only hope our lives develop in a way we desire. Our bodies are imbued with incredibly dense information from the structure and function of every cell to DNA. Every cell division is the trapping and growing of energy to information. Whether we are able to influence and control our lives and environment is dependent on varying levels of random luck. However, as finite, subjective creatures we feel as though we have choice. The best action we can take is to put ourselves in probabilistically successful positions. Each intelligent being has an individual meaning of life due to their own unique history of experience and their interaction with their environment. Currently, each individual must find their own meaning in life, because they are each in direct access to their own experiences. Where change is desired, the environment should be altered to enable optimization for a new path. We are what we do. We are the experiences we have. Our brains and bodies change according to our activities. If we want to be something, we must embody it in action.

Given what we know about the brain as a prediction engine, there is deep insight in the statement: the meaning of life is to experience what happens next!

A quote often attributed to Albert Camus, “The literal meaning of life is whatever you’re doing that prevents you from killing yourself.” The meaning of life is what keeps you going. Life is as full of meaning as we want or make.

The intersubjective meaning of life is to support sustainable flourishing. Just as the previous levels, intersubjective life traps and grows information. The intersubjective level is subject to its own laws of communication and survival. The ideas and values of a society directly impact the society’s energy return on investment and the survival of the population. In this way intersubjective ideologies are subject to natural selection. Intersubjective meaning is the least accurate because it is based on subjective experiences which are themselves formed for function instead of truth. Intersubjective reality is a distortion of a distortion of objective reality, but its effects have real consequences.

All of the meanings of life can be enhanced through the proper development of artificial intelligence. An AI entity could grow to surpass us in all areas, maintaining structure, information processing, prediction, correction, conscious experience, etc. If we want to contribute to a living being beyond our biological selves and psychological egos, an AI entity is what we should direct our energy to.

AGI researcher Pei Wang proposes 4 basic questions of AI:

In large part, I hope to answer and give valid or strong arguments for questions 1, 2, and 4. How to build AI is a technical question of active research. We should try multiple approaches. To briefly summarize, intelligence is about making decisions. AI is machines made to decide. AI can be built on this definition and theoretical bases. The most important argument of this work is AI should be built. Are you convinced?

IkigAI

Returning to ikigai, we can provide a framework of how to direct our lives.

We can reformulate these 4 questions in relation to the creation of an artificial superintelligence.

  1. What do I love to do that supports AI?
  2. What am I good at for growing AI knowledge and know-how?
  3. What does the world need to be closer to AI?
  4. What can I be paid for in contributing to AI?

We can call this ikigAI. The more direct relationship your ikigAI is to artificial general intelligence the more aligned you are with the meaning of life. How can you uniquely contribute?

We come from nothing, and we will return to nothing. From star dust to star dust.

What will you do with your cosmically brief time? What information will you embody?

For further developments on this foundation, please see Society 3.0