Despite popular misunderstandings about chaos theory, dynamical historiography offers both futures studies and game designers a powerful tool to interrogate the "social psychological, political, economic, social, or cultural implications" (Bell, vol. 1, 1997: 182) of simulation worlds and scenarios. Embedding the design of games like Civilization III within a "critical"/"epistemological" framework can counteract the blind-spots and defects inherent in the world-building philosophy of many programmers, which model specific aspects of reality but caricature others (Poole, 2000: 61). For Sohail Inayatullah, "epistemological boundaries--languages, structures, and practices--define the significant and the trivial, the negotiable and immutable, and the real." (Galtung and Inayatullah, 1997: 163). So how significant or trivial is Civilization III?

While the God game perspective of Civilization III erases the 'mutant' individual from history, its artificial intelligence engine can model when "a single, often trivial change, has momentous consequences." (Ferguson, 1997: 12). This reflects geopolitical history: a single decision during China's Ming dynasty to pursue an isolationist policy changed the world system's trajectory and technological flows (Wright, 2000: 163-164; Fernandez-Armesto, 2000: 265).

Another extremely useful epistemological lens to study Civilization III is macrohistory: "the study of histories of social systems, along separate trajectories, in search of patterns." (Galtung and Inayatullah, 1997: 1). Its nomothetic focus (Galtung and Inayatullah, 1997: 2-3) counterpoints William McNeill's influence. In principle the God game perspective that upholds societies over the individual (Galtung and Inayatullah, 1997: 175), time as the unit of analysis (Galtung and Inayatullah, 1997: 182), and the ability to study macro-processes across time-space (Galtung and Inayatullah, 1997: 167) should be a 'core competence' of games like Civilization III. The reality is different.

Civilization III does not overtly implement insights from macrohistory but by understanding the perspectives of individual macrohistorians, we can develop a useful critique of its game-play. The game's model draws upon many historical precursors, including Adam Smith's progression from agricultural to exchange economies (Galtung and Inayatullah, 1997: 42-43), Auguste Comte's positivistic society (Galtung and Inayatullah, 1997: 57), Karl Marx's techno-economic superstructure (Galtung and Inayatullah, 1997: 62), Herbert Spencer's social stages (Galtung and Inayatullah, 1997: 72-73), Vilfredo Pareto's degeneration of elites (Galtung and Inayatullah, 1997: 80), the player as a Weberian charismatic leader (Galtung and Inayatullah, 1997: 87), and Oswald Spengler's cultural lifecycle (Galtung and Inayatullah, 1997: 100-101). Each perspective has its epistemological insights, biases and blind-spots. Therefore it's unfortunate that Civilization III's AI engine does not draw on this macrohistorical knowledge base and key factors (Galtung and Inayatullah, 1997: 161) more effectively. Perhaps this is a sign that Meier and his design team are caught in the cultural transition from Prabhat Rainjan Sarker's vipra (intellectual) to vaeshya (capitalist) mode (Galtung and Inayatullah, 1997: 135). Spengler might have contended that games like Civilization III personified the stifling inwardness of declining 'civilization' elites (Galtung and Inayatullah, 1997: 239).

Often the player is forced into the "grab-what-you-can mentality" that Pitirim Sorokin warned against (Galtung and Inayatullah, 1997: 117). Economic cycles and political crises created the cultural homogeneity critiqued by Antonio Gramsci (Galtung and Inayatullah, 1997: 130). Yet this dynamic, which is closer to biological and cyclical models than linear ones, also captures Spengler's central insight that "there are many cultures, each with their own patterns within a general overall pattern-birth, growth, decay, and death." (Galtung and Inayatullah, 1997: 180). The player's leadership skill influences subsequent next historical stages (Galtung and Inayatullah, 1997: 189).

The macrohistorical strength of Civilization III is that its isomorphic viewpoint accidentally embodied a basic understanding of macrohistory as sociography (Galtung and Inayatullah, 1997: 225). The game's macrohistorical simulation may stimulate game-players to critically reflect upon their individual microhistory and the emergent sociocultural mentalities of their historical era (Galtung and Inayatullah, 1997: 227). Johan Galtung and Sohail Inayatullah's in-depth summary of macrohistorians offers Civilization's designers an innovative way to develop immersive game-play in future editions.

Civilization III and Digital Game-Based Learning

While Peter Schwartz was creating a scenario about 21st century adolescents (Schwartz, 1991: 126), several trends were converging to create a market for digital game-based learning. Early epistemic roots included Seymour Papert's exploration of microworlds at MIT (Horton, 2000: 574-575), David Kolb’s learning loop (van der Heijden, 1996: 37), and social psychology experiments (Prensky, 2000: 42). The late 1980s fascination with MTV editing (Prensky, 2000: 13) foreshadowed Generation X's hypertext minds (Prensky, 2000: 44). Changes in the early 1990s business ecosystem, notably the growth of corporate universities (Prensky, 2000: 333), design for doing (Prensky, 2000: 59; Schrage, 1999), knowledge management (Beer, 2000: 103), collaborative action learning (Beer, 2000: 42, 165; Prensky, 2000: 20), and communities of experts (Beeer, 2000: 57; Prensky, 2000: 107) laid the groundwork for digital game-based learning models. Today such models are increasingly used as recruiting tools (Prensky, 2000: 337) and teaching strategic thinking (Prensky, 2000: 359). Digital game-based learning highlighted the truism that "new knowledge develops at the fringes" (Schwartz, 1991: 73).

Researchers found that simulations were intrinsic games that linked internal goals and gameplay (Prensky, 2000: 164). They replicated cognitive models of learning and organizational behaviour (Beer, 2000: 75). Simulations taught players to assume multiple roles by "breaking the rules and experiencing the consequences." (Horton, 2000: 572). They provided "natural case-based learning" opportunities that were not possible in centralized learning programs (Beer, 2000: 170).