Decentralism: A Theory of Informational Order

Introduction

In this article, we present the theoretical foundations of decentralization, exploring its philosophical bases and its relationship with the evolution of social, economic, and technological systems. It will be demonstrated how decentralization is a comprehensive philosophical principle, with high explanatory power and applicability in various areas of philosophy, being a framework that contains several other principles as its logical and natural unfoldings.

Decentralism

Decentralism asserts that all of reality, in its different realms, can be explained based on the principle of decentralization, resulting in a multiplicity of applications of this principle. From the perspective of decentralism, decentralization is the foundation for the basic structure of reality in all its domains: from metaphysics, forms and abstract objects, cosmology and physics, evolution and history, the formation of individuals, cultures, societies, technology, and even the evolution of capitalism. For this reason, we begin with the definition of “decentralization” used here, and then proceed to show its implications and applications.

Decentralization

The definition of “decentralization” that we use here is more comprehensive and abstract than the common definition. In the common sense, “decentralization” refers to the distribution of power, control, or authority among multiple parts of a system, rather than its concentration in a central controller. Applied to governments, it means the transfer of competencies from central power to local or autonomous entities; in technology, it refers to the distribution of processes and data among independent nodes; in economics, it implies the replacement of intermediaries with direct interactions between agents. Thus, decentralization is understood as the distribution of control.

In the philosophical definition of “decentralization” used here, the distribution of control is an essential aspect and a natural tendency of decentralization, but it is not the most basic definition of the word. In every decentralized system, there is distribution of control, but not all distributed control is decentralized, as we shall see later. We make this description more abstract by basing it not only on autonomous and distributed control, but on all system determination based on non-arbitrary (natural) rules. Thus, the principle of decentralization can also be applied to other realms, such as the epistemological, ontological, physical, evolutionary, and, in fact, to the structure of any type of system. In other words, decentralization, as understood here, is a basic principle of all reality. It is not merely a type of control in societies and machines, but it encompasses all of nature.

Control and Determination

Control, in general, can be understood as the way in which systems, organisms, and machines exert influence and determination over themselves, over their parts, and over the environment. When a system is centralized, this relationship of determination is asymmetrical, with a decision-making core concentrating the power to influence the other components through arbitrary and unilateral criteria. On the other hand, in decentralized systems, determination occurs in a distributed manner, with multiple parts exerting mutual influence and being regulated by the same criteria of positive and negative feedback. For example, in a neural network, whether biological or artificial, decentralization occurs because the nodes or neurons interact directly with each other, transmitting signals in a distributed manner, without relying on a single controlling center. Intelligence and learning emerge from this decentralized process of adaptation, where complex patterns are recognized and adjusted through feedback. In the case of humans, neural connections that produce certain outcomes, such as obtaining food and pleasure, are reinforced, while connections that produce outcomes such as pain, suffering, lack of resources, or social rejection tend to be inhibited. Instinctive adaptations, and the very criteria for reinforcement or inhibition of learning, in turn, are adjusted by natural selection. Adjustments that reinforce certain information are positive feedbacks, while those that inhibit are negative.

Control, in this way, is decentralized to the extent that its parts are all determined by the same criterion, but not just any kind of criterion. For example, in a statist system, where the state is a centralized controller, one might say that everyone is following the same positivist laws described by that government's penal code, yet the system is still centralized. What makes this system centralized is the arbitrariness and inconsistency of the rules, whereby each part of the system (in the case of society, the individuals) is controlled by a different criterion, with the state and the individuals who are part of it having a monopoly over certain rights, such as imposing rules by force and collecting money through taxation. For a system to be decentralized, its parts, or nodes, need to be controlled—or control themselves—all based on the same rules or criteria, in a consistent manner.

In the case of neural networks, this criterion is the types of feedback that reinforce or inhibit neural connections. The most fundamental feedback criterion in neural networks is the persistence and reproduction of those very neural networks: ultimately, natural selection, which regulates which feedbacks will have a reinforcing or inhibiting effect on the parameters. In the case of artificial neural networks, the most useful and profitable networks tend to have their code and architecture reproduced, while the less useful ones tend to be abandoned, which is also an example of both natural and artificial selection (both are not mutually exclusive). In the case of organisms competing under natural selection, the capacity to adapt for survival and reproduction, within the same laws of physics, is a criterion that all organisms follow and, in this way, natural selection is a natural criterion, and the process of evolution, in general, is decentralized.

Criterion of Conformity to Natural Laws

In this way, decentralization essentially consists of the determination of a system’s parts by the same consistent rule, whereas in centralization, one part is determined differently from the others. In a statist society, the state is a centralized controller, as it can use military weaponry and impose rules through the use of force, while the citizens cannot do the same.

For a rule to be consistent, it must be capable of determining all parts of the system under the same criteria, without exceptions or arbitrary hierarchies. For example, the laws of physics are consistent rules, as they determine the behavior of the entire space-time without any arbitrariness. Similarly, logic, mathematics, and natural selection are examples of consistent rules or sets of rules. These rules are also natural, as they emerge naturally—either as a necessary structure of reality, such as logic and mathematics, or as a natural tendency over time, like natural selection.

Natural rules are the consistent rules that are not determined arbitrarily, but that underpin or emerge in a natural, necessary, and universal manner within a given realm, such as the physical, biological, or social realms. Just as natural laws are consistent, rules that violate natural laws are inconsistent, for they are arbitrary. Consistent rules, therefore, are only those natural rules and laws, or those that are based on natural laws and do not violate them. This leads us to the following conclusion: only the rules that conform to natural laws are consistent. Therefore, conformity to natural laws is the essential criterion of decentralization.

Philosophical Definition of “Decentralization”

Therefore, this is the most abstract and consistent philosophical definition of “decentralization” used here: the conformity of systems to natural laws. It is important to note that this applies not only to any concrete system but also to systems that can be considered abstract, such as logic, mathematics, epistemology, metaphysics, and physics. These systems can be understood as decentralized, given that their structure is based on natural rules or laws.

Informational Order

The word “order” is often used to refer to two different things, namely low entropy and evolutionary order, making it important to disambiguate and clarify the meaning used here.

In physics, the degree of entropy in a system is a measure of “disorder” in the sense of energy dispersion, relative to which order would be the concentration of energy—for example, in the singularity of the Big Bang, where entropy is minimal. Entropy can also be interpreted in an informational sense, where the greater the entropy, the larger the amount of information required to describe the states within a system. In terms of entropy, order decreases from the past to the future, as energy disperses and becomes unavailable for work, and the amount of information needed to describe the system increases.

Evolutionary order, on the other hand, refers not only to the complexity of organisms but primarily to the efficiency of these systems in propagating their genetic or memetic information, whether analog or digital, through survival, reproduction, or communication. Based on the transmissibility of information, evolutionary order can also be called “informational order.” Unlike the order described in terms of entropy, evolutionary order increases from the past to the future and is accumulated through the process of natural selection. However, these two forms of “order” do not contradict each other, as they are, in fact, two different concepts that share the same name.

Natural selection is based on the propagation of information. The positive and negative feedbacks through which systems adapt are mechanisms of information transmission and processing, ranging from the basic mechanism of selecting physical and genetic information to more evolved and complex forms of adaptation and information processing. Success in surviving and reproducing is the success in transmitting one's own information to the future and to new descendants. Even the mere survival of a simple organism is a form of information transmission, in which information is passed from one moment in time to the next. This information is then processed, refined, and perfected by competition, by natural selection, among different capacities for transmitting various pieces of information. Therefore, every ordered system, in the evolutionary sense, is an information transmission system that is part of a broader system of information processing. For this reason, it is also correct to refer to evolutionary order as “informational order.”

Thus, every order related to the efficiency of systems in propagating their genetic, memetic, cultural, social, technological, digital information—or any other system derived from organisms and memes—originates from natural selection. Consequently, when we refer to all these types of order, we are referring to evolutionary order, and not to the thermodynamic order related to entropy.

These evolutionary processes that form order are derived from the basic laws of physics, which allow the continuity of information throughout space-time, and therefore every natural rule established in an ordered system over time depends on the natural laws of physics. Considering that these laws are reducible to mathematics and logic, and that mathematics itself can be reduced to a set of propositions that follow the laws of logic, we can conclude that logic is the most basic natural rule, with all other rules ultimately derived from logic. The entirety of physical reality can be understood as an informational network of logical and mathematical relationships. In informational cosmology, physics emerges from abstract rules.

Logic is not only the basic rule of the structure of propositions, but also of the facts denoted by them. If true propositions denote real facts, and true propositions must be logically consistent with each other, then real facts must also be logically consistent with each other, with reality conforming to the natural rule of logic.

Emergent Order

Emergent order is the organized structure that arises spontaneously from the decentralized interaction among the elements of a system, without the need for central control. This order, which emerges from the bottom up—from the interactions among the parts of the system to the whole—arises from the transmissibility of information. The most transmissible information (stable, resilient, and reproducible) reaches ever greater quantities, while the less transmissible becomes less prevalent or is extinguished. This is nothing more than the process of natural selection, which here is not understood as merely biological, but as a process of forming informational order that occurs over time, with physical, genetic, memetic, economic, social, cultural, and technological information. Every order that arises in a physically decentralized universe, based on natural laws, is emergent and derived from natural selection, which is itself based on decentralization.

Decentralized organization occurs without a central controller, and order emerges from the interactions among its individual components. This implies that no individual part of the system possesses total control over the entire system; instead, each component has its own sphere of influence, and the interaction of these individual spheres of influence is what composes the overall order. However, in an emergent order system, the criteria for generating, transmitting, selecting, and processing information are not random; there must be a criterion that all parts of this system follow so that their interaction can create and sustain this order. Natural selection, which is the most basic mechanism for all emergent order formation, is a process in which the organisms or memes subject to selection must share the same environment, governed by natural rules that apply equally to all, based on which selection is made. In the case of natural selection, it occurs based on the laws of physics, through which information can be propagated, transmitted, or reproduced. In the market, natural selection occurs based on the law of private property and voluntary exchanges. Similarly, natural selection—and therefore any formation of emergent order—must always occur based on conformity to natural rules. At the most abstract level, all natural rules depend on the natural rule of logic.

Since decentralization is conformity to natural rules, and since all formation of informational order is based on the natural selection of information based on natural rules, it follows that decentralization is the basis of all informational order.

Decentralization as Natural Order

Decentralization is not chaos or disorder, but rather the determination of systems based on natural laws. This conformity to natural laws, in turn, generates a tendency toward an ever-greater informational order, as it is the criterion that allows for the natural selection of information. Genetic, memetic, economic, and digital information become increasingly transmissible and processed by systems as they develop a more resilient decentralized order. This decentralized and resilient order is favored by natural selection for one simple reason: decentralized control—by allowing competition and avoiding informational blockages—is more efficient in enabling evolution by natural selection, not only of organisms but also of societies, cultures, economies, and technologies, all of which are based on competition under a natural rule, that is, decentralization. Decentralized systems are more efficient in their functioning, in the transmissibility and processing of information, and in the speed at which they evolve.

Therefore, there is an evolutionary tendency, through the natural selection of the most transmissible and resilient information, toward increasingly decentralized and resilient informational systems.

Decentralization as a Natural Tendency

Not only is the decentralization of systems the basis for the formation of informational order, but informational order itself tends toward the increase and strengthening of systems’ decentralization. Decentralization emerges as a tendency because decentralized systems offer structural advantages over centralized systems. For example, in a dictatorial society, the state interferes in the economy by imposing taxes and appropriating what is produced. This hinders the reinvestment of profit in further production and decreases the total production and wealth of society. More decentralized societies, therefore, produce more and foster greater growth in their means of production. Over time, natural selection leads to an increasing prevalence of economic freedom and a greater tendency toward collapse and stagnation in dictatorships and centralized economies.

Not only do decentralized systems grow more through feedback, but their resilience also increases through natural selection. For instance, more decentralized and productive societies can invest more in technological development to boost the productivity of their means of production, including their techniques and technologies. These technologies enable new forms of security: weaponry, walls, locks, long-distance communication, knowledge, critical thinking, freedom of expression, and even cryptography and digital security. Thus, decentralization, favored by natural selection, also becomes increasingly resilient over time. However, this process is gradual and coexists with centralization, which will continue to exploit new vulnerabilities at each evolutionary stage.

The centralized controllers themselves are also formed by emergent order. For example, a dictator is an intelligent being whose neural network is decentralized, with each neuron functioning autonomously, without a central neuron. In this way, all order is fundamentally originated in an emergent and decentralized manner, and it tends toward ever greater decentralization.

Decentralization as Informational Structure

Just as natural selection is an informational process—applying to genetic, memetic, social, technological, and digital information—decentralization is also an informational process, and the two are interdependent. Natural selection is accelerated by decentralization, and decentralization becomes more resilient to central controllers as systems undergo natural selection, becoming more efficient in transmitting their information and preventing it from being blocked. The transmission of information includes survival (transmitting one’s own information to the future), reproduction (transmitting one’s own information to descendants), and action and behavior (using resources for specific purposes). Productive efficiency is also informational, as the creation of economic value enables feedback within systems: it supports survival, reproduction, and expands the reach of a system’s behavior—be it organisms, machines, individuals, or algorithms. Systems that are effective at increasing their available resources are also more efficient at propagating information.

Social and Legal Decentralization

In the social and legal realm, the natural rule is private property, as it is the only law capable of avoiding all conflicts, provided it is always followed. The deduction of this law was detailed in the article “Fundamentos da Ética Libertária.” The (free) market is the social and economic application of decentralization, where competition and the absence of regulations and taxes—characteristics inherent to a decentralized social system—make it more efficient than its centralized counterpart, which is based on a state monopoly.

Decentralized and Distributed Control

Since determination by natural rules is the criterion of decentralization, it follows that in some systems control can be distributed yet still remain centralized. For example, in societies controlled by multiple militias that compete among themselves, control may appear to be distributed among groups of bandits, but it is actually a form of centralized control, because the natural social rule—private property—is systematically violated, potentially more frequently and intensely than in large territories where a single state prevails. Therefore, not every distribution of control is decentralized. However, respect for the natural law of private property increases each owner’s control over their property and over themselves, and in this way, all decentralization implies the distribution of control.

Is All Reality Decentralized?

This concept of decentralization may lead to the following question: if all systems that are completely determined by natural laws are decentralized, then wouldn’t all systems be decentralized, since all of them follow the laws of logic, mathematics, and physics? In the physical realm, where all events follow the same basic laws, yes, all physical systems are decentralized; however, in the social realm, for example, individuals and societies follow multiple different and arbitrarily imposed rules and laws. Therefore, in the social, economic, and other realms of equivalent or greater complexity, systems can be more centralized or more decentralized.

Even in the physical realm, suppose a digital simulation of a universe were created in which a central controller could, exclusively, violate the physical laws of that universe. It is, therefore, even questionable whether the laws of physics are necessarily decentralized for all universes.

Autonomy

In a decentralized system or network, the parts, or nodes, determine their own behavior based on natural feedbacks, independently of a central authority, and are, in that sense, autonomous. In contrast, in a centralized system, nodes have their behavior determined by other nodes or groups of nodes. Autonomy, in this way, is decentralized control. Feedbacks can be considered natural when they are not controlled by specific entities, organisms, or machines, but are transmitted through decentralized interactions with the environment or other parts of the system. Systems can have different degrees of centralization or decentralization, depending on how autonomous their parts or nodes are. Decentralization is a spectrum or a gradual process, in which different systems or organizations have varying levels of decentralization in their structures and control processes. It is not a binary condition, but rather a continuum that ranges from completely centralized to completely decentralized, with many intermediate stages. Moreover, these degrees apply separately to multiple realms, such as the physical, economic, ethical, computational, etc.

Centralization

A system is centralized to the extent that it does not conform to natural laws—for instance, when organisms or machines control parts of the system in a way that violates some natural rule. For example, if the natural rule of the market is private property, and a monopoly imposed by force breaks that rule, then that monopoly is a centralizing system in the market and society. Every system whose order is generated and maintained emergently—and through the interaction among autonomous behaviors—has decentralization as the foundation and origin of its order. In other words, order emerges based on the competition of forms within some common natural rule, such as logic or the laws of physics. However, this fundamental-level decentralization does not necessarily apply to all realms of the system, and a system based on decentralization may contain partial centralization. For example, a social system can be decentralized at the level where all its parts follow the natural laws of physics, competing through natural selection, but be centralized in the social sphere, where the state controls individuals by violating natural social rules.

Directed Order and Hierarchy

It is important to note that although all emergent order is decentralized, not all directed order is centralized. Centralization is not the same as direction and hierarchy. For example, a company operating in the free market may be managed by a director; however, if that director is—or was chosen by—the owner of the company, then that direction and hierarchy is decentralized, as it conforms to the natural rule of the social system, based on private property, and allows free competition with other companies, owners, and directors. In contrast, a state that establishes a state-owned enterprise as a monopolist in a sector is a centralized hierarchy. Similarly, a society in which small militias fight among themselves and control small territories is also socially centralized, because even though there is emergent order from the competition among the militias, that competition systematically violates the natural rule of private property, even if it does not adhere to a unified hierarchy.

Informational Decentralization is Accelerated

Just as in evolution in general, there is a tendency for the decentralization process to accelerate, since the more informational efficiency is accumulated through natural selection, the more efficient systems become at increasing their own informational efficiency. In this sense, evolution itself evolves, reaching new paradigms ever more quickly.

Relevant theses on the evolution of evolution itself were explored by Daniel Dennett, Ray Kurzweil, and Nick Land from different perspectives. Dennett (2017) describes how evolutionary systems, by developing increasingly sophisticated layers of adaptation, become capable of accelerating their own evolutionary processes. Kurzweil (2005) formulated the Law of Accelerating Returns, according to which each new evolutionary advance, including technology, reduces the time required for the next innovation, making progress exponential. Land (1992), in turn, characterized technocapital as an intelligent informational network that expands and reconfigures autonomously, driven by economic and computational feedback circuits that intensify its own processing and adaptation capacity. In this sense, decentralization—as a trend in technological evolution—finds its extension and continuity in technocapital, which dissolves institutional and political barriers as it accelerates, transforming into an irreversible decentralizing force. Technocapital embodies the process of natural selection in a concrete and intelligent system.

Hayek (1945) described the market as a distributed system for information processing, in which prices act as signals that condense and transmit dispersed knowledge about supply and demand. Mises (1920) argued that the price system, adjusted through voluntary exchanges among owners, is necessary for economic calculation, which is legally based on private property. In a free market environment, each economic agent, by reacting to variations in supply, demand, and prices, contributes to a continuous process of adjustment and efficient reallocation of resources without the need for centralized coordination. This mechanism, fundamental to economic calculation, ensures that relevant information flows in a decentralized manner, allowing the economy to evolve organically. Thus, free markets function as adaptive cybernetic systems, in which the decentralization of information and economic self-organization become catalysts for the evolutionary acceleration of the capitalist system itself, which processes ever more information and develops a form of intelligence in a decentralized network.

Technological Decentralization

As informational systems evolve and transmit ever more information, intelligence, language, and tool manipulation emerge as natural tendencies, leading to the development of technology. In human civilization, information technology is achieving increasingly greater degrees of decentralization, as evidenced by the internet, blockchains, and artificial intelligence. These systems are capable of transmitting and processing ever more information in a manner that is increasingly efficient and resilient to regulations, interferences, and censorship.

Technological Singularity

Technological singularity (Kurzweil, 2005) can be considered here as a point at which the time between two evolutionary paradigms becomes so short, due to acceleration, that it tends toward zero, or where evolution simply reaches its limit. Based on the thesis that decentralization is an accelerated evolutionary trend, it can be concluded that, provided there are no intrinsic limits to technological and evolutionary acceleration, technological singularity will be a moment of maximum decentralization and autonomy, in which informational order will tend toward complete independence from central controllers.