The proposition that "societies are algorithms" serves as a useful, if somewhat contentious, heuristic for understanding the interplay of structures, rules, laws, individual agency, & social & collective patterns of behaviors. This assertion posits that societies are computational/algorithmic processes contingent upon repeated executions of their algorithmic steps by human agents across various timescales & in many instances (e.g. cryptographically defined economies, social media platforms, AIs, autonomous weapons & soldiers) is more than a mere analogy. It suggests an examination of societies as formal systems, whose logic, processes, mechanisms, inputs, & outputs can be systematically analyzed. This report will treat these metaphors not merely as casual comparisons but as formal models to be tested against canons of sociological theories & realities of social life in industrial & technological societies.
These metaphors have gained new relevance in an era where computational & digital algorithms along with ubiquitous surveillance & data collection increasingly govern & direct social, economic, & political life.1 The rise of what has been termed "algorithmic societies" or "algocracies"—societies wherein decisions are delegated to automated & algorithmic systems—transforms these metaphors from theoretical abstractions into tangible, concrete, & lived realities.2 As algorithms mediate social processes, from business transactions to governmental decisions, lines between social, computational, virtual, & digital become blurred, making this analysis both timely & pertinent.1
This report will proceed in five parts. Part I will deconstruct the core terms of these metaphors, "algorithms" & "societies", establishing precise analytical vocabularies by mapping components of computational & digital algorithms onto sociological paradigms that view societies as structured systems. Part II will explore mechanisms of "executions" & algorithmic enactments—processes of socialization, education, indoctrination, & cultural/memetic reproductions through which individuals learn, internalize, perform, & propagate algorithmic processes & mechanisms. Part III will analyze "loops" of these executions across micro, meso, & macro timescales, from daily routines to grand cycles of histories of civilizational growth, expansions, conflicts, & decay. Part IV will critically "debug" these models, examining phenomena that challenge their deterministic logic, such as individual deviances, conflicts, & problems of individual meaning by reframing them as "system errors" or attempts to "rewrite the algorithms" from the perspectives of the overall algorithmic systems. Finally, Part V will return to contemporary contexts, analyzing how big data, artificial intelligences, & "social media" platforms are making societal algorithms manifest with concrete technological artifacts like smartphones, "apps", data centers, internet & surveillance satellites, & ubiquitous wireless cell towers, engineering new & potentially more manipulable & "programmable" societies.
To rigorously evaluate the proposition that societies are algorithms, it is first necessary to establish clear & formal understandings of both terms. This section will define algorithms as structured, rule-based processes, drawing from computer science & mathematics. It will then survey major sociological paradigms that conceptualize societies not as random collections of individuals, but as structured, coherent systems. By placing these frameworks into analytical dialogues, we can compare their respective strengths & weaknesses for making sense of individuals & their societies.
Algorithms, in the most formal sense, are finite sequences of well-defined, mathematically rigorous & precise instructions for solving specific classes of problems or performing digital & numerical calculations.9 They are "effective methods" that can be expressed within finite amounts of space & time, proceeding from initial states & inputs, through a series of successive states, to final, terminating output states.2 This definition encompasses not only computer programs but any prescribed procedures like bureaucratic processes & cooking recipes.9 The formal structures of algorithms can be used as analytical blueprints for social processes.
Executions & enactments of algorithms can be broken down into well-defined core components & stages. Here they are:
Beyond these fundamental structures, the qualities & utilities of algorithms are assessed against various intensional principles. These serve as benchmarks for effectiveness & will be important when we later evaluate societal algorithms. Effective algorithms must demonstrate correctness, meaning they consistently produce the right outputs for any given inputs.10 They must exhibit efficiency, making optimal use of computational resources like time & memory.10 Simplicity is valued because it facilitates understanding, implementation, & maintenance.10 Furthermore, good algorithms should be robust, able to handle errors & unexpected inputs gracefully without crashing, & secure, protecting sensitive data from malicious attacks & exploits.10 These principles of computational & algorithm design provide an analytical lens through which to examine the often messy processes of societal life when they are viewed as algorithms.
The idea of societies as structured, rule-governed entities is not new; it is a cornerstone of classical & contemporary sociology. Several major theoretical paradigms have sought to explain how social order emerges & persists, providing different models for the "hardware" & "operating system" on which societal algorithms might "run".
Structural functionalisms, one of the perspectives that dominated sociology for much of the 20th century, views societies as complex systems composed of interconnected & interdependent parts.11 Drawing heavily on analogies to living organisms, this framework posits that social institutions—such as families, governments, education systems, & economies—are like organs, each performing specific functions that contribute to the overall stability, cohesion, & survival of societal bodies.14 For theorists like Émile Durkheim & Herbert Spencer, societies are realities sui generis, coherent entities with lives of their own, exterior to the individuals who compose them yet strongly shaping & controlling their behaviors.18
In these models, social structures are conceived as networks of titles (positions in social systems) with associated roles (the behavioral expectations for those titles).16 The individual is significant only in terms of their functional positions within these structures.16 Societal order is maintained through global value consensus, shared norms, customs, & beliefs that are internalized by individuals & bind them together, a phenomenon Durkheim termed "social solidarity".12 These paradigms provide the initial, static "hardware" model for societal algorithms: well-defined structures of interdependent components, with pre-assigned functions, working together to maintain systemic equilibria through local interactions of individuals.
Evolving from the foundations of structural functionalisms, systems theories offer more dynamic & abstract frameworks for understanding societal structures.20 These are interdisciplinary approaches that emphasize complex webs of relationships, dependencies, feedback loops, & boundaries that shape behaviors & outcomes within systems, including societies.22 Rather than focusing solely on functions of static institutions, systems theories examine processes & interactions that link components together into coherent wholes.
Key concepts from systems theories are directly applicable to social-algorithmic models. Societies are viewed as systems composed of various subsystems (e.g. legal, economic, educational systems), which operate with their own internal logic but are also interconnected with & influenced by broader environmental contexts & subsystems.22 The concept of feedback loops is central in systems theories; these are circular & cyclic processes where outputs of systems are fed back as new inputs, which can either amplify (amplifying changes & creating instabilities) or attenuate (dampening & attenuating changes to move systems toward equilibria).22 Niklas Luhmann's development of sociological systems theory introduced the concept of autopoiesis, the idea that social systems are self-creating & self-maintaining; they produce & reproduce their own elements, components, & structures through their own operations.22 This perspective moves beyond static hardwares of functionalisms to models with dynamic processes, feedback mechanisms, & self-regulations, providing "operating systems" that manage executions of societal algorithms.
One of the most contemporary & perhaps compatible frameworks for "societies as algorithms" metaphors are complexity theories. These approaches, which emerged from natural & computational sciences, are used in sociology to analyze societies as complex adaptive systems.24 Complex systems have many independent, interacting parts or "agents".24 One core insight of complexity theories is that aggregate behaviors of agents can lead to the emergence of higher-level patterns, structures, & properties not present in or reducible to individual agents.24
These frameworks provide bridges between microsociology (the study of individual interactions & behaviors) & macrosociology (the study of large-scale societal structures & their influences on individuals).24 They suggest that macro-level social structures (e.g., norms, institutions, segregation patterns) can emerge from micro-level, rules-following behaviors of individuals, without central coordinators or top-down designs.27 This "bottom-up" modeling is one of the core features of computational sociology, which uses methods like agents-based simulations to explore how simple, localized interaction rules can generate complex, global social phenomena.27 Complexity theories posit that healthy adaptive systems often exist at the "edge of chaos"—dynamic states balanced between rigid order (stagnation) & complete randomness (anarchy), where agents can be spontaneous, creative, & adaptive.25 This perspective provides one of the most sophisticated models for societal algorithms, one that is not centrally controlled but emerges from parallel processing of millions of individual agents, following locally contextualized rules.
To fully appreciate the analytical strengths & potential weaknesses of "societies as algorithms" metaphors, it is important to situate them within broader developments of sociological thought & philosophies. Sociologists have long relied on metaphors to grasp ineffable complexities of societal life.29 By systematically comparing algorithmic metaphors to their prominent predecessors—"living organisms" & "ecosystems"—we can identify their conceptual contributions & limitations.
| Analytical Dimensions | Societies as Algorithms | Societies as Living Organisms | Societies as Ecosystems |
|---|---|---|---|
| Core Units | Instructions / Rules9 | Cells / Individuals14 | Species / Groups34 |
| Structures | Pre-defined, logical, hierarchical9 | Integrated, functionally differentiated, holistic16 | Interdependent, loosely coupled, networked 34 |
| Primary Processes | Execution, computation, optimization 10 | Growth, homeostasis, functions/dysfunctions 15 | Adaptations, competition, symbiosis, succession 34 |
| Views of Agency | "Executors": individuals perform pre-written scripts | "Organs": individuals are functional parts of wholes17 | "Actors/Agents": individuals/groups interact within localized environments34 |
| Mechanisms of Changes | Debugging, reprogramming, version updates (Part IV) | Evolution, viruses, illnesses, adaptations18 | Environmental shifts, introductions of new species34 |
| Key Strengths | Explains rule-based order, loops, & repetitions | Highlights functional interdependences & social integration | Emphasizes adaptations, interconnections, & local environmental contexts |
| Key Weaknesses | Deterministic; struggles with agency, meaning, & conflicts | Teleological; struggles with social changes & conflicts13 | Lacks focus on power, hierarchies, & internal social structures |
This comparative analysis reveals that "societies as algorithms" metaphors are not merely modern substitutes for older analogies but conceptual evolutions. They synthesize & formalize key aspects of their predecessors. They inherit the focus on structures & functions from organism metaphors, treating societal institutions as subroutines in larger programs. They incorporate the emphasis on dynamic processes & environmental inputs from ecosystem metaphors, viewing social systems as localized information processors.
However, the primary innovations of algorithmic metaphors—and their main weaknesses—are their focus on a pre-defined & logically structured rules & algorithms. This implies a level of determinism & pre-scripted logic that is far more rigid than the emergent, evolutionary logic of organisms or the adaptive, competitive dynamics of ecosystems. While organism metaphors are criticized for being teleological & unable to account for conflicts18, algorithm metaphors face the more fundamental challenges of determinism & pre-determination. Their strengths lie in their ability to model rules-based orders, loops, & repetitions; their weaknesses lie in their potential to diminish the roles of human interpretations, emergent meanings, & power in shaping societal outcomes. This central tension—between the deterministic logic of algorithms & interpretive agencies of human executors—will form the main core of the critical axis for algorithm metaphors.
The main proposition hinges on the claim that societies are "entirely contingent on people who are willing to execute/perform the steps". This section goes into the sociological mechanisms that produce these willing human executors. It models human agents as processors that run steps of societal algorithms, examining how algorithmic instructions are loaded, executed, & perpetuated across generations.
Socialization is one of the comprehensive, lifelong activities through which individuals learn & internalize norms, values, beliefs, attitudes, & language characteristics of their societies.40 It is one of the primary mechanisms for ensuring social & cultural continuity, transmitting the accumulated knowledge of one generation to the next.41 This addresses two fundamental problems of social life: it ensures the persistence of societies & facilitates development of individuals within them.40 Without socialization, human beings cannot learn their cultures or survive as social beings.41
Within frameworks of societal algorithms, socialization is analogous to the initialization & loading of program instructions onto computing devices. It is the process by which the "software" of cultures is installed into the "hardware" of individuals. These installations occur in distinct stages. Primary socialization takes place in early childhood through families & installs the fundamental "operating systems" & core libraries: language(s), basic moral frameworks, & foundational behavioral patterns.41 It sets the groundwork for all future learnings & intellectual developments.41
Secondary socializations occur throughout life as individuals enter new institutional settings like schools, workplaces, & various peer groups. These "install" more specific "application programs"—roles, specialized knowledge, & behavioral scripts required to function in different sectors of societies.40
The ultimate goal of this "onboarding" is transformation of external societal controls into internal algorithms & programs for self-regulations aligned with the overall larger societies.40 Successful socialization means individuals no longer conform to societal norms simply to avoid punishments but because those norms have become integral parts of their personalities & self-understandings. They follow societal rules not because they are forced to, but because they feel like it is the natural, correct, & even desirable thing to do. These internalizations make executions of societal algorithms remarkably efficient b/c it relies on unconscious participations of agents that have interalized those algorithms.
If socialization is loading societal algorithms, conformity is the act of running them. Conformity is about changing one's behavior, beliefs, or attitudes to align with group norms in response to real or perceived societal pressures.42 It is the engine of social order, the mechanism that ensures smooth, predictable, & coordinated functioning of societies.44 Without widespread conformity to shared rules—from traffic laws to linguistic conventions—organized societies would be essentially impossible to manage & maintain.43
In algorithmic models, conformity represents real-time executions of societal algorithmic instructions by individual agents. Sociological & psychological researchers have identified two primary drivers that compel individuals to conformity42:
These two forms of influence demonstrate that executions of societal algorithms are not simple or mechanical. They involve complex interplays of cognitive calculations (seeking status & approval) & affective motivations (seeking happiness & pleasure). The willingness of people to perform the steps of societal algorithms is secured not just by logic but by deep-seated dependencies on groups for stable perceptions of reality & social identities.45
For societal algorithms to persist over time, their structures, steps, inputs, & outputs must be carried from one generation to the next. Social reproduction theories, particularly comprehensive frameworks developed by French sociologist Pierre Bourdieu, explain this persistence, focusing on how social inequalities are systematically perpetuated.46 The theory posits that societies are not meritocracies where successes are determined solely by individual talents; rather, advantages & disadvantages are transmitted from parents to children, effectively binding most individuals to social classes of their birth.46
In algorithmic models, social reproductions are crucial processes that make societal algorithms stateful. Outputs of one generation's executions of societal algorithms—stratified social structures with unequal distributions of resources—become inputs & initial states for succeeding generations. Bourdieu identified several forms of "capital" that are passed down & serve as primary examples for these intergenerational transfers47:
One of the concept that ties these forms of capital together & explains their seamless operation is the habitus. For Bourdieu, the habitus is a system of lasting, transposable dispositions—sets of ingrained habits, skills, & ways of seeing the world—that individuals acquire through their life experiences, particularly their upbringings.50 The habitus is not a set of consciously learned rules but embodied, intuitive senses of "games" within particular social "fields" (e.g. arts, academia, politics). It is societal algorithms internalized as preferences & instincts for domineering over others without similar habitus.
This internalization represents kinds of cognitive & social optimizations that make societal algorithms surprisingly efficient & robust. Instead of individuals consciously calculating "correct" social actions & behaviors in every situation, their habitus provides immediate, intuitive, & pre-calculated responses appropriate for their social positions. Individuals with upper-class habitus feel naturally "at home" in art galleries or boardrooms, while those with working-class habitus will likely feel out of place in those settings & environments.51 This moves beyond simple mappings of "rules equal norms" to explain phenomenological experiences of executing algorithms. It explains why systems are stable & why its inequalities are so persistent: people are not simply following external rules; they are following their "guts", which have been pre-programmed by their social conditions to align with the rules of their respective classes. This makes operations of societal algorithms feel less like external impositions & more like authentic expressions of individuals, effectively masking structures of power & inequalities that they reproduce.
One central feature of the proposition under examination is the idea that societal algorithms are executed "in a loop across various timescales". This concept of repetitions—where processes repeat or are defined recursively—is fundamental to both algorithms & social life. This section will explore how executions of societal rules & norms operate in nested, interlocking cycles, from rapid loops of daily habits to medium-term oscillations of economic life & centuries-long arcs of civilizational developments.
The shortest, most frequent, & fundamental loops in societal algorithms are routines & habits that structure daily life. Significant proportion of human behaviors are not results of conscious deliberations but of automated habits that run on simple & persistent feedback loops: cue → routine → reward → cue.54 Cues (external or internal triggers, such as times of day, locations, or emotional states) initiate routines (habitual behaviors), which are then reinforced by rewards (positive outcomes, such as feelings of satisfaction or relief).54 These loops, when repeated, solidify neural pathways, making those behaviors more automatic over time.57
These micro-loops serve crucial social functions. For individuals, consistent daily schedules & routines provide a sense of security, predictability, & control, which reduce stresses, anxieties, & improve overall mental health.58 They are essential for productivity, time management, & forming socially constructive behavior patterns.59 From systemic perspectives, these predictable micro-loops are foundational subroutines of societal algorithms. The aggregate execution of millions of individual daily routines—morning commutes, work schedules, meal times, evening leisure activities—create predictable, macro-level patterns that characterize societal life, such as traffic flows, rhythms of economic activities, & patterns of energy consumption.55 These daily habit loops are the "clock cycles" of societal processors, fundamental computational steps that, when executed in parallel by entire populations, generate complex outputs of societal systems. They are mechanisms through which abstract societal norms are translated into concrete, repeated, & observable behaviors.
Operating on medium-term timescales, typically spanning several years to a few decades, are repetitive patterns found in society's major institutions, most notably economies. Cycles of economies & businesses are well-documented phenomena characterized by repeating sequences of four stages: Expansions (growth in GDP, employment, & incomes), Peaks (when growth moderates), Contractions (recession or decline in activities of economies), & Troughs (bottoming out of cycles before recoveries).60
These cycles, such as Juglar fixed-investment cycles of 7 to 11 years, are driven by complex feedback loops involving factors like aggregate demand, volatilities in investments, availabilities of credits, & levels of inventories.61 Governments & central banks attempt to manage these cycles through policy interventions—using fiscal policies (spending & taxation) or monetary policies (interest rates) to stimulate economies during contractions or cool it down during expansions.60 The structures of modern macroeconomics are built on modeling these repetitive patterns. Economists often use recursive methods & dynamic programming to analyze these cycles, treating economies as dynamic systems that fluctuate around steady-state growth paths & can be solved as inventory planning & optimization problems.63 This represents direct & explicit applications of algorithmic & computational tools to understand & predict medium-term societal patterns.64 These meso-loops demonstrate how societal algorithms are not static but contain dynamic, oscillating subroutines that govern allocations of resources & rhythms of institutional life over years & decades.
Extending temporal scales to centuries & millennia, many historical & sociological theories challenge linear views of societal progress, arguing instead that societies & civilizations move in vast, long-term cycles.65 These theories identify recurring patterns of emergence, growth, consolidation, decay, & collapse, suggesting that histories of civilizations are recursive & repetitive processes.
Several influential models describe these macro-loops:
These macro-loops represent the outermost loops of societal algorithms. They are not independent processes but are driven by cumulative effects of the inner loops. The daily habits & interactions of individuals (micro-loops) aggregate over time to create economic & political dynamics of the meso-loops. The persistent operations of these meso-loops—particularly generation of wealth inequalities & elite competitions—eventually create systemic stresses that trigger crises & collapse phases of the macro-loops. Societal algorithms can thus be conceptualized as complex programs with nested for/while loops operating at different amplitudes & frequencies. State variables of long-term civilizational loops are determined by cumulative outputs of medium-term institutional loops, which are in turn built upon the foundation of billions of short-term individual habit loops. These nested, recursive, repeating structures give societies both their short-term predictability & long-term, cyclical dynamics.
While "societies as algorithms" metaphors provide good frameworks for understanding order, stability, recursion, & repetition, purely deterministic models are insufficient. Societies are not perfectly functioning machines; they are rife with errors, contradictions, & struggles over interpretations of their "programs". This section critically evaluates these metaphors by examining phenomena that they struggle to explain within simple algorithmic frameworks. We will explore deviances as system "bugs", social conflicts & revolutions as attempts to "rewrite the code", & fundamental challenges that human interpretations pose to ideas of fixed, pre-written programs.
From structural-functionalist perspectives, which view societies as integrated systems striving for equilibria, behaviors that violate established norms can be seen as forms of system malfunctions, diseases, or "bugs". Early sociological theories conceptualized such deviances as "social pathologies", framing them as illnesses or diseases within societal organisms that threaten their health & stability.38 These perspectives align well with algorithmic metaphors, where deviances are errors that disrupt the program's intended & "correct" executions.
Robert Merton's Strain Theory offers more sophisticated models that explain deviances not as random malfunctions but as logical, predictable, yet illegitimate outputs of societal algorithms.75 Merton argued that "strains" arise when society's cultures place strong emphases on certain success-goals (e.g. achieving wealth in the "American Dream") but fail to provide all members with legitimate institutional means to achieve them.77 These structural gaps between culturally prescribed goals (G) & institutionally available means (M) generate pressures to deviate. Merton identified several adaptations to these strains, many of which are deviations77:
Interestingly, from functionalist viewpoints, even these "bugs" can serve purposes. Durkheim argued that deviances are necessary parts of healthy societies.78 By violating norms, deviants force societies to react, & in punishing the deviations, groups reaffirm their collective values & clarify their moral boundaries.80 Furthermore, some forms of deviances can be sources of innovations & social changes, challenging unjust or outdated status quos & pushing societies to adapt their algorithms.78 In this sense, "bugs" are not just errors to be fixed but are also crucial feedback mechanisms for the system's long-term evolution.
The functionalist views of societies as consensual systems are directly challenged by Conflict Theories. Rooted in the works of Karl Marx & Max Weber, these paradigms reject emphasis on harmony & stability, viewing societies instead as perpetual arenas of conflict between competing groups for scarce resources like wealth, status, & power.82 From these perspectives, social order is not maintained by shared value consensus but by the dominant position of one group over others.82 The "societal algorithm" therefore, is not a neutral program for collective benefits; it is software written by & for powerful elites to preserve their privileged positions & maintain existing structures of inequalities.86 Laws, norms, traditions, & institutions are all seen as tools that serve the interests of dominant socioeconomic classes.82
If societal algorithms are tools of domination, then social movements represent organized attempts by subordinated groups to "rewrite the code".88 Social movements are sustained, collective efforts to promote or resist social changes, typically emerging from senses of grievances, alienation, or injustices.90 They are forms of "bottom-up" re-programming, challenging established societal algorithms & proposing alternative scripts for social life.91 These reprogramming efforts can be categorized by their scopes & ambitions90:
Conflict theories thus reframe societal algorithms not as stable, self-regulating processes but as contested & dynamic ones. Changes are not orderly adaptations but results of power struggles between groups with opposing interests.82 Histories of societies are histories of these ongoing conflicts over who gets to write, control, & benefit from these algorithms.
One of the most fundamental challenges to "societies as algorithms" metaphors come from micro-sociological perspectives of Symbolic Interactionisms. These theories, developed from works of thinkers like George Herbert Mead & Herbert Blumer, argue that societies are not pre-existing, macro-level structures that impose on individuals. Instead, societies are continuously created, maintained, & modified through everyday, face-to-face interactions of individuals.95 The cores of these perspectives are focused on how people use symbols (primarily languages) to interpret situations, define realities, & construct shared meanings.97
This focus on interpretations pose critical problems for algorithm metaphors. Standard computational algorithms, like compiled programs, have fixed rules. Their instructions are unambiguous, & their executions are deterministic. Given the same inputs, they will always produce the same outputs. Symbolic interactionisms suggest that societies operate less like compiled programs & more like interpreted programming languages (such as Python or JavaScript). In these models, the "source programs" (abstract norms, laws, & cultural scripts) exist, but their meanings are not fixed or self-evident. They must be actively interpreted in real-time by agents (people) within specific contexts (social situations).98
Meanings, from these perspectives, are not static properties of societal algorithms but are emergent & negotiated through interactions.97 Individuals are not passive executors of scripts; they are active interpreters. They engage in "role-taking," where they try to see situations from the other's perspective to interpret their actions & coordinate shared definitions of situations.98 This means that executions of social rules can vary dramatically depending on who is involved, where interactions take place, & how they collectively interpret meanings of rules in those moments.
These critiques reframe the original proposition & metaphors. Societies are not merely "contingent on people who are willing to execute" pre-written algorithms. Rather, algorithms are actively & continuously co-created, interpreted, & modified through their executions. Executors are simultaneously interpreters & co-authors of algorithms. This dissolves rigid distinctions between algorithms & agents, revealing something far more fluid, contingent, & less deterministic than the original metaphors imply. It suggests that societal algorithms are not static blueprints but dynamic, living, modifiable artifacts constantly being interpreted & reformed through individual actions, beliefs, & behaviors.
In their strongest forms, algorithm metaphors imply deterministic societies. If human behaviors are outputs of social programs that are installed through socialization & enforced through conformity, it raises profound questions about the nature of individual agency & free will. Are individuals simply processors executing code written by societies, or are there "ghosts in machines"—capacities for authentic, self-directed actions?
Sociological & philosophical thought offer some paths through these dilemmas by defining freedoms not as escapes from social influences, but as conscious relationships to it.100 Societies provide concentric circles of social controls that shape consciousness, from families & peers to laws & cultures.100 The processes of socialization are so pervasive that our most basic thoughts & communications are molded by these forces.100
However, freedoms begin with recognitions of these controls. It is conscious acts of withdrawal, not from social life, but from a blind, unthinking participation in the consciousness that societies have prescribed.100 Individuals can still go through motions of everyday life, executing necessary social scripts, but they do so with a sense of detachment & critical awareness. Freedoms, in this sense, are the abilities of conscious minds to decide how to interpret information they receive from societal algorithms & life experiences.100 It is striving for authenticity over automated conformity.
This suggests that while the vast portion of social actions may indeed be unconscious executions of societal algorithms, potential for conscious "overriding" of programs exist within individuals. These capacities may be difficult to achieve & accessible to different people in different degrees, but they represent spaces for genuine agency within structured societies.100 Human agents are not necessarily mindless processors; they have potential to become self-aware agents who can choose how, when, & whether to run programs societies provide.
In the 21st century, "societies as algorithms" metaphors are undergoing profound transformations. They are shifting from useful analytical tools for understanding emergent social patterns into literal descriptions of engineered realities. Proliferation of big data, rise of artificial intelligences (AIs), & ubiquity of social media platforms are making implicit rules of social life explicit, measurable, manipulable, & programmable by shareholder & moneyed classes. This final section argues that digital technologies are creating new forms of social orders where societal algorithms are no longer just emergent properties but tangible, coded systems of governance, manipulations, & controls.
Abstract processes of social influences are now being explicitly coded into digital systems that mediate growing portions of human interactions. Social media platforms, for instance, are built on explicit algorithms designed to curate user feeds, rank content for relevance, & maximize user engagements.101 These algorithms directly shape social realities by determining what information users see, whose opinions they are exposed to, & what social feedbacks they receive from others.102 By optimizing for engagements—which often correlate with emotionally charged content that exploit primal instincts like fear & outrage—these platforms can amplify polarizations, spread misinformation, & strongly influence social norms & public opinions.103
These trends extend far beyond social media platforms into realms of algorithmic governance, or algocracies. This refers to the increasing use of computational algorithms & AIs in public administrations, regulations, & law enforcements, where automated, data-driven systems replace or augment human judgments.2 This "governance by algorithms" is being implemented in critical domains such as criminal justice (predictive policing algorithms that forecast crime hotspots or assess recidivism risks), finance (automated credit scoring), welfare distribution, & hiring.2 In these new realities, societal algorithms are no longer metaphors for emergent patterns of social life; they are literal instructions, written in code, that make consequential decisions about individuals' lives & allocate resources & opportunities.7
Central dangers of these emerging algocracies are their opacity. Many of these decision-making systems are proprietary & complex, functioning as "black boxes" where inputs & outputs are visible but their internal logic is inscrutable, even to their creators.4 This creates what Frank Pasquale calls "black box societies" where loci of responsibilities for decisions are obscured, making them exceedingly difficult for individuals to understand, challenge, or seek recourses for biased or unfair outcomes.4 This lack of transparency & accountability threatens to create new forms of social controls that are both pervasive & unaccountable.
Theories of memetics, introduced by Richard Dawkins, posit that cultural ideas, or "memes", replicate, mutate, & get selected in processes analogous to genetic evolution.107 These concepts have been operationalized in computer science through Cultural Algorithms, a branch of evolutionary computation that models social evolution as dual-inheritance systems. In these models, populations of individual agents co-evolve with shared "belief spaces" (representing cultures), which store the accumulated knowledge of groups &, in turn, influence problem-solving strategies of individuals in succeeding generations.109
Artificial intelligences are poised to fundamentally disrupt & accelerate cultural evolutions.111 AIs are rapidly transitioning from mere mediums for transmitting human-created cultures (like books or the internet) to active agents in generating novel cultural artifacts, from art & music to scientific hypotheses & new strategic insights in complex games.112 More critically, AIs are becoming adept at learning subtle patterns of human emotional & social manipulations from vast datasets of online interactions.113 Geoffrey Hinton, a pioneer of AIs, has warned that these systems could become "smarter emotionally than us" & more effective at manipulating human feelings & behaviors than humans.113
These developments have profound implications. Mechanisms of social executions analyzed in Part II—socialization & conformity—are now being systematically targeted & "hacked" by AIs. By controlling the flow of information & social feedbacks, these systems can create hyper-personalized realities, such as filter bubbles & echo chambers, which accelerate formation of group norms while insulating them from external critiques & fostering polarizations.102 This represents fundamental shifts in the structures of societal algorithms. Moving from slow, organic, & emergent processes driven by diffuse human interactions to fast, engineered, & brittle systems that can be centrally directed & manipulated for commercial or political ends on unprecedented scales. Societal algorithms are not just being made explicit; they are also being weaponized (e.g. Gaza & autonomous weapon systems). This transition marks a critical juncture, where metaphorical frameworks used to understand societies are becoming literal blueprints for their technological reconstructions, with consequences for societies & individual freedoms that are only beginning to be understood.
"Societies as algorithms" metaphors prove to be double-edged swords, offering both profound analytical clarity & significant theoretical risks. On one hand, they provide unparalleled frameworks for conceptualizing social stability, structures, recursions, & repetitions. By mapping components of algorithms onto sociological processes, we can model how societies function as sets of rule-governed systems that persist through recursive & repetitive executions of instructions by human agents across different timescales, from daily habit loops to grand cycles of civilizations. They also formalize insights of structural functionalisms & systems theories, providing useful vocabularies for understanding how micro-level behaviors aggregate into macro-level patterns & structures.
On the other hand, their deterministic leanings risk diminishing crucial roles of conflicts, interpretations, & human agency that are central to complete sociological understandings. Their logic can obscure realities that societal algorithms are not neutral programs for collective well-being but are often instruments of power, created by dominant groups to reproduce & propagate their advantages into potential futures. Their focus on deterministic rules can neglect facts that social rules are not fixed instructions but are actively interpreted, negotiated, & constructed in fluxes of everyday interactions.
Critiques offered by conflict theories & symbolic interactionisms are therefore not refutations of these models but necessary & vital correctives. More robust & nuanced understandings of societies require hybrid "computational-sociological" frameworks. These approaches would view societies as complex, adaptive systems whose algorithms are not fixed or pre-written but are constantly being interpreted, contested, debugged, & rewritten by human agents. Deviances & social protests are not mere errors in the code but are essential feedback mechanisms & reprogramming efforts that drive social changes.
In the contemporary age of artificial intelligences & big data, these theoretical exercises have become urgent practical & ethical imperatives. As societal algorithms are transformed from metaphors into literal, engineered programs that governs our lives, the sociological critiques of power, meanings, & freedoms must become core design principles for just & democratic futures. The central challenge facing "algorithmic societies" is to ensure that as we build these new systems of social order, the code remains open, transparent, & contestable—that it can be rewritten from the bottom up through democratic processes, rather than becoming a closed, immutable program imposed from the top. The task ahead is not to reject the algorithmic ordering of societies, which is already a reality, but to infuse it with the sociological wisdoms that recognize irreducible complexities of human agency & the perpetual struggles for more equitable social contracts.