It’s 907 AD. In Eoferwic, former stronghold of the Norsemen, the King of the Britons of Strathclyde, Run, marshals his warbands. In the twenty years since Alfred, King of Wessex, and Guðrum, King of East Anglia, agreed a temporary peace after the Battle of Eddington, Run has led the Britons from their fastness at Govan to smash the invading Norsemen of Northumbria and their English vassals, forge an alliance with the Vikings who now rule in Alba, and break the back of the Mercians. Now, he turns his gaze further south, directing his sons and brothers – for he is seventy winters old, and unable to lead a band himself – to prepare to strike at the oldest foe of the Old North, the last, strongest bastion of the English in these islands: the Kingdom of Wessex.
This never happened.
The name of the King of Strathclyde after the sack of Dumbarton in 870 by Vikings is unknown. While there was a Kingdom of Strathclyde, it did not overcome the new settlement of Norsemen in Northumbria, or conquer its way down to York (Eoferwic). And it certainly did not engage in a long struggle for overall control of what is now called England; while that struggle did take place (although to call it a singular struggle is somewhat misleading), it was between the English of Wessex and the Danelaw.
What I described above is not history. It is a counterfactual history produced by the mechanics of a video game – in this case, Creative Assembly’s Total War: Thrones of Britannia. I’ve alluded previously to the way that historical fiction is ‘inherently inaccurate’ by its nature as fiction, but in this post I want to explore the ways in which video games’ greatest asset – the agency they can afford players – affects their presentation of history, how that can distort the picture of the past – and how designers can leverage it to produce a greater understanding of a historical worldview and the process by which history is made.
Almost all historical games present a historical counterfactual. In some, it is the idea of a fictional character influencing key figures and events of a particular place and time (a la Assassin’s Creed), in others, it’s the idea that players can refight historical battles and come to a different result (historical shooters generally, but in terms of games this blog has analysed, the Battlefield series is a good example). In the case of strategy games, though, particularly the open-ended kind, the counterfactual runs deeper, and has more possibility: not just that one can refight battles and change the result, but that a player can step into the shoes of a historical leader and radically alter the course of history by prevailing where others failed, making peace where others warred.
This is a far deeper possible counterfactual than other games produce. In a historical shooter one can refight an engagement, but the course of the historical war on which other maps are based does not change, so the counterfactual the game can present is fundamentally limited. In contrast, something like Total War, in by and large abandoning historical particularity for the construction of a historical sandbox, allows a much greater array of alternate pasts to be constructed.
The series generally constructs a historical start position in which the player can take control of several different factions, gives the player objectives (sometimes simply to expand, sometimes to capture specific regions in the game map), and lets them have at it. Computer-controlled factions will follow their own logic, objectives, and long-term strategies in a dynamic fashion, resulting in a situation that is never the same twice. In a game set in the ancient Mediterranean, Sparta might forge an alliance with Athens and expunge the kingdoms of the Diadochi while Rome seized the Western Mediterranean. A crafty Daimyo might forge a Republic of Japan from the Boshin War. The Kingdom of Strathclyde might come back from the brink.
This is, on the face of it, bad history. It’s inherently wrong: the past is a series of events that happened, and allowing radical change of this nature means that the game is inevitably becoming less of a history and more of a fantasy. It’s also somewhat difficult to reconcile the player’s position in a game like this with any actual historical figure. Many other historical games have the player taking control of an actual individual, and even if they’re a fictional one, this is easy to reconcile with historical reality: historical fiction imagines non-existent people all the time. In a strategy game such as Total War or Civilization, though, the player takes on the role of an almost all-seeing, all-knowing commander: immortal, untouchable, capable of exactly marshalling the resources of an entire group at once, giving orders that are conveyed instantly across thousands of miles, exactly controlling multiple characters in the game. Clearly, such an entity did not exist: it’s an abstraction of gameplay.
However, what such abstractions and counterfactual achieve is producing a greater understanding of historical contingency than might otherwise be achieved. I said in the last paragraph that the past is a series of events that happened. This is a popular theory, and one that seems sound, but I believe it obscures an important facet about the past. we can characterise things that happened – major storms, volcanic eruptions, earthquakes and meteors – as events that happened, outside of any human choice and therefore without possibility for change.
But History is not about the study of things that had nothing to do with humans. History is about the study of the human past, and humans have agency. They make choices, according to their worldviews and circumstances. They are unpredictable. Every human action is taken in a web of possible actions, and in choosing one thing to do every person subtly changes the universe from one where all those other possible actions could happen into one in which they did not. History is the study of decisions humans have made. Often, these are small, or unconscious, or only visible in aggregate. It’s difficult to categorise a plague, for instance, as a result of human decisions, but without trade networks or urban centres or an appreciable laxity of pest control or hygiene, plagues would not occur – and these are all the result of human action and therefore human choices.
In this light, then, sandbox strategy games in which the player can send history careening off around a different path become a valuable tool for analysing the potentialities in historical situations. Is it likely that the Kingdom of Strathclyde would surge back and retake Britain? No. But it’s possible, given certain choices and circumstances by many different people. Total War presents a gameplay sandbox within which the player can make choices and produce these counterfactuals.
Total War: Attila allows players to hold the Western Roman Empire together against the collapse of the fifth century, something that would have drastically changed the course of history – but could have happened.
However, even in the production of historical counterfactuals there are gradations of historical plausibility and possibility. Total War, as its name suggests, is focussed mainly on conflict. Raising and commanding armies is the series’ primary mechanic. Diplomacy is often perfunctory, and there is no way for a state or faction to project ‘soft power’. There is no way to re-enact Gaius Popillius Laenas’ circle in the sand around King Antiochus, even if you’re playing as Rome. Thus the view of history the series projects is one where the main driver of history is force of arms. Even if one ultimately subscribes to this view, it’s clear that Total War’s view of history can seem a little myopic.
This brings us neatly onto the topic of how designers can shape the historical worldview their game imparts by carefully tuning its mechanics, and how the sandbox of tools offered to the player can help the player understand the predominant worldview of the time. Here I’m going to use two games by Paradox Interactive as examples, to contrast with Total War: Crusader Kings 2 and Europa Universalis IV.
Like the Total War series, these two games present a historical startpoint from which systemic gameplay can begin, and like Total War, both assume an ‘overseer’ player character who acts through individuals or institutions in the game world. However, both games take great care (arguably greater care than Total War does) to tailor their mechanics to the time period they depict: in the case of Crusader Kings, the middle ages, and for Europa Universalis the Early Modern period.
Crusader Kings can be quickly described as a ‘feudal politics simulator’: the player controls a historical figure and their dynasty, switching to the next heir when the avatar character dies. Rather than interface with armies or command nations, the player interacts with the world strictly as a specific individual. When the player conducts diplomacy, it is with other characters. Rather than selecting ‘generals’ to command forces, players must cultivate relationships with their households and select marshals. Instead of simply pressing a button to marry, players can woo potentially spouses and educate their heirs (although arranged marriages are simpler to achieve). In this, the game eloquently communicates the fusion of the personal and political that characterised feudalism as a form of political organisation: if you’re at war with the Kingdom of France, the likelihood is that it’s because of your relationship with the King, not because two arbitrary states wanted each other’s territory.
Europa Universalis, on the other hand, takes the opposite approach. Individuals exist in Europa Universalis, but unlike Crusader Kings’ full characters with wants, needs, foibles, and virtues, they’re mostly abstractions used as tools or providing statistical bonuses to the player. What the player truly controls in Europa Universalis is the institutions of a state (again, as an ‘overseer’ rather than a specific individual). The game focusses on trade, diplomacy, networks of treaties, and the development of political structures. In Crusader Kings you might be concerned with making sure your heir doesn’t grow up a weakling or a monk, but in Europa Universalis you’re more likely to worry about balancing the flow of goods through your centres of trade or developing an institutional memory in your navy so they can more effectively hunt pirates. Where Crusader Kings centres the personal, Europa Universalis puts the institutional in the spotlight.
Both of these games use their mechanics to represent a historical worldview – indeed, to present a historical argument about what was important at the time. Should the player wish to ally with, say, Castile in both games, they will go about it in different ways: in Crusader Kings they’ll send gifts to the monarch, or support his heirs and siblings in usurping him. Power flows from individuals and their support among other individuals. In Europa Universalis, the player will propose an alliance with the Castilian state, perhaps offering territories or mutual defence against other states. Power flows from institutions and political structures: discontented nobles are a statistic to manage, not individuals to keep sweet. Both of these arguments have their flaws: politics was intensely personal throughout the Early Modern period, and Medieval states often had sophisticated institutions which could function almost divorced from the individuals operating them. Both games take their view of history from their mechanics; in defining the basic methods of interacting with the world, the designers have encoded a belief system and worldview into the game.
But what these games, and Total War, reveal is a possibility of using video games as a powerful pedagogical tool. I talked in my post on Discovery Tour about how we can use video games to show people the past as it happened, but with games like this we can help people relive the past as it was made. It’s one thing to have the personal nature of politics explained – how, say, overlords could use symbolic titles to bestow status on vassals to keep them loyal – and quite another to have to navigate the subtle politics of rival vassals by carefully distributing the positions of Royal Falconer and Cupbearer. It brings history alive by guiding the player into thinking like those in the past – and acknowledging that history was not inevitable, but was in fact the result of choices that those acting upon and within those worldviews made.
That’s extremely valuable.
 For example: Battlefield 1 offers a maps depicting the 1918 Spring Offensive in which the German Army pressed the British Army back to Amiens, and maps depicting the Allied counteroffensive through the Marne, Meuse, and Argonne sectors that followed the Allied victory at Amiens – but it does not offer maps depicting a counterfactual German breakthrough at Amiens. Even in a game that takes a possible counterfactual as its basis, the scope for changing history is limited.
(For good reason: making maps for every possible counterfactual would be an absurd demand on developers).
 Each of these examples comes from a campaign I’ve played in Rome 2, Shogun 2, and Thrones of Britannia respectively.
 An anecdote from the apogee of the Roman Republic’s power: a senator, Pompillius brought a decree to a foreign king demanding he end his war against Egypt, and when the King stated he would discuss this with his council, Pompillius drew a circle around the King in the sand, stating that he King would not move until he had given Pompillius a reply to give the senate. Livy and Polybius relate this as an example of the power and hegemony Rome held throughout the Mediterranean, that a Roman could issue this decree to a King in front of his army and be obeyed merely by invoking the threat of Roman displeasure.
 In fact, they present several: each has a series of ‘bookmarks’ from which players can start which approximate the state of the game world at a given date.
 Mostly. Commanding armies and levying taxes is done through ‘impossible’ means and features similar mechanical abstractions as in Total War.
 To what extend ‘Feudalism’ as a form of political organisation ever existed, and how far one can describe the political settlements of various medieval states as ‘feudal’, and even what the word ‘feudal’ means is an extremely contested topic among medievalists; the word is used here as a shorthand.