So a design I'm working on (thinking about while I toss boxes) was giving me some trouble. It's about the American Southwest during the Spanish colonial period and the relationships between the New Mexican settlers, Spanish colonial administrators and the heads of different bands of Indians. It's going to be super small-scale and turn based, and it will involve war but mainly things like slavery, trade, agriculture, seasonal migration and so on. Most of my ideas for historical games come from specific monographs, and this one in particular came from Ned Blackhawk's Violence Over the Land which is a good book as long as you have a stomach for digesting typical academic history. Which I do.
Part of the challenge of a historical setting for a computer game is getting people to behave in hisorical ways. The behavior is measured in a weird way. Wargamers want two things out of a historical game: fidelity to history measured in terms of the result, and fidelity to history measured in terms of the process. Normally they emphasize the first (see this) and ignore the second. Developers like those behind the Magna Mundi atrocity for Europa Universalis III try to game the process by plugging all orifices with arbitrary values, to the point where anti-semitism is quantified as a plus five percent modifier to revolt risk once the Jewish population of a province passes a certain threshold.
At least that's better than the event-driven approach of other mods for other Paradox games, like the Vicky Improvement Project for Victoria, which 'simulated' the rush for Africa not by diminishing the expansionary capacity of Europe but by outright forbidding colonization before a divinely ordained date, as if everyone was really just holding back and toeing the starting line for fifty years before 1880 out of a sense of sportsmanship.
Of course both of these mods and the games they were designed for are on a much higher level than a specific event or a specific intra-province dynamic. Its really not that bad, in the great scheme of things, if anti-Semitism is modeled in an anti-Semitic way in a few provinces if the whole thing kind of holds together on a macro level and produces believable results. Wether or not the Vicky Improvement Project or Magna Mundi (which is now spiraling completely out of control in the form of a stand-alone game) actually do produce believable outcomes or if they do it in a believable way is another question.
But it is important to think about these small things, especially if your game is going to be tiny and built around nuances that grander strategic games would abstract out of existence. Making people think and act like people did in the past doesn't demand that you place restraints on them in a system that should logically not have that restraint (for example: Vicky is built around conquest and colonization, why restrict it?). It demands that you make the conception that people had of their time and their role in the world real by designing your game around it. Wargames can learn from RPGs in this instance. One example that comes to mind is Darklands.
The designers of Darklands said that the starting point for the game was asking the question "what if the world was actually what the average German peasant thought it was?" From this premise adding monsters and magic was entirely logical. And players in the game accept the premise unthinkingly and adapt to rituals (praying to saints, for instance) that would be arbitrary and ridiculous if the game were modeled on real life but required the player to pray anyway because that's what people did back then. Making a game requires giving the average citizen of your imagined world some credit. You have to treat the things he thinks as real as if they were real. It makes for a better game and maybe designing and playing it gives you some insight into the absurd rituals in your own time and place.
Anyway, this led me back to my design. Modeling things like seasonal migration and slavery is easy enough- put resources in different areas of the map, vary their peak availability times and let players figure out when the best time to hunt small game is and when to visit the piñon groves. But modeling something like the relationship between the settlers and the Indians is tougher. Why, for instance, did the Spanish bother baptizing Indians, and why did Indians not immediately kill any settler they came across? From the perspective of the 21st century the existential danger that European colonialism posed to native peoples is blindingly ovbious. But it wasn't back then (Utes initially invited the Mormons to settle near them, for instance), otherwise violence against settlers whould have been the first resort of native peoples and not the absolute dead last (after long periods of hospitality, accomodation, friendly trade, credulous treaty-negotiation, and tolerating increasingly hostile impositions). To natives at the time settlers appeared as a far more neutral force. They occupied some land and deprived natives of access to some resources, but they had trade goods like metal pots, horses and rifles that natives found immensely useful, and in some cases the paternalism of an imperial power was preferable to slave-raiding by your neighbors: when the Escalante expedition passed through Utah, for instance, they promised to establish missions and settle among the Indians they encountered (which is a promise they never kept) and many bands of Indians, who by that time had endured decades of raids from more powerful tribes to the east, were very happy to hear it. That's why the design emphasizes trade and co-habitation and overall adopts the viewpoint of a group of people who did not (and could not) see with perfect vision several hundred years into the future.