Gaming the Past

Disclaimer: I decided to post this immediately on finishing and I wrote it quickly. It is a discussion piece, not even remotely a polished essay. Would love your feedback.

We Should Move Beyond Evaluating a Historical Game Primarily on Whether it Portrays a List of “Facts” Historically Accurately

When I first started shifting from my independent work on video games and history education into the fledgling community of Historical Game Studies fostered by playthepast.org which was launched under founding editor Ethan Watrall’s guidance in 2010 I began as an educator and historian approaching games. I suggested in The Happiness Metric in CivCity:Rome and the Critique of Simulation Games (September 2010)

Asking whether a historical simulation game is accurate as if that were an all-or-nothing quality seems to me to be missing the point—the accuracy of any historical interpretation is not something that can be determined with any certainty. One historian’s common sense convention is another’s faulty construct to be dismantled. One generation’s conventions are the next’s biased assumptions. What really matters in historical interpretations is the extent to which any particular one is constructed based on the strongest, most defensible readings of evidence and the best supported and culturally sensitive understandings of human behavior. So, a far better criterion than accuracy when critiquing a historical simulation game is whether its core gameplay offers defensible explanations of historical causes and systems. So for example, it is not a question of whether a civilization building game allows a player to develop nuclear fission in the 17^th century, but whether the game reasonably models the factors, including constraints, that lead to the development of such technologies. This focus on defensible models of causation is absolutely critical when one’s interests, like mine, center on using simulations as models to aid student (most often high school, but middle school and college as well) in understanding historical systems and learning to critique interpretations of the past. Let me be clear on the term interpretation here. We do not access the past directly nor do we present the past directly; we can only access and construct interpretations. This is true whether the interpretation comes from the Roman historian Livy, the modern historian McCall, or the game designers at Firefly. In this light, so long as a game’s core gameplay is historically defensible, any, even many inaccuracies serve as highly useful targets for getting students to launch evidence-based critiques. Evidence-based critiques is the operative term; simulation games should be critiqued using abundant references to the contents of valid sources of historical evidence.

Most of my thinking about studying historical games has progressed–or at least moved–since 2005 when I started talking to others about them, and 2010 when I first started writing about them. So I have developed guidelines for what quakifies a game as having defensible models. But before we get to that, I’m hoping you will indulge me–since this is a blog and not a formal article–in a bit of philosophizing about history and a bit of history of Historical Game Studies. I’ve put headers in so you can skip straight to the defensible-models piece.

The Need for Historical Game Studies to Move Beyond the “Is the game accurate” focus

Why should anyone interested in critically approaching historical games care about whether they have defensible models? Understand, caring whether a game has defensible models does not mean excluding those without as unworthy of play or consideration or study: not at all. But I do not agree with the extreme position (which I will not pin to anyone, and indeed, I have no one in mind who takes this extreme position) that, outside of an educational context, the connection of a historical game to documented (with valid evidence) history is unimportant. Certainly, in the earliest days of Historical Game Studies (say 2005 with William Uricchio’s article; coincidentally also the year of the first national talk I gave on using videogames in high school history education and the start of my connection with UW-Madison’s Games Learning and Society Conference, which was foundational to my journey) the fledgling field hoped to move beyond the simplistic and largely unhelpful exercise of historians and history educators (many, many, many are both at all levels of education btw) who looked at historical games and simply listed all the ways they were not accurate, by which they usually meant the game presented phenomena to the player or discrete details that went against documented text accounts of the past.

I summarized the problem of nit-picking a game’s “historical accuracy” in a 2020 article introducing Historical Problem Spaces in more detail to the HGS community.

Analyses of historical video games that are sensitive to the medium must move beyond an itemized list of the discrete atomic facts in a video game that are deemed accurate or inaccurate. Such listing treats the game as merely the sum of its parts, a series of independent propositions about the past to be judged by adding up the total of accurate and inaccurate ones. This approach boils down to an ad hoc, piecemeal system. It also risks ignoring what Kapell and Elliott rightly indicate are the most interesting questions of historical game studies, not whether “a given product deviates from the historical record, but rather for what reason it does so and what effect this might have” (Kapell & Elliott, 2013, p.8). It ignores the fact that a functioning game must be a closed working set of formally — that is mathematically — defined systems needing player input to function. It is a historical medium different from a book, lecture, morality tale, painting or film. It is decidedly not a historical monograph, nor do its consumers wish it to be. Instead it is a powerful, sometimes moving, highly engaging medium that has garnered millions of fans through what it offers.

Analyses of historical video games that are sensitive to the medium must move beyond an itemized list of the discrete atomic facts in a video game that are deemed accurate or inaccurate. Such listing treats the game as merely the sum of its parts, a series of independent propositions about the past to be judged by adding up the total of accurate and inaccurate ones. This approach boils down to an ad hoc, piecemeal system. It also risks ignoring what Kapell and Elliott rightly indicate are the most interesting questions of historical game studies, not whether “a given product deviates from the historical record, but rather for what reason it does so and what effect this might have” (Kapell & Elliott, 2013, p.8). It ignores the fact that a functioning game must be a closed working set of formally — that is mathematically — defined systems needing player input to function. It is a historical medium different from a book, lecture, morality tale, painting or film. It is decidedly not a historical monograph, nor do its consumers wish it to be. Instead it is a powerful, sometimes moving, highly engaging medium that has garnered millions of fans through what it offers. (McCall, Game Studies – The Historical Problem Space Framework: Games as a Historical Medium, 2020)

Of course, I was focused more on historical videogames at the time of this writing, so one could certainly object that analog games (including TTRPGs) are not always a closed set of formal mathematical models (though I would argue that even analog games almost always are mathematical models — consider a basic like Monopoly–but that’s for another day and another discussion). I quoted in full here because I agree that HGS needed to move consideration of historical games beyond a piecemeal list of discrete facts in a game that are or are not “accurate”.

Abandoning the Historical Accuracy Obsession Does Not Mean Abandoning Interest in How a Game Portrays Evidence-Based History

It does not follow, however, that a rich and sensitive consideration of how a game connects to valid historical evidence, how it portrays a historical world is problematic or indicative of a wrong-headed aspect to consider when thinking about historical games or any agential game referencing the “real” world

(and pace readers who want to challenge the idea of a real world. I get it. I know. But we have to have some common ground for discussion to get anywhere at all)

Rather I suggest that assessing how a historical game does and does not conform to valid evidence, to the real world, should be a critical part of historical game studies. Again, though, the point is not to condemn games for their connection to evidence. Rather it is simply to keep in mind what valid evidence can and cannot support about the world. Let me unpack this.

First of all, all history–the “curated representation of the past” (McCall 2020) or, as is often more helpful with my students, “selected” or “selective” representation of the past (McCall 2024, p. 75 n. 13)–or rather claims about history, if sincere, are fundamentally didactic or educative. Or more precisely, they all have the potential to educate. I would define the core of what makes something didactic or educative is that it is has the possibility–sometimes even the intention–to change (or affirm) other people’s thoughts and ideas and practices, ways of being and knowing and problem-solving. Saying 1+1 = 2 can educate, to persuade or confirm a way of thinking about the world. If the listener does not believe it, the claim as claim asserts it should be believed. To bring people to understand, analyze, believe, think, be in new ways or confirm them in those ways, that is what education fundamentally is. And of course–of course, of course–responsible educators try to base that persuasion/explanation/direction to new ways of being and knowing and doing in valid evidence based responsible practices. And of course, intending something to be educative does not mean it successfully educates.

So if we apply that to historical claims/statements in any medium, I think we still have to understand them as at least potentially didactic and educative in effect. All historical claims have the potential to be educative and very often the intention. So a historian that claims, as Ken Burns, Sarah Botstein, and David Schmidt have relayed in their 2025 American Revolution, that preserving enslavement of Africans and African Americans was a vital motive for some rebels in the 1770 and 1780s, can persuade and/or confirm others to agree with that point of view. The Holocaust denier’s claims could potentially persuade others to agree with this odious idea.

Why am I going on about this? Because historical claims are functionally educative and very often intentionally educative. So there is no bright line between the functionality of what a teacher suggests in a classroom about the past, some academic completely separated from classrooms says about the past and a historical game says about the past. All are functionally educative. This suggests to me that it does matter how all these sources educate about the past and whether the claims each makes, teacher, book writer, history game, history film, are consistent with the evidence is a valid question to keep in mind, even if evaluating a game’s arguments about the past is not the primary goal of the person considering the game. In other words, we should not jettison any concern or interest with how evidence-based a game’s depiction of the past is in the process of avoiding nit-picking.

For anyone more formally an educator in history, I think it is even easier to make the case that some consideration of whether a game’s portrayal of the past fits valid evidence / our understanding of the “real” historical world is important. Or to put it another way, I think educators are considering that all the time, but maybe not even realizing it. Fundamentally, it seems to me impossible for educators to avoid assessing whether a particular account, a particular film, a particular game fits the historical evidence if that medium is being used to educate students. If the educational goal is at some point to get students to discern what is valid, what is evidence-based, what is best practice, those discernments have to be based, I think, on some standards of evidence (facts if you like) about the world. It seems to me more than a stretch to suggest that educators analyze historical games with no consideration of the historical evidence.

Also, one cannot, it seems to me, consider what a game “does with history” or “how it portrays history” without some baseline of what one considers the historical evidence and record to suggest. So focusing on how a game interprets or presents history definitionally requires considering the historical evidence / historical record.

But we may not always be processing when we do these things. I suspect knowledgeable academics tend to feel less concerned about the how a game or some other model measures up to evidence because we are in command of a lot of evidence and the body of knowledge of our discipline, so we don’t even realize we are constantly referencing the evidence-based world (as we individually perceive that evidence of course) automatically and instinctively as we look at any medium presenting the world. We are making judgments about sensible and silly all the time and don’t think twice about it. But we are often content experts. Our students, or the people we make history for may not be.

Tl;dr summary of the above

• We should consider as a matter of course whether a historical game has historical evidence-based validity
• I strongly suspect those interested in the discipline of history and in history–not least of all educators and academics and academic educators– do consider this even when they are not consciously focused on it.
• Any HGS analyst consideration of how a game presents/interprets/shapes history presumes some base of historical evidence the HGS analyst is operating from formally or informally.
• BUT absolutely, nit-pickng a game into a list of discrete factual phenomena and declaring each one accurate or not accurate is NOT a meaningful or productive way to do this.

Evaluating the History in A Game: Defensible Models instead of Historical Accuracy

I began this musing with my first mention of defensibility in public writing. The term was not meant to be militaristic, but rather to reflect my experience in my formal defenses of my Master’s Thesis, Doctoral written exams, and Doctoral Dissertation (in Greek + Roman History, the latter 26 years ago this July). A defense, for those who have not partaken of one, consists of defending one’s written work, whether massive test essays or formal written essentially mini-books and books) to a set of experts, usually one’s academic advisor at their University, a few other relevant experts and often a professor outside the department to provide departmental accountability. The scholarly panel asks questions and seeks explanations and reasoning and support and commentary on the parts of the defending grad student’s written work that interest or trouble them. So the defending grad student’s task is to explain, justify, recant in a way that satisfies the examiners that the student has a strong grasp on the evidence, analytical frameworks, and so on in the field (for the written exams) or on the topics (for the Thesis and Dissertation). (I am speaking only from my experiences as a grad student and an examiner, and they are limited; the process varies I am sure).

For me, defenses were a flexible moving experience of discussion of evidence, theory, scholarship ultimately assessed by the examiners as successful (i.e. their positions were generally defensible; this was definitely not an assessment that obsessed about every single “fact” uttered or not uttered in a 2 hour defense) if the examiners were satisified the grad student had shown the necessary skill and insight in interpreting the past, unsuccessful (their positions proved not really defensible or insufficiently) otherwise.

So that’s what I meant when I first presented the idea that assessing whether a historical games’ in “core gameplay offers defensible explanations of historical causes and systems.” is very important for the history educator. By now I hope I’ve persuaded some of you that this consideration is important (fundamental though not necessarily central) for any historical consideration of a historical game. Defensible gameplay can have valid evidence readily rallied that suggests that explanation, as I called it in 2010 is reasonable to assert. Not a perfected statement of absolute truth. Not “accurate” in avery atom. Rather, reasonable to assert based on the valid evidence. Defensible.

I’ll spare the reader from tracking the term through my writings but note that, for a chapter essay that came out in 2020 “Digital Legionaries: Video Game Simulations of the Face of Battle in the Republic” I naively toyed with using the term “authentic” as meaning what I meant by defensible: generally supportable by the valid evidence, with no claims to point-by-point comprehensive “accuracy” in every atomic claim. Phew — when the peer-reviewers saw that (thanks as always, fantastic editor Christian Rollinger for navigating me through that!) they roasted me. Authenticity had been and has been claimed by HGS folk to mean something like “feels legitimate to the player”. Fair enough. So I continued on with defensible.

As I started to publish more on my Historical Problem Space framework and come to see that historical games are fundamentally about agents in systems, I started to use the term “defensible models” more and that is generally how I use the term today: I am interested as educator and academic in the extent to which any historical game defensible models, or has defensible models, of the past. Gaming the Past Second Edition (2022/23) is the most recent full explanation of the concept (p. 30). Check out the book for a longer discussion; here is a critical excerpt:

Historical games present the past as historical problem spaces. Accordingly, it is often not enough to simply judge a historical game’s consistency with the historical record as if it were a written text, with all the precision, linearity, and citation that entails. One certainly could critique a historical game by listing all the details it “gets wrong” or its lack of citations. Richer and deeper historical analysis work, however—and thus critical thinking opportunities—are available for students. Namely, it is important to consider, in addition to the factual accuracy of this or that detail, the extent to which any particular historical game’s basic problem space and player agent generally function as defensible models and provide defensible analogies to historical actors and systems. In other words, to what extent can one reasonably argue that the player agent, goals, action-choices, and gameworld elements and systems (agents, resources, tools, obstacles) of the historical problem space agree with the historical record? For each game, this is absolutely an open-ended question: reasonable people can and will disagree on what interpretations the historical record supports. But the idea of a historical game offering analogies that are more or less consistent with historical evidence, more or less defensible, is important for any educator considering games for use in class. It is also important for orienting productive class-learning activities. After all, assessing the connection between historical record and historical game, through research, analysis, and discussion, is a central critical learning activity for students working with a historical game, as the anecdote beginning this chapter illustrates.

Again, one could conceivably generate a sizable list of discrete details in a historical game that cannot be supported by the historical record. And this could be a useful experience for lower grades and a warm-up for higher grades. But a richer critique—and arguably one employing more sophisticated critical reasoning skills—will also consider the historical problem space presented by the game more holistically. This includes considering which aspects of the game offer arguably better or worse analogies to historical agents and systems. So, taking into account that—as all historical media do—even historically well-grounded games will have factual inaccuracies and simplifications, one can ask questions like these to investigate the basic relationship between a historical game and the historical record:

To what extent are the game’s player agent, designed goals, and available action-choices defensibly analogous to historical agents, their goals, and available action-choices?
To what extent are the gameworld elements and systems of elements in the problem space (agents, resources, tools, obstacles) historically defensible?
And always, how can students engage historical evidence with a teacher’s guidance to critique these games?

Again, “defensible” is the operative term. In the discipline of history, arguments offered by one historian may be rejected by the next. The propositions that endure become historical convention, the arguments that have withstood criticism due to the strength of their explanatory power and supporting evidence. Historical convention can always be revised; indeed, challenging conventions is a time-honored tradition in history. By these standards, then, a rich, media-sensitive critique of a historical game for pedagogical purposes should consider the extent to which the game’s core problem space components are reasonable based on the available evidence. This standard also promotes the idea so critical for training flexible, creative thinkers, that when it comes to humans interpreting and making meaning of the past, shades of gray and maybes outnumber certainties. This also encourages students to think about what can and cannot be sustained by historical evidence. (McCall 2023, p. 30)

Here I set out the principle that a historical games’ defensible models can be assessed by considering the historical problem space the games presents (for more on games as historical problem spaces and the Historical Problem Space framework for analyzing, teaching, and designing historical games, see The Historical Problem Space Framework for Game Analysis and Design | Gaming the Past. If you are serious about HPS please read BOTH the 2020 article and the 2024 article. People often seem to skip the second one, but it is a critical methodology piece). Because a historical problem space (all history games present the past as historical / agential problem spaces) is a set of systems, assessing the defensibility of that system gives us that essential consideration of how a history game models the past without requiring the distracting and less helpful breaking apart of a game into a list of claims that are accurate or inaccurate.

And now, with the upcoming Designing Historical Games for the Classroom: A Practical Guide for Educators, I have taken the standard set out above in GTP 2.0 and just made it even more concrete. A game has more defensible models of history to the extent the valid historical evidence supports the whole or parts of the game’s historical problem space (HPS). In other words, a has more defensible models if the valid evidence suggests:

Some agents existed in a place like THIS, and had goals like THIS, in a world system like THIS, and were able and did at times make action-choices like THIS to achieve those goals.

Where THIS means “the ones depicted in this particular game”. I find this standard allows for inaccurate discrete details without categorically proclaiming a game is “not historical.” It even allows some parts of the model might be less defensible. Perhaps the player agent and goals are defensible but elements of the world system or action-choices are less so. But, overall, since historical (agential) games are about agents acting within systems, this standard seems to me to get at heart of evaluating game’s modeling of evidence and “reality”. And that, as I have suggested, is something we really should do if we think (as I suggests all history educators likely do) that there are representations of the past that are better supported and worse supported by the available valid evidence.

This is rough and quick and perhaps on the way to an article depending on what you think. Please reach out and share your feedback.