Gaming the Past

This is the second part in a planned series of dialogues with historian, educator, and historical game designer, Vinicius Marino Carvalho, about the Historical Problem Space Framework as set of concepts and tools for analyzing and designing historical games in any medium. (Read Part I)

Vinicius: So, Jeremiah. I think I owe your readers an apology. I ended the first part of our collaborative blog post on a *massive* cliffhanger, promising to introduce a perspective that could in theory render the entire HPS framework moot.

(On hindsight, I really should make a note to tone down on the hubris…) But, more to the point, I think it’s fitting that I explain what I was talking about.

This might sound trivial, but there’s an explicit assumption in the “A.R.T.M.O.”. typology (Agents, Resources, Tools, Metrics and Obstacles) (Jeremiah’s note: it changed from A.R.T.O. to A.R.T.M.O.s for the upcoming book) that these four elements are different things. Meaning: there are entities that act, and entities that are acted upon. Moreover, in history, specifically, we generally assume that agents are human, groups of humans, or at least human-like in some capacity.

Herein lies the rub: there are some theorists who have been trying to problematize the distinction between ‘agents’ and ‘things’. You can find them in perspectives like posthuman theory, actor-network theory, animal history, interspecies history, archaeology (which deals with the agency of things), new materialism (which deals – among other things – with the agency of inanimate matter), etc.

Now, the principle that human history is impacted by non-human phenomena isn’t exactly new. The idea that material affordances shape how we make sense of and interact with the world is also a well-known problem in game design. It notably appeared in your blog conversation with Luke Holmes, in which you discussed Chris King’s principle that historical interpretations should be dictated by gameplay elements. In a way, it’s the return of the old ‘if your only tool is a hammer, everything looks like a nail’.

But these recent perspectives are trying to go a step further, arguing that humans and non-humans co-create the world together. Consequently, animals and things are not necessarily ‘objects’, and we’re not solely ‘agents’. If the human world is a collection of social ‘construct’, we’re not necessarily the ones doing the ‘constructing’.

Some games have had fun with this idea. In Nier: Replicant’s 2021 remake, for example, player character Kainé asks a machine intelligence why every interaction they have is violent, even though they are purportedly just trying to get to know one another. The machine replies (I’m paraphrasing): ‘because violence is the only word in your vocabulary’ – a metanarrative nod to the fact that ‘to kill’ is the default interaction in the game.

The visual novel Slay the Princess runs even wilder with the idea. One of the only objects are our disposal is a knife that we have to use to kill the eponymous princess. On a surface level, the blade is a tool: we need it to kill our target. However, picking up the knife also spawns the princess as a target worth killing. If we decide to get armed, the princess will appear as a villain or monster. If we approach her unarmed, she will be ‘generated’ as a much more sympathetic entity.

These are fantastic non-historical games, however, which have a lot of leeway to get away with this sort of things. But recently I’ve been wondering what this revision of the human-non-human divide means for history. What becomes of the A.R.T.M.O. typology  if we blur the hard distinction between agents and objects? Can the resulting framework still be called a HPS, or have we moved into a different kind of problem – one that is at odds with history itself?

Jeremiah: Brilliant as always, my friend! One of the reasons I wanted to write this dialog with you is because you test these ideas against the nature of reality itself, which is so important but not always my strength in my daily work with ninth and tenth graders. Let me try to respond to these points and see where we line up.

So this philosophical disintegration or breakdown of our understanding of the “REAL” world (yep, I know that real is a setup) as agents and things is fascinating. It reminds me of something I read somewhere suggesting that perhaps all matter has consciousness. Being a headline blurb, however, I never got to see what consciousness means if a rock has it and why some philosophers think this. But anyway, I’m on board with the idea that this subject – object binary is not how the “REAL” world really functions and exists, and is created.

But if I have understood correctly, I’m not sure that these cutting edge ideas actually change our core practical daily conscious operations as we engage the world. Let me set up a parallel to simple Free Will / Determinism debates that I have spent more time thinking about and you tell me if it’s not a useful parallel. Some philosopher(maybe Searle, but maybe Rachels?) once pointed out that a hard determinist in the classical sense … (as I understand it, that would be a person who argues that the current state of all the matter in the universe is pre-determined by previous states and the laws of classical physics. In other words, we have no free will but are just physical bodies dictated by physics). Anyway, someone pointed out that even a hard determinist would still perceive themselves to have free will. Even a hard determinist does not go into a restaurant and refuse to order because there is no need because the outcome is set through the laws of classical physics.

The reason this seems like a good parallel for me is that you are talking, I think more about the fundamental nature of reality. And so if our fundamental nature of reality is not what we think it is, history as it is usually thought of does not really work. To which I say, “maybe!” but I don’t think we’ll be giving up on seeing ourselves psychologically as the ego in our universe.

To get back to game design and historical problem spaces, your Nier example and Slay the Princess (love that game!) make perfect sense to me. We see the world as we are, we use the vocabulary we have and it shapes the world just as much if not more than it reflects the world. And definitely, the fact that historical games shape the evidence of the past and our interpretations into HPSs influences us and our reality. That’s one of the reasons why it is so important that we debrief and discuss with our students the limits on seeing the world as nails to a hammer, seeing the world as problems for a player agent to solve.

But I feel like maybe you are loading HPS with a status that I don’t  think I have claimed for it. Again, if I follow (this is so interesting Vinicius; wow you’ve made me think this Sunday morning!) you well you seem to suggest that HPS IS History HPS = History. Because you suggest, I think, that undermining seeing the world as full of A.R.T.M.O. (nonplayer agents, resources, tools, metrcis, and obstacles) undermines history itself? But I think HPS ≠ history is not the case, Rather HPS is a type of history, a subset of history. It is not necessarily a type limited to games: we have a lot of people (some quite harmful) that seem to see the world as a problem space in which they are the protagonist player agent and the rest of the world is A.R.T.M.O.s for their goal-seeking. I would argue that, for all the reasons you have suggested, that is a problematic understanding of the world. But however that may be, HPS  is absolutely the type of understanding wholly promoted by games at their core. So HPS does not reflect reality. Rather it is an organizational structure overwhelmingly found in historical games. And I think that is probably because games themselves (at least if we go with a working definition from Salen and Zimmerman and others of a rules based artificial conflict with a quantifiable outcome) suggest an ego – objects split.

Vinicius: This makes perfect sense to me, and it resonates with some of your previous writing and practice. Namely, about how history is a broad genre of selective representations of the past that can be found outside books and textbooks as much as within them.

I think that what exasperates me about historical games, in particular, is that there are tons of interesting stuff currently going on with historical theory, but historical games – and, to some extent, historical game studies – are being somewhat reluctant in keeping up with developments.

The more-than-human perspectives I mentioned earlier are a great example. This is a *huge* topic of research within many fields – and there are plenty of very talented people in other corners of game studies doing stellar work on that. But I don’t hear as much about that coming from historical game studies (even though history itself is slowly but gradually incorporating these perspectives).

Would there be, I wonder, a way to tweak the frameworks with which we study and design historical games to incorporate these approaches?

Regarding the HPS, one idea I have been tinkering with ever since I’ve presented my paper for the 2025 DiGRA conference, is relationality. I personally arrived at the concept via the work of anthropologist Eduardo Viveiros de Castro on Amerindian cosmology, but there are plenty of people from different fields and backgrounds conceptualizing it in slightly particular ways.

Basically, relational beings are not inherently stable, and ‘change’ their nature (or acquire new ones) based on their relationships with other beings. The more – and more different – connections a given entity establishes, the more it ‘multiplies’ its potential agential space. The bottom line, however, tends to be a decentering of (historical) human agency.

There are many ways to develop these ideas, but, in terms of the A.R.T.M.O. framework, what strikes me as useful is that it allows entities to be all four letters at once – and possible multiple variations of the same letter.

Some possible examples:

• ·   In relation to their vassals, a player-controlled faction leader in Crusader Kings III is a sovereign human agent. In relation to the plague, it is arguably part of the gamespace – as a network node it can use to hop onto new characters and spread, much in the way of the ‘tile’ of a board game. (Consequently, if we interpret a 14^th c. playthrough of CKIII as an ABM on the Black Death, the player-agent loses its centrality, even though nothing in the HPS technically changed.)
• ·   In relation to humans, “oil” can be a resource. Yet, in relation to the oil industry, *humanity* is the actual resource (or tool, if you will): we are subject to oil to the extent that we can’t get rid of it, lest our economy crashes and our society stops functioning. 
• ·   If we extend the category of ‘personhood’ to non-humans, then humans can be both sentient agents and whatever these non-humans, with their non-human sensibilities, perceive us to be (Although, as Viveiros de Castro notices in respect to Amerindian cosmology, which does operate with an expanded concept of personhood,  not at the same time)
• ·   If we take ‘witnessing’ as an action, then any game element that can ‘witness’ another (even if only in terms of tags in the code) can technically be thought of as an agent. And to the extent that the act of witnessing affects the reality of what is witnessed, each of these elements is simultaneously creating their own reality. I play with this notion a bit in my environmental history game, Galar. Its first chapter, now available as a prototype, features  a ‘double act’ of witnessing: the player-agent ‘watches’ the effects of their actions, as ripples in the ecological weave, and the system ‘watches’ the player in turn, and records their actions via progress variables.  

This can get very crazy really quickly, so I think I’ll just pass it over to you. Have you ever thought about exploring this dynamism within your framework (either at the level of analysis or game design, properly speaking)? Is the HPS really compatible with the sort of conceptual fiddling I’m attempting here, or am I just taking your ideas out of context? 

Jeremiah: I’m going to raise the question of HPS validity of existing games at the end. For sake of argument, here, (and I truly say this with humility) let’s suppose that I nailed something essential in historical game studies with insisting we understand game histories as historical problem spaces. Then you questions become broader. Not only does HPS allow for your more-than-human perspective but can historical games allow this? I think in both cases the answer is yes, but it may require us to expand what is covered by the terms of the analytical framework. Again, this is why I learn and think so much when I dialog with you Vinicius, because you push me in such interesting directions. So let’s jump in to your first part.

So first of all, I do not want to trivialize your insights (and Viveiros de Castro’s that inspired you) on relationality. And this idea, if I understand it, that relationships change in meaning and function according to connections with other entities is fascinating and, I think, has some real truth. It reminds me that just the other day after I gave a talk on historical game design using HPS that a fellow (Wilko Hardenberg) was working with his students to design a game about “environing” as he termed it  where the environment was in flux and constructed symbolically and intellectually, perhaps even physically by agents and their world views and interactions (in my effort to understand him, I brought up Scott’s work on Seeing Like a State where the state reduces forests to certain quantifiable lumber types and ignores the rest or reduces fields to cadastral maps. He said that was a good example). If I am connecting the dots well, both of you are investigating the nature of reality as a social construct dependent on the sentient entities engaging. 

Anyway, the question you pose, I think, is that the A.R.T.M.O framework in HPS categorizes but does it allow for the possibility that a gameworld element can functionally be more than one of these at once. Absolutely! And your example of CKIII and Oil in reality are great ones. I have written this somewhere (maybe GTP second edition, maybe EOLT, maybe in the upcoming book. Personally, I find this comes up the most, and your examples point to it, in how non-player agents can be functionally implemented and classified in a historical game. Non-player agents in historical games are very rarely, if ever, functioning as independent agents with their own agentialities. Rather they are resources, tools, and obstacles for the player agent, the other three major elements in the game space. This often becomes very problematic because non-player agents very often if not 99.9% of the time are  instrumentalized. Because they are modeled in game mechanics, they have functions. And since they have functions (which as far as I can see, fall into one of the A.R.T.M.Os but with the caveat I just brought up that Agents have one or more of the mechanical functions of Resource Tool Obstacle) they can be reduced by an unreflective player to their function. Turned into instruments, turned into means for the player agent ends, to riff on Kant. 

To me then, it seems, that your insights on relationality and the fluidity of elemental roles (just a vocab reminder, folks. I call all the parts of a historical problem space components and the stuff in the game world that is NOT the player agent gameworld elements: (non-player)Agents, Resources, Tools, Obstacles) provide rich philosophical depth and guidelines for the ramifications of games as historical problem spaces. Because they remind us that HPS forms a perspective around the player agent. But that is not the only perspective one could take. As a matter of course, I talk about this all the time with my students when we discuss a game we have played. Whose perspective is taken? Whose is not? What would happen if we shifted the player agent to one of the gameworld elements (the most fascinating example of this I know, which was part of the initial inspiration for my initial HPS essay is Trevor Owens and Rebecca Mir’s PlaythePast essay:  if (!isNative()){return false;}: De-People-ing Native Peoples in Sid Meier’s Colonization. He gets into the scripting for Colonization and makes Native Americans playable agents in the game.) For me those questions generally focus on picking other agents in the game to be player agents. You expand that to picking other traditionally not-tough-of-as-agential entities as agents: the other gameworld elements (plague, oil, etc.). I think that’s exciting and HPS can do it. And hopefully my case is at leas supported by my use of HPS terms in all of this.

But I do think there may be limits to how a historical game can simulate the complexities you introduce. Let me say that differently. Of course there are limits: all histories are simplifications of the past (those making them generally want them to be abstractions, i.e. simplifications that preserve essential features). Games are no different. At this point, video game developers have, I think, no meaningful way to code and design a game that does what you say, without simply turning that not-usually-thought-of-as-sentient element into the player agent (like PLague Inc.). And they can certainly do this with more flexibility. Arguably every time a marginalized person that was a simple obstacle in an earlier game is made a player agent in a new game, that is being done. So we could extend this to any entity and still analyze the game meaningfully as an HPS.

Let’s look at your examples. I think HPS enables us to meaningfully analyze the complexities you suggest in CKIII.  I completely agree with you that vassals become different things depending on the situation, Plague too. It is an obstacle and an opposing agent (there is probably little meaningful way to distinguish plague behaviors from enemy army behaviors for example) if it harms or obstructs  the player agent. It is a tool and an allied agent if it helps the player agent by harming opposing agents And since I can use HPS concepts (I bolded them above) to understand the functional mechanical behavior, I suggest that the framework holds up. But going further, HPS even seems to allow us to coherently analyze your Plague example.

See, if we do interpret CKII as an agent-based model of the plague, as you mused, I would argue the CKIII Plague “Game”  is either:

1.  No longer a game as we understand it because there is no player agent at all and no interaction once the play button is pressed

OR

2. Still a game centrally around the player agent because the “player” got to set the initial variables for the plague (even if that was the coder who hard coded them)

In case #1 HPS works as an analytical tool because its focus is on games not “pure” agent based models. In case #2 HPS works as an analytical tool because the PLague game is still a game with a player agent, just a player agent representing a non-human being. Arguably though, assuming for sake of argument that plague is not sentient, the player makes it sentient by making choices about the starting values of plague variables and behaviors (in this hypothetical CKIII turned into a plague ABM).

And then the same is true from there. The possibility of  non-human sentient agents and expanded personhood that Viveiros de Castro proposes and you have developed in games. Humans as resources. Trees as witnesses or the earth itself (and haven’t games like Sim Earth done some of this – note: I never played it)

Does that provide a reasonable game version of the relationality you are seeking to explore and express, Vinicius? So, one can make all these games about plagues and oil and human resources. But, I would argue, that to fit what we understand as a game, there really can only be as many player agents simultaneously as there are human players. And in a single player game, there is only one player agent at a time. We’re only human as players, right? And even if our fields develop into something more expansive like you suggest, I don’t think it will change that many people will want and understand games with gameworlds to focus a single perspective at a time. So HPS perhaps is limited by this parameter. It is possible, for example, for the computer playing the game to have thoughts on the matter but the only perspectives we are currently able to be aware at the moment are the ones we play as or discuss playing as? I put a question mark because this is definitely more a question than a certainty.

Okay, let’s shift a little but stay in the same area. You posed another set of questions along the same lines at the same time as your part above, but we separated them to better keep dialogue. 

So now let’s go back to the question of structure vs. change in historical games and share with readers your current (as of the time you wrote this of course) thinking on this.

Vinicius: Okay, about historical games: I admit that thinking about the structure vs. historical change conundrum has made me somewhat disillusioned with them – or, at least, with a certain way they tend to be invoked and designed.

Let me see if I can explain how I feel. As I see it, history is not just about models, but also about how we interact with them. Or, as post-structuralists would say, structures don’t exist independently of our observations. We got to take ourselves into account too.

We don’t simply chronicle facts or summarize systemic features. Sometimes we move backward, returning to the causes or origin of a phenomenon after assessing its consequences. Sometimes we ditch chronology almost entirely and divide our texts by theme. We zoom in and out of the historical fabric, either focusing on structures, either on local associations of people or even individuals. We move up or down the scale of abstraction, mix-and-matching concepts with things that exist experientially.

 All of this, too, is “history” – or “historying” as Adam Chapman once called it. And the possibilities this kind of operation offer us are immense. I dont think philosophers of history are being hyperbolic when they write that there’s a multiplicity of historical pasts, of futures, and of temporalities in general. Our snapshots may be limited in number (and in complexity), but there are so many ways we can assemble them that the possibilities are almost limitless.

Now, games *can* do all this, and there are some who argue, quite convincingly, that they are the medium par excellence in which these experiments can be tried out.  

But I have the feeling that historical computer games often treat this as an exploit, not as a feature. E.g. We scoff at the idea that time travel can be historically valid, but save scumming (which is literally time travel) is allowed (albeit tempered by Ironman modes and the like).  We don’t factor immortality or resurrection into our explanation schemes, but respawning doesn’t raise as many eyebrows.

Yet, if we all agree accuracy is overrated and games are constructed systems, why not go wild and incorporate these player agencies – which can have immense analytical value – into our simulations as well? And if so, would there be space in the HPS for them?

Jeremiah: I completely agree with how you’ve put this. As much as I have designed a structuralist approach to analyzing game mechanics ( it just sort of happened that way), I have always thought HPS ultimately is descriptive and subjective because it ultimately talks about how “we” generally observe, perceive and design historical games. You do a great job problematizing those conventions here, I would note! In the upcoming book, I shift to prescriptive because my hope is to show how researching a historical problem space blueprint before game design can help craft a historical game with defensible models of the past. But still that prescriptive turn is based on describing how the current corpus of historical games seem to work, and that is based on my perceptions.

So just jumping into your big question in this part: can HPS accommodate and meaningfully help work with these kinds of play? So things like save-scumming and respawning? When you said this, I immediately thought of Jesper Juul and Half-Real (2005), where, if I recall correctly, he refers to Mario having three lives in Donkey Kong as “incoherent”:

“We call this type of fictional  world incoherent,  meaning that there are  many events in the fictional  world  that  we  cannot  explain  without  discussing  the  game  rules”  (Juul 2005,  p.130)<- and I took this quotation from a discussion of Juul here I just found in Hogenbirk, van de Hoef, and Meyer 2018->  View of Clarifying Incoherence in Games

Now, I have to read this article because they offer a critique, I gather, and I’m fascinated. And they may already point this out. But I do see why save-scumming, for example, could be seen as incoherent with the game world, following Juul’s thoughts on Mario. And I don’t think Juul is mistaken here – there really is no within-game effort to explain Mario’s three lives in Donkey Kong. But there could be. You propose, Vinicius, that the save-scumming could in fact be not only part of the game (which it ultimately is in a problem space if it becomes a part of our goal-seeking in the game) but be a coherent part of a historical game and how that game approaches its topic. As you say, save-scumming = time travel. regenerating health = resurrection or immortality (or super pharmaceuticals from the future). Fascinating! (also reminds me of the undo button that Soren Johnson put into Old World that is part of gameplay but he apparently felt no need to narratively integrate it. I didn’t need him to either; I just appreciated the chance to take back my in-game screw-ups).

Honestly, when I first started musing about historical games as historical problem spaces (Historical Simulations as Problem Spaces: Some Guidelines for Criticism), I was not proposing a framework. Rather, I commented on the functional cohesive nature of  historical games (optimally) and, therefore, the close relationship between how each historical phenomena is represented mechanically in a game with the functioning of all the other mechanical phenomena in-game. As I’ve developed that idea into something more useful, I have definitely listened to scholars like you pushing and prodding on it in constructive and powerful ways. So, at this point, I think that the HPS framework does allow for the kinds of analysis you are suggesting. For me, this all depends on the scoping of the gameworld, its limits. Many of us have thought of the scoping of the gameworld as delimited to the parts rendered onscreen or in the boardgame components on the table. This would also fit, I think, Adam Chapman’s (I think he coined this term for games ?) diegetic level of the game: the game features that are part of the narrative world of the game. HPS terminology already enables analyzing at least three ways that the diegetic level and the gameworld as just what is onscreen or on the table are not coterminous. The first is the division into explicit and implicit space: space actively portrayed and able to be occupied by gameworld elements vs. space that must exist diegetically for the coherence of the gameworld but is not “playable”. So, the terrain hexes and their borders on Settlers of Catan and the tiled terrain between cities in Civilization I & II are explicit spaces: both can have game elements placed on them. Then, the actual architecture, streets, people, and space, which is not represented or able to be interacted with by game elements (units etc.)in Catan or Civilization I and II, etc. The other is explicit and implicit agents: agents that are actively portrayed and have mechanical functions in the gameworld and those that must exist diegetically for the coherence of the gameworld. To use the city examples again, one cannot identify or directly interact with the denizens of the cities in either game. They must exist, though; a city must have city dwellers. The final distinction is between embodied and unembodied player agent. Embodied player agents occupy explicit space (i.e they have a spatial avatar). Think of the player tokens in Monopoly, or the high command unit in wargames that have them (historical board games actually don’t have embodied player agents as frequently as video games). Or think of the avatar in any fps or 3rd person game etc. Unembodied player agents, on the other hand,  do not occupy explicit gamespace. Think of the commander in Risk, or the Colonizer leader in Catan or, in video game, real time strategy leaders, civilization leaders in Civilization and so on. 

Since explicit/implicit spaces, explicit/implicit non-player agents and embodied/unembodied player agents are all analytical concepts in HPS, I think the framework already allows for regenerating health. It is a mechanic in the gameworld and an element in the problem space. I would maybe  classify it as as a tool: an in-game element that enables the player to do something.  And regenerating health is worked in, in a well designed game, to the user interface and is part of the user experience. It’s just not usually incorporated by the designer when describing the gameworld. So when describing Call of Duty: World War II the focus is (usually) on Normandy and infantry, and regenerating health is not spoken of, though it is a part of the game. So there, a designer would just need to lean into the regenerating health, if they chose, and work it in diegetically. I’m not sure what I think of that as a historical game with defensible models, but HPS would support it.

I think it also allows for save-scumming similarly. It is an out of explicit space and embodied agent action, usually considered out of game altogether,. But if the designer leaned into it as part of the gameworld (scoped the gameworld to include the save-scumming as a power like pointing and clicking on units from outside the gameworld in an RTS), it already has a mechanical function and could be explicable in HPS terms. It would function as a tool, I suggest, an ability that allows the player agent to do something they could not do without it.

***

So it seems to me, beyond just reading and learning about all the fascinating thoughts and thinkers you’ve been mulling over, Vinicius, that your question is whether the Historical Problem Space Framework can accommodate this kind of more-than-human perspectives? And to tackle that question, I start to wonder about the connection between HPS and historical games. As you know, I developed this framework from a simple observation some time ago that critiquing discrete elements of historical games as if they were discrete elements was problematic because, in fact, all the mechanics of a game are interconnected in a functioning system. (The original inspiration was my colleagues Trevor Owens and Rebecca Mir’s  excellent analysis of Colonization 2008 and the portrayal of American Indians and avoidance of enslaving African Americans – Sid Meier’s Colonization: Is it offensive enough?, Playing at Slavery: Modding Colonization for Authenticity ). So to fully understand designed in-game  phenomena around the historical portrayal in mechanics of this or that thing (American Indians, avoidance of plague and enslavement, or in the case I gave the portrayal of enslavement in Hegemony: Philip of Macedon), one has to appreciate that:

1. This or that particular phenomena expressed in mechanics is part of a cohesive functioning system (at least in what we tend to term a successful game).
2. That functioning system is ALWAYS (?????) in the form of a historical problem space: a player agent pursuing designed goals by making action choices in a gameworld.

I still keep coming back to this, and it’s a problem that is hard to solve and, honestly, may not be important. Would we understand a historical game to be a historical game if it were not a problem space? Would it matter? Is the HPS tied to conventional definitions of game so that it really will be the case that any historical game that meets a certain definition of game will be in HPS form? I mean, as far as I can tell, all existing historical games are problem space (but inductive reasoning can fool me!). Is that a Western bias toward what a game should be? These questions fascinate me.

What do you think about all these claims of flexibility for HPS, Vinicius? You pointed out once when editing a chapter of mine, that the actions one most commonly takes in historical problem spaces are quite colonial in baggage (occupying, exploring, contesting, controlling, exploiting, developing). Are you persuaded that HPS can accommodate these innovative directions you would like to see history games (and history) go? I think some of this question is caught up in the limitations of any media in conveying lived experience. No medium can do so perfectly. HPS is certainly tied up in conventional definitions of a game (because that was clearly important to me at a level I did not even feel the need to articulate much original). Can we keep those conventions and do “good” history games? Is there a medium like games that allows for continual player interaction in a gameworld but is “better” at capturing the areas you have talked about in this section? Or is it, like an agent-based model after any user values are entered and no goals, not really a game? Are historians better off abandoning games when something better comes along? I’m looking forward to your thoughts on these and whatever else strikes you as we talk more!

End of Part II