Tuur Ghys

Tuur Ghys is a cultural scholar, trained as a researcher in science & technology studies at Maastricht University. He is currently doing his PhD at Antwerp University, in the field of social innovation targeted at poverty and social exclusion.


Technology Trees: Freedom and Determinism in Historical Strategy Games

by Tuur Ghys


Technology trees are evolutionary tree diagrams that simulate the progress of technology in historical strategy games in a deterministic manner. Having the double function of representing the history of technology while being a core mechanism in the game itself, the tech tree plays an important role in the overall game design. This article covers the question of how the tech tree determines the possibilities for representing the history of technology in historical strategy games, using the concept of technology determinism. In doing so it tries to bridge the fields of STS and game studies. I discuss and compare four important games in the genre (Age of Empires, Empire Earth, Rise of Nations and Civilization IV); and present insights from interviews with the lead designer of each game.

Tech trees come in the shape of linear upgrade paths or interlocking vines structures, and fulfill various strategic and narrative functions in the games. Technological determinism is present in tech trees under three forms: by forcing a set sequence, by influencing social changes in history and by characterising (thus determining) eras and civilizations. On the level of individual technologies it becomes clear that besides the broad concept of technology, there is a lot of freedom in how the tech trees are filled in. Both the selection of technologies and the effects attributed to them are different in all four games, which makes four interesting case studies possible. The conclusion explains how tech trees determine the design of technological history in a general way, but that there are many ways for both designers and players to interpret history in their own way.

Keywords: technology, tech tree, determinism, strategy, history of technology, comparative research 


In historical strategy games, in which the player controls a nation that progresses throughout history, the developers implemented a technology tree to help simulate the progress of technology from stone tools to nanotechnology. A technology tree, or ‘tech tree’, can loosely be defined as “a structure that controls progress from one technology to a better technology, enabling the player to create better facilities or more powerful units” (Morris & Hartas, 2004, p. 141). It is a rule set of certain premises that have to be fulfilled to unlock a technology, which then has certain consequences, often including unlocking the path to newer technologies. Little has been published on the topic of technology trees, although their technological determinist aspects have attracted some scholarly attention (Macdougall, 2009). Despite losing credit in the academic world, the idea that technological innovation drives social progress and follows an inevitable course can still be found in popular culture (Wyatt, 2008), including digital games. Although scholars from the field of science and technology studies are aware of this presence, they have not yet laid bridges to fields like games studies to see how the determinism operates in concrete cultural artifacts like videogames.

Determinism is more than a pitfall in historical thinking, when embodied in a mechanism like a tech tree it can form a script that influences the design and content of popular culture. This article covers the question of how the tech tree determines the possibilities for representing the history of technology in historical strategy games. The technology tree plays a double role in these games: First it is a mechanism or technology itself, functioning as a system of upgrades within the game. Second the tech tree represents technologies and technological history. Using the terminology of Salen & Zimmerman (2004, p. 6), I will view the technology tree both on the level of ‘rules’ and that of ‘culture’, both important to how the game is designed. This article is not player or ‘play’ oriented, and focuses more on the design and production than the reception of the concerned games. It will show how at least three forms of technological determinism are present in the structure of the technology trees, but on the level of individual technologies a great freedom for historical interpretation is still possible. 

Selection, Methodology and Literature

I compare four iconic historical strategy games: Age of Empires (Ensemble Studios, 1997), Empire Earth (Stainless Steel Studios, 2001), Rise of Nations (Big Huge Games, 2003) and Sid Meier’s Civilization IV (Firaxis, 2005). The first reason for comparing multiple titles is that it broadens the discussion to historical strategy games as a genre instead of just one case. The second reason is that it allows many interesting comparisons, among others the difference in which technologies are selected to feature in the tech trees. For selecting my four titles I used the following criteria: 1) the game must represent the progress of history over multiple time periods; 2) the complexity and importance of the technology tree; 3) popularity and relevance to the history of the genre (all selected games are ‘bestsellers’). I focus on the offline game experience. The four titles can be sorted in many different ways, for example Age of Empires (AoE), Rise of Nations (RoN) and Empire Earth (EE) are Real Time Strategy (RTS) games in terms of gameplay, while Civilization IV (Civ4) is a Turn Based Strategy game. Furthermore there is a distinction between games that tend towards the simulation of history (Civ4, RoN) by counting on historical mechanisms to drive (alternative) history, and games that tend to re-enact specific historical events (AoE, EE), although all are simulations to a certain extent.

For this research I drew upon two important sources next to general background literature on game design and the history of technology. The first source is my own extensive analysis of the games themselves by playing them offline (160 hours in total) and studying the structure of the tech tree. The second source consists of e-mail interviews with the lead designers of the games in question. In chronological order, I interviewed: Soren Johnson, who was lead designer of Civilization IV (2005) at Firaxis; Rick Goodman, lead designer of both Empire Earth (2001) at Stainless Steel Studios and Age of Empires (1997) at Ensemble Studios, who was separately interviewed for each game; Bruce Shelley, who worked on many big strategy titles but was interviewed in the context of his position as co-lead designer of AoE (with Goodman); and Brian Reynolds who lead the design of Rise of Nations (2003) at Big Huge Games.

This research is of an exploratory nature as a consequence of its subject matter: outside the blogosphere little to nothing has been published on the combination of technology trees in strategy games and the history of technology. The studies that come closest to the research presented here are those on the educational value of games (like the Civilization series) as macrohistory simulations (Burns, 2002; Squire & Jenkins, 2003; Squire, 2004). Other research investigates the Civilization series from an ideological perspective; its imperialist ideologies have been critisized by, for example, Douglas (2002), Kapell (2002), and Poblocki (2002). Henthorne (2003) discusses the utopian ideological aspects of various strategy games including the Civilization series, and the political implications of Age of Empires and Civilization are also discussed by Weib (2007). It is not because little has been published on the subject of tech trees that an academic debate is nonexistent. As he recounts on his blog, Rob MacDougall (2009) explicitly addressed the determinist assumptions of the games in his lectures. Other scholarly bloggers before him have highlighted the latter point (Goodfellow, 2005;  Owens, 2009). I want to refresh this debate by comparing multiple games and by paying attention to the technologies in the tech tree themselves instead of just looking at their relation.

The functioning of technology trees

Apart from some isolated cases (see Basalla, 1989) there is no real tradition of depicting technological progress in evolutionary tree diagrams. In 1980, a technology tree first appeared in the board game Civilization (Hartland Trefoil, 1980). Game designer Sid Meier was the first to create a technology tree mechanism in a video game with Sid Meier’s Civilization (MicroProse, 1991). Obtaining new technologies by investing resources, thus ‘moving up the tech tree’, can increase the attributes of troops and buildings (speed, defense, etc.), allow the construction of new units and buildings, or give the player new abilities. Besides this, technologies are often linked to each other, so researching a technology can give access to a more advanced one. What is generally called a tech tree consists of both buildings, units and technologies. I call the latter the upgrade tree, and from here onwards reserve the term technology tree for the dependencies between technologies. Mathematically speaking tech trees are directed graphs. Either they have a ‘vine’ structure in which multiple technologies are required for a new technology, and/or a technology can lead to multiple new ones. Or they have a structure of separate branches that follow linear ‘vertical’ upgrade paths, each technology leading to a new one. With vertical links I mean connections between old and new upgrades of the same type, by horizontal links I refer to connections between different upgrades but within the same timeframe.

The Age of Empires (1997) upgrade tree actually looks like a real tree diagram, because a building tree underlies the tech tree (new buildings allow new technologiesclick here to see Figure 1)). The tech tree itself however consists of linear, vertical upgrade paths. The same is true for Empire Earth (2001), which has fully separate branches of technology, all vertically linked (click here to see Figure 2). Rise of Nations (2003) has a very interlinked tech tree (click here to see Figure 3), with cross-branch links (horizontal) and upgrade path link (vertical). A distinction must be made between the 40 library techs researched at the library and the 43 other technologies. The five branches of library technologies are independent of buildings, only having vertical links to each other, but serve as (horizontal) conditions for the other technologies and buildings. In Civilization IV, technologies are linked to each other in various ways (click here to see Figure 4). The first way they can be linked is directly (vertically), like bronze working leading to iron working. A second type of link is when technologies have two requirements (‘and-ports’), like the compass needs iron working and sailing. Then there are ‘or-ports’, technologies that have multiple possible premises. For example animal husbandry can be reached by either hunting or agriculture. Lastly there are and/or ports, in which technologies have one vast requirement and one additional optional one. Noteworthy, in Civilization IV all buildings, units and other options come directly from the tech tree, meaning that they can only come into play when a new technology is researched (thus the tech tree is the upgrade tree). The Civilization IV tech tree looks more like a vine and doesn’t have any clear branches.

What is all this botany good for? Technology trees fulfil a number of functions that make them interesting for game designers, of which I will only cover the most important ones here. First of all the tech tree forms a sort of decision tree, giving the player choices which add strategic depth to the gameplay. Tech trees that have a vine structure of interlocking requirements, like in Civilization IV, are strategically interesting because they force the player to plan far ahead to reach certain technological goal. However (linear) technology trees do allow specialization within one game session, as Rick Goodman argues in defence of AoE: “I wanted to enable players to have the choice to go further down one path and, even ignore the other paths. This gives players meaningful choices in strategy” (interview Goodman, May 12, 2011). One of the most important design purposes of a tech tree is to act as a release mechanism: new units, buildings and abilities slowly become available through playing when the player researches technologies. This comes in handy when building a campaign, as each scenario can introduce new parts of the tech tree (in line with moving further down the timeline). Connected to this is the function of tech trees as reward systems (see Jakobsson & Sotamaa, 2011): they give the player something new every now and then if he/she keeps playing. Technology also helps to give one player an advantage over the other, providing him/her with momentum to decide the game. Morris & Hartas point at another virtue of the tech tree: “...it gives structure to the story of the level” (2004, p. 73). The tech tree functions as a narrative tool on three levels in skirmish matches, that would otherwise lack the scripted story of campaign scenarios. On a macro level the player advances throughout the whole of history in one skirmish match, and the tech tree contributes to this experience by telling the story of technological progress. Especially in games that deal with alternative history this is important, as the tech tree offers a supportive framework that keeps things in parallel with real world history and recognizable for the player. On a meso level it provides each game with a clear structure of a humble beginning (a few root techs), eventful middle (many choices) and mighty end (accumulated abilities). On a micro level, the tech tree creates immediate goals (and drama) for the player.

So far I have kept the concept of technology unspecified, referring to it as upgrades. To avoid theoretical discussions on the nature of technology, I will adopt an ‘actor's’ concept of technology, which means that if the games (do not) call something a technology in their tech tree, I will (not) consider it to be a technology. When one looks at the tech trees of these games it becomes clear that they all use a very broad concept of technology, far surpassing that of mere machinery, which only makes up 30% of all the technologies featuring in the four games. We find machines (steam engine), techniques (sailing), sciences and bodies of knowledge (physics), abstract and religious ideas and rituals (polytheism, philosophy) and forms of social organisation (guilds, feudalism). The latter examples can be called ‘social technologies’ (no relation with recent network technology), and are quite prominent in all games, especially in Rise of Nations. Lead designer Brian Reynolds defends his selection: “You want to have a game be about social & economic developments as well as scientific ones” (Interview Reynolds, May 27, 2011). Soren Johnsen from Firaxis shares this thought:  “I believe that social developments are no less important than scientific ones and wanted that to be reflected in the tech tree” (Interview Johnson, May 10, 2011). The broad concept of technology in historical strategy games comes close to a constructivist approach of technological systems consisting of objects, processes and knowledge (Bijker et al, 1987, pp. 3-4). This broad inclusion of technologies helps to soften the technological determinist aspects of tech trees that are examined here.

Three sides of technological determinism exposed

There is some recent use of the term technological determinism in videogame studies (Bogost, 2009), but it originates from technological history and generally refers to: “the belief that social progress is driven by technological innovation, which in turn follows an ‘inevitable’ course” (Smith, 1994, p. 38). Nowadays this belief is considered a fallacy in the academic community, but is still widely diffused in popular culture (Wyatt, 2008). The technology tree can be seen as a representation of this idea because it is a driving force of history within the games and the course of technological progress is predefined.  We will explore the technological deterministic aspects of the tech trees in the four games from three angles. The first two come from Heilbroner’s (1967) classic two sided conceptualization of technological determinism. One side is the fixed sequence of technological development and therefore a necessitous path of progress, and the other side is that technology imposes certain social and political features upon society. A third perspective is the idea of technological determinism as characterisation (Wyatt, 2008, p. 167), based on Hannah Arendt’s remark that “tools and instruments are so intensely worldly objects that we classify whole civilizations using them as criteria” (1958, p. 144).

Technological determinism as the set sequence of technology is clearly present in the structure of tech trees in all games, and is made explicit in some: “Tools and other technologies were cumulative in nature. Cultures had to master the preceding technology to proceed and advance” (Age of Empires manual, Ensemble studios, 1997). The exact selection of technologies can differ per civilization or follow the same track for all, as is the case in Empire Earth. However the strength and set sequence of these links is not as strong in all games. In Civilization IV a technology can have multiple prerequisites and form meaningful connections, while in EE there are thin and seemingly non causal links between technologies, only preventing the player from simultaneously researching technologies with the same effect. Soren Johnson has no problem recognizing that Civ4 follows a deterministic model of history, but replies that:

“When making a game, players need a strong sense of deterministic progression. If they invest X turns researching gunpowder, then they can spend Y turns making musketeers, which will give them a big edge in combat by a certain year. These logical chains form the backbone of the gameplay, by giving the player a reason to choose one path over the other instead of flailing about randomly” (Interview Johnson, April 14, 2011).

Sometimes Civilization IV breaks its own determined order by having ‘or’ ports in its tech tree, for example both flight and artillery can lead to rocketry. To get an idea of what this deterministic orders implicate, it is worthwhile to look at the chains of necessity beyond the next step, for instance with mysticism in Civilization IV: mysticism is required for meditation; meditation is required for philosophy; philosophy (or divine right, which also needs mysticism) is required for nationalism; nationalism for constitution; constitution for corporation; corporation for replaceable parts; replaceable parts for industrialism; industrialism for plastics and plastics is required for robotics. No robotics without mysticism?

It is hard to talk about a real social sphere in strategy games, starting with the fact that the player has godlike control over his ‘serfs’, but there are still ways to examine technology as driving social change. Civilization IV comes closest to representing society, having a system of 20 ‘civics’, which are forms of organisation in the domains of government, labour, etc. These civics are unlocked by technologies, for example bronze working (tech) allows slavery (civic). Rise of Nations also has a branch of civic technologies, but these do not lead to new forms of government. Another sort of technological determinism lies in the inclusion of social and political principles in the tech tree, and the links in which ‘mechanical’ technologies lead to ‘social’ ones. Some links are quite subtle, for instance in Civilization IV there is the seemingly strange requirement of needing either plastics or fission to research environmentalism. I asked lead designer Soren Johnson about this puzzle:

“It is a cause-effect relation. Both of those connections are supposed to signify developments which would push people to consider being more ecologically aware. For example, plastics led to a boom in disposable items, which eventually made people more sensitive to how wasteful we are as a civilization” (Interview Johnson, May 10, 2011).

The connections in the tech tree do not always go from mechanical to social technologies. For example paper, a mechanical technology (because it refers to an artefact) in Civ4, has civil service or theology as its conditions. Instead of having another technique (‘woodworking’?) as a prerequisite, this link seems to imply that either the expanding needs of a bureaucracy or the need for religious texts could stimulate the demand for paper to be produced.

Besides the names of the ages (tool age), technological determinism can be exposed in the characterization of civilizations in Age of Empires. In AoE each civilization has a different tech tree, which has the same structure as the others, but with limited options: the Assyrians cannot research chainmail and Greece cannot have monotheism. This was an early way of creating diversity between factions, before Starcraft (Blizzard Entertainment, 1998) introduced completely different sides to the Real Time Strategy (RTS) scene. This ‘culture = destiny’ locking of tech trees was done “based on historical research in an attempt to match each civilization’s unique abilities” (Interview Goodman, May 17, 2011). However some obvious mistakes can be observed in these tech trees, for example Persia not having coinage, despite being historically known for introducing its use on a large scale. When confronted with this Bruce Shelley acknowledges that Persia should have had coinage, but that the technology “worked better for another [civilization]” (Interview Shelley, May 19, 2011). It turns out that historical accuracy was sacrificed in the playtest phase of development.

I hope to have shown here how determinism can surface in different forms, and games can be more or less deterministic depending on what criteria they are judged: for example unlike Civilization IV, the Age of Empires tech tree has more loose connections and no social determinism, but is unique per civilization, thus characterising them by technology. Not all orders are strictly determined nor causally intended, as alternative routes are sometimes possible and within the tech tree can also be spotted influences going from social innovations to mechanical ones.

The fruit on the Trees

“I used a cork board (...). I made a list of what I felt were the 80 most important human innovations and tried, as best I could, to make cause-and-effect connections. I wrote all the names down on slips of paper, tacked them to the board, and used rubber bands to show the connections. I would revisit the tree every few days until I was happy with the arrangement” (Interview Johnson, April 14, 2011).

So far the focus has been on the internal relations in the system of tech trees. This part will zoom in further and explore the objects in the system and their attributes. It is about the fruit on the trees: what specific technologies are selected and what do they do?

It is not an easy task to represent the technological history of the world in about 80 innovations. Soren Johnson based his decisions on his own background in history: “I tried to spread them out across a number of different eras and to be ‘culturally blind’" although that is hard to accomplish in practice. I also tried to identify a number of important ‘root techs’ which would open up large new areas of the tree” (Interview Johnson, April 14, 2011). Rick Goodman used many books from which he made lists of “technological advances and the years in which they were invented. This list had 100s of technologies and, of course, only a small fraction could be included in the game” (Interview Goodman, May 12, 2011). He later explains how he made his final selection: “Technologies were sorted based on technological importance. That was the first criteria. Second criteria was fitting them to their game play function. Many of the most significant technologies did not fit a game play function. So, in the end it was a compromise between historical significance and game play function”.

One merit of doing this research with four games is that it has allowed me to search for technologies that overlap between the games, thus drawing up a list of technologies that are generally considered to be important as multiple designers selected them. I limit myself to mentioning those technologies that are in at least three games. Note that this is partly subjective, because sometimes the same technology is indicated by different names.

And the winner is...monotheism. The worship of one god is the only ‘technology’ adopted in every tech tree, a surprising result as few people would call this idea a technology. In general religious ideas are dominant, as we find polytheism and mysticism present in three out of four games. There are six other technologies that overlap tree times: engineering, writing, coinage, astronomy, scientific method and the printing press. Note that the last one is the only machinery that is in all games (Age of Empires could not have had it, as its timeline stops in 500 AC).

If the selection was compared to historical narratives of technology, we would find that some pockets of technology are left out altogether. For example there is not one technology in all four games related to textile production. There is no mentioning of weaving techniques, and steam-powered spinning machines are not included despite their key role in the industrial revolution. One reason might be that it is hard to attribute an effect to textile production other than the profit gained from it, and that only Civilization IV has a money economy that could reflect this. Since technology trees form paths of technologies (and thus should lead somewhere), it is logical that certain famous ‘technological dinosaurs’ are left out, like zeppelins. From a feminist point of view, one could remark that certain domestic innovations are totally overlooked despite their historically important role in emancipating the daily lives of women. In general there are few civil/consumer technologies in the tech trees, as the player plays from the perspective of the state.

Comparing four games also allows for examining the size of the overlapping percentage of technologies themselves, which means the percentage of technologies in a certain tech tree that are also used in other games. Civilization IV  has an overlap 34%; Rise of Nations an overlap of 30%; Age of Empires overlaps for 39% despite only going to 500 AD, and Empire Earth only has an overlap of 17%. So it can be noted that, on average, the games share 1 out of 3 technologies and fill in the other 2/3rd with unique technologies.

Systems of effects

The function of tech trees was examined earlier in this article, but it remains to discuss in detail what technologies themselves do within the games. The effects of technologies are mostly metaphorical, representing what their usage on a large scale would do to the society that operates them. In general, technologies can a) lead to other technologies, b) boost the statistics of units and buildings, c) allow the production of new units and buildings and d) grant the player new abilities.

In order to understand how a particular effect is ascribed to a certain technology, or why a particular technology was selected to fulfil a certain function, what I call the ‘system of effects’ that each game uses, must be studied. With this term I mean that each game has reserved a certain role for technology upgrades to affect the gameplay, and a way in which technologies relate to each other and their effects accumulate.

Rise of Nations has a complex system of effects, in which most technologies effect more than one thing but in a predictable way. Library technologies have a quantitative effect of their own and can serve as a condition for other technologies, plus allow the construction of new buildings and units. For example chemistry (science tech) makes researching other technologies cheaper and 10% faster, and leads to new technologies (medicine) and allows the construction of the smelter building. Non-library technologies do not influence the upgrade tree, and only have a quantifiable effect which is typically stronger than that of the preceding technology in the upgrade path. Thus we can see that technologies in RoN can have multiple effects, but they are predictable and always consist of numerical increases or decreases of something, making the technology-effect relation rather arbitrary.

Since all new things in Civilization IV come from the tech tree, the system of effects was designed to give technologies a wide variety of functions. Technologies can allow new buildings, units, improvements, civics, religions, and actions. Furthermore they can reveal new resources, allow new diplomatic options, grand prizes to the first civilization that researched them and obsolete the effects of buildings. Lastly technologies can give quantifiable bonuses themselves, like engineering allowing units to move one extra tile per turn over roads. This system of effects allows the designers to chose more freely what effect they want to attribute to a technology, therefore allowing a more meaningful connection between technology and effect than in RoN. One limitation of this system is that each upgrade should be different, thus the Civ4 tech tree cannot include many technologies of the same kind.

Empire Earth in contrast has a simple and rigid system of effects. With the exception of two cases that add new abilities, technologies have two types of effects: either they give a range, attack, hit point (health) or speed bonus to units or buildings, or they give an economic productivity bonus of 15%. This system of effects is very limiting, as its rigid numerical pattern (+15% for production) does not allow the designers to determine the scale of effect of a particular technology. A second limitation is that there are only five branches in the tree to hold 86 technologies, which makes that there are many technologies of the same kind.

In Age of Empires the effects of technologies are fairly flexible. Some technologies have a clear quantitative effect, like domestication gives farms a 75% production bonus. Other technologies boost units or serve as conditions for very powerful ones, while yet other technologies have unique effects. Finally some technologies are combinations of the above, for instance craftsmanship gives a +1 range bonus to missile units, +2 to woodcutting and serves as the condition for building a heavy ballista. This system of effects allows both to connect simple effects to technologies or to turn them into something unique. One restriction is that AoE only has four resources, limiting the economic effects that can be handed out.

Concluding that each system of effects allows the designer to attribute different effects to technologies, the chicken or egg question arises: what came first in the design of the tech tree: the technology or the effect? There is not one answer that fits all cases. Rise of Nations suggests that effects came first, while Soren Johnson did the opposite for Civilization IV:

“I defiantly built the tech tree first and then found locations for everything (units, buildings, civics, religions, etc.) afterwards. My reasoning was that if my tech tree was a reasonable model of human history, every new game option should fit on it somewhere (Interview Johnson, April 14, 2011).

Bruce Shelley remembers how they “started with the technologies and tried to fit the game to them but then reversed ourselves and decided what effect we needed for the good of the game. Then we assigned what seemed like appropriate technology names to the effect” (Interview Shelley, May 19, 2011).

Identifying overlaps in the selection of technologies between the four games allows for a comparison of the effects of shared technologies. Out of the 35 overlapping technologies, only 5 have similar effects (architecture, coinage, genetics, irrigation and the wheel). Since the systems of effects are different, ‘similar’ here means that they loosely fulfil the same function in the gameplay or hint at the same principle. For example genetics gives a 40% health boosts to citizens in EE and giving a +3 health bonus to cities in Civ4, both suggesting an increase in civil health. If only 16% of the overlapping technologies have similar effects, this means that the games not only mostly selected different technologies, but also attributed different effects to the ones they share. Next, I illustrate this creative freedom in detail by looking at two case technologies.

Bringing in specific technologies

In this part I use the previous insights to see how historical strategy games can show us different sides of two selected technologies.

The first technology is writing, which is present in three games. Written word is the first technology of the science branch in RoN, and in Civilization IV it is an important ‘root tech’ that opens up large new parts of the tech tree. In this case it will be interesting to see how there is consensus on the function of writing. The Age of Empires manual follows suit: “The key importance of writing is that it allowed information to be stored and passed on easily, thereby accelerating the accumulation and spread of knowledge” (Ensemble Studios, 1997). All the games attribute the effect of spreading knowledge to writing, but implemented this in different ways. In Rise of Nations written word decreases the research costs and time of other technologies (thus hinting at the acceleration of knowledge), and allows the construction of the temple building. In Age of Empires writing has a unique and inventive effect, as it allows shared exploration with allies, which means that you can see what your allies can see (the rest of the map is veiled in the ‘fog of war’). Note that this is a rare literal implementation of knowledge sharing, as the field of view of each player is the only kind of real time knowledge that could have been shared in the game. In Civilization IV writing leads to the ‘open borders’ diplomatic deal, which allows players to scout and move inside each other’s borders (a lighter version of shared exploration), and allows the construction of libraries (which boost research). Although the function might be common, writing as a technique greatly differs between civilizations, which can lead to problems if the tree includes cultural specific technologies: “For years, we annoyed people by getting the Alphabet and Writing backwards on the tech tree (you can (...) have the latter without the former, but not vice-versa). We finally fixed this issue with Civ4, but it’s a bit embarrassing it took us 4 iterations to get it right!” (Interview Johnson, April 14, 2011).

Writing shows that even if there is agreement on the specific importance of a technology, creative game design allows designers to capture the effect of a technology differently in three equally interesting strategic options.

The printing press can be found in the tech trees of EE, RoN and Civ4. The difference between this case and writing is that each game gave a different interpretation to the effects of the printing press, showing three aspects of the same technology. In Rise of Nations the printing press is connected to the spread of knowledge, increases the knowledge (resource) output of scholars in the university buildings by +10. In Civilization IV the printing press has a whole different effect on first sight, increasing the wealth production of towns and villages (improvements) by +1 gold. This is a fairly modest interpretation that stresses the commercial possibilities of the industry that developed around printing, and possibly paper money. However the research capacity of a civilization in Civ4 is calculated on the percentage of commerce distributed to research, so the printing press indirectly boosts both economy and knowledge production. The printing press is one of the two (out of 86) unique technologies in Empire Earth that doesn’t have a quantifiable effect, but instead is part of the religious tree branch and allows priests to convert other priests. This effect may seem weird on first sight, but historically the printing press is connected to the rise of Protestantism and Calvinism, which started as a revolt within the Catholic church. The printing press allowed the mass production of religious texts (including the bible), which allowed priests to convince others and form a new church. So far the games showed three different yet all correct impacts of the printing press: “Printing had a major impact on the Renaissance, the Reformation and the scientific revolution that transferred European society” (Pacey, 1990, p. 85). The different visions on the role of the printing press are also reflected in the order of the tech tree, as printing press leads to reformation in EE, and to scientific method in both RoN and Civ4. In the latter the printing press is also a requirement for democracy, giving us another example of technological determinist reasoning in tech trees.

The case of the printing press makes clear how designers can highlight different facets of a technology they all deemed important, each showing one side of the history of technology, and giving it a unique function in the gameplay of each game.

When we look at technologies on the level of objects and attributes, we find that designers not only (even if it is only in the exceptions) enjoy a large freedom to highlight specific aspects of technologies, but also have creative ways of employing them in their games. Comparing the printing press to paper, the former shows three different aspects of the same technology, and the latter three different gameplay implementations of the same aspect.  


Writing allows shared exploration in Age of Empires, and so it did when exploring how the tech tree determines the possibilities for representing the history of technology in historical strategy games. The tech tree is both a representation of technological history and a mechanism or technology itself within the game. As a mechanism it has many functions that support the gameplay, and on a general historical level it offers a background meta-narrative of technological process that informs the bigger historical narrative or simulation.

The ordering of technologies in evolutionary tech tree diagrams is a controversial model for simulating the progress and connections in technological history. The controversy is tempered because it is limited to structuring the order of technology in a functional way without promoting explicit ‘single factor’ deterministic theories (like White, 1962). This article has demonstrated how the tech tree model can embody technological determinist assumptions in (at least) three ways: by forcing a set sequence, by influencing social changes in history and by characterising (thus determining) eras and civilizations. Especially in games that count on simulating historical mechanisms the first point can lead to problems, as the technology is often the invariable and determined mechanism among more dynamic elements. This is not just a matter of alternative history, but of placing all civilizations on the same track of evolution, which can lead to errors:

“There are some unavoidable issues though, many civilizations did develop Writing but never developed the Alphabet, yet according to Civ4’s tech tree, they never would have gotten to Literature or Drama. Clearly, this is not true, look at China, for example, but the tech tree could only be so flexible” (Interview Johnson, April 14, 2011).

Games that re-enact history like Age of Empires can solve this by adjusting the tech tree of each civilization. However balancing these individual tech trees against each other can threaten their historical correctness (like leaving Persia without coinage). It also directly leads to determinism by characterisation as it locks civilizations to what we in retrospect impose as their cultural destiny. But when this is not done, like in Empire Earth, the problem arises is that each civilization follows the same (in this case western) track of progress.

Although classical mythology advises us not to, cutting of the hydra’s head so that three new ones appear has one advantage: it makes each face of technological determinism smaller. In other words: by dissecting technological determinism the argument becomes more nuanced and relative to each game. Besides this, the two major shapes a tech tree can have (the interlocking vine or the individual upgrade path branches) each offer their own degree of freedom. In the tech tree of Civilization IV (and to a lesser degree RoN) the player gets to choose alternative paths at certain moments in the evolution of technology, either with so called ‘or’ ports or (if possible) by taking a long detour around a technology. In Empire Earth this is not possible, but the autonomy of the technology paths allow the player to neglect branches of technology altogether. One example of this is food production: the player can build farms and invest in +15% agricultural upgrades, but there is also a path of technologies that boosts hunting and gathering all the way up to the digital age.

There are also ways around determinism that lie in the blind spots of the game design, sometimes allowing the player to bypass the logic of the tech tree diagram. As Bimber (1994, p. 89) remarks, unintended consequences can work against determinism. In his Age of Empires strategy guide Jim Chamberlin describes the evolution in strategy of the early competitive community:

“Phase 3a:  A Revolutionary Discovery
It was learned that farming for food was much less efficient than foraging for berries or hunting.  People stopped making farms in the Tool Age and started focusing on natural food.  Bronze [Age] times across the board dropped about a minute with this revolutionary discovery” (Chamberlin, 2004).

What this means is that in Age of Empires it is possible to achieve faster growth by picking berries than adopting agriculture, given there is sufficient supply of berries (which depends on how the map is generated). This early form of ‘theorycraft’ (Paul, 2011) shows how some games allow the player to dodge the normal order of moving up the tech tree. Lastly one could interpret the inclusion of social technologies and their links to other technologies as countering the technological determinist assumptions of the tech tree.

These social technologies can only exist in the tech tree because all game designers opted for a very broad definition of technology, which buys them some space on the level of the individual technology. The selection of effects and technologies is not totally free, since we have seen that besides historical importance the overall structure of the tech tree and the systems of effects serve as restrictions. Still there is a large degree of freedom for interpretation as can be learned from comparing the four games that they share on average 30% of their technology selection, filling in the remaining two third with unique choices. Add to this that of the 35 overlapping technologies only 5 have similar effects in all games, an even greater liberty is found to exist for designers to tell the story of technological history in an unique way, highlighting different aspects of one technology. Working in tandem with the broad concept of technology adopted in the historical strategy games, this allows technology to connect to more themes than just the increasing mastery over nature. This was demonstrated in the cases of the writing and printing press technology, two technologies which had different gameplay functions or historical interpretation in each game.

In this research I have attempted to bridge science and technology studies and games studies, with the realm of strategy games. It is possible to delve deeper into this. I have only briefly touched upon themes like how designer incorporate academic knowledge in the design of their historical narrative. Another largely untouched theme is how the tech tree determines not the content but the pace of historical time in these games. Because of space limitations these topics remain unexplored in the current article1, but deserve exploration in future research.

The tech tree mechanism clearly forces the games to represent the history of technology in a deterministic way, leaving out the real diversity and experimentation that characterises this history. Still each games finds ways to tells a different history of technology, leaving some liberty to the player to steer how it will be played out.


1 The full research report which explores these questions is available from the author upon request.


Arendt, H. (1958). The Human Condition. Chicago: University of Chicago Press.

Bimber, B. (1994). Three Faces of Technological Determinism. In: Smith, M., Marx, L. (eds). Does technology drive history? The Dilemma of Technological Determinism. (pp. 79-100). Cambridge: MIT Press.

Basalla, G. (1989). The Evolution of Technology. Cambridge: Cambridge University Press.

Big Huge Games (2003). Rise of Nations. Microsoft game studios.

Bijker, W.E., Hudges, T.E., Pinch, T.J. (eds). (1987). The Social Construction of Technological Systems:. New Directions in the Sociology and History of Technology Cambridge: MIT Press

Blizzard entertainment (1998). Starcraft. Blizzard entertainment.

Bogost, I. (2009). Videogames are a Mess. Electronic version, retrieved October 15, 2011, from http://www.bogost.com/writing/videogames_are_a_mess.shtml

Burns, A.  (2002). Civilization III: Digital Game-Based learning and Macrohistory Simulations. Electronic version, retrieved Octrober15, 2011, from http://www.alexburns.net/Files/CivilizationIII.pdf 

Chamberlin, J. (2004). Age of Empires guide. Retrieved October 15, 2011, from http://www.gamefaqs.com/pc/90380-age-of-empires/faqs/2270 

Douglas, C. (2002). “You Have Unleashed a Horde of Barbarians!” : Fighting Indians, Playing Games, Forming Disciplines. Postmodern Culture 13(1).

Ensemble Studios (1997). Age of Empires. Microsoft game studios

Firaxis (2005). Sid Meier’s Civilization IV. 2K games.

Goodfellow, T. (August 20, 2005). Bunk, Progress and Process �" Games take on History. (blog). Retrieved October 15, 2011, from http://flashofsteel.com/index.php/2005/08/20/bunk-progress-and-process-games-take-on-history/

Hartland Trefoil (1980). Civilization

Heilbroner, R.L. (1967). Do Machines Make History? Technology and Culture, 8, 335-345.

Henthorne, T. (2003). Cyber-Utopias: The Politics and Ideology of Computer Games. Studies in Popular Culture 25(3)

Jakobsson, M., Sotamaa, O. (2011). Game Reward Systems. Game Studies, 11 (1).

Kapell, M. (2002). Civilization and its Discontents: American Monomythic Structure as Historical Simulacrum. Popular Culture Review, 13(2), 129-35.

MacDougall, R. (March 18, 2009). Technology grows on trees. (blog) Retrieved July 15, 2011, from http://www.robmacdougall.org/blog/2009/03/technology-grows-on-trees/

Morris, D., Hartas, L. (2004). Strategy Games. Lewes: Ilex Press Ltd.

Owens, T. (February 10, 2009). Science Grows On Trees: History of Science and Technology Acording to Video Games. (blog). Retrieved October 15, 2011, from http://www.trevorowens.org/2009/02/science-grows-on-trees-the-history-of-science-and-technology-acording-to-video-games/

Pacey, A. (1990). Technology in World Civilization. A thousand-year history. Cambridge: MIT Press. 

Paul, C.A. (2011). Optimizing Play: How Theorycraft Changes Gameplay and Design. Game Studies, 11(2). 

Poblocki, K. (2002). Becoming-State. The Bio-Cultural Imperialism of Sid Meier's Civilization. Focaal -- European Journal of Anthropology, 39, 163-177.

Salen, K., Zimmerman, E. (2004). Rules of Play. Game design fundamentals. Cambridge: MIT Press. 

Smith, M.L. (1994). Recourse of Empire: Landscapes of Progress in Technological America. In: Smith, M., Marx, L. (eds). Does technology drive history? The Dilemma of Technological Determinism. (pp. 37-52). Cambridge: MIT Press

Squire, K. (2004). Replaying History: Learning World History through playing Civilization III. Ph.D. dissertation, Indiana University. 

Squire, K., Jenkins, H. (2003). Harnessing the Power of Games in Education. Insight, 3, 7-32.

Stainless Steel Studios (2001). Empire Earth. Sierra Entertainment.

Weib, A. (2007). Computerspiele als Aufbewahrungsform des Politischen: Politische Theorie in Age of Empires und Civilization. In Bevc, T. (ed.). Computerspiele und Politik: Zur Konstruktion von Politik und Gesellschaft in Computerspielen. Berlin: Lit Verlag

White, L. (1962). Medieval technology and social change. Oxford: Oxford University Press.

Wyatt, S. (2008). Technological Determinism Is Dead; Long Live Technological Determinism. In Hackett, J.E., Amsterdamsta, O., Lynch, M., Wajcman, J. (Eds.) The Handbook of Science & Technology Studies. Cambridge: MIT Press.

©2001 - 2012 Game Studies Copyright for articles published in this journal is retained by the journal, except for the right to republish in printed paper publications, which belongs to the authors, but with first publication rights granted to the journal. By virtue of their appearance in this open access journal, articles are free to use, with proper attribution, in educational and other non-commercial settings.