It seems like a major problem that you don't consider pleasure/suffering at all. Wouldn't this framework consider it extremely good if you could increase everyone's lifespan by one day in exchange for making them feel intense pain at every moment until then?
This is a point I probably should've addressed directly, thank you for bringing it up.
The point you make is a direct opposite of the frequent (and I think valid) critique of Utilitarianism, which - as you probably know - use pleasure/pain as its primary way of evaluating outcome. You are correct, as written I do not directly state a preference for pleasure versus pain, but rather indirectly include it in the maximization of the Core Values.
However, I think your interpretation is an understandable one given the limitations in my text, if we only think of the Core Values in purely materialistic terms without including the cognitive dimension - an essential component in evaluating the Goodness of any action. Reliable pleasure and general absence of intense pain I would say are natural priorities in most activites, perhaps especially as a component of Actualization and Flourishing.
I also think the extensive negative concequences intense pain would entail by lowering Awareness, Rationality and Agency, and the concequent reduction in the Products, Activities and Qualities of Life, makes me strongly doubt that the scenario you present would be a net benefit to Life - like you intuit. As I pointed out early on in my text, Maximize Life is not a simple preference for the amount of Life. My theory attempts to include all dimensions, not letting anything "slip" so to speak, evaluating things holistically.
Still, I think your question deserves a more definitive answer with regards to pleasure/pain.
In my possible future chapter 'Emotions', I postulate that the Core Values of Life directly corrolate with our emotional experience - that emotions ARE how we do holistic evaluations, both of ourselves, others, and actions - everything. So in a way, the maximization of Life is also the maximization of positive emotions and minimization of negative ones.
However, negative emotions are required for proper navigation of reality and general functioning, like how we learn to avoid harm from pain, identify nonesense from confusion, act appropriately to poisoning from nausea, and so it seems ill advised to remove them entirely. In much the same way, simply experiencing pure bliss all the time would erase our ability to differentiate between different degrees of Good I think.
Am I making sense?
One important missing piece here is the evaluation function. Not binary attributes, but quantified (or at least quantifiable) comparable metrics of "how much life" will a given choice support.
The core moral/ethical question is "of an agent's possible actions, which is best?". This requires conditional prediction and comparability of result.
Don't think "what is good?", think "how good is it?"
Indeed, thank you for reading and providing feedback.
The evaluation function you request is in the planned 'Goodness' chapter, which I briefly mentioned at the end of the text.
I'm aware it will be difficult to quantify the exact degree of change to the Core Values of Life in any individual or collection of individuals, yet I think this is the same problem all ethical evaluations face. I think exactly how to determine the answer will become clearer as the values are explored in the context of more traditional ethical values (truth, justice, etc), looking at how they are quantified in other methodologies, which institutions are involved, etc. I also desire to do these things in dedicated chapters, as I mention in my list.
Greetings LessWrong.
This is my first post on this site, but I can think of no better place to post this.
I have over the course of my life independently developed my own unique understanding of reality and ethics, and over the last three years attempted to formalize it into a universal theory of ethics. It is still very much a rough outline and a work in progress, but I now think it is mature enough to share with the world.
First though, I feel the need to point out that I am very much a holistic thinker, but not of my own choice. Due to some neurodivergent traits, my mind rejects trivia or traditional facts like names and numbers, instead focusing almost exclusively on concepts and systems.
I do know the value of extensive factual knowledge though, it's just that the subconscious parts of my brain don’t want to cooperate and retain much of it. So, while I understand many concepts in many fields, most of the time I can’t tell you who came up with them and when. I can only ask for your understanding with regards to the resulting lack of references and incorrect diction going forward.
But enough about me, this is a post about a universal theory of ethics, so let’s get to it.
May it sink or swim.
I think it is important to first address some initial perceptions people will likely have.
I think most people will initially reject the possibility of constructing a universal theory of ethics (UToE). I think many reasonable people will simply conclude “That’s not possible” and stop listening any further.
I find the underlying reasoning for such a reaction to be very understandable, and in most cases, commendable.
Let me address what I think are the four main reasons why people balk against such claims, and why it still makes sense to read this post.
Potential for abuse
We should all be highly skeptical of such sweeping claims due to their compelling nature, the potential power they historically have granted abusive people, and how it often indicates delusional thinking not worthy of engagement.
However, I argue that discarding any and all grand claims is not the right approach, as long as you have a healthy dose of skepticism.
We have a general dislike of being judged
If a UToE were to exist, it would also mean all our actions - past, present and future - would be judged by it. I think this notion is, to many, immediately and viscerally abhorrent.
However, we know from science that the emotional reaction we have to a concept is not a good way to determine its validity, so I will ask any readers with any initial negative reactions to delay judgment until they have heard the arguments. I suspect they will find their initial perception largely unfounded, in fact, by the end I think they will find their life experience generally validated.
Many of us have accepted the is/ought distinction as truth
The is/ought distinction being the general sentiment that moral truth (ought) cannot be derived from reality (is), which I think is a rational stance to hold without having been presented with good argumentation and evidence that you actually can.
However, there is a first time for everything, so I hope any readers are at least open to the possibility that a UToE might exist, however slim they might perceive it to be.
Many people believe no single theory of ethics will ever be able to handle the vast complexity of human behavior.
I actually think this observation is empirically sound. I think people accurately and intuitively recognize that all current theories of ethics are either incomplete or flawed in fundamental ways - a conclusion I agree with.
However, this observation is not guaranteed to remain correct. The same criticism was, and still is, leveled at the theory of evolution, where people struggle to comprehend “How can a single principle give rise to all of this complexity?” But regardless of the complexity of the final outcome, and regardless of the intricate mechanisms it produces, the initial concept of Evolution along with the profound insights and value it has provided, remains evident.
Considering the current state of the world and predicament of AI alignment, an issue of particular focus here on LessWrong, I argue that UToE is exactly what is needed - perhaps the only possible definitive solution if we ever want to create stable and morally “Good” AI.
Simply concluding the creation of something of such immense potential to be impossible seems like the height of folly, so in my opinion it's worth the attempt - even if the odds are a thousand to one.
-
While there are no doubt more objections to the concept itself, I’ll proceed in the hope that I have addressed the major ones.
Let’s start by defining what a theory of ethics is, and then define the additional requirements a universal theory of ethics must meet.
In essence, any theory of ethics must: Provide a methodology to evaluate behavior with; use the methodology to differentiate between good and bad actions; and claim that the good actions are preferable to the bad actions. This is very general. Any attempt at crafting a UToE though, will need to deliver on far more.
A UToE needs to be able to:
Evaluate any form of Life
Most ethical theories seem to limit themselves to human behavior, but a true UToE needs to be more comprehensive than this. We already seem to value Life other than only that of humans; pets being prime examples of this.
In the future, we will also have to contend with digital consciousnesses and AI, perhaps even alien lifeforms, so for it to truly be universal the theory should be able to be applied to all forms of Life; current and future ones; natural and artificial; from single-celled organisms to superintelligent AI.
Evaluate any scale or organization of Life
A truly universal theory needs to be able to be applied and evaluate not only single lives in isolation, but also collection of individuals and whole societies. This should include hives, bacterial colonies, and whole ecosystems - the complexity of which may not be readily apparent.
Evaluate in a comprehensive way
As far as is practical, it should take all future consequences in mind when evaluating.
Evaluate regardless of complexity
Although impossible, given perfect knowledge the theory it should be able to give a single answer regardless of the complexity of the situation evaluated.
Rational
It should be constructed using our current best understanding of Rationality, and be modified and improved as our understanding of Rationality improves.
Empirically true
It should match up with and correspond to empirical observation of Life as it exists now, and thus also have predictive power.
Useful
Additionally, for the UToE to be useful, it needs to fulfill several additional requirements, like being easy to understand, efficient in application, translatable to many different fields and contexts, and so on. I will skip describing these in detail as they are pretty self-explanatory.
-
While the things I’ve listed above are not comprehensive, I think it gives a good impression of the scale of what needs to be in place.
A distinct requirement, setting it apart from most ethical theories, is that it needs to be empirically true. This indicates that a universal theory should not be something that is formulated from nothing, but instead should already be in use by all of Life - a discovered truth, much like Evolution was.
Of course, much like evolution, there is no guarantee that this truth is ethically worth pursuing - we rightly do not accept “survival of the fittest” as our primary ethical principle - but it might be, and I will argue that it is.
First though, we need a guiding principle. A question all of Life would answer the same way, and I have a proposal.
After thinking about it for a long while, I think I have found the single thing all Life can agree on:
We all want to live.
If not individually, then at least those we care about. If not our specific genetic line, then at least our species. If not Life this planet, then least Life somewhere in our universe. Surely, a UToE that does not at minimum seek to ensure the survival of its users, is both self-defeating and self-invalidating. It would simply die out along with its users.
And so, the foundational presupposition for this theory is:
Life is preferable to Death.
As a prescriptive statement, it becomes:
The Meaning of Life is to Live
However, just stating a preference does not suffice for a robust universal theory of ethics; we need one or more value-functions to maximize, so let’s restate it as such:
The Meaning of Life is to Maximize Life.
I think it’s important to immediately point out that Maximize Life does not mean any simplistic instruction like "just have as many kids as possible”. No, it needs a broader, more profound definition than only the amount of Life.
A proper UToE should include its requirements, what it produces, what it does, and what are its qualities - all of it. After all, we are attempting to create a theory capable of ethically evaluating any situation and all the complexities that come with it, which needs to be reflected in our definition.
It needs to be said; all the choices of words and their definitions going forward are individually somewhat arbitrary. By that, I mean that the exact choice of words to use - the number of them, how they distribute qualities, and how they are defined - are all necessarily limited by our vocabulary and how we currently conceptualize things.
However, this doesn’t invalidate the choices made, nor the entire endeavor, even if anyone saying many other combinations of values could be picked are technically correct.
What is truly important is that they are internally consistent, and when combined, complete. This is in addition to also meeting all the general requirements for a UToE.
Let’s start with the simplest possible approach.
What are the most basic traits we require something to demonstrate for us to call it Life? Looking at other fields for inspiration might give us what we need. Computer science uses input, processing, and output as the most basic way to describe a computing system, which seems as simple as it gets, so let’s convert these into terms applicable for Life.
Input seems best described as Awareness.
Without Awareness, and thus access to information about reality, there can be no evaluation nor output. If we define it as broadly as possible, it becomes: Awareness is the ability to perceive reality.
While it is easy to conceptualize it in a human context; seeing, hearing, smelling; it might be more difficult for many to consider how a single cell can be aware of anything. For example, we do know that bacteria can react differently when close to nutrition and toxins, so even if the behavior is purely instinctual and as basic as can be, it can still be described as a type of Awareness.
While this may be a more abstract interpretation, I think it's a valid one. If we do not, we risk not including systems like the greater collective awareness a colony of ants exhibit compared to any individual member - much like we are multi-celled organisms, yet function as we do.
Evaluation seems best described as Rationality.
It might be tempting to call it thinking, but we have to keep in mind that all traits we include in our definition of Life are also traits we seek to maximize, so Rationality seems like the better choice. It not only indicates a general capacity, but also a descriptor of quality. If we define it as broadly as possible, it becomes: Rationality is the ability to evaluate reality.
As with Awareness, Rationality is usually used with regards to human behavior, but using the more generalized definition above, we can also see how it might refer to the general efficiency of structure and behavior. With it, we can describe even the most simple forms of Life as more or less rational.
To give an example: We can say that a more successful strain of bacteria is more Rational, if it makes better use of and is more successful within the same environment compared to others.
Output seems best described as Agency.
Agency is the capacity to perform actions, which necessarily includes having ways to do so. This can be anything from a body to a bacterial flagellum, but can also be highly limited, like only text output. Defined broadly: Agency is the ability to change reality.
A critical component to Agency is the ability for Life to change not only its surroundings, but also itself. Without this ability, evolution would not be able to function. While in simple life forms like bacteria this is done by simple trial and error, more complex selection processes like sexual selection are used by more complex life, further increasing efficiency. In the future, genetic engineering and programming will likely be able to speed this up dramatically.
-
All three traits seem essential, as defining Life without any of them doesn’t make sense. Life without Awareness is oblivion, Life without Rationality is madness, and Life without Agency is captivity.
Who can say they would want to (or could) live without any sensory organs, without any grasp on reality, or without any way of a influencing the world around them? We intuitively know how devastating the removal of these traits are, after all, we broadly consider them methods of torture: Sensory deprivation, brainwashing, and solitary confinement.
While they are all interconnected and interdependent, they also clearly demonstrate a sequential relationship of importance: Awareness is foundational to Rationality, and Rationality is foundational to Agency.
Together, the three traits are the most basic requirements and the fundamental building blocks of Life, and so it seems fitting to call them the Foundations of Life:
Awareness - Rationality - Agency
While I think it is possible to boil down all of Life to these three core traits and to describe the purpose of Life as their maximization, let’s see what we can build upon them.
If we consider The Foundations of Life as primary colors, and combine pairs of them to create secondary colors, what do we get?
The combination of Awareness and Rationality seems like a good description of Knowledge.
Evolution gave rise to RNA and DNA as means to store and pass along knowledge embedded in the structure of the organisms themselves, enabling its continued improvement and increase in complexity.
With more complex brains and the eventual ability to externalize knowledge, humankind was able to further speed up the process. The combination of Awareness and Rationality seems like a good description of our best methodology for creating knowledge: Science. When done correctly, science is a seamless dance of practicing Rationality while seeking empirical evidence.
Combining Awareness and Agency seems to be a good description of Opportunity.
We only call something an Opportunity if we both observe its existence with Awareness and consider it possible to seize with Agency.
Opportunities might exist without needing our direct hand, but we also actively create Opportunity. As Life adapts to its environment, it also shapes the environment, effectively creating opportunities for itself to thrive.
Humankind does this directly by organizing its surroundings, like buying groceries to put in our fridge, ensuring we are able to satisfy our nutritional needs with fresh food whenever we want. In a way, you could say that doing so is making our surroundings more Rational.
The combination of Rationality and Agency seems like a good description of Ability.
Evolution developed physical bodies to increase Ability, like large and powerful frames; built-in tools like beaks, teeth and claws; and eventually powerful brains capable of consciously enhancing its own capabilities.
In human affairs, the most obvious example of Ability are our tools: At first only pointed sticks and clay pots, but soon enough bombs and battery-powered drills.
-
The Products of Life differentiate themselves from the Foundations by the fact that they can be created and exist separately as latent capacity. Life is still ultimately essential for them to make sense, validating their inclusion in our definition of Life.
Similarly to the Foundations, they also demonstrate a sequential relationship of importance: Knowledge is foundational to Opportunity, and Opportunity is foundational to Ability.
Without sufficient knowledge, we cannot recognize Opportunities. Without Opportunity, we cannot enhance our Ability.
As they are produced and used by Life, it seems only fitting that we call them the Products of Life:
Knowledge - Opportunity - Ability
If we reintroduce the the third missing Foundation to each of the Products, we produce the next set of three.
Knowledge (product), when applied through Agency (foundation) to change and adapt, seems like a good description of Learning.
Evolution primarily learns through trial and error and various selection processes, which is inherently slow, so complex neural networks with plasticity developed.
At first, this simply allowed Learning to happen within each individual, but over time intergenerational transfer of knowledge was made possible by humans through the use of language, followed by pictograms, writing, multimedia and software.
Opportunity (product), when acted upon using Rationality (foundation), seems aptly described as Actualization.
Seizing opportunities as they present themselves are crucial to survival. Evolution has provided us with strong instincts and incentives to capitalize on opportunities, such as eating food when we find it and securing mates.
Ability (product) wielded with Awareness (foundation) seems best be described as Assertion.
Sometimes, mere Actualization is not enough. When different societies, species, and individuals clash, being able to assert oneself and avoid becoming a victim becomes necessary. Although it can be seen as a form of external Actualization, its distinct role in conflict scenarios and how it seems connected to the feeling of fear, justifies its categorization as a separate activity.
-
Again, they seem interconnected and interdependent, and demonstrate a sequential relationship of importance: Learning is needed for Actualization, and Actualization is needed for Assertion.
We can see this progression all around us in the animal kingdom, which includes our own behavior: We first play, socialize, and go to school to learn what we can, we then use the gathered knowledge to find a job and a suitable mate, and finally we assert ourselves by securing both and demanding to be respected as an individual by society at large.
As only Life is capable of performing these three activities, it seems fitting to call them the Activities of Life:
Learning - Actualization - Assertion
The last set of three is made from observing the outcome of the three Acts of Life when performed over time, and deducing what seems to be their likely ultimate goals.
Learning over time produces Understanding.
Effectively, Understanding is maximized Awareness (foundation) - not only able to interpret raw sensory information, but also deeply seeing the causal chains, patterns and interconnectedness of events.
Awareness, Knowledge, and Learning are all pursued because of how they enhance our Understanding.
Actualization over time produces Flourishing.
Effectively, Flourishing is maximized Rationality (foundation) - not only able to think logically, but also to synthesize effective solutions and implementing them to maximize desired outcomes.
Rationality, Opportunity, and Actualization are all pursued because of how they enhance our Flourishing.
Assertion over time produces Control.
Effectively, Control is maximized Agency (foundation) - having the freedom and ability to take action when needed, being capable of asserting oneself when threatened.
Agency, Ability, and Assertion are all pursued because of how they enhance our Control.
-
Again, while they seem interconnected and interdependent, they also demonstrate a sequential relationship of importance: Understanding is needed for Flourishing, and Flourishing is needed for Control.
Together, they seem to do a good job describing the qualities of a good Life I think most can agree with: Living a rich life fully engaged with reality, able to improve and reliably fulfill our needs and desires, in control of our own destinies, and so, I think it is justified to call them the Qualities of Life:
Understanding - Flourishing - Control
Combining the Foundations, Products, Activities, and Qualities of Life, we get the Core Values of Life:
Qualitites
Understanding - Flourishing - Control
Activities
Learning - Actualization - Assertion
Products
Knowledge - Opportunity - Ability
Foundations
Awareness - Rationality - Agency
They are what we all should agree to maximize, individually and collectively, now and for all future time. A new definition of Good.
Of course, there is still a mountain of work yet to be done in order to prove their validity and usefulness. I have at the end of this post included a list of potential future chapters to do what I can to this end.
I’ve decided to include one more chapter in this post though, because I think it fits best here and complements the Core Values of Life: I want to redefine Evil.
As with the act of proposing a UToE, redefining Evil - a concept with almost as many interpretations as there are people - will cause many to object. However, I argue it is both necessary and ideal to do so.
If we look for the red line cutting through all the interpretations of Evil, it would be: Our ultimate universal enemy. It is what we all think everyone should agree to identify and destroy. From my own observations, people are heavily motivated to do exactly this, as it has an ability to mobilize our anger and hatred like nothing else.
I can think of few things capable of doing more good than to redirect this intent, from arbitrarily chosen targets based on vague and often superstitious definitions, to its true nature and origin. I think it's worth the attempt.
So, what is Evil?
Lucky for us, by already having defined Life, we have what we need.
Using the Core Values of Life, we can define Evil as their opposite.
Awareness becomes Ignorance, the lack of ability to perceive reality.
It is easy to see how someone can act badly if they either do not know they are doing so or mistakenly believes they are doing good. When given this description, most do not assign blame to the individual, even as we recognize the harm that has been done, because we recognize the lack of any real Agency in making the correct choice. As such, the only correct thing to blame is Ignorance itself, and everything that contributed to it.
Rationality becomes Irrationality, the lack of alignment with reality.
Without being able to reason, essentially lacking needed cognitive structures to make the right choices, we cannot justly blame someone for their erroneous actions. We intuitively know this with regards to children, but consistently fail to extend the same understanding to adults - an essentially arbitrary categorization.
While we should acknowledge the pragmatic appeal of assigning personal responsibility to people’s actions in a world full of non-essential and non-comittal interactions, it is still a dangerous illusion to cling on to when guiding our general reasoning and approach to building a better world.
Agency becomes Inagency, the lack of ability to change reality.
On the circumstantial side, we can make a quite simple argument: It is easy to be a good person if you are never given the opportunity to do bad. It is also easy to be a bad person if you never get the opportunity to do good. Of course, we don’t operate as light switches, all good and all bad, and the same is the case with opportunities given and experiences forced on us. Without the ability to change our circumstances, we are victims to it.
Again, like with each set of the Core Values of Life, while they seem interconnected and interdependent, they also demonstrate a sequential relationship of importance: Irrationality is built upon Ignorance, and Inagency is built upon Irrationality.
Together, they seem to be exactly what we should expect, matching with our current best scientific understanding. As they are the opposite of the Foundations of Life, it only seems fitting to describe them as the Foundations of Evil:
Ignorance - Irrationality - Inagency
As we did with the Products of Life, we combine the Foundations of Evil in pairs like primary colors to produce another set of three.
Combining Ignorance and Irrationality, we get Misinformation - the opposite of Knowledge.
Combining Ignorance and Inagency, we get Threats - the opposite of Opportunity.
Combining Irrationality and Inagency, we get Restrictions - the opposite of Ability.
While interconnected and interdependent, they also demonstrate a sequential relationship of importance: Misinformation create Threats, and Threats create Restrictions.
As with the Products of Life, all three can exist separately from Life as latent capacities, and so it seems appropriate to call them the Products of Evil:
Misinformation - Threats - Restrictions
If applied deliberately, we can see what each Product creates.
Misinformation create Deception - the opposite of Learning.
Threat create Compulsion - the opposite of Actualization.
Restriction create Oppression - the opposite of Assertion.
While interconnected and interdependent, they also demonstrate a sequential relationship of importance: Deception create Compulsion, and Compulsion create Oppression.
Although these activities can arise from unintentional material conditions and organizational structures, they are often the result of intentional harmful actions, inflicted both interpersonally and upon oneself, and so it makes sense to call them the Activities of Evil:
Deception - Compulsion - Oppression
The end result of each of the Activities stand as opposites to the Qualities of Life:
Deception create Delusion - the opposite of Understanding.
Compulsion create Suppression - the opposite of Flourishing.
Oppression create Subjugation - the opposite of Control.
While interconnected and interdependent, they also demonstrate a sequential relationship of importance: Delusion produce Suppression, and Suppression produce Subjugation.
Together, they represent the end states achieved if acted upon by Evil, and can justly be called the Qualities of Evil:
Delusion - Suppression - Subjugation
Combining the Foundations, Products, Activities, and Qualities of Evil, we get the True Nature of Evil:
Qualities
Delusion - Suppression - Subjugation
Activities
Deception - Compulsion - Oppression
Products
Misinformation - Threats - Restrictions
Foundations
Ignorance - Irrationality - Inagency
Taken together, we see that Evil is simply different manifestations of Ignorance, Irrationality and Inagency, and that they are what we fundementally should direct our animosity against and attempt to correct.
Of course, it is important to state that Evil is sometimes a necessity in order to maximize Life. However, the more we improve the Core Values of Life for all, especially the Foundations of Life given their importance, the less this compromize should be needed.
Thank you to anyone having read this far - I’m honored.
Although this is very much a work of passion, everything I do is ultimately geared towards contributing to and improving the world in some fashion, so I hope you found it useful.
Below is a list of the things I’m working on to substantiate and expand on this theory. Some I’ve already written, some barely started, and everything in between.
I’m by no means ego-bound to this theory. Feel free to improve upon or correct any mistakes it might have. Contributions and collaborations are also very welcome.
Whether I post more though, is dependent upon the reception of this one. If I discover I'm demonstrating high levels of DK, then at least I can be appropriately humbled and move on to more productive endeavors.
-
List of future topics I’m working on to expand this theory:
Of course, all these are completely unsubstantiated claims as they stand, only intended to give a preview of what I hope to produce.
Goodness
Define Good and Bad actions by their total contribution or reduction of the Core Values of Life, across all current and future Agents, for all future time.
Rationality
For completeness, include brief descriptions and justifications, of different aspects and tools of Rationality while stressing their importance. These include:
Emotions
How emotions ARE how the Core Values of Life are implemented and experienced by Life.
Institutions
How we have established societal functions and institutions to provide, safeguard, and regulate the Core Values of Life.
Ethical Values
How commonly held ethical values correspond to the Core Values of Life.
Equality
How equality maximizes the Core Values of Life through the mitigation of existential risk and maximization of future potential.
Behavior
Using the Core Values of Life, we can rationally explain preferences and behaviors, including:
Framework
How the Core Values of Life can serve as a framework with which other ethical theories can be understood and contextualized.
Implementation
Explore how the Core Values of Life can be practically applied.