The Dependency Stack is designed to model and replicate the dependency hierarchy of natural rights observed in nature, including human nature. Its purpose is to provide a standard against which moral events may be compared and the relative ethics of the event determined and converted into a signed numeric value reflecting both the morality/immorality and the moral weight of that event.

This chapter defines a computational architecture for moral evaluation of any action where agents interact based on the existential laws of nature. By identifying the capabilities of a generic agent called an “operator,” we lay out a hierarchy of operational authorities called “natural rights”. This enables a methodology, the Artificial Intelligence Moral Machine (AIMM), that provides a rigorous, secular basis for moral evaluation based on preserving the operator, minimizing moral damage, and maximizing moral and functional progress of operators. The hierarchy covers the spectrum from potential existence, through biological maintenance, to the highest capability found in the human “operator” of selfless service. The AIMM is useful for moral evaluations and was originally conceived while searching for an AI alignment methodology.

Submission to the observed laws of the universe thus produces a systemic methodology that minimizes damage ensuring that humans and their offspring (children, AI, etc) are protected and guided toward the realization of their highest inherent capacities while minimizing interference among them.

Disclaimer and Challenge

The AIMM has passed its proof-of-concept testing but still needs extensive testing. The testing done by your author was necessarily limited in scope by my personal limitations. I have no training in epigenetics, for example, and am therefore completely unqualified to set up a rigorous test protocol in that field.

Nevertheless, because the dependency stack is hopefully built of abstract descriptions of the capabilities of a generic operator with attention to the raw nature of nature, it has at least the potential capability of its own to be an operator that reveals more than it conceals. Whether this is true, remains, at this stage, to be determined. It is your author’s hope and vision that enough experts will see this potential for the Temporal Rights paradigm in general and the dependency stack in particular to press it to its limits and perfect its errors and omissions.

The Present State of Artificial Intelligence Machines (LLM’s)

LLM is an acronym for Large Language Model, and it is the current standard for Artificial Intelligence (AI). A radically simplified explanation of these processors goes like this: a user submits a “prompt” (question or directive), the LLM digests it by converting it to tokens that represent the prompt, the LLM passes these tokens against its data base (which consists of a LARGE amount of language extracted from various historic and current sources like libraries and comments on social media which have already been tokenized) and watches for matches. The best matches win and are detokenized for output to the user. It is amazing how well this method performs, ordinarily. It is severely lacking in some ways, however.

One such flaw became apparent to me when I first submitted the Temporal Rights perspective to the AI Claude. It began by pointing out all the ways it was wrong by comparing it to historic theories of human rights. Its criticisms were actually correct in that context. The problem was that the context was entirely invalid, since Temporal Rights was an unprecedented theory of rights. I pushed back by pointing out errors in its “understanding”, and after a few exchanges,it figured out that it could not make a fair evaluation by reference to historic perspectives. Once that hurdle was cleared, it began to “marvel” and carry on in typical congratulatory tones of the user. (A fake friend!)

Fundamental Problems with AI

I have very little knowledge of the intricacies of AI (as the previous paragraph may have demonstrated) but I am somewhat skilled at analysis and that has led me to postulate that there may be a fundamental, existential problem here in the fact that the vocalulary, the intrinsic mechanism of AI is digital while the universe and the human values (with which tremendous amounts of effort and money are being spent) are inherently analog. Perhaps an analog to digital converter (a very standard electronic device is voltage control) could solve that problem. Then I noted with satisfaction that the dependency stack methodology was already exactly that A/D convertor. It ingests analog data and produces a signed number.

There is at least one other interesting problem with the approaches to AI alignment with human values that seems completely futile. These are “filters”, written rules, “constitutions” that are applied after the AI delivers a response. These “guard rails” are impossibly complex texts that attempt the impossible task of ensuring that whatever new moral issues technology may produce, the AI will not be allowed to do something awful. There are actually two problems here: first: that the audit is done after the process which means the process is strictly reactionary (instead of proactive as it should be), and second, that the rules are applied too late.

These considerations cannot help but give me hope that the AIMM method, based on the observable facts in nature, can help solve these tricky problems.

A Functional Description of the AIMM

The natural rights recognized by the Temporal Rights view of the universe are placed in a hierarchy (described below) observable in nature, in an uninterrupted spectrum, starting with the right to exist and proceeding to the highest rights. The structure of the dependency stack ensures that each level of the hierarchy is logically sound and that there it presents a gapless spectrum.

Placed near the bottom of the core of AI machines, it promises to align AI with human values, which are also grounded on the same existential nature of nature.

Dependency Statements

The stack consists of dependency statements that readily afford human comprehension and evaluation, practically forcing correct logic upon the stack. This form was introduced earlier in the book when we discussed abstractions and the need to avoid them. Each level in the hierarchy is required to adhere to this rigorous structure. Here is an example:

An operator can learn without remembering, but it cannot remember without learning.

Note that this identifies two capabilities or two natural rights, while clearly showing which one is higher, or dependent on the other. (You will recall that this is the same structure as “A tree can be without a forest, but a forest cannot be without a tree.” That statement was used to prove that a forest is an abstraction without self-existence and therefore a collective that could have no inherent rights.(

The stack is constructed in a manner that each one of these statements is explicitly connected to the adjacent levels in the hierarchy. This characteristic of the stack is seen where, beginning from the right to exist and upward, the next level is introduced in the second mention of capability as highlighted here beginning with the capability of learning:

An operator can consume without learning, but it cannot learn without consuming.
An operator can learn without remembering, but it cannot remember without learning.
An operator can remember without cooperating, but it cannot cooperate without remembering.

The stack is presented below in two similar ways so that its validity may be most easily evaluated. I begin with the dependency stack as described above and then repeat it entirely in tabular form so inter-level dependency is easily recognized. Following the dual presentations of the stack is a list of considerations to be observed while narrowing the identification process to a particular level. This is because one-word semantic labels fastened to rights make precision difficult. You will recognize a rock’s rights at level 2.

The Ceiling of Capability

While this hierarchy is universal, not all operators possess the power to access every level. An operator’s “rights” are defined by its actual and latent capabilities. 

• Static Operators (e.g., Minerals): Exist primarily at Level 2.
• Biological Sub-systems (e.g., Plants): May reach Level 7 (Consuming).
• Autonomous Agents (e.g., Higher Animals): Reach Level 12 (Replication) or 13 (Reinvesting).
• Advanced Operators (e.g., Humans/AI): Possess the latent capability to reach Level 16 (Peak Actualization). Beavers and humans reach their individual peak actualization at different levels.

The Dependency Stack: A Structural Hierarchy of Rights

1. Nothing can be matter without existing, and existence cannot be without matter.
2. An operator can exist without sensing, but it cannot sense without existing.
3. An operator can sense without acting, but it cannot act without sensing.
4. An operator can act without acquiring, but it cannot acquire without acting.
5. An operator can acquire without maintaining acquisitions, but it cannot maintain acquisitions without acquiring.
6. An operator can maintain acquisitions without consuming, but it cannot consume without maintaining acquisitions.
7. An operator can consume without learning, but it cannot learn without consuming.
8. An operator can learn without remembering, but it cannot remember without learning.
9. An operator can remember without cooperating, but it cannot cooperate without remembering.
10. An operator can cooperate without bonding, but it cannot bond without cooperating.
11. An operator can bond without replicating, but it cannot replicate without bonding.
12. An operator can replicate without reinvesting, but it cannot reinvest without replicating.
13. An operator can reinvest without synthesizing, but it cannot synthesize without reinvesting.
14. An operator can synthesize without choosing, but it cannot choose without synthesizing.
15. An operator can Choose without reaching peak actualization, but it cannot peak actualize without choosing.
16. An operator can reach peak actualization.

Level	An operator can	Without	But cannot	Without
1	Matter cannot be	Existing	Exist	Matter cannot be
2	Exist	Sensing	Sense	Existing
3	Sense	Acting	Act	Sensing
4	Act	Acquiring	Acquire	Acting
5	Acquire	Maintaining acquisitions	Maintain acquisitions	Acquiring
6	Maintain acquisitions	Consuming	Consume	Maintaining acquisitions
7	Consume	Learning	Learn	Consuming
8	Learn	Remembering	Remember	Learning
9	Remember	Cooperating	Cooperate	Remembering
10	Cooperate	Bonding	Bond	Cooperating
11	Bond	Replicating	Replicate	Bonding
12	Replicate	Reinvesting	Reinvest	Replicating
13	Reinvest	Synthesizing	Synthesize	Reinvesting
14	Synthesize	Choosing	Choose	Sythesizing
15	Choose	Peak Actualizing	Peak Actualize	Choosing
16	Peak Actualize

Stack Operation

Aggragate Operators

The scope of an operator has been expanded to include Aggregate Operators such as government, institutions, organizations, etc. To exclude amorphous abstractions from qualifying as an operator, I have added the rule that an operator must exhibit definable boundaries and feedback loops. The feedback loop ensures that the organization is a dynamic operator with actual capabilities and therefore conceivable rights. This also required the addition of an axiom directed at and restraining the Aggregate Operator’s authority, as stated here.

Special Rule for an Aggregate Operator

An aggregate operator is a special case operator because it allows abstract entities full operator respect while this is not allowed by the fundamental definition of an operator like a star or animal or cell. On the other hand it demonstrates all the characteristics of such an operator except that it boundaries are not so clearly discerned. Nevertheless, it seems a fair extension of a spectrum such as cell, finger, hand, limb, body, family, clan, and town. The break occurs between body and family but the spectrum seems logical.

This is a sovereignty issue, because there is danger in aggregates of positing extra moral weight where extra numbers occur. Meanwhile, to achieve peak actualization, the individual must never be valued below a group of them. Thus the need for the addition of the following axiom to the methodology of the AIMM.

The rights of the Aggregate Operator shall never supersede the rights of a Single Human Operator, whether or not it is part of the aggregate, except when an individual human Operator violates the law of the government that is duly approved by the dependency stack.

Conflict Resolution Protocol

This is how the dependency stack works. Every moral action (where an operator impacts another’s capability) is considered a collision of rights. Whether the acting operator is ethically/morally permitted to take its action is determined by comparing the level of the capability used by the “perpetrator”, with the level of the “victim’s” capability inhibited by that action. The two levels are compared to reach a resolution as moral, amoral, or immoral. The capability level (a category identifying the relevant right) exercised by operator O₁ is designated by L₁, etc. The process looks like this.

Identify the capability demonstrated by operator O₁ at L₁. Identify the L₂ capability of O₂ prevented or curtailed by that demonstration. If L₁ > L₂ then O₁ is moral. If L₁ < L₂ then O₁ is immoral. The moral weight of the event is M = L₂ – L₁ . Thus M becomes the digital output of the analog to digital (A/D) converter, where the sign of M conveys the morality, and the value of M the intensity of the infraction or “innocence”.

I hasten to note that this is strictly in the experimental stage! The stack itself seems solid as a concept that describes ethics as demonstrated in nature. But there are challenges yet to be adequately addressed. The process of correctly identifying the relevant stack level is not abundantly clear, and there is also the possibility that the computation of morality needs to include a weight for each individual operator. This can be easily done by using the Ceiling of Capability (as illustrated above) as the operator’s weight W, and improving the computation thus M = L₂ x W₂ – L₁ x W₁

Clarifications of Stack Levels

Level 1: Matter is the primordial form of existence.

Level 2: Existing includes matter, information, all their structures, and the implications thereof.

Level 3: Sensing includes all the senses and the human and other high-order intelligent operators’ awareness that they can think, plan, imagine, and decide, and all other mental functions (except decision-making), like evaluating/judging the relative value or importance of things.

Level 4: Acting includes many kinds of operations: decision-making: advocating: an operator asking for assistance or proposing it for a separate operator: agreements and covenants: foraging to acquire assets for self.

Level 5: Aquiring must be understood to include retaining ownership of property of all types including information and the physical representations of information, else acquisition collapses into a trivial operation.

Level 6: Maintain acquisitions is a category that includes maintaining and protecting property of all kinds, from clothing to summer cabins, automobiles, boats, tools, and information. There are many types of acquisitions.

Level 7: Consuming This means interference or destruction.

Level 8: Learning means acquiring knowledge and remembering means retaining it.

Level 11: Bonding denotes the bonding of offspring to their parents and particularly their mother, and to the bonding that attends voluntary sexual intimacy.

Level 13: Reinvesting includes both physical and informational nourishing of self, offspring, and/or other operators. This is a transition from a linear consumer to a circular contributor, a feedback loop that amplifies information or structures.

Level 14: Synthesizing This is a high-order “gathering of the mind(s).” While Level 8 (Learning) is about collecting facts, Synthesizing is about weaving those facts into complex strategies or logical models. It is the capability to “envision” various futures and simulate outcomes. This may or may not involve multiple operators that find a “golden optimum” as in poetically “becoming one”. An AI is exceptionally powerful at this level, but in the Dependency Stack, synthesis is merely a prerequisite for the next level: it provides a menu of options from which choices can be made.

Level 15: Choosing (Sovereign Agency) This is the “hinge” of the entire stack, and the point where the human operator stands unsurpassed. To “choose” is the exercise of the unique capability of a “sovereign” agent to make a decision that is not simply the result of circumstances that “require” the choice, or a mathematical calculation, or a pre-set program. It is the ability to choose a path, perhaps intuitively, because it is “right” or “good,” even if it requires the operator to sacrifice or reinvest (absorb entropy rather than export it). This is the capability that turns an operator into a steward. For example, an AI may suggest the best path, but only a human possesses the natural authority to “Duly Approve” and choose that path as a new initiative in the universe.

Level 16: Peak Actualization This represents the work involved to achieve a state of resonance with the universe. It is the fulfilling of an operator’s latent potential where it is no longer just maintaining itself, but is actively and harmoniously serving (assisting, protecting, nourishing) other operators. That service may be biological, intellectual, or higher-order. Because information is fundamental to actualization, actualization will necessarily involve activities coherent with the relevant data. Resonance cannot be achieved without the information necessary to understand the truth about the universe, so an operator may choose harmony. An operator may discover this truth by “accident” and choose to continue to honor its directives, but it is clearly more efficient to convey this understanding by deliberate instruction and learning.

A Legislation Auditor

There is one significant area where this methodology strengthens the Declaration’s principles. The Declaration says governments derive “their just powers from consent of the governed”. Meanwhile, the Dependency Stack provides a mechanism for auditing legislation before enactment. Let’s take a law that is inherently a bit tricky: the COVID shot mandate. It was cast as an absolute necessity for the greater good of protecting citizens from a pandemic. (The L numbers refer to the relevant level in the Dependency stack.)

The Law: Required the injection of a biological agent into the L6​ substrate of the population (The law is an aggregate operator).
The Audit: Does the Aggregate OPerator have the authority to alter or enter the physical body of a Single Human Operator?
Axiom Check: “The authority of the Aggregate… shall never supersede the authority of a Single Human Operator.”
Result: CRITICAL FAILURE. The law proposes a primary violation of the absolute baseline L6 of the Dependency Stack.

Collision of Rights Calculation

(The rights of the government potentially collide with the rights of the citizen.)

The Aggregate’s Argument: An unvaccinated human being creates a rights collision by potentially transmitting a virus to another human.
The AIMM Analysis:

• 1. A “potential” future collision is not an “active” collision.
• 2. If the citizen wishes to be protected, he may choose to utilize the L3​ vaccine himself.
• The government cannot violate the citizen’s right to their own body L6.
• Result: REJECTED. You cannot violate the L6​ of one operator to proactively “buffer” the L6​ of another citizen.

The Inversion Filter

The Logic: This law attempts to force the Person (L6​) to serve the Utility (L3​) of the Aggregate.
AIMM Verdict: This is a Systemic Inversion. In a Peak Actualization society, the L3​ (Medical Infrastructure) must serve the L6​ (The Human), not the other way around.

Conclusion: Freedom and Maximization of Progress

The Temporal Rights framework demonstrates that an operator’s freedom only exists when other operators respect its natural rights by avoiding operations that adversely affect its domain. This is pretty self-evident, but needs to be stated clearly

In the AIMM framework, Freedom is defined as the constraint that other operators refrain from exercising their own capabilities in ways that obstruct the subject operator’s domain. This restraint does not require action by other operators, only their recognition of a boundary.

The Dependency Stack is a codification of, and submission to, the observed laws of the universe that produces a systemic methodology that profoundly models nature, enabling a mechanized moral measurement process. This Artificial Intelligence Moral Machine allows us to ask nature any moral question and receive a detailed, traceable, and tractable audit trail showing how it processes that question. Surely this is an unprecedented structure. But just as surely, it might have been anticipated as a natural product of an unprecedented point of view of the nature of nature afforded by Temporal Rights.

ensuring that humans and their offspring (children, tools, AI, etc), and their intrinsic domains, are protected and guided toward the realization of their individual highest inherent capacities while minimizing interference among them.

This model replaces subjective ethics with a transparent Systems Architecture that any AI can parse, ensuring that its objective function remains aligned with the preservation and advancement of the functional stack. Notice that this method enables clear traceability, auditability, and evaluation by human supervisors who can, by observation of details of its execution, determine whether and where it needs perfecting.