Entropy and the Idea of God(s): A Philosophical Approach to Religion as a Complex Adaptive System

Abstract

While a universal definition of religion eludes the field of religious studies, it certainty seems that people are becoming differently religious rather than a-religious, especially since the latter half of the twentieth century. To explain the enduring relevance of religion in human experience, this article expands on recent evolutionary and sociological research in the systems theory of religion and develops a philosophical approach to understanding religion as a complex adaptive system. Frameworks of meaning and beliefs communicated by religious systems emerge and adapt in relation to interpretive selection pressures communicated by individuals-in-community relative to entropy’s role in one’s contingent experience as a “teleodynamic self” in the arrow of time. Religious systems serve an entropy-reducing function in the minds of individuals, philosophically speaking, because their sign and symbol systems communicate an “anentropic” dimension to meaning that prevents uncertainty ad infinitum (e.g., maximum Shannon entropy) concerning matters of existential concern for phenomenological systems, i.e., persons. Religious systems will continue to evolve, and new religious movements will spontaneously emerge, as individuals find new ways to communicate their intuition of this anentropic dimension of meaning in relation to their experience of contingency in the arrow of time.

1. Introduction: Religion as a Complex Adaptive System

One of the problems in the field of religious studies is the lack of common agreement on the concept of “religion” or “religious”, supposedly the very phenomenon it studies. There are various reasons for this: that the etymological foundations of the concept emerged in European culture without analog in most languages; that many cultures do not distinguish between religion and cultural identity or praxis; that there was/is a Eurocentrism in the initial study of religion that defined the concept relative to its own religious history—to name a few examples. Moreover, functional theories, while helpful in providing cross-cultural insights, are often viewed with suspicion by religious practitioners who understand their beliefs to reflect what are to them aspects of reality irreducible to social–scientific explanations. Wouter J. Hannegraff notes that the specialization of the field of religious studies has lost touch with the original ambition to understand religion in a general sense, and that such an understanding is “essential for any adequate understanding of the history of society and human culture as such” (Hanegraaff 2020, pp. 72–73). What is needed, he says, is a new, minimal narrative (not the narrative) for formulating a theory of religion that takes seriously both functional and experiential domains of explanation that can serve as a baseline for accurately predicating “religion” to psycho-social phenomena observed in the past, present, and future. A synthesis of different scholarship in systems theory of religion (Luhmann 2013; Pace 2011; Purzycki and Sosis 2022; Coltea 2023) offers a framework for bridging—but perhaps not overcoming—the functional–substantive divide in explanation due to system theory’s methodological commitment to emergent complexity. Religion is irreducible to a single cause because it is emergent from the interpenetrating communication shared between the individual (psychic system) and her community (social system). The benefit of the theory is that it proceeds from the very system dynamics that we all observe and participate in every day, and it understands ontology (being; what is) to be operational and dynamic, rather than substantial and static. A systems theory of religion is congruent with David Chidester’s call for a material approach to religious studies that considers “embodiment and the senses, objects and their social lives, exchange and power relations, media and mediation, and all the forces and fluctuations in the production, circulation, and consumption of things” to be “the stuff of religion that demands the attention of the study of religion” (Chidester 2017, p. 74). This allows us to remain on the empirical side of theory construction regarding religion, and to adjust the theory according to new observations.

A systems theory of religion was pioneered by Niklas (Luhmann 2013, [1998]) near the end of his life (1927–1998) as an extension of his sociological theory that societies (social systems) result from the communication shared between individuals (psychic systems), rather than from actions conducted by individuals (see Arnold 2013; Beyers 2022, pp. 792–3; Coltea 2023, p. 7). Luhmann himself was influenced by the semiotics of Talcott Parsons (see Murphy 1982) and the systems theory of Ludwig von Bertalanffy, an early twentieth-century biologist who is credited with coining the term in General Systems Theory (Von Bertalanffy 1968). Luhmann drew from systems dynamics in biology and argued that social systems maintain their identities over time through the continued communication of meaning in a way similar to biological autopoiesis (Maturana 1975, 1980). Autopoiesis is the way biological systems, from cells to our bodies, take in energy from the environment, incorporate it into their system, and so “self-create” over time. The literal translation of autopoiesis is inaccurate, however, because the biological organism does not so much self-create but self-organize and transfer the energy outside its system into its own. Luhmann laid the foundation for understanding how the communication of meaning, discerned through the observational act of a psychic system distinguishing itself from its environment in communication, results in an analogous self-organizing of the social system such that it not only persists in time but feeds its sign–symbol system back into its constituent parts, the psychic system, in the form of meaning (see Coltea 2023, pp. 7–8; Reese-Schäfer 2013; Mavrofides et al. 2011). Enzo Pace further developed Luhmann’s communication-based approach, showing how systems of religious belief are “expert systems” relative to other subsystems of communication in society like law, politics, economics, etc., with a particular focus on the power dynamics involved in the emergence and evolution of religious systems (see Pace 2011, p. 227; Pace 2016).

Benjamin Grant Purzycki and Richard Sosis’ Religion Evolving: Cultural, Cognitive, and Ecological Dynamics (Purzycki and Sosis 2022) takes an evolutionary approach to the systems theory of religion, describing religion as a complex adaptive system (CAS) that “serves to support extensive human cooperation and coordination and social life as we know it” (2022, p. 139). Purzycki and Sosis’s evolutionary approach to the systems theory of religion complements a communication-based analysis, by providing an explanation for the way the communication between individuals establishes a religious system that contributes to the adaptive fitness of individuals through facilitating group cooperation. Their research provides an emergentist and adaptationist explanation of the role of religious systems, as an alternative to byproduct theories in the cognitive science of religion that consider phenomena associated with religion to be unnecessary extrapolations from cognitive modules that developed according to natural selection to help animals survive, such as a belief in supernatural agents being attributed to the module of hyperactive agency detection (2022, pp. 66–68, 126–28). Instead, Purzycki and Sosis advocate for going “beyond the mind” (2022, ch. 8) to examine the ecological influences that give rise to religious systems. Adaptationist approaches assume the fundamentals of the cognitive approach (Purzycki and Sosis, p. 129), that mental states are dependent on cognitive and neurological states, but go further than byproduct conclusions by “giving equal attention to the nature of the human mind, the natural and social environments, history, and all of the forces that exert influence upon them and their relationships…. In order to understand why humans are religious” (2022, p. 138). Religions are not static nouns, but dynamic verbs (processes) involving the individual and her community that constantly change over time, ever eluding our efforts to label and classify as various “-isms” (Purzycki and Sosis 2022, p. 159). Religious systems are not explainable by only looking at beliefs and their underlying cognitive influences, nor only by explaining the way religion acts at the level of society, but it is instead necessary to understand how cognition is employed in a social environment to recognize how religious systems evolve as complex adaptations that help us navigate the world.

Ritual has traditionally served as a fundamental criterion for predicating “religion” to social phenomena. Purzycki and Sosis discuss how ritual is a pre-human evolutionary adaptation that promoted group cooperation in the face of environmental pressures over time, and “these traits served to maximize the potential resource base for early populations, thereby increasing individual fitness” (2022, p. 139). Robert Bellah locates the foundations for religious ritual in mammalian play dynamics, where the conceptual boundaries are drawn that make play possible (2011, pp. 91–97, 111–12). If those boundaries are crossed by one animal biting too hard, for example, the play comes to an end and may not occur between the animals in the future. Bellah posits that the conceptual boundary-setting required for play may serve as a socio-cognitive foundation for the emergence of ritual, symbolic thinking, and mythmaking in human evolution. Ritual is therefore the foundational “building block” of a religious system’s complex adaptivity, and Purzycki and Sosis identify seven other “recurring core features” that contribute to the adaptive fitness of religious systems: taboo, authority, myth, sacred, supernatural agent, moral obligation, and meaning (2022, pp. 142–9). Other features of religion such as “music, spirit possession, afterlife beliefs, prophecy”, etc., “are better understood as secondary forms of one of the essential building blocks” (2022, pp. 142–3). Eliminating any one of these building blocks risks the collapse or transformation of the religious system:

Remove one of the building blocks, and the adaptive functionality of the religious system will be compromised. A religious system that lacks organization (authority), fails to impart significance (meaning), forsakes ceremonial activity (ritual), maintains no limits on activity (taboo), is unable to sanctify anything (sacred), offers no beings capable of transcending the natural world (supernatural agents), does not establish social commitments (moral obligation), or provides no narrative that can link all of these elements together into an explanatory framework (myth) will ultimately falter or transform into some other kind of social institution. Notably, religious systems are impressively resistant to the elimination of one of their core building blocks…

(Purzycki and Sosis 2022, pp. 146–7)

These eight building blocks subsist in a network of complexity, interpenetrating one another, and in so doing function to produce the religious system which contributes to cooperation, survival, and the reproduction of the individual via the group—i.e., evolutionary fitness. It is in this sense that they describe a religious system as an “extended phenotype”, an extension of human embodiment that is selected for, or against, relative to the natural and social environment (Purzycki and Sosis 2022, p. 152). Purzycki and Sosis recognize that the question, “why these eight?” can only be answered scientifically through “a phylogenetic account of the emergence of the religious system”, but admits that we have not sufficiently developed an understanding of religion as a complex adaptive system to conduct such a study (2022, p. 146).

Language serves as the “glue” holding the building blocks together, facilitating their function over time to produce adaptive changes that contribute to the maintenance of the religious system (Purzycki and Sosis 2022, p. 148). Energy enters the religious system through social information encoded in ritual behavior as emergent from interactions between various elements in the religious system (Purzycki and Sosis 2022, p. 149). Rituals tend to occur spontaneously when a community is under threat, and that “once such [supernatural] agents become linked to a ritual, desires to please or appease the agents can proximally motivate the ritual performance [of participants]” (Purzycki and Sosis 2022, p. 151). There are five effects produced by the interaction of the eight core building blocks on individuals within the religious system: physiological, emotional, cognitive, neurological, and technological (Purzycki and Sosis 2022, p. 151). Where the first four are internal to the psychic system, the fifth recognizes that “the religious system also produces a primary external effect, or what can be considered an extended phenotype… in the form of ritual objects” (Purzycki and Sosis 152). It is this fifth effect at the level of the system, the extended phenotype, that constrains the interpretation of meaning by individuals by communicating “shared cognitive schemata, ethos, symbolic meanings, material culture, historical memories, and group identities” that “can powerfully shaper individual lives” and “produce societal order by creating structured and stable social worlds–often fantastically imaginative–that individuals inhabit and navigate” (Purzycki and Sosis 2022, p. 152). Beliefs, ideas, sentiments, emotions, etc., at the level of personal experience are both structuring and being structured by the semiotic environment of the social system, contributing to the adaptive fitness of individuals who participate in the ritualized performance of the building blocks, establishing a common social identity.

Andrei-Razvan Coltea’s Complexifying Religion (Coltea 2023) builds off this evolutionary understanding of religion as a CAS and considers religions to be “solutions to the complexity of the environment” (p. 13). Coltea’s sociological case studies show how religions are entropy-reducing technologies that reduce information complexity in the environment (uncertainty) by producing the perception of a system of meaning in psychic systems (individuals) through the communication of ten building blocks rather than eight: rituals, supernatural agents, myths, taboos, the sacred, moral obligations, afterlife beliefs, religious identity, superstition, and authority (2023, pp. 16, 29–52). For Coltea, it is meaning, rather than cooperation, that is the adaptive output of the building blocks constituting the religious systems (2023, p. 28). Particularly relevant for the present analysis is Coltea’s observation that psychic systems interpret a meaningful “signal” amidst the “noise” of information constantly encountered in experience, and that meaning can be understood as that information which makes reality predictable and intelligible by selecting for courses of action available to the individual (2023, p. 14). I shall return to this dynamic below in my discussion of the role of evolutionary epistemology in the interpretation of meaning, but taken together, the two approaches to understanding religion as a CAS show that internal states must matter to the psychic system in some meaningful way (if only to avoid pain), as a form of “symbolic capital” that can translate that meaning into the social energy (see Mavrofides et al. 2011, p. 364) needed to continue participating in the building blocks necessary for reproducing the social and semiotic order that guarantees one’s own experience in time.

What is needed is a philosophical turn in the multidisciplinary framework of the systems theory of religion that builds off the how of religion as a complex adaptive system to explain why the communication of beliefs through socio-cognitive structures like the building blocks motivate cooperative involvement in the form of meaning. Pace implicitly recognizes a need to explore the philosophical side of the systems theory of religion when he observes that religious systems are influencing and being influenced by the act of interpretation (see Pace 2011, p. 226; Cf. Coltea 2023, p. 56). There is a need to better understand the way the contents of symbolic information motivate individuals to participate in the feedback of their experience of meaning through ritualized participation, however it is expressed. My effort here is to begin to apply systems theory philosophically, to better understand those “internal forces” hinted at by Pace that drive religious communication (2011, p. 226). We need to return to the mind, to understand why a particular semiotic system, cultivated by the ritualized expression of the building blocks, is considered relevant to individuals, such that they continue communicating the framework of meaning to other psychic systems, thus promoting group cooperation and fitness through the production of meaning on the complex adaptive system model. To do so, we need to better understand the role of entropy in complex systems, a property left relatively unexplored in systems theory in general (see Mavrofides et al. 2011) and in relation to the systems theory of religion in particular (see Coltea 2023; Fisher 2017, 2019).

Where the systems theory of religion has thus far assumed an emergent relationship between psychic and social systems, my goal in the following sections will be to further develop our understanding of the interpenetrating relationship of these systems through an analysis of the role of entropy in human experience. What I offer is not to be interpreted as an attempt at a scientific theory, but rather as a philosophical approach to explaining religion that draws on the science of systems and the role of entropy, which will serve as a “centroid” for interpretation (see Cho and Squier 2013, p. 22) that may serve as a minimal narrative for establishing a cross-cultural baseline for understanding what we mean by religion. Methodologically, then, I am bracketing theological, metaphysical, and essential questions and implications and looking only at religion “from below”, as a phenomenon emergent from psycho-social system communication (Beyers 2022, p. 790). This will help us better understand how that bracketed content emerges and sustains itself as an autopoietic system of logic and symbolism (a semiotic system/framework of meaning) capable of exercising constraint on the interpretation of meaning by individuals (see Mavrofides et al. 2011; Coltea 2023), thus contributing to their adaptive fitness per the CAS model.

2. Entropy, the Arrow of Time, and Uncertainty

The concept of entropy emerged during the industrial revolution to explain the energy an engine loses over time in the form of heat (Carroll 2010, pp. 33–35). Building off observations made by Sadi Carnot (1796–1832) in his experiments with heating and cooling a chamber to force the movement of a piston, Rudolph Clausius (1822–1888) first posited the idea that entropy was the measurement of usable energy in a system and observed that in a closed system entropy only increases. While the total energy remains constant in a closed system (first law of thermodynamics), the usability of the energy declines over time, as some of that energy is lost in the form of heat (second law of thermodynamics). The second law states that in a closed system, energy goes in one direction from high temperature to low temperature and will not reverse “direction” without influence from energy outside the system. While physical processes are reversable, mathematically speaking, the thermodynamic processes that impact our everyday lives are very much directional, which we describe according to the arrow of time (see Carroll 2010, p. 38; Davies 1974, pp. 29–30; Martyushev 2017). A hot cup of coffee will radiate heat into the room in which it is placed until it reaches the temperature of the surrounding room in thermal equilibrium, for example. While this process is theoretically reversable, we never observe the cup of coffee spontaneously warming up. Instead, the cup remains at room temperature unless one reheats it (outside energy). Similarly, cold things “warm up” to room temperature, but this is not because they are radiating energy into the room, but rather because the cold object is absorbing heat from the room, the wider environment. Entropy is considered low when there is more useable energy than not, and vice versa, with maximum entropy (unusable energy) representing thermal equilibrium. While time is reversable in the laws of physics, cosmologists describe it as a measure of the global increase in the entropy in the cosmos as a closed, thermodynamic system. Assuming the standard inflationary “Big Bang” model, there is a finite amount of energy being used and rearranged in the space of an inflating cosmos, leading to the conclusion that the universe will end with “heat death” over billions of years, when it has reached a uniform temperature and maximum entropy. Martyushev defines time psychologically and cognitively as “the measure of [change] of some phenomenon and the number of ways of its representation during the [change measuring process] in the brain” (Martyushev 2017, p. 4). Time is the experience of this irreversible direction of energy expenditure, and the probabilistic distributions of various subsystems throughout the universe toward equilibrium. Martyushev concludes that time should be considered isomorphic, with increases in the thermodynamic entropy of an isolated system (Martyushev 2017, p. 6), and so we can conclude that the arrow of time, though subjectively perceived, corresponds to the overall increase in entropy in the cosmos.

Entropy is also familiar to people from its articulation in the statistical mechanics of Ludwig Boltzmann (1884–1906), who wanted to describe the distribution of gas particles in a closed system and noticed that those particles tend to spread out evenly within the system and remain distributed, if not acted upon by external forces (Carroll 2010, pp. 36–38). This dynamic is familiar to anyone who has smelled a scent disperse throughout a room; or observed a recently cleaned house become messy again; or observed a mass of sports fans leaving a stadium and emptying out into the parking lot and wider community. In all these cases, the individual members of the system—perfume molecules, household items, people—go from a low-probability configuration of elements (i.e., in the bottle; clean and organized room; or a community in a stadium) to a higher-probability system state (i.e., anything other than that arrangement) within the environment (room; wider community). Phenomenologically, we usually describe increases in statistical entropy as increases in “disorder” relative to a previous state the observer considers “ordered”. However, Boltzmann’s mathematical articulation of statistical entropy pertains to the probability of the distribution of elements within a system, not an evaluation of the distributed state. Although it is possible that there is a statistical fluctuation that “reorganizes” the system in a low-entropy configuration, it is not probable without external influences on the statistical system. Though there is nothing in the reversibility of the laws of physics to dictate that the perfume particles could not re-concentrate in the corner of the room from which they originally spread out and go back in the bottle, it is so improbable that it is a de facto impossibility.

Finally, there is entropy in information studies (see Pierce 1980, pp. 78–106). In A Mathematical Theory of Communication (Shannon 1948), Claude Shannon described entropy as a measure of the probability of reproducing a given message sent across a communication medium by its receiver, and the formula he arrived at is an analogous expression of the Boltzmann equation in statistical mechanics (Mavrofides et al. 2011, p. 3; Deacon 2011, pp. 378–9). Working for Bell Labs, Shannon was concerned with the physical limitations of communication technology, and in his honor, I will explain information entropy by using the example of the children’s game, “telephone”. In this game, participants sit in a circle and whisper messages to each other, one after another, in a series. The fun happens when the final participant in the series fails to accurately reproduce the original message whispered by the first participant, because the final, reproduced message is usually far from what the original message was, often inducing laughter. The next part of the fun is the debrief amongst the participants: where in the channel did the information diverge from the original message and why? Did someone misunderstand the original message, did they intentionally distort it, or something else? According to Shannon’s articulation, the entropy in the communication between two children is high when the receiver is unable to predict what message will be communicated. As the sender speaks, information is contained in the speech, and the entropy of the message decreases because words signify meaning that filter out the possibility of other interpretations—“hello” does not imply “goodbye” in English, for example. The probability of accurately replicating the message from the sender increases according to one’s understanding of the meaning of the message. Information entropy is a measure of uncertainty, and equilibrium occurs when there is equal probability of different information being sent in the information channel. Information reduces uncertainty, and so the players could lower the entropy in their communication channel by agreeing on a topic, such as “pets”, before the game begins. So, a child may hear “fog”, but infer “dog”, because the latter corresponds with the meaning of the category contained by the information “pets”. Most importantly, even if I am entirely confused on what was said, “stadium” was not intended, because that lies far outside of the semiotic information constraint, “pets”. However, the introduction of such a category would serve as external information directing communication and work between psychic systems. The category “pets” concept is now an informational constraint that selects for or against a framework of meaning (a semiotic system state) relative to what is heard via the whisper (cf. Deacon 2011, p. 389).

This dynamic of reducing uncertainty (Shannon entropy) in the minds of individuals is what religious systems appear to be adapting to, through their communication of a framework of meaning as information that reduces uncertainty as to how to interpret one’s existential experience, relative to information in the physical and social environment (cf. Coltea 2023, pp. 13–16; Pace 2011, p. 206; cf. Mavrofides et al. 2011). Coltea describes this as the psychic system distinguishing the signal of meaningful information that increases our predictive capacity from the noise of “useless” information that does not (2023, p. 14). Purzycki and Sosis touch on different types of information in their discussion of Rappaport’s (1999) distinction between indexical and canonical information communicated in ritual (2022, p. 149). Indexical information pertains to the physical activities of ritual participants in time, date, location, movements, etc., and signals for participants the canonical level of information that provides the meaningful framework for why that indexical information is happening. Canonical information links “past and future” when “performers…perceive rituals to have enduring or even eternal referents” (Purzycki and Sosis 2022, p. 149). The more canonical information provided to the members of the community about the necessity of maintaining the community, the higher the probability of replicating a message in communication that ensures group cooperation, contributing to individual fitness in an evolutionary sense. This is why “successful ritual behaviors—that is, those that are performed and passed on to future generations—require cognitive support” (Purzycki and Sosis 2022, p. 150). A religious system can only be adaptive if it can select for and against elements of its semiotic system that militate against uncertainty in the minds of individuals, who then adapt that uncertainty to the type of symbolic capital that can be used as social energy, through interpreting uncertainty according to the operating framework of meaning in the social system, thus subsuming the difference of interpretation into the system, allowing it to achieve greater complexity (Pace 2011, p. 221; cf. Coltea 2023, p. 56).

A paradigmatic example of a religious system selecting for and against meaning that mitigates uncertainty can be seen in the way the Nicene–Constantinopolitan creed functions to establish “orthodoxy” in theological interpretation concerning the relationship between Jesus of Nazareth and the God of Abraham in Roman Catholicism and other Christian traditions. Because Christians could openly practice their faith without fear of persecution after the Edict of Milan (313 c.e.), bishops and their presbyters began sharing theological interpretations in letters and public sermons, which injected new information into the framework of meaning in the nascent religious system. Arius’s claim that the Son, and therefore Jesus, was the “first born of creation” was a novel interpretation of the relationship considered erroneous by many bishops, who thought it would nullify the possibility of salvation. Differences in theology notwithstanding, there were also practical consequences for the socio-political economy of Constantine’s newly unified empire, as the open rhetorical conflict between church leaders threatened its stability. Arius’s interpretation can be understood as a “perturbance” in Coltea’s model, defined as “informational inputs of increased magnitude, continually threatening to falsify the core propositions around which systems are organized, as well as dissolving the established relationships between the system’s elements, thus affecting its emergent aspect and output (meaning) and endangering the very existence of the system by increasing the difficulty of fulfilling its entropy reducing endeavor” (2023, p. 56). In the language of systems theory, Arius’s interpretation was perceived to threaten the very autopoiesis of the system of meaning that made salvation through faith in Jesus possible, because it introduced an articulation of meaning that increased, rather than decreased, different avenues of interpretation concerning who Jesus ultimately is. To resolve the theological conflict, Constantine called the first ecumenical council of bishops to Nicaea in 325 c.e., tasked with the goal of determining the “right faith/teaching” of the scriptural understanding of Jesus’ relationship to God the Father and the Holy Spirit. While controversy over the theological language to describe this relationship would last for decades and require multiple councils to resolve, the Nicene–Constantinopolitan creed formulated at the Councils of Nicaea (325 c.e.) and Constantinople (381 c.e.) finds its first authoritative reference in the historical record at the council of Chalcedon (451 c.e.), albeit unintentionally (Ayres 2004, pp. 255–6; Cf. Kelly 1983, p. 31). After Chalcedon, the creed served to distinguish the conceptual set upon which “orthodox” theology would proceed in the life of the “one, holy, catholic (universal), and apostolic church”. Any interpretation that lay outside the boundaries of meaning established in the creed was “anathema”, as the many schisms after Nicaea attest, with the formation of Christian communities that rejected the interpretation of Jesus as homoousios (of the same being) with the Father, preferring instead the language of “similarity” (homoiousios).

A philosophical approach to the systems theory of religion using the centroid of entropy is not concerned with which of these interpretations is true, but rather with the evolution and adaptation of a tradition’s symbolic capital over time. What is important is that the creed is perceived by an individual to serve as an interpretive constraint on one’s experience of the Gospel message (Cf. Coltea 2023, pp. 16, 56). This perception in turn reduces the Shannon entropy of possible interpretations in the minds of individuals concerning who Jesus is, and the continued communication of the meaningful information contained in the creed feeds back into the religious system via the building blocks of liturgy and theology and maintains the identity of the Church as an anti-fragile semiotic system resistant to collapse (see Coltea 2023, p. 58). It is this adapting to increases in information entropy (uncertainties), introduced by Arius, by leveraging Greek philosophical concepts and language found outside of scripture to maintain the religious system’s autopoiesis of meaning, that leads us to recognize the development of the Nicene Creed (and credal statements in general) as an interpretative constraint, and therefore a complex adaptation of a religious system relative to informational stimuli communicated in the social environment.

3. Entropy and Religious Systems

In the CAS model, the autopoiesis of social systems emerges from communication between psychic systems, and psychic systems emerge from the biological the autopoiesis of our bodies, which are in turn emergent from various subsystems in human physiology. We ended the last section by discussing the way religious information reduces entropy in the form of meaning, and so here, we need to better understand the way entropy conditions the interpretation of meaning in experience, as emergent from the psychic system’s embodied cognition. Our bodies are continually vacillating between increases and decreases in entropy, as we take in food from the external environment to be translated into caloric energy to survive as a self-organizing (autopoietic) body in space and time (Maturana 1975, 1980; Undgaokar 2001). What is more, most of “my” current cellular structure is not the same as my cellular structure from when “I” was a child, yet I have memories sufficient to predicate “myself” over my experience of the arrow of time in a continuous present-tense moment (see Augustine 2006, p. 271 [Confessions, XI.iii.26]; cf. Deacon 2006). And so, we feel for ourselves and others, and intensely so, as time marches on, in ways that no EEG or CT scan could ever really fully describe, because it somehow misses those qualia, or “what-it’s-like-ness”, of my experience. If we look at the data gleaned by methodological naturalism from the disciplines of evolution, biology, and cognitive linguistics, we recognize that the possibility of self-conscious experience is dependent on a nested hierarchy of emergent complexity, evolved over millions of years of hominin evolution, that keeps our bodies in a low-entropy configuration until we eventually age to a point where autopoiesis is unsustainable. Anthropological neuroscientist Terrance Deacon has developed the concept of the “teleodynamic self” as emergent from the homeodynamic (Deacon 2011, pp. 206–34; cf. Deacon 2006, pp. 126–30), morphodynamic (Deacon 2011, pp. 235–63; Deacon 2006, p. 131), and goal-directed (teleodynamic) processes (Deacon 2011, pp. 264–87; cf. Deacon 2006, pp. 136–7) in our bodies that work together to make the phenomenological unfolding of experience possible. For Deacon, the “[s]elf is, in all cases, the origin, target, and beneficiary of functional organization, whether molecular or mental” (2011, p. 466). The teleodynamic self emerges from the low-entropy maintenance of our bodies in time such that self-reference, i.e., conscious experience, is the organizing structure that supervenes on our body, capable of directing it this way or that way, adapting it to a dynamic information flow from the environment of nature and society as an operational, emergent, dynamic, yet non-substantial, “self”.

The neurological underpinnings of our conscious experience have evolved from millions of years of hominin evolution (Purzycki and Sosis 2022; Bellah 2011), leading Robert C. McCauley (2011) to conclude that religion is fundamental to human thinking, with science arriving later in social system evolution. Science is in fact socially fragile and took a long time to develop, as it is historically dependent on the emergence of literacy and abstract thinking developed from the emergence of theoretic forms of religious explanation like theology, after popular forms of ritualized religious expression sufficiently saturate a given culture. In short, humans are not “forced” to be religious where they would otherwise think rationally and scientifically, but religiosity instead appears as the default mode of human knowing. McCauley’s study shows how Western science emerged from religious communication as a different mode of knowing that developed later in human social evolution, distinguished by a commitment to a probabilistic form of explanation established through observation, methodological naturalism, and peer-review. This socio-evolutionary understanding of the development of science is supported by Donald T. Campbell’s evolutionary approach to epistemology, which recognizes that evolution is “a knowledge process, and that the natural-selection paradigm for such knowledge increment can be generalized to other epistemic activities, such as learning, thought, and science” (Campbell 1974, p. 413). The mental representation of memory that serves as the organizing principle of experience itself emerges from selection dynamics that “remember” a multi-billion-year input–output relationship with the wider environment as recorded in an organism’s genome. William Harms argues we should think of natural selection as an information transfer process, conceived not as information going from a source to a receiver, but rather where “selection sort of induces a probabilistic tracking relationship between the population distributions and features of the environment which affect the fitness of the types in the population. Information is the tracking relationship, ‘transfer’ is a metaphor for this induction process” (Harms 2004, p. 147). In other words, information is the individual’s observing something and internalizing it (induction) by constructing it as a semiotic “rule” that one adapts to, to navigate the world or society. Harms’ understanding of evolution as an information transfer process bears a resemblance to Pace’s understanding of information as “the elementary unit of the system-environment relational process” (Pace 2011, p. 223), and so we see how evolutionary epistemology is congruent with one of the main insights of Luhmann’s systems theory, the idea of “double operation” or “double contingency” (Pace 2011, p. 220): the act of distinguishing establishes a contingent system in experience, and necessarily so. By distinguishing this from that, one draws boundaries around where this (system) interacts with that (environment), based on whatever criteria one is drawing on to make the distinction (cf. Coltea 2023, p. 6).

We draw distinctions according to a symbolic understanding of meaning structured through language (Deacon 1997), and so the perception of meaning by the psychic system can be selectively advantageous if it contributes to the survival of the individual, by reducing uncertainty as to how to interpret experience in a way that reinforces the autopoiesis of the community, which in turn reinforces the autopoiesis of the teleodynamic self and his experience through access to cooperation and resource allocation on the CAS model. Masson’s work in applying cognitive linguistics to theological language shows how the use of metaphor and symbolism reduces epistemic uncertainty by producing “new, often unanticipated, meanings” to make sense of two or more often conflicting inputs (2014, p. 102). This conforms with Harms’ observation that selection stabilizes the possibilities for interpreting information, and thus “[w]e get the emergence of meaning conventions in a way that does not require their previous existence” (Harms 2004, p. 196). Meaning is therefore “biologically basic”, because it is a signaling system that emerges from communication between sender and receiver, interpreted according to a social system’s operating framework of meaning, that also changes over time. Harms understands moral codes and taboos to be “stabilizing conventions” that serve as historical patterns of the functional history of the religious system that have been preserved as epistemic sub-systems, contributing to the religious system’s autopoiesis over time. Meaning can be understood as a recursive referencing of previously selected information that serves as a constraint on future interpretations of environmental stimuli, to reduce uncertainty and make sense of our experience. “Making sense”, on this model, would be a reduction in the probability of the natural, social, or religious system state diverging from what is expected, as held according to the conventions accrued over one’s experience, via the distinction between indexical and canonical information and its consequences (cf. Deacon 2011, p. 393; Coltea 2023). The religious system, by definition, reinforces its semiotic framework in communication with phenomenological systems to constrain future interpretations of uncertainty by psychic systems to a sufficient degree to maintain the autopoiesis of meaning (cf. Pace 2011, p. 221). Because meaning is that which is ultimately produced by the symbolic communication of religion as a complex adaptive system, Coltea prefers the concept of “allopoiesis” rather than autopoiesis to describe religious systems, because it is “a sub-category of autopoiesis which includes systems that are generated by other (external) systems” (2023, p. 2). Meaning, of course, cannot be located in the genome, and so what is reproduced and sustained by the religious system is not just the autopoiesis of biology, but the emergent allopoiesis of the semiotic system of meaning that constitutes the religious system in communication.

I agree with Coltea that Richard Dawkins’ concept of the meme, as a memory unit of culture passed down through communication, while eschewed by evolutionary theorists, should be revisited in light of our understanding of religion as an emergent and complex adaptive system (2023, p. 3). The genealogy of religion gains new importance when religions are viewed as complex adaptive systems promoting the allopoiesis of meaning among psychic systems, and we can explain the conceptual relationships that exist between some religious systems and not in others to be the result of individuals reacting to the semiotic selection pressures in one’s “social ecology”, as populations in certain geographic regions interact together over time (see Purzycki and Sosis 2022, pp. 183–9). The evolution of religious belief shows something like a “memetic drift” in symbolic thinking, clustered according to geographic constrains on human interaction and technological constraints on human communication that contribute to the instability of religious systems over time (cf. Coltea 2023, pp. 63–67). Of course, social power dynamics guide those interactions, and so systems of belief that survive semiotic adaptation are usually operating because they belonged to a victor somewhere down the historical line (see Pace 2011, pp. 207–10).

To summarize, if religion is a complex adaptive system, then it is adapting to the communication between people that is emergent from biological and social pressures that are transformed into semiotic information in the form of meaning. The teleodynamic work of cognitive information processing in communication helps explain how we get from the psychic system’s experience of meaning to the ritualized energy that maintains a religious system over time. Much like how the energy efficiency of plant growth is described according to the “golden ratio”, the building blocks of religion are considered as such because they are the most energy efficient psycho-social mechanisms for facilitating the reduction in the degree of Shannon entropy necessary to facilitate the emergence of meaning in the minds of enough psychic systems, in the form of symbolic capital, to maintain group cooperation and identity over time (see Mavrofides et al. 2011). Just as living beings take in caloric energy from systems of flora and fauna in the environment to survive, so too does a reduction in uncertainty at the symbolic level of meaning allow the phenomenological system to survive as a person: a non-fungible and irreducible dynamic of autopoietic self-reference emergent from the physical and informational teleodynamics that instantiate the psychic system (cf. Luhmann 1995, p. 315). In a phenomenological sense then, the person is a psychic system for whom one’s autopoiesis has become an issue in a sense analogous to Heidegger’s conception of Dasein, for whom non-being is a persistent horizon of which one is aware, but that one never experiences (Heidegger [1928] 1962, pp. 32–33). We therefore seek to secure for the person-in-community the possibility of preserving this constancy of self-reference (i.e., existential concern) found in and through our relationships, and language is our organ for doing so (Deacon 1997; Masson 2014).

4. A Philosophical Turn in the Systems Theory of Religion: Entropy and the Idea of God(s)

If social systems emerge from communication between psychic systems, and psychic systems are emergent from the teleodynamic system as a physical system in function (i.e., subject to the second law and natural selection), and the psychic system needs the social system to survive (i.e., cooperative fitness), then the semiotic tools for developing those frameworks of meaning within religious systems, such as theology and philosophy, must likewise be subject to the constraints the second law places on one’s neurology and cogitation, as well as the religious system’s ability to adapt to maintain the framework’s allopoiesis of meaning as an information system, as we saw in the case of the Nicene–Constantinopolitan creed above. What, then, drives the adaptation and evolution of religion from a philosophical point of view?

For a person, the autopoiesis of oneself as a phenomenological system is self-evident in the mental self-reference of the present moment, constructed linguistically and dynamically through our socialized relationships with others as part of a community over time. The autopoiesis of the psychic system is already grounded “beyond” itself in the cognitive linguistics of the community’s symbolic framework of meaning that is communicated since infancy in some way, facilitating the emergence of regular self-awareness in the ensuing years. In the words of the Jesuit philosopher, Bernard Lonergan, “prior to the ‘we’ that results from the mutual love of an ‘I’ and a ‘thou,’ there is the earlier ‘we’ that precedes the distinction of subjects and survives its oblivion…. it is as if ‘we’ were members of one another prior to our distinctions of each from the others” (Lonergan 1971, p. 57). As one grows, the phenomenological system refers to oneself through community in the interpretation of experience (cf. Pace 2011, p. 221), and so one intuits one’s framework of meaning to be immune from entropy’s probabilistic decline, and necessarily so, because it serves as the logical condition establishing the allopoiesis of the system of meaning perceived to make possible one’s autopoiesis as a living being. To use respiration as an analogy, through symbolic language, the phenomenological system “breathes” a framework of meaning that goes beyond mere functional or indexical information, in a way analogous to the way our lungs circulate oxygen through our blood, keeping our biological autopoiesis going in a low-entropy, organized state, as a psychic system. The consumption of meaning constructs and reinforces identity, belonging, and existential significance in the face of second-law contingency, where time only ever moves from before to after, uncertainties induce arresting anxiety, and loneliness is of existential concern (cf. Coltea 2023, p. 25). It is in the allopoiesis of meaning discerned in community—that feeling of “belonging”, broadly speaking—that we find respite from these concerns, learning that the fragility of one’s autopoiesis is not the end, but that one’s “spirit” of experience will live on in the lives and memories of others, at a minimum, and possibly in a metaphysical state in some form, depending on the framework of meaning.

Religions are complex adaptive psycho-social systems, that cultivate for phenomenological systems (persons) the inference of a logical set, the assumption of which establishes the possibility of the perception of meaning as promoted, yet constrained by, the symbolic framework communicated via dynamic socio-epistemic structures like the adaptive building blocks. However, we need to distinguish that dimension to meaning that qualifies as “religious” in the social system that allows it to operate as an “expert system”, speaking to the existential concerns of individuals-in-community, relative to what we would describe as non-religious systems operating in society (see Pace 2011, p. 225). Otherwise, all communication would be considered religious, which would mean the concept is unhelpful for explaining what we mean by religion, and the phenomenon should instead be reduced to explanations offered by other disciplines like cognitive science, sociobiology, anthropology, or economics. Using Coltea’s insight that religion is a “complexity reducing technology” that sifts through the noise of information to isolate the signal of meaning in communication (2023, pp. 13–14), we can say philosophically that the religious dimension of meaning in communication is perceived by the phenomenological system when it distinguishes the anentropic character of a signal indicating canonical information, as distinct from the “noise” of entropically applicable indexical information. By anentropic, I mean “not subject to entropy”, rather than negentropic, or “entropy reducing”. Indeed, Coltea shows how meaningful information communicated by the religious system reduces entropy (uncertainty) and so contributes to the adaptive predictability of the psychic system, but why that information would be considered meaningful at the level of human experience requires a philosophical approach.

What is perceived when communication is understood to contain “religious” meaning is the abstract representation of the interpenetrating dynamic of the allopoiesis of meaning, constituting the community-of-individuals and the individual-in-community, depending on one’s starting point in communication (cf. Pace 2011, p. 221). The referent grounding this interpretation is intuited to be in some way unconditioned by the very conditions set on experience that drive the need to secure for experience such an unconditioned referent. Otherwise, there would be no meaning, but rather uncertainty “all the way down”, resulting in persistent anxiety. This is not to say that the theory requires the symbolism of a religious system delivered through the building blocks, such as supernatural agents, must exhibit only anentropic characteristics to be considered the building blocks of meaning. This would betray a Western bias in religious thinking that understands divine agency to necessitate anentropic ontology, and therefore the existence of some sort of ultimate reality “beyond” the world in order to explain the existence of the world and the source of meaning. As we see in the various myths of oral and written traditions, however, the category of supernatural agents is often described as subject to the demands of the second law, and so not always supernatural. Hence, the anentropic characteristics of omnipotence and omniscience, describing the idea of God in Western philosophy of religion, is itself a symbolic development in the evolutionary epistemology of phenomenological systems communicating their intuition of the anentropic over time in those geographic regions of the world we describe as “western”. What I mean by the intuition of the anentropic dimension of meaning is not the character of the referent itself—i.e., intuiting a being that is ”everlasting” as a necessary condition—but the way communication establishes a conceptual null set in symbolic cognition, which makes possible the interpretation of meaning that grounds our sense of self and social identity beyond the lifespan of individual psychic systems interpreting this information. The intuition of the anentropic can be considered the perception of a signal that an individual interprets to be contained in communication, because it ultimately refers to the possibility of continued psycho-social autopoiesis in time (i.e., identity), despite the persistent experience of life in the arrow of time to the contrary. The autopoiesis of our self-reference or social reference is merely the analogical starting point from which this referent is apophatically inferred in the phenomenological encounter of self-referential constancy in the arrow of time.

Pace says something similar to what I mean by the intuition of the anentropic dimension of meaning as rooted in an inference dynamic, in his description of the way different traditions appeal to “the notion of spirit” as an allusion to an “idea of an event that was sought after and dreamed of, but that ultimately takes place as if the individual had not really been expecting it”. According to Pace, this dimension to experience “guarantees its endless reproducibility” because it functions as a “communication code”, the logic of which is universal and so beyond the control of any one upstart or system of belief (Pace 2011, p. 216). But where terms like “spirit” or “supernatural” are problematic in describing many new religious movements in our secular social system today, given that there are anentropic referents securing identity that do not invoke metaphysical postulates, the idea of the “anentropic” can serve as a negative descriptor because it signifies the intuition of something that is necessarily persistent across the experience in time. This lets us describe anentropic agency like Gods and spirits apophatically via linguistic negation, rather than predicated ontology. We predicate “religion” or “religious” to a given communication not in terms of what the communication describes, such as the supernatural character of an agent, or the importance of a ritual, or ethical principle, but rather in the way those building blocks reproduce the intuition of the anentropic dimension of meaning in the minds of individuals over time, promoting one’s ability to see oneself in the symbolic system of meaning produced by their communication. This affords us the opportunity to compare systems of belief without privileging a cultural conception of the anentropic dimension of meaning—e.g., as omniscient, omnipotent, etc.—because the anentropic represents an epistemological move to explain why meaning is the output of religion as a complex adaptive system, rather than a metaphysical claim that explains that for beliefs to be religious they must refer to the anentropic or negentropic character of “supernatural” agents—though that may be the easiest way to identify the intuition of the anentropic in some religious systems. As an epistemological claim, the intuition of the anentropic dimension of meaning helps distinguish the religious character of communication in general, but it cannot be elucidated beyond its functional articulation without the ability to recognize the symbolic representation of that intuition in communication. Put simply, meaning may be the output of religious systems, but why a particular framework of meaning is considered meaningful remains a “black box” to outsiders unfamiliar with its symbols and metaphors. Religions produce meaning as adaptive semiotic systems, emergent from adaptive biological systems, and so an “inside-the-faith” understanding is required to recognize the articulation of the anentropic dimension of meaning in a social system’s history of symbolic communication. In addition to insights gleaned from cybernetics, sociology, and evolutionary theory, the traditional tools of religious studies and the humanities, such as linguistic, textual, and cultural analysis, will remain relevant for an interdisciplinary approach to the systems theory of religion.

Social systems theory shows that religious systems evolve over time due to “improvisations” by individuals in the interpretation of meaning, often due to social exclusion (Bognár 2017), and their ability to galvanize followers around those interpretations and dominate as sociological upstarts (see Bellah 2011, pp. 175–82). According to Pace “This personality is capable of shifting the symbolic boundaries of previous systems of belief and showing that the established meanings they contain can be interpreted differently” (Pace 2011, p. 218). By improvising “on a theme already known” in the semiotic environment, the religious system evolves into a new “symbolic universe”. Pace’s idea of improvisation may be better explained in the context of the second law, given its relationship to time, and therefore the phenomenology of dissatisfaction that appears throughout the human record. Oliveira et al. (2024) show the role of entropy in the “opinion dynamics” that lead to cooperative behavior in a small group experiment, and that the potential for an interpretation against cooperation is often high and never zero. Cooperative dynamics emerge spontaneously, then, as a constraint on the potential for disorder produced by a noisy social environment, thus mitigating social entropy in the system by “inducing exuberant collective phenomena in complex systems” (Oliveira et al. 2024, p. 1). Religious systems often emphasize that the upstart can emerge from within, with how we think and feel. Buddhist traditions, for example, speak of the root of the psychological anguish of suffering to be dukkha—“dissatisfaction” or “unsatisfactoriness”—that spontaneously arises in the mind (Botha 2022, pp. 23–34). As the narrative of Siddhartha’s youth shows, even when someone has everything one needs to survive and live well, ennui still creeps in, leading one to quest after meaning, when the established forms are insufficient to address one’s existential concerns. Since there are more ways to be dissatisfied than satisfied when it comes to material needs, and since there is a teleodynamic correlation between our bodies and our experience, we could say that dissatisfaction is a phenomenologically more probable system state than happiness. We see the world’s religious systems expending a great deal of cognitive energy developing semiotic systems to manage the highly probable increases in dissatisfaction one will encounter in one’s life, especially as we gain more autonomy as individuals in a secularizing world where we are often left to deal with the anxiety induced by uncertainty on our own (see Coltea 2023, pp. 76–94).

Furthermore, Coltea’s research on the fragility and resilience of religious systems in relation to entropy production shows that while the communication establishing a religious system proceeds from individuals, perturbations in the interpretation of meaning are often motivated by pandemics and other stressors in the environment that have an anxiety-inducing impact on human experience (2023, pp. 67–75). Whether in response to a plague, poor authority, or a natural disaster, by providing a challenge or alternative interpretation to any of the building blocks mitigating the production of Shannon entropy in the religious system, uncertainty increases in the minds of individuals regarding the referents on which the framework of meaning depends. The injection of this information increases the probability of interpretations in the religious system diverging from the operating system state, which has the potential to de-couple psychic systems from an epistemic correlation between the framework of meaning and their felt need to participate in those building blocks, thus destabilizing the communication of meaning in the religious system, leading to its evolution.

Religious systems seem to evolve and even collapse, then, when expressions of dissatisfaction from individuals, motivated by various causes, are unable to be managed by the operating framework of meaning in a religious system, because those who are dissatisfied can sustain the continued communication of their dissatisfaction with others. Moreover, we appear to observe tipping points in the evolution of religion that correspond to advances in communication technology (Cf. Coltea 2023, p. 61). Bellah’s analysis of the evolution of religious systems in Axial Age (ca. 800 b.c.e.–ca. 200 c.e.) India, China, the Middle East, and Ancient Greece shows that large-scale populations with access to a novel communication technology—e.g., writing—led to the emergence of “moral upstarts” who critiqued the status quo by appealing to anentropic referents of causation (a first cause) or value (a highest good) understood to apply to all people, regardless of community and social status (Bellah 2011, pp. 265–566). With access to writing, religious theory in large-scale societies tilted toward the abstract, both to preserve the system of dynastic state power, and to afford the dominated a means to communicate their understanding of experience in a world where only the few live well and the rest toil and labor until death. When a society reaches a critical mass such that sociological factors create tension within a community, paradigms of religious thought and practice will be interpreted and re-interpreted in such a way as to justify the allopoietic, self-evident significance of those who do not see the source of their autopoiesis in the operational framework of meaning (see Bognár 2017, p. 28; cf. Fischer-Lescano 2012). Bellah’s point is that the theoretic culture that emerged from the Axial Age, made possible using writing, encouraged the upstart to move into the abstract space of ideas, appealing to moral and causal principals applicable to all, regardless of social status, rather than the use of force. If people are unable to find fulfillment in this life, for example, then perhaps fulfillment will be found in frameworks of meaning “beyond” the entropic condition, as Bellah shows is the case in the evolution of religious systems in these regions during this time. Hence, systems theory is helpful in explaining the ways religious systems evolve in relation to the degree of communication complexity afforded by various technologies—writing, printing, mass media, the internet, social media, AI, etc.

Religious systems adapt and change over time, because we continue to improvise on our interpretation of the anentropic dimension of meaning, and we cannot help but share our interpretation with others, particularly in the face of suffering or dissatisfaction. Whether conceived as a supreme reality, supernatural or metaphysical agency or principle, people and land, political utopia, or even a personal “happy place” or feeling of flow state, the anentropic dimension of meaning is symbolically constructed as an inferential reference to that which makes the autopoiesis of meaning possible. Nevertheless, our process of believing is rooted in embodied cognition; the inference dynamic underlying the intuition can only draw on data from the entropic condition to formulate its conception of the anentropic dimension of meaning. This is why references to the anentropic are often structured symbolically in terms of anthropomorphically conditioned imagery, because it reflects what we know best: ourselves. Belief is dynamic, everchanging, and is more than a rational choice calculus. People are not religious because they believe in propositions, but because they feel and experience something. The framework of meaning communicated by the religious system, with its symbolic expression of beliefs and values, helps them articulate it (see Purzycki and Sosis 2022, pp. 110–21). As a teleodynamic self, we are constituted by an interpenetrating dance of biology and linguistic information, and so our conceptions of the anentropic are forever constrained by an experience conditioned by language, the second law, and the arrow of time. Communication is considered “religious” when information is intuited to construct for the phenomenological system an anentropic dimension to meaning that facilitates the possibility of meaning as meaningful, i.e., as guaranteeing for the person both her causal autopoiesis, as well as the interpretive allopoiesis of meaning at the root of one’s sense of identity as an individual-in-community (cf. Luhmann 1995, p. 74). This intuition would be circular if it were not for the arrow of time, so we can think of the logic more like a spiral, with memory connecting the nodes of the spiral together in a perpetual present that will one day fade. In this sense, then, a systems theory approach to philosophy of religion arrives at the idea of God(s) not through an analogy of being, but through an analogy of meaning.

5. Conclusions: Why Religions Persist

The problem for systems theory, based in non-reductive emergent descriptions of ontology (being), is to explain the connection between system micro-states and macro-states, where epistemic limit domains present themselves across different levels of system analysis (Batterman 2002; Hooker 2004; Mavrofides et al. 2011). Given the complexity involved between psychic system and social system communication, Pace concludes that “we therefore cannot answer the question of why religions persist”, but argues that interdisciplinary approaches are the only way we will be able to provide anything resembling an effective description of what we mean by religion (2011, p. 224). I believe Pace’s epistemic humility is warranted, given the way meta-theories of religion function at the popular level—usually, ineffectively at best, and harmfully at worst. However, and remembering Hannegraff’s call, I proposed here that we consider entropy’s role in both sociological and evolutionary interpretations of the systems theory of religion, and when we do so, we obtain a more refined understanding of what we mean by religion as a complex adaptive system, and we get closer to answering Pace’s unanswerable question. However, we will have to be content with understanding religion as an emergent and complex phenomenon, requiring both functional and substantive domains of explanation, operating together in a system of explanation (see Coltea 2023, pp. 9–13, 16).

Religious systems persist, functionally speaking, because they are the most efficient mechanism for reducing Shannon entropy in the minds of psychic systems, facilitating group cooperation, as described by Purzycki and Sosis’ articulation of religion as a complex adaptive system that contributes to the fitness of the individual phenotype and her genome. I showed how communicating a framework of meaning in the form of language and symbolism supervenes on the statistical arrangement of psychic systems, keeping them in a statistically low-entropy population distribution over a geographic space (community), facilitating group cooperation and access to the resources necessary to mitigate increases in thermodynamic entropy. The perception of the supervenience of the signal of meaning on individual minds can be understood as the foundational way psychic systems adapt to the cooperation needed to access the resources necessary to maintain their physiology in a thermodynamically low-entropy state, which correlates with one’s continued experience as a phenomenological system emergent from a teleodynamic self. In summary of the three approaches to the systems theory of religion discussed here, we can say that religious systems are semiotically, sociologically, and evolutionarily adaptive and complex, because the communication of the building blocks is the fundamental psycho-social mechanism for the way people-in-community manage the experience of increases in informational, statistical, and thermodynamic expressions of entropy.

Religious systems persist substantively, then, because they are “complexity-reducing technologies” (Coltea 2023, p. 13) that reduce Shannon entropy in the minds of individuals (uncertainty) by articulating and constraining the symbolic parameters of the intuition of the anentropic dimension of meaning for psychic systems in communication, thus providing one with a perception of what matters, existentially speaking. The autopoiesis of self-reference experienced in the teleodynamic self cannot help but demand its continued self-referencing in time at the level of symbolic thought, and so death becomes an issue for the phenomenological system. A person adapts to the horizon of non-being through the cogitation and communication of anentropic meaning that inferentially establishes the possibility of autopoiesis extending beyond one’s own experience of self-referencing-in-community as a phenomenological system in time. The anentropic, therefore, is never experienced or grasped as such; it should instead be understood as an epistemic operation in time that interprets a given communication, to be encoded with information intuited to be in some way fundamental for the autopoiesis of meaning through which one constructs an identity for oneself and one’s community. Therefore, communication that qualifies as religious reflects an implicit or explicit reflection of the intuition of the anentropic dimension of meaning. Religion seems to have emerged as humanity’s initial tool for reducing uncertainty, and it stands to reason that developments in technology that facilitate greater degrees of communication complexity between psychic systems will contribute to the evolution of religious systems over time. The emerging fields of digital religion, implicit religion, and consumerism and religion show that new religious movements are taking unanticipated forms. We can conclude, therefore, that while the symbolic imagery articulating the intuition of the anentropic dimension of meaning may change in form, it will never disappear, so long as the arrow of time is perceived to be an existentially relevant issue for the autopoiesis of individuals-in-community.