CHAPTER 4

Computer Simulations and Creative Theory

Decades ago, several computationally sophisticated social theorists believed that computer simulations, many of which employed artificial intelligence principles, could be used to test and improve the rigor of theories in the social sciences. Examples include books like World Dynamics (1971) by Jay Forrester, Micromotives and Macrobehavior (1978) by Thomas Schelling, and The Evolution of Cooperation (1984) by Robert Axelrod.1 Although a European journal exists in this area, The Journal of Artificial Societies and Social Simulation, and other journals sometimes publish social simulation studies, this methodology has not become a common tool for theorizing in sociology, political science, and anthropology, despite some popularity in economics. Emergence of a coherent yet creative cultural science will require tools for theorizing as well as for analyzing data. This chapter will consider the options in two ways: (1) consideration of how suitable many classic sociological theories are for computer simulation and (2) exploration of how an intellectually sophisticated commercial computer game can represent multiple ideologies.

Simulation of Religious Culture

The term artificial intelligence has many meanings. Currently, much of the research excitement concerns forms of machine learning, including deep neural networks in which a complex pattern of real-number weights on the connections between nodes in the memory structure adjusts the network’s output on the basis of the input of complex data. But very different definitions include rule-based reasoning, in which a structure of predefined formal relationships allows abstract analysis of different situations. I have found both approaches to be of value in the development of sociological theories and in testing their logical coherence and predictive rigor, in addition of course to empirical tests of the extent to which a theory describes the real world. Here we summarize a compatible set of social theories that were originally framed as sequential lists of formal rules, thus very suitable for rule-based simulations. However, we begin by noting that different AI methods can be combined, just as an individual’s mental processes are a fundamental component of social influence.

Social theory development by means of neural nets does not necessarily mean getting accurate predictions, because the goal is to model human thinking which often incorporates errors or imaginative additions to reality, as in the case of literature and drama. In a 1995 issue of The Journal of Mathematical Sociology, I presented results from a program I jokingly called MIND: Minimum Intelligent Neural Device, showing how an intentionally limited neural net could model human ethnic prejudice:

A simple but effective neural network algorithm illustrates common principles of this new class of computational tools. Designed for use in a range of simulation studies, this minimum intelligent neural device is capable of learning which of a complex set of stimuli to avoid, and large numbers of these devices can be assembled in programs to explore the development of prejudice and of various interaction strategies. Neural nets are error-reduction algorithms with the potential to perform a wide range of useful tasks, including modeling theories of the social consequences of human error.2

This somewhat defective MIND was based on Gordon Allport’s theory of cognitive effort that avoiding prejudice requires mental work. In this simulation, misbehavior by a few people we interact with need not force us to avoid all members of their group, but to make more careful distinctions among them.3 The neural net could solve the problems presented to it, but it was specifically designed to do so only with difficulty.

Beginning with journal articles published in 1979 and 1980, Rodney Stark and I had been developing a general sociocognitive theory of religion that concerned both how individual human minds conceptualized the supernatural and how people interacted to create religious movements and organizations. In the 1960s, Stark had worked with senior sociologist of religion, Charles Glock, doing research primarily using questionnaire data. One of their books focused on a specific kind of cultural prejudice, Christian Beliefs and Anti-Semitism.4 In a 1965 article, Stark collaborated with John Lofland to offer a theory of the emergence of religious subcultures, integrating concepts from a well-established theoretical tradition.5 Their theoretical model could be applied to a wide range of subcultures, but was presented as a series of steps an individual must go through in order to convert from conventional religion to a radical cult or sect. Each step could be modeled as a procedure in a computer program, although the authors presented the model as a structure for human cognition. The recruit to a radical religious group must:

1. 1. Experience enduring, acutely felt tensions,
2. 2. Within a religious problem-solving perspective,
3. 3. Which leads him to define himself as a religious seeker;
4. 4. Encountering the group at a turning point in his life
5. 5. Wherein an affective bond is formed (or pre-exists) with one or more converts;
6. 6. Where extra-cult attachments are absent or neutralized;
7. 7. And where, if he is to become a deployable agent, he is exposed to intensive interaction.

The first step in the Lofland–Stark theory came from the strain theory of Robert Merton and Neil Smelser that people will deviate from standard norms and join or create new cultural movements only when they are dissatisfied with their current life situation.6 A neural net can represent “acutely felt tensions,” by having a suboptimal model of external contingencies, such that the variable measuring success is declining or negative. Step 2 in the theory observes that people may function within a broader cultural context, for example, conceptualizing social problems in religious, political, or psychiatric terms. An especially sophisticated program could represent these alternatives in a neural net, allowing a simulated individual to abandon one problem-solving perspective and learn another, through social interaction, although that was not a feature of the Lofland–Stark model. Steps 5 through 7 of the model are a simplification of one version of differential association theory, developed by Edwin Sutherland. It could be applied to many kinds of human behavior, perhaps all kinds that could be described as culture, but was aimed at criminology. As summarized by Sutherland in 1947, it described a series of symbolic interaction steps that would result in commission of crimes:7

1. 1. Criminal behavior is learned from other individuals.
2. 2. Criminal behavior is learned in interaction with other persons in a process of communication.
3. 3. The principal part of the learning of criminal behavior occurs within intimate personal groups.
4. 4. When criminal behavior is learned, the learning includes (a) techniques of committing the crime, which are sometimes very complicated, sometimes simple; (b) the specific direction of motives, drives, rationalizations, and attitudes.
5. 5. The specific direction of motives and drives is learned from definitions of the legal codes as favorable or unfavorable.
6. 6. A person becomes delinquent because of an excess of definitions favorable to violation of law over definitions unfavorable to violation of the law.
7. 7. Differential associations may vary in frequency, duration, priority, and intensity.
8. 8. The process of learning criminal behavior by association with criminal and anticriminal patterns involves all of the mechanisms that are involved in any other learning.
9. 9. While criminal behavior is an expression of general needs and values, it is not explained by those needs and values, since noncriminal behavior is an expression of the same needs and values.

This is a marvelously rich theoretical statement, but it has often been interpreted too narrowly to state that people become criminals simply because their friends are already criminals. It is too easy to read associations as friendships or social relations more broadly. However, as clearly explained in a later edition of Sutherland’s criminology textbook, associations can mean mental associations between concepts.8 Successful criminals must have skills required to commit their particular crimes, the techniques mentioned in step 4 of the theory, and those include cognitive abilities the individual can learn from other people. Skills are significant features of most subcultures, such as criminal gangs, classical music orchestras, and the Chicago School of Sociology to which Sutherland belonged. The learning also involves “the specific direction of motives, drives, rationalizations, and attitudes,” each of which can be one of the “definitions” that the subcultures employ to understand and exploit the world.

Thus, while focused on social groups, this theory is highly cognitive in nature and does not actually require the individual to learn through face-to-face interaction. It may be that criminal behavior is so difficult and so dangerous that only strong social influence, perhaps in the context of wider social disorganization, can complete the process of learning. But for many other kinds of subculture, the mode of communication could be reading novels, watching movies, and scanning online social media. We could restate step 6 in terms of any subculture, such as: A person becomes a sci-fi fan because of an excess of definitions favorable to science fiction over definitions unfavorable to science fiction. How would this happen? “Differential associations may vary in frequency, duration, priority, and intensity.” Watching Star Wars in impressionable childhood (priority meaning earlier in life) and happening to see all the other unusually exciting (intensity) sci-fi films of that period (frequency and duration) could explain becoming an avid adult reader of science fiction novels.

When Stark and I collaborated, I naturally added my own theoretical background to his. At that point, I had written my first two books. My doctoral dissertation was a primarily historical study of the social movement that caused governments to invest in developing space rockets, and I still tend to see science and technology as cultures that become social movements in their more dynamic stages.9 My second book was more like cultural anthropology, an ethnographic field study of a radical religious movement that had formed from a pre-existing network of college students and artists. Having the Sutherland and Lofland–Stark models in mind, I described the formation of the group as a social implosion: “In a social implosion, part of an extended social network collapses as social ties within it strengthen and, reciprocally, those to persons outside it weaken. It is a step by step process.”10

Also in mind was the theory of my main teacher in graduate school, George Homans. He was a personal friend of B.F. Skinner, and accepted the fundamental Behaviorist principles in Skinner’s 1938 book The Behavior of Organisms.11 He was also inspired by the 1939 book Frustration and Aggression, by John Dollard, Neal Miller, and their colleagues, that considered the cognitive consequences of failure to achieve one’s desires, but in a manner more compatible with Behaviorism than the Merton-Smelser framework.12 Homans developed his own form of Sociological Behaviorism, beginning with his 1950 book, The Human Group, where he postulated that people who interact rewardingly with each other repeatedly will come both to value each other and to become more similar to each other.13 Note the emphasis on rewards rather than the associations in Sutherland’s theory. In its fully developed form, communicated in the 1974 edition of his book Social Behavior: Its Elementary Forms, Homans began with a small set of axioms he called propositions:14

1. 1. For all actions taken by persons, the more often a particular action of a person is rewarded, the more likely the person is to perform that action.
2. 2. If in the past the occurrence of a particular stimulus, or set of stimuli has been the occasion on which a person’s action has been rewarded, then the more similar the present stimuli are to the past ones, the more likely the person is to perform the action, or some similar action, now.
3. 3. The more valuable to a person is the result of his action, the more likely he is to perform the action.
4. 4. The more often in the recent past a person has received a particular reward, the less valuable any further unit of that reward becomes
   for him.
5. 5A. When a person’s action does not receive the reward he expected, or receives punishment he did not expect, he will be angry; he becomes more likely to perform aggressive behavior, and the results of such behavior become more valuable to him.
6. 5B. When a person’s action receives reward he expected, especially a greater reward than he expected, or does not receive punishment he expected, he will be pleased; he becomes more likely to perform approving behavior, and the results of such behavior become more valuable to him.

The first proposition merely states the starting point for Skinnerian Behaviorist theory. The second proposition twice uses the word similar, which refers to a cognitive judgment by a human mind, and provides a starting point for a theory of art, although I do not recall Homans developing this line of analysis. A crude example is that people learn to seek fruit to eat, because it is tasty and satisfies hunger, so they may come to like pictures of fruit as well, despite their lack of nutritional value. A more complex example may explain the popularity of computer games. In real life, people learn they must compete against each other in order to gain a variety of material rewards and also the instrumental reward of social status that helps them gain rewards of many kinds. The games are somewhat realistic, including gaining status with simulated human beings and a bank account of simulated money. Thus popular games tend to be somewhat similar to reality.

The fourth proposition relates to the arts in a different way. When a person frequently succeeds in gaining a particular reward, it becomes less valuable, at least in the short term until the person runs out of it. That liberates the person’s mind to seek some other reward. While rich people may consume more costly food than poor people, they are also able to afford rewards of lower practical value, such as artworks. This option can illustrate itself through a cute pun rather than an ugly proposition: truffles versus trifles. An elite cooking school has even named itself Truffles and Trifles.15 However, all of the arts deserve to be taken seriously, and practically all of them have traditionally been connected to religion, from literature like the bible, to Gregorian chant music, to Salvador Dali’s surreal painting of the crucifixion. Or should that word be crucifiction?

Even without invoking the term sacred, religion is among the very most significant aspects of human culture. Debates about the truth of particular religious beliefs are both socially antagonistic and intellectually profound. For example, Harvard philosopher and psychologist William James argued for a pragmatic definition of truth: “The true is the name of whatever proves itself to be good in the way of belief, and good, too, for definite, assignable reasons.”16 “‘The true’ to put it very briefly, is only the expedient in the way of our thinking, just as ‘the right’ is only the expedient in our way of behaving.”17 If religious beliefs serve valuable functions for human beings, then for that reason they are true. When Stark and I began collaborating in the development of a formal theory of religion, we gave each other some room to hold different views on the truth of religious beliefs, but the resultant theory is consistent with the view that religion is an artform, comprised of metaphors and imagination, rather than being a system either of firm beliefs or statements of fact. We began with a slightly different set of axioms than those proposed by Homans18:

1. 1. Human perception and action take place through time, from the past into the future.
2. 2. Humans seek what they perceive to be rewards and avoid what they believe to be costs.
3. 3. Rewards vary in kind, value, and generality.
4. 4. Human action is directed by a complex but finite information-processing system that functions to identity problems and identify solutions to them.
5. 5. Some desired rewards are limited in supply, including some that simply do not exist.
6. 6. Most rewards sought by humans are destroyed when they are used.
7. 7. Individual and social attributes which determine power are unequally distributed among persons and groups in any society.

In the form presented in our 1987 book, A Theory of Religion, these seven axioms provided the basis for 344 hypotheses we suggested (but did not fully demonstrate) could be derived as theorems from these axioms, within a framework of 104 definitions of concepts. It is a highly cognitive theory, and the last three axioms place the human mind in a world where rewards are limited, thus presenting people with dynamic challenges in their attempt to gain rewards. Because rewards vary in “vary in kind, value, and generality,” so do human plans to gain this diversity of rewards. The original statement of this theory used the term explanations to refer to such plans, although in my computer simulation research the term algorithms seemed appropriate as well. Here is the subset of definitions, originally presented in a chain of theorems, which will not be quoted here, that led to the emergence of religion in human culture:

Def.10: Explanations are statements about how and why rewards may be obtained and costs are incurred.

Def.18: Compensators are postulations of reward according to explanations that are not readily susceptible to unambiguous evaluation.

Def.19: Compensators which substitute for single, specific rewards are called specific compensators.

Def.20: Compensators which substitute for a cluster of many rewards and for rewards of great scope and value are called general compensators.

Def.21: Supernatural refers to forces beyond or outside nature which can suspend, alter, or ignore physical forces.

Def.22: Religion refers to systems of general compensators based on supernatural assumptions.

Def.23: Religious organizations are social enterprises whose primary purpose is to create, maintain, and exchange supernaturally based general compensators.

Def.52: Magic refers to specific compensators that promise to provide desired rewards without regard for evidence concerning the designated means.

Def.57: A sect movement is a deviant religious organization with traditional beliefs and practices.

Def.58: A cult movement is a deviant religious organization with novel beliefs and practices.

Def.67: Cults are social enterprises primarily engaged in the generation and exchange of novel compensators.

In the context of cultural science, “Cult is culture writ small.” The theory is not limited to primitive magic or traditional religion, but extends to all forms of artistic culture. Watching Luke Skywalker wield his light sabre in a Star Wars movie provides vicarious rewards, perhaps a feeling of self-esteem if we identify with him, but does not require belief that his powers are real. In an earlier book about how religion is depicted in online virtual world, I observed:

Religion has always been deeply implicated in the creative arts, but the relationships among them are changing. Perhaps we shall come to see religion merely as an especially solemn artform. Suspension of disbelief is the essence of art, according to Samuel Taylor Coleridge, and electronic games are a new and powerful artform that often depicts religion. Yet we may wonder whether suspension of disbelief is really very different from belief itself.19

In the most complex of my religion simulations, I assembled several similar procedures that learned how to achieve different rewards, with a simple but higher-level procedure that would decide which reward the simulated person wanted at the moment. When rewards were unavailable, the model could become very imprecise, no longer able to distinguish fact (obtaining a reward) from fiction (responding as if a reward had been received).

Simulation of Sociocultural Influences

In several earlier studies of cultural dynamics, using computer simulation methods, I showed how the combination of Sutherland’s differential association theory and an alternative theory offered by Fritz Heider could model the emergence of complex subcultural structures in large populations. In his 1958 book, The Psychology of Interpersonal Relations, Fritz Heider offered what he called balance theory.20 A cultural application of balance theory would be the observation that people tend to harmonize their social ties and their cultural orientations. If Person A is friends with Person B and Person B is committed to Politics X, then Person A may tend to adopt Politics X as well. However, if Person A is already committed to Politics Y, then there is an imbalance. It can be solved by Person A adopting Person B’s politics, Person B adopting Person A’s politics, or the two people breaking off their friendship. My 1987 book and software collection, Sociology Laboratory, included a rule-based simulation combining Sutherland and Heider, but also a far more cognitive neural net simulation charting the social discovery and sharing of religious compensators.21 Here I will rerun a much more recent version using the multiagent rule-based software I developed for my 2006 book, God from the Machine: Artificial Intelligence Models of Religious Cognition, focusing on the Sutherland–Heider theory.22

Cyburg is a town of exactly 44,100, living in one-person homes arranged as a square 210 on a side. Obviously, this is an abstraction, and the number 44,100 was selected not only because it was a good size for simulations of this kind, but also a perfect square evenly divisible by all integers, 1 through 7, plus 9, 10, 12, 14, 15, 20, 25, 50, and 100. Each person (except those unfortunate enough to have a home on the edge of town) had eight neighbors, with each of whom a friend relationship could exist. The software offered a graphics display of little colored squares, with lines representing the friendships, and as the computer ran the colors could change and lines could appear and disappear. The colors gave a visual impression of the competing subcultures, but after each run in which each of the 44,100 has a chance to act, a table of data could be opened giving the current census of group membership.

The software gave the opportunity to make many choices for a given run of the simulation, allowing subsequent comparison of results from different preconditions. For the one run reported here, in Table 4.1, the first decision was that there should be five approximately equal subcultures in Cyburg—call them religious cults—so the simulation began by giving each resident in turn a one-fifth chance of starting in each subcult. The Sutherland and Heider theories were given equal probability of being applied by each simulated person on each turn. Under Sutherland’s differential association theory, the simulated person would keep or adopt the majority cult membership among friends. The Heider theory was applied just to the friendships, adding ties to members of the same cult, and removing ties to members of other cults. On a modern desktop computer the 44,100 actions of one turn took a few seconds, and 50 turns were run.

Table 4.1 Simulation of competing cults with different outreach cultures

Turn	1.0 Cult	0.8 Cult	0.6 Cult	0.4 Cult	0.2 Cult	Density
0	19.85%	20.12%	20.17%	19.80%	20.05%	50.07%
1	20.28%	20.51%	20.30%	19.42%	19.49%	63.56%
2	21.39%	20.85%	20.12%	19.06%	18.58%	70.36%
3	22.08%	21.42%	20.13%	18.71%	17.66%	74.67%
4	22.96%	21.87%	20.13%	18.24%	16.79%	78.17%
5	23.87%	22.27%	19.90%	17.90%	16.07%	80.45%
6	24.52%	22.58%	19.79%	17.59%	15.52%	82.55%
7	25.21%	22.87%	19.72%	17.30%	14.91%	84.16%
8	25.83%	23.05%	19.65%	16.97%	14.50%	85.30%
9	26.37%	23.21%	19.55%	16.71%	14.16%	86.22%
10	26.75%	23.37%	19.44%	16.56%	13.88%	87.06%
15	28.21%	23.83%	19.24%	15.89%	12.83%	89.35%
20	29.03%	24.18%	19.07%	15.51%	12.21%	90.68%
25	29.45%	24.41%	19.06%	15.29%	11.79%	91.38%
30	29.79%	24.58%	19.00%	15.13%	11.50%	91.65%
35	30.02%	24.73%	19.01%	15.03%	11.22%	91.98%
40	30.21%	24.83%	18.95%	14.95%	11.05%	92.15%
45	30.29%	24.90%	18.96%	14.95%	10.90%	92.27%
50	30.34%	24.95%	18.97%	14.94%	10.79%	92.50%

Why was the first cult able to grow from 19.85 percent of the population up to 30.34 percent? Because the cultures of the cults were set differently at the beginning of this simulation run. An additional opportunity was given to each simulated person to develop a friendship with a neighbor, regardless of that neighbor’s cult membership. That was the outreach and recruitment feature of the Lofland–Stark model, in which members of a cult became deployable agents for it. Members of the most successful cult would always engage in outreach; thus it is called the 1.0 Cult. In the 0.8 Cult, members had just an 80 percent chance of reaching out on a given turn. The cult that shank most in members, from 20.05 percent of the population to 10.79 percent, was the 0.2 Cult with just a 20 percent outreach chance.

This not only illustrates how additional rules can be added to a rule-based simulation, but also how they can be used to analyze the effect of the existing rules. I ran the simulation twice more, once with Sutherland’s theory operating all the time, and Heider’s not at all, and the other time with only Heider’s in effect. The results were similar to each other in that the populations of the groups changed hardly at all, in contrast to Table 4.1. Thus, the Sutherland and Heider factors were both necessary to model competition between subcultures, and outreach was a factor that would increase their differential effect. An important secondary effect was that the density of the social network—the fraction of possible social binds that actually exist—increased from 50.07 percent to 92.50 percent, a strengthening of social structure that would have many consequences had it been experienced by real people.

Simulation of Societal Ideologies

As illustrated by Neverwinter in Chapter 2, massively multiplayer online games are art forms that converge movies and literature with artificial social intelligence. Occasionally, their designers incorporate sophisticated social theories, and here the best example is Fallen Earth, which depicts the emergence of a collection of competing utopian communities, in the wake of a devastating world war. A reviewer of the manuscript of this book offered a better summary of this section than I had written: “The complex cultures produced in some MMOs allow researchers to play through possible social relations. That playthrough might help them conceptualize their conventional problems differently and might help them rethink how group affiliation works.”

To provide the cultural background, Table 4.2 begins by listing four rather famous MMOs that emphasized living in an imaginary environment based on rich ideas, but that no longer exist: Star Wars Galaxies (2003–11), City of Heroes (2004–11), Tabula Rasa (2007–09), and WildStar (2014–18). Star Wars Galaxies was set in galactic history right after the original Star Wars movie, but the action is separated from the main story line of the movies, and emphasized building homes and communities in collaboration with other players. City of Heroes drew upon the established super-hero genre of comics and movies, but chiefly explored and dwelled within a marvelously huge city with a diversity of neighborhoods. Tabula Rasa imagined that a few human beings have escaped Earth after aliens conquered our world, and are desperately setting up colonies on two distant planets, interacting with extraterrestrial cultures including the relics of an advanced civilization that had completed scientific discovery. WildStar, like a few other MMOs but giving more emphasis to building a home and economy, contrasted two forms of society, anarchic rebels competing with religious imperialists.

Table 4.2 Computer games that simulated complex cultures

	Launch date	Wikipedia pageviews	Metacritic professionals		Metacritic user scores
Name			Metascore	Critics	Mean	Raters
Multiplayer extinct:
Star Wars Galaxies	2003	477,201	71	33	5.8	104
City of Heroes	2004	520,194	85	58	8.4	112
Tabula Rasa	2007	143,255	78	34	7.7	111
WildStar	2014	298,112	82	52	7.5	951
Solo player:
Fallout	1997	1,680,778	89	12	8.8	1,045
Fallout 2	1998	926,374	86	15	9.1	1,187
Fallout 3	2008	2,515,470	91	48	7.8	3,966
Fallout New Vegas	2010	2,021,109	84	39	8.7	3,513
Fallout 4	2015	7,646,883	84	38	5.5	8,124
Multiplayer surviving:
Fallen Earth	2009	43,844	71	11	6.9	163

The relatively high rating for City of Heroes reflects the fact that many people wish it still existed. Ordinary videogames and computer games produced years ago can be played today, merely using the software the player had purchased, but that is not true for massively multiplayer online games, because they require server-based software and a database maintained by the company that published them. Back in 2010, the computer science magazine, Communications of the ACM, had published an essay written by my avatar, Rumilisoun, who spoke to readers from Elrond’s Library in Lord of the Rings Online, a game that has survived:

The library where I work in Rivendell is 1,000 years old, and I have trouble imagining all the difficulties you might face if you were to try to build a Digital Library of Virtual Worlds. Yet what a shame it would be if the glorious creativity of the first generations of virtual worlds were truly gone forever. The first great grand opera, l’Orfeo, composed by Claudio Monteverdi in 1607 is still performed today, and anyone may buy a recording for a few dollars. Four hundred years from now, I hope your descendants will still be able to visit me so I can introduce them to Frodo, Bilbo, and Gandalf … and perhaps all go Orc hunting together.”23

In 2019, a frenzy of communications in gamer forums and blogsites rejoiced when the world discovered that several groups had resurrected versions of City of Heroes, triggering active debate about how they could become legal, given that the company that closed the virtual city down still held the intellectual property rights and how they could adapt to changing technological standards.24

Table 4.2 shows the number of pageviews of Wikipedia articles about these virtual worlds, data beginning July 1, 2015, and ending February 1, 2019. These are considerable numbers of views, given that the first three MMOs had shut down before Wikipedia began tabulating views on their pages. Table 4.2 also reports Metacritic data, the metascores which Metacritic calculates through a secret algorithm from the various ratings given by professional critics, and the user scores on a quite different scale. After these four MMOs, we see data for a sequence of five postapocalyptic solo-player games set in various parts of the United States, as the Fallout summary article with 4,466,003 pageviews explains:

The series is set in a fictionalized United States in an alternate history scenario that diverges from reality following World War II. In this alternative atompunk “golden age”, a bizarre socio-technological status quo emerges, in which advanced robots, nuclear-powered cars, directed-energy weapons, and other futuristic technologies are seen alongside 1950s-era computers and televisions ... More than a hundred years before the start of the series, an energy crisis emerged caused by the depletion of petroleum reserves, leading to a period called the “Resource Wars” in April 2052—a series of events which included a war between the European Commonwealth and the Middle East, the disbanding of the United Nations, the U.S. annexation of Canada, and a Chinese invasion and subsequent military occupation of Alaska coupled with their release of the “New Plague” that devastated the American mainland. These eventually culminated in the “Great War” on the morning of October 23, 2077, eastern standard time, a two-hour nuclear exchange on an apocalyptic scale, which subsequently created the post-apocalyptic United States, the setting of the Fallout world.25

The fundamental Fallout principle has the quality of prophecy: Faith was misplaced that a New World Order would unify humanity in peace and prosperity. Once the world disintegrated, following the principle enunciated by Frederic Thrasher in his 1927 book, The Gang, large-scale social disorganization triggered small-scale social organization.26 Numerous gangs and social movements emerged in distributed locations, each focused on a person or an idea. The independent MMO, Fallen Earth, is frequently compared with Fallout, and more than once I have seen players discuss the comparisons in the Fallen Earth game chat. However, the diffuse postapocalyptic subculture contains many other influential examples, such as the movies Things to Come (1936), On the Beach (1959), Mad Max (1979), and more recently Divergent (2014) that also had utopian characteristics. Wikipedia provides this background:

The Fallen Earth story begins in the 21st Century, when the first in a series of natural disasters hits the United States. As Americans struggle to recover, an investment tycoon named Brenhauer buys a controlling stake in a mega-corporation named GlobalTech. By 2051, he moves his headquarters to the Grand Canyon Province, where GlobalTech eventually creates a self-sufficient economic and military mini-state. Meanwhile, in India and Pakistan, the Shiva virus, named for the dance-like convulsions that it caused in its victims, appears among the human populace. As the infection starts to spread, countries accuse each other of engineering the virus. Political paranoia turns to open aggression and nuclear conflict. The nuclear conflict combined with the virus devastates the planet. Less than one percent of Earth’s population survived the Fall, and the Hoover Dam Garrison and Grand Canyon Province are the only known outposts of human civilization. Outside the protective confines of the Hoover Dam Garrison, the player encounters ruins of the old world, genetically altered creatures, strange technology, and six warring factions. Some factions seek to rebuild the old world, others wish to build a new one in their own image, and some simply desire chaos and anarchy.27

As described on Fallen Earth’s website, although many story-based quest arcs are available, the emphasis is on freedom:

Explore, harvest and stake your claim to over 1,000 square kilometers of harsh and mysterious terrain. The classless advancement and non-linear gameplay allows you to play the character you want ... Fallen Earth gives you the freedom to do exactly as you want. The world may be a shadow of its former self, but there’s no limit to what’s possible for you to accomplish.28

Each of the “six warring factions” has a coherent, competing utopian ideology, and a major goal of my several periods of research inside Fallen Earth was to document their thinking. In an online guide, William Usher explained:

You do have the option of playing in a neutral organization within the game, but the tradeoff is a depth versus breadth experience. Neutrals can experience a lot more in the game since they aren’t being targeted by an enemy faction, while engaging in a faction provides a richer, deeper play experience as players explore the inner politics, specialized Knowledges and community bonds of playing Fallen Earth in a group.29

Table 4.3 reports the “reputation” status of each of my four avatars. Bridgebain was my main avatar that represented myself and fully explored this virtual world, while the others offered different perspectives based on the thinking of authors whose identities I temporarily adopted. Oswald Spengler was famous for predicting the fall of western civilization.30 Pitirim Sorokin, founder of the Sociology Department at Harvard, offered a more complex theory in which a culture arose in ideational (ideological) mode, then transformed into sensate (sensual) mode before falling.31 Mary Shelly suggested in her novel Frankenstein that harmful technologies could degrade humanity. The three alt (alternate) avatars halted exploration when they had completed exploring the social groups situated in the Plateau region, where all avatars begin and become familiar with the factions, and Northfields, where their faction membership consolidates.

Table 4.3 Faction reputations of four survivors of world war

Faction	Cultural value	Bridgebain Level 55	Spengler Level 20	Shelley Level 20	Sorokin Level 30
Factions which players may join:
Lightbearers	Spirituality	6,513	–50,260	27,773	–1,537
Vistas	Balance	82,085	–84,373	0	–2,950
CHOTA	Anarchy	4,950	–42,896	0	–2,009
Travelers	Profit	–13,026	25,126	–55,563	–1,503
Techs	Technology	–164,190	42,184	0	0
Enforcers	Order	–9,900	21,448	0	–2,001
Friendly nonplayer factions:
Townspersons	Protection	11,309	1,176	1,125	1,244
Bankers	Finance	18,782	1,800	2,300	19,287
Franklin’s Riders	Delivery	30,280	6,780	5,530	6,250
Shiva’s Favored	Pollution	17,564	0	0	0
Hostile nonplayer factions:
Clerics of Gates	Cybercultism	–323	–73	–84	–62
Gully Dogs	Raiding	–114	–47	–116	0
Blade Dancers	Violence	–151	–331	–283	–24
Shiva’s Blessed-Favored	Pollution	–372	–47	–24	–61
Human League	Purity	–149	–3	–3	–124
Devil’s Own	Evil	–462	–66	–83	–277
Gaunt’s Raiders	Looting	–116	–99	–13	0
Night Wolves	Thievery	–95	–44	–78	0
Judges	Zeal	–304	–74	–38	–62
White Crows	Russia	–255	–24	–17	–5

Table 4.3 serves multiple functions. First, it is the basis for charting the social structure comprised of factions that inhabit this virtual world. Second, it suggests the concepts the factions represent, each a subculture within a larger system that is rather coherent. Third, the very different numbers in the four columns illustrate how people may choose different paths within a multicultural system, combining political and artistic principles in somewhat personal ways. Each column of data could be conceptualized as the responses to a questionnaire from an individual person, although based on behavioral data rather than opinions. The gross structure of the factions arranges them in three categories: (1) the six factions to which both nonplayer characters and the avatars of players belong, (2) four nonplayer factions with which avatars may develop positive reputations, and (3) a number of low-culture gangs of enemies. The reputation scores for the first two groups are somewhat abstract and can be gained in complex ways, the negative scores representing enemy status, meaning that an NPC in the faction would attack the player’s avatar if it came near. The scores for the potentially hostile nonplayer factions simply tabulate how many members of each the avatar has killed, with the minus sign emphasizing enemy status.

Originally, the six joinable factions were arranged in a complex system of three pairs of arch-enemies, with each faction having two potential allies and two potential enemy factions as well. However, in 2012 this system was simplified, to become just three hostile pairs without alliances connecting some factions.32 As described in the wiki devoted to Fallen Earth, the arch-enemy pairs are Children of the Apocalypse (CHOTA) versus Enforcers, Lightbearers versus Travelers, and Techs versus Vistas:33

• • The CHOTA are a defiant heretical group who vehemently oppose order.
• • The Enforcers are the militaristic body of Fallen Earth. They are based upon strict values of organization and control, and despise anything chaotic or unorganized.
• • The Lightbearers are a group of spiritual sages who are dedicated to the practice of martial arts and enlightenment.
• • The Travelers are the gypsies of the new world. They travel from place to place, peddling their wares, trying to make as much money as possible.
• • The Techs are a group of scientists and engineers who are primarily interested in restoring the world as it once was through technological knowledge.
• • The Vistas are advocates of the natural ecosystem.

Originally, there were nine starter towns in the Plateau region, locations where avatars could begin their adventures in Fallen Earth, but the number was reduced to three, the other six remaining as places low-level avatars could visit. This major shift represents a common problem for MMOs, but also illustrates a more general challenge faced by subcultures and social movements in the real world. Given great publicity, a new MMO will begin with a flood of players, all of course at level one of experience, who will need room to perform their early missions without getting in each other’s way. That requires multiple starter locations, to spread the surge out. But it is also important for new players to encounter other new players, at first cooperating spontaneously, then forming guilds to enable enduring collaboration. Therefore, after the initial surge, a smaller number of starter locations is appropriate. In a study of the Transcendental Meditation movement, Daniel Jackson and I observed a flood of new members who trained to become teachers, but the rate of recruitment of newcomers was not sufficient to keep them busy in that prestigious role.34 Thus the population dynamics of multiplayer games may, at least abstractly, offer insights into evolutions of social structures more generally.

Of the three current starter towns, Boneclaw is dominated by members of CHOTA, and the emphasis is on teaching combat skills to new avatars. Given their anarchistic tendencies, the CHOTA of Boneclaw are rather disorganized, with low morale and partly divided into subfactions. By completing several quests, the player can unify the CHOTA of the town, but this does not give the player’s avatar an improved reputation with CHOTA, because faction affiliation does not begin until the second main region, Northfields. The Midway starter town is dominated by Travelers, and the training emphasis is on crafting, which makes sense because the travelers are merchants and crafting produces products that may be sold. Like the CHOTA of Boneclaw, the Travelers of Midway are divided, but into the equivalent of two competing crime families, and one quest has the player select which of the two family leaders to murder, also not gaining faction reputation. The third starter town, Clinton Farm, emphasizes support missions and does not have a dominant faction.

The six other original starter towns were similarly arranged in two trios, in each of which one town specialized in teaching combat, crafting, or support. Mumford was the crash site of a radioactive military satellite and remains conflict-oriented, as the wiki page for its missions reports:

You have a choice to either support Jon Dawkins and the Vistas with cleaning up the radioactive waste from the Mumford Crash Site or supporting Mercury Reynolds and the Techs with salvaging the valuable technology. The rewards do vary based on which side you support, but supporting one side or the other plays no part in your eventual faction alliance.35

Nearby South Burb belongs to the Vistas, has flourishing farms, and emphasizes crafting, while North Burb belongs to the Lightbearers and is the region’s main center for production of medical supplies. In the third triad of starter zones, Zanesville belongs to the Enforcers and emphasizes combat; Depot 66 belongs to the Travelers who teach crafting and have set up a Shakespeare theater, and Terance where several missions involve a ruined LifeNet facility and an insane artificial intelligence named TETRAX.

Upon returning to Fallen Earth to complete the research for this section, I created my fourth avatar, naming her Mary Shelly after the author of Frankenstein, and deciding she should emphasize medical training, become a Lightbearer, which required her to be an enemy of the Travelers, and unlike the three other avatars avoid entanglement in any of the other four primary factions. She stayed for a while in North Burb, doing as many Lightbearer missions as she could, and marveling at what she learned. In an online guide to the factions, William Usher wrote:

The Lightbearers are wandering disciples who teach peace and emphasize spiritual, mental and physical development. They believe in protecting their “light” within and fighting for peace if necessary. They follow the teaching of Shakti, a woman who traveled the Province healing and protecting the innocent in the years after the fall, and excel in martial arts and close combat techniques.36

Wikipedia reminds us:

Shakti ... is the primordial cosmic energy and represents the dynamic forces that are thought to move through the entire universe in Hinduism and Shaktism. Shakti is the concept or personification of divine feminine creative power, sometimes referred to as “The Great Divine Mother” in Hinduism.37

Thus, we might conclude that the Lightbearers are adherents of Hinduism. However, the main architectural feature of North Burb is a pagoda, a kind of structure more fully developed in China and Japan, where martial arts are especially traditional. Just east of the pagoda, Mary Shelly found Master Henry Solomon sitting in an eastern meditative pose, prepared to teach her not only martial arts but also self-mastery. His proclamation of faith superficially suggests an Asian religion:

In the spirit of our ancestors, we pride ourselves in harmony of body, spirit, and mind. Through our body, we connect ourselves to the earth, taking what we need from it, and giving back that which the earth requires. Through our spirit, we connect with one another, forming a community of life that weathers the greatest of storms. Through our mind, we connect to ourselves, becoming honest to our abilities and the true meaning of our existence. Wisdom lies not in the mastering of the mind, but the combined strength of body, spirit, and mind.38

However, that proclamation really came from a very different source, Luther Halsey Gulick Jr. (1865–1918), “an American physical education instructor, international basketball official, and founder with his wife of the Camp Fire Girls, an international youth organization now known as Camp Fire.”39 I knew this because I and my sister had attended two of the summer camps he founded in Maine, Timanous, and Little Wohelo, where everybody sang, “He has welded a symbol, Body, Spirit, and Mind”—sung to the melody of the first “Pomp and Circumstance” march by Sir Edward Elgar.40 The Timanous symbol is a triangle that has also played a role as the symbol of the Young Men’s Christian Association (YMCA), and an outline of its history states,

Luther H. Gulick, who revolutionized sports and physical fitness at the YMCA, proposed a red equilateral triangle as a symbol in 1891. It was adopted immediately by Springfield College. The sides of the triangle, Gulick said, stood for “an essential unity: spirit, mind, and body: each being a necessary and eternal part of man.”41

In a history of the college, Laurence Doggett described Gulick as “a combination of virility with a great deal of the theoretical” and speculated he might have derived the triad from Buddhism.42 However, his source of the idea that Buddhism was trinitarian was a 1940 tourism book, but not a scholarly study, and the mystique of a trinity is certainly central to Christianity.43

At the risk of unleashing endless theological debates, we might speculate that: Jehovah = mind; Christ = body; the Holy Spirit = spirit. In much earlier centuries, wars were fought between the Homoousions and the Homoiousians, who disagreed literally by one iota of difference over whether Christ was the same as God or merely similar.44 Perhaps in the Age of Artificial Intelligence there will be similar bloodletting over the exact identity of the Holy Spirit. Gulick was not a partisan in such debates and gave spirit a far different meaning, close to joy:

When I speak of the “higher life of the spirit,” do not apprehend that we are drifting into a religious discussion. A higher liveliness of the spirit would have expressed my thought even more adequately. The “play of the spirit” is not an empty phrase. It is always the spirit that plays. Our bodies only work. The spirit at play is what I mean by the higher life. Play is the pursuit of ideals. When released from the daily work, the mill we have to tread in order to live, then we strive to become what we would be if we could. When we are free we pursue those ideals which indicate and create character. If they lead us toward wholesome things—literature, music, art, debate, golf, tennis, horseback riding and all of the other things that are wholesome and good, then our lives are rounded out, balanced and significant.45

Within the universe of massively multiplayer online games, Fallen Earth has sufficient verbiage and intellectual content to be considered literature or philosophical debate, but its spiritual play element is also achieved by the horseback riding Gulick mentioned, indeed the riding of many different steeds and vehicles. His profound comment, “Play is the pursuit of ideals” may describe many of the subcultures explored in this book, not merely the Lightbearers.

My four Fallen Earth avatars never debated the contrasting merit of the factions, but chose four very different paths for themselves, as four different players might have done, thus illustrating multiperspective phenomenology conducted by a single researcher. Bridgebain was the original explorer, entering the virtual world December 30, 2011, at a time when there were six available starter towns, of which he selected Clinton F.A.R.M., aware it was under siege by the Blade Dancer gang and that it was not really a farm but a school for “Fire, Alpine Rescue and Medical.” Initially, he aligned himself with the two more nonviolent and spiritual factions, by experience level 23 reaching a reputation score of 43,914 with the Vistas and 36,230 with the Lightbearers. As an experiment, on May 8, 2012, he obtained a Faction Reset potion from an NPC named Disgruntled Outsider and erased all six of his main faction reputations. In his first phase of research, the unified system of factions was enforced by the software, but he continued to follow that structure after it became optional, in part because many of the quests had been designed on the basis of the original faction structure. When he completed exploration of the entire territory and reached the experience cap of level 55, he had positive reputations with three factions: Techs (187,594), Travelers (174,542), and Enforcers (75,686).

A new research project reset Bridgebain’s reputations again, so he could explore the environmentalist Vista faction more thoroughly, and two more avatars were added, Spengler and Sorokin, based on two theorists of the decline of civilization, Oswald Spengler and Pitirim Sorokin.46 Spengler reached level 20 without doing any crafting, while Sorokin reached 20 without unnecessarily killing any human NPCs. For this new research in 2019, Sorokin’s affiliations with the six main factions were reset, and he decided he would oppose all of them. Meanwhile, Mary Shelley constantly affiliated with the Lightbearers as she ascended the experience levels. Unlike some of the most popular virtual worlds, Fallen Earth does not record how many hours each avatar has been played, but the total for this study was certainly more than 300 hours and more likely 400. MMO players often are active in a particular game in a series of episodes, as illustrated by the periods I operated one or more of the four avatars: December 2011 to April 2012, December 2012 to March 2013, August to September 2013, August to September 2014, January to February 2019. This complex history suggests how real people in the real world may affiliate intermittently with subcultures over the course of their lives.

Conclusion

Computer simulations have always seemed to me to be an efficient but also creative way to explore the meaning of social theories, and thus today very applicable to cultural science. By “always” I mean since I obtained a Geniac “computer” in 1956, the educational system designed by Edmund C. Berkeley, a founder of the Association for Computing Machinery. As Wikipedia recalls: “Widely advertised in magazines such as Galaxy Science Fiction, the Geniac provided many youths with their first hands-on introduction to computer concepts and Boolean logic.”47 At that point I had read every monthly issue of Galaxy since it was first published in 1950 and was especially impressed by Alfred Bester’s novel The Demolished Man, which had been serialized in Galaxy in 1952, and was the first novel to win the prestigious Hugo award.48 Decades later, when Bester and I were both panelists at a science fiction convention, I jokingly praised him for having invented a new religion, but really his novel was a deep exploration of social-psychological theory, assuming the existence of telepathy in order to explore some of the deep implications of psychoanalysis. After his death in 1987, it was fascinating to see him simulated as a character revealingly named Bester, in the Babylon 5 television science fiction series.49 What can we today learn by telepathically reading Bester’s mind? Fiction, whether in the form of traditional printed literature, television programs, or computer games, is the most extensive simulation of human reality. Thus, the main topic of literature studies in the humanities has always been mental simulations of human experience, today largely published through computers.

---