Simulating Life

Political Game Design

Pierre is a political scientist and programmer. After studying at the Institut d’Études Politiques in France, he pursued an MFA in Game Design at NYU before starting to write code for money and fun. His work stretch from purely formal experiences to a close analysis of how programming defines both users and developers.


It is useful for me to think of games as simulations. Although not all games are explicitly simulations, they all come to life by representing some activity as a prerequisite to being able to afford the player to partake in that activity. While this activity implies some sort of closed environment–be it through physical ability or problem solving–it is also possible to use this simulated aspect as a way to open up to the world around us and, as we will take a closer look here, to politics and political systems. This article aims to expose the inherently political aspect of game development. Having gone to school for both political science and game development, I have seen political questions appear over and over in the design and development of both my games and those of my colleagues. By taking the particular topic of political philosophy of the Enlightenment, I hope to show how computer simulations can help us in acknowledging the politics in designing and developing games.


An agent-based model (ABM) is a way of, if not solving a problem, at least putting together several recognized features into agents to see how they interact together and give us global, emergent patterns to interpret, without hard-coding these patterns in the first place. Some of the current research is, for example, in sociology (1) where it is the interaction between agents themselves that is being looked at or in international relations (2) through aggregates of agents.

Along with the development of computers, ABMs allow us to think about situations on a larger and larger scale, as they prevent us from missing certain steps between the original facts and the ‘emerged’ result, not being able to imagine these ‘emerged’ results or to figure out how slight changes to our starting assumptions could lead to changes in the ‘emerged’ result. Those simulations, through computing power, allow us to test out these assumptions. They allow us to directly assess the consequences of our choices without doing the heaviest lifting, but also provide a fresh look on those assumptions as we set them before running the simulation.

As developers and designers, when we first set out to build a simulation, we have to focus on what it is we want to explore in it. When doing this, we deconstruct it so that its definition is a bit closer to an objective, machine-readable statement, and a bit further from a human, subjective formulation. The definition that is then embedded in the machine is now irrevocably subjective; it is the objective expression of the developer’s subjectivity. For example, defining democracy as a concept for everyday use and democracy as a concept meaningfully embedded in a piece of software yields interesting results:

democracy, n.: a system of government by the whole population or all the eligible members of a state, typically through elected representatives.


public static int decisionPower = 1;
    for(int i = 0; i < society.size(); i++){

        Agent a = society.get(i);

        a.decisionPower = decisionPower;}

Once the simulation runs, it presents us with a heartless, uncompassionate result of the interactions we wanted to explore. We established rules and these rules gave us results. And, because we externalise the thought process, we can look at the results from a different perspective. These emergent results, then, can be traced back to our assumptions until they become the logical, coded results of a given process. They shed a new light on our assumptions, by showing us the consequences of these assumptions.


This isn’t about politics in general, because politics in general is both a very vague and misunderstood notion as well as actively despised by those who associate politicians with politics. The politics of the Enlightenment are both a very specific subset of political studies but also what I consider to be the most important aspect of politics. Politics of the Enlightenment is arguably the next very relevant branch of political thought that came after the classic philosophers, starting around the 16th century until the 18th century. It’s not a coincidence that they are contemporary to the formation of modern Western democracies as political systems. It is also the branch of political philosophy which has had the most impact on our lives. Since digital games are consumed more massively than any other media product, they also have the opportunity–and all of them do–to have an impact on our lives. If all game design is political, then looking at them through the lens of Enlightenment philosophy is the easiest way to make that feature stand out.

Thinkers such as Hobbes, Locke, Smith or Rousseau are all at the beginning of our modern understanding of society. This is because they were the first ones to start thinking about what became known as social contract theory, which boils down to these two questions:

How is it that we manage to live together?

Is this the best way we can live together?

As a matter of fact, John Locke, considered the father of classical liberalism, had a significant influence on the US Declaration of Independence (3) , while, Rousseau’s ideas provided a specific framework for the French Revolution (4) and the drafting of the declaration of rights of Man and of the Citizen, which itself is a fundamental document in the history of human rights (5). Political philosophers of that era heavily influenced the way we define living together today.

However, they only proceeded through thought experiments, and thought experiments imply a subjective start, leading to an objective end, formulating assumptions about individuals, and deriving laws to reach optimums. Even though logic is a significant branch of philosophy, it is stuck in dealing with very abstract, very high-level concepts, simply because our brains cannot consciously process the global outcome of so many details. Kant himself, as he wrote in Idea for a Universal History with a Cosmopolitan Purpose (6) , was aware of the problem of theorizing without empirical basis, which only ends up being speculative psychology.

We start to see how these thought experiments can easily be replaced by computer simulations. They might lose their theoretical weight, but gain in practical application while allowing designers to shift the focus of their thought process.

In parallel, the work of political philosophy closely relates to game design, in that whenever you design a game, whenever you design a system, you create a world of politics for players to live in. If you agree that our institutions–the ones that are currently in place–dictate our lives not only on a functional level, but also on a moral level, then that implies that you, as a designer, are imposing certain moral behaviors onto your players, consciously or not.


One of the things artists do (and I count designers among them) is they use the most recent technological tools as a medium, usually subverting them to reveal a previously overlooked part of the world. In my opinion, games, re-appropriating simulations and extracting them from their scientific environment, go beyond that mere subversion, and engage on a more aesthetic and emotional level with their audience, that is, their players.

The transition from scientific to artistic allows for more freedom in terms of the framework in which it is being used (instead of being limited to the strict standards of academia and the research community). What I understand by artistically representing a simulation is removing it from peer reviews, references to previous work, or how it fits in the current paradigm within a given community: all constraints to the statement an author would wish to make. How, then, can we represent a given set of human behaviors in a game? And the corollary: what consequences do those representations imply both during development and as a finished game?

By forcing very loosely defined concepts (survival, wealth, culture, which were all at the heart of political philosophy) into a hyper-rigid framework (code) there emerges a tension that yields interesting results. For example, when dealing with ABMs, we have to think about what constitutes an individual as a self-reliant agent (i.e. without external or artificial structures). Using object-oriented programming, we create a new object named Agent, and what arguments do we give it? What are the core components of what constitutes a human? That is, what arguments represent a human in theoretical terms as well as in practical, game design terms? None of those choices can be neutral, and yet all have to be made for the player to interact with a game. Some of these representations might be obviously political (do characters from different genders have the same stats?), some might not, but they are all statements made by the designer and developer.

So instances of the Agent class need a position, a spatial representation, two floating numbers for a 2D space, but they also need a shape. What shape are they? Are humans essentially all the same, or all different? Do they inherently have different sizes and aspects, or do their sizes and aspects start the same at the beginning, and then change over the course of the simulation? All these questions are political questions, as they are in themselves forcing the designer/developer to take a stance on a political issue (i.e. ‘Who are we?’).

It would be easy to answer all these questions based strictly on a game design point of view, but designing something as fun as possible usually comes to the detriment to designing something as politically conscious as possible, because it demands more time and effort than just designing a fun game (and, incidentally, politically-conscious games sometimes end up being not fun at all).

Another example would be seeking wealth. In the simulation, what are the conditions that must be met for an agent to gather a resource? Once he or she has located this resource, what is the interaction that happens between the agent and the resource? Is the resource acting upon the fact that it has been found by an agent or is the agent acting upon the fact that it has found a resource? In this particular instance, we can see a clear stance between an ecological line of thought and phenomenological thought (ie. what is the most important: the human or the resource?), This debate, alive and well in modern philosophy, has found in games a natural medium for expression, with as many arguments as there are designers.

There is an extensive amount of literature in the field of software studies, and yet this is one of the most under-represented aspects of politics in game design and development. As an example, the way the AI works in The Sims 3 is explained in a very well-titled paper: Dumb People, Smart Objects (7). This approach is very much a political representation of people as dependent on their material environment and their material belongings.

Scientific simulations, the ones used in research laboratories in universities around the world, are differentiated from game simulations in that they have a pretension to truth. Even though they rely on previous assumptions, as well as on the personalities and biases of the researcher, they try to make the results of these simulations appear as reliable and explicit as possible. While the scientific simulation is expected to make a rational statement, that is, as independent as possible from the scientist’s assumption, designers are free to use these tools and imbue them with their particular worldview to communicate to their players through the phenomenon of computer-facilitated emergence. As an example, the faking of data, or variable tweaking, if greatly frowned upon in the scientific community, is a well-accepted practice in game design. Variable tweaking is a subtle way to change a virtual world in order to conform it to certain expectations, either from the developer or the player. As a consequence, it is easier for the developer to open up a debate about that worldview than it is for the scientist, where only scientists can refute that view through equally scientific methods. A game simulation is an explicit opinion, even though it might create a whole world based on that opinion.

To finish, let’s take a moment to talk about elements of game design that are not strictly-speaking simulations, but still very much political. The one big opposite (but not excluding) is the use of story and narrative in games. The artistic concept of simulation gives us the possibility to integrate a more intimate component to the scientific, logical process. Art, and design as art, is about interpretation, and the thing most prone to interpretation is the story, as written or spoken word. The simulation as a game, and the simulation as an art form, not only allows for different modes of expression, be it systematic of visual, but also semantic, multiplying the number of perspectives a designer can provide on a particular issue. Stories in games are not limited to their self-sufficient existence, but can also provide a gateway to understand other rhetorics at play in interactive software.

By exploring our own assumptions regarding political matters, we can shift the focus of that exploration from the consequences, computed for us by the machine, to what it actually means to define these assumptions, and to allow players to interact with them. Because everything you design, code, and present to the world is undeniably political, whether it is obvious or not.



***This paper has been written in parallel with the development of Social Contact, a digital simulation of political philosophy, downloadable here:***

(1) EPSTEIN, J., Modeling Civil Violence: An Agent-Based Computational Approach, Proc Natl Acad Sci U S A. 2002

(2) CEDERMAN, L.-E., Emergent actors in world politics: how states and nations develop and dissolve, Princeton Press, 1997

(3) PANGLE, T. L., The Spirit of Modern Republicanism, University of Chicago Press, Chicago, 1988

(4) ROBISCO, N.-B., J.-J. Rousseau et la Révolution française : une esthétique de la politique (1792-1799), Annales historiques de la Révolution française, 1993

(5) FLOWERS, N., Human Rights Here and Now: Celebrating the Universal Declaration of Human Rights, Amnesty International, 1999

(6) KANT, I., Idea for a Universal History with a Cosmopolitan Purpose, Yale University Press,

7 TIRRELL, J., “Dumb People, Smart Objects: The Sims and the Distributed Self”, The Philosophy of Computer Games Conference, Madrid, 2012