Enough of me talking. Have a read of the draft thesis introduction (if you fancy). At the very least you’ll get a better understanding of my personal flavor of archaeogaming.
PS—This post’s title includes the word “leveling” which can be read as either “leveling up” MMO-stylee, or as in “knocking down.” Probably both at once.
Digital games and virtual worlds are both archaeological sites and artifacts. These born-digital built environments contain their own material culture, sometimes created by people directly, and at other times created by algorithms triggered via code one step removed from the programmer. Many games, like many traditional archaeological sites, can show evidence of occupation, contain their own economies and currency, allow for the crafting and trade of objects, and evolve over time. In most instances, games create landscapes that can be settled by human players, interacted with as both obstacles and resources. In some games, however, these landscapes and the things they contain are procedurally generated through code that establishes a set of rules: every visit to the game is different, and the code gives rise to “machine-created culture,” something wholly new to archaeology. How can archaeologists begin to deal with these new, born-digital spaces, which are (or have been) used/inhabited by millions of real people? What does machine-created culture look like, and how do we as humans (and archaeologists) observe and document it?
This thesis seeks to apply/amend current archaeological tools, methods, and theories to these incorporeal spaces while searching for a greater understanding of how hardware/software begets culture through both human and non-human agency, artificial intelligence, and computational complexity. Any rules derived from the study of games/worlds might then be applied to real-world sites as predictive models for settlement patterns and artifact distribution.
“Archaeology is the science of new things.”
–Gavin Lucas, “Archaeology and the Science of New Objects”
Since 2002, video game archaeology has appeared in peer reviewed archaeological publications starting with Ethan Watrall’s essay, “Interactive Entertainment as Public Archaeology,” in the SAA Archaeological Record. Known more widely as “archaeogaming,” this subdiscipline focuses on the archaeology both in and of video games. In blogs such as archaeogaming.com and playthepast.org, archaeologists and historians turn a critical eye to understanding video games within their broader context as interactive media featuring material culture, and of video games as examples of material culture in their own right. While much of the published literature over the past 15 years has focused more on the reception of archaeology and archaeologists by video game developers and players (see Holtorf 2005, 2007; Lowe 2013; Mol 2016), or on games that simulate historical places/events and provide digital reconstructions of real-world historical spaces (see Chapman 2016; Copplestone 2016; Gardner 2007), it has been this author’s intent to study video games more as examples of “interactive digital built environments.” As this thesis will explain, video games are archaeological artifacts. They are also archaeological sites. This is more than a clever analogy, and begins to consider born-digital, virtual spaces of agency and activity as the new frontier of 21st-century archaeology.
As the corporeal and incorporeal continue to blend for people by virtue of increasingly affordable, omnipresent digital technology, we progressively descend into our own entertainment. Through these games, we gain instant satisfaction of fundamental human needs of making and exploring. The feedback is immediate and increasingly haptic. What once frustrated ancient people in their curiosity about their environment—lack of time, materials, and understanding—vanishes during contemporary gameplay. Play is pure creative expression, and games continue to provide that interactive outlet to players who create and explore, whether these games are set in a real or imagined past on Earth or elsewhere.
Over the past 45 years and billions of copies of thousands upon thousands of titles since the creation of the first publicly available video game, Computer Space (Nutting Associates, 1972), entire economies and gaming cultures have evolved (King and Borland 2004). At this writing, there are more players of the massively multi-player online role-playing game World of Warcraft (Blizzard Entertainment, 2004) than ever lived and died in the City of Rome when it was the sole capital of the Roman Empire [CITATION]. Video games comprise a larger share of the entertainment market than movies, music, and books, with global sales forecast topping US$100B in 2017 according to a February 15, 2017 cnbc.com report. Players invest hundreds of dollars a year in game subscriptions and purchases of standalone titles. Popular games engender huge followings of players and fans, and create their own subcultures outside of the game. Within some games, in-game cultures thrive, both real (e.g., guilds/groups of human players) and imagined (races and lore created by the developer to assist in creating a rich world and to propel game-narrative). This is something that Champion calls “cultural presence”: “a feeling in a virtual environment that people with a different cultural perspective occupy or have occupied that virtual environment as a place” (Champion 2016: 64). Listening to some of these players recount their adventures and explorations of far-flung, fantastic worlds, describing cities and monuments and history in exquisite detail is not unlike hearing a Romantic recounting of one’s first visit to Rome, or to hearing an archaeologist explain the finer points of Dressel amphora types to a captive conference audience.
Games have long graduated from being the sole province of young people (and arguably never were). They are serious business, and are taken seriously by both players and creators. It makes sense then to treat video games as interactive digital built environments, to see them archaeologically, and to begin to understand their entanglements with the past and present, the real and virtual, the social and economic, as well as with each other within a wider context. With the advent of new video games that can create their own environments without direct human intervention, we now have the chance to witness the birth of a question not yet asked: what does a culture look like when it is created independently of human intervention?
This thesis is the first formal attempt to understand digital games archaeologically as sites to be surveyed and excavated. Games, unlike all other media counterparts (i.e., printed material, audio, images, and film), allow for complete action and interaction by people (players) in fully immersive spaces. Games are built environments, constructed like houses (and occasionally like cathedrals), designed, planned, and built, featuring later additions and occasional modifications made by the player-inhabitants. Games sprawl like cities as well, often spilling out from their tidy plans into a wider landscape to be explored and interacted with (as seen in open world games), which allow people to wander where they will, occasionally allowing them to build for themselves, or to find evidence of past cultures (or past players). In building a community of practice, the video game archaeologists create their own media culture. As media archaeologist Huhtamo explains, “media forms and their uses are constantly negotiated, tested, and contested. Material applications meet discursive ideas” (Huhtamo 2016: 123). We are finding our way to describe this new-ish medium through archaeological practice. Chateau and Moure agree in their introduction to Screen: From Materiality to Spectatorship–A Historical and Theoretical Reassessment: “We need a new vocabulary with which we can fully account for the various levels according to which the filmic universe can be defined” (Chateau and Moure 2016: 15). Substitute “filmic” for “gaming.” Our discipline needs new words. Brittain concurs: “We must explore new ways to appropriately articulate a ‘lifeworld’ (an archaeological topology of place)” (Brittain 2013: 258).
Games occupy real-time and contain their own time(s) and chronologies within them. A game such as Raiders of the Lost Ark (Atari 1982), has its own specific date of creation, which places it in an absolute chronology of digital games. It post-dates Adventure (Atari 1979) and pre-dates Joust (Atari 1983). It is of a specific type—an Atari 2600 cartridge—made of plastic and a silicon wafer. The game encoded on the wafer is also of a specific type: an “action RPG.” Games are played for a year or two after their release, the most popular of which entertain millions of people, prior to a precipitous drop in interest as new games come to market. The games, however, do not disappear; they perdure (to borrow a favorite phrase of Ingold), continuing to be games whether or not they are being played. Thirty-five years after the fact, one can revisit these old games, which respond as they were programmed, as if no time had passed at all. Other time is at work within digital games as well, where “years” can pass during normal play. Players can observe changes in the landscape within a game, can see cultures rise and fall, and can have a direct hand in the success or failure of empires, centuries reeling off in a matter of real-world hours or days.
These virtual environments are largely simulations in which players interact with things explicitly placed by the game’s maker(s). As such, they can be used for modeling human behavior, for visualizing structures or landscapes and how people interact with them. Most games include artificial intelligence (AI), which is designed to interact with players as they proceed through the game’s environment. The AI in more recent games also includes machine learning, where the game, through observing the player, adapts to player-behavior. Game-controlled elements (typically opponents or non-player characters), react in new and unexpected ways to player agency/action, which makes the game more interesting and more challenging to the player. A story unfolds in play, adding to the internal history of the game, and in some instances, becoming part of history outside of the game as well.
The interaction of player and digital space is not unlike the interaction of a person in a corporeal space. Both spaces and interactions are governed by rules, whether they be mundane laws of physics, or by more complicated social constructs. As people interact with their environments, new/different behaviors emerge. This emergent behavior is a residue of complexity, a product of rules-based behavior. In a game, a player’s interaction with the internal environment gives rise to other in-game actions. Borrowing Ingold’s analogy in his book Making (Ingold 2013: 61) regarding how a site is “made,” in the construction of a game by the game’s maker(s), digital materials are manipulated in such a way as to create a visual space in which players can operate. In the real world, a carpenter works with wood in order to frame a house within which occupants can live. In either case, real or virtual, space is constructed for others to inhabit (including archaeologists). The spaces, Ingold argues, are created in movement; they are “performed” (Ingold 2013: 85). In his article “The Temporarilty of the Landscape,” Ingold states that “the practice of archaeology is itself a form of dwelling” (Ingold 1993: 152). Players certainly perform within the games they play.
The archaeological study of digital games, however, takes a new step in understanding built environments. Ingold wondered: “Could certain practices of art, for example, suggest new ways of doing anthropology? Could not works of art be regarded as forms of anthropology, albeit ‘written’ in non-verbal media?” (Ingold 2013: 8). Video game design certainly speaks to these questions. This study is post-materialist, especially when compared with excavating the foundations of a real-world house. In the real world, the archaeologist finds evidence of stone, a material with which a builder created foundations, the base of walls. The stone might be local, sourced specifically for wall-building, or imported because of various desirable properties based on the needs of the builder or the desire of the owner. Further excavation might yield wood that framed the house, or slate that shingled it, or glass from the windows, each material used for specific purposes based on past experience. In Ingold’s famous essay, “Materials Against Materiality,” he proposes: “might we not learn more about the material composition of the inhabited world by engaging quite directly with the stuff we want to understand: by sawing logs, building a wall, knapping a stone or rowing a boat? Could not such engagement—working practically with materials—offer a more powerful procedure of discovery than an approach bent on the abstract analysis of things already made?” (Ingold 2007: 3).
With the digital, one deals with a single “material”: the pixel. The pixel itself is not a material, per se, but is instead the product of electricity, light, and thought. When constructing a digital simulation of a house, a pixel is given properties by the maker, which then joins with other pixels with similar properties to make the “stone” to make the “wall.” One material transforms into an infinite number of other materials in the creation of new, incorporeal spaces. The resulting house can be built upside-down. It can be built in the air, defying gravity (in games, both earth and sky are imagined constructs). It can be as big or as small as desired without the worry of the expense or rarity of materials. Despite the “immateriality,” the virtual house is still built. It is designed and constructed and ultimately used within the virtual environment. By being built, it becomes a thing, a space. And through its history of use by both maker and player, it attracts the archaeologist’s eye.
In some digital games, however, the maker-to-house scenario as described above is removed one step from human agency. With larger, more complex games, it is too time-consuming for the maker(s) to build every house in a town, or to create every tree in a forest. For the maker(s), time is as much of a material as pixels. To solve the problem of how to rapidly fill a space with houses or trees, the maker(s) turn to code. Maker(s) might draw several different kinds of walls, roofs, doors, windows, chimneys, as well as textures for stone, brick, wood, and glass, and then create rules to indicate how all of these pieces should be assembled. These algorithms are called “procedures” and the result is a town that is “procedurally generated.” The game-maker(s) did not create the town full of houses of various shapes and sizes, but instead created the rules for what a town might look like, leaving it to the game to put the pieces together much faster than any human could. The software created the space. And in some ways, the software begins to create “culture.” Perhaps most startling is the fact that these cities and forests are created fully realized in the blink of an eye. Rome built in a day indeed.
Digital games featuring procedural generation of content (“content” meaning things and landscapes) are beginning to make their own cultures. The maker(s) create the initial set of rules, which the software then “interprets” to create objects, designs, and even language and world-views in the most robust instances. These are cultures never-before-seen. They are new, even to the people who created the rules governing their emergence. It is at this moment that digital games cease being simulations of a kind of reality, and instead bring forth something completely new that operates according to its own rules whether or not a player is present. These games (and their archaeological study) are post-humanist (Lucas 2013: 377). This “machine-created culture” (MCC) begs for its own archaeological understanding of how the world was created, as well as the things that populate the space, even (or perhaps especially) if humans are not present.
This feels as exciting as finding a “lost” civilization in the real world, but through MCC, the new cultures and civilizations are infinite. For archaeologists then, when confronted by new digital constructions created by encoded rules, the main issue is how to deal with something on such a scale. As with the archaeology of the recent past, in light of the presence of mass-produced things, strategies emerge, and the archaeologist is able to divide and conquer (or at least choose the battles), working backwards from the artifacts into an interpretation of the site (and rules) which produced them. For this new endeavor, the archaeologist requires tools. More importantly perhaps, the archaeologist also requires questions and hypotheses, plus method and theory to guide one towards answers manifested in the artifacts and data recovered by tools. For this new archaeology of digital spaces, we start with what we know and use in the field, and adapt as needed. Everything is iterative, including archaeology.
In Edgeworth’s 2014 article “From Spade-Work to Screen-Work: New Forms of Archaeological Discovery in Digital Space,” he observes that “the term ‘site of discovery’ might usefully be taken to refer to virtual on-screen realities as well as off-screen ones” (Edgeworth 2014: 44). Archaeologists discover things in the data as well as in the ground. In his assessment, “the virtual landscape can potentially yield an almost infinite number of new discoveries, each one giving rise to further paths of exploration that can be followed towards further discoveries and insights” (Edgeworth 2014: 59).
This thesis serves as version 1.0 in how to arrive at those discoveries and insights, through surveying, excavating, and documenting interactive digital built environments (i.e., games).
This thesis will apply established archaeological tools, methods, and theories to interactive digital built environments. A baseline case study of one game (with which the author is familiar) will follow largely the guidelines for excavation as established by Prof. Steve Roskams in his 2001 work, Excavation, but will be ported to the digital environment (hence the quotation marks surrounding some terms that will need to be better defined for fieldwork in virtual spaces):
- Pre excavation strategies: research objectives, aerial photography, fieldwalking, “shovel testing,” documentary material, previous “excavations,” ground-based remote sensing, “evaluation trenches”
- Background preparation: finance/administration, staff/support facilities, safety
- Site preparation: site clearance, site grid, spoil removal, shoring, de-watering, finds retrieval
- Recording: stratigraphic unit, numbering systems, recording process and sheets
- Photographic record: Reasons to photograph, preparation and technique
- Spatial record: techniques, equipment, and drawing conventions, types of plan, techniques of measurement, types of section, piece-plotting finds
- Stratigraphic record: types of stratigraphic relationship, representing these, calculating these
- Descriptions: who records and when, computer storage of record, deposit and non-deposit descriptions
- Excavating the stratigraphic unit: sampling strategies for finds, methods of collection, “troweling” methods, making stratigraphic distinctions, completing the record, checking the record
- Stratigraphic analysis: tidying the record, on-site interpretations, correlating between units, stratigraphic nodes and critical paths
Based on the success or failure of the application of traditional archaeological tools/methods as described by Roskams and others (see Carver 2009; Hodder 2005; Johnson 2010; Raab and Goodyear 1984; Renfrew and Bahn 1991; Roskams 2001; Trigger 2006), these will be revised or built from the ground up to better facilitate virtual fieldwork. Two additional case studies will be conducted on games new to the author. One game will be “traditional,” designed with standard gameplay in mind (pre-defined culture, history, lore, buildings, etc.). The other game will contain examples (all or in part) of machine-created culture. The intent is to see if the same tools/methods can be used to understand each digital space archaeologically. The purpose of studying one game known to the author with current archaeological tools/methods is to establish a baseline, or a control. The two additional case studies of games not previously played by the author will serve as variables. This mirrors real-world archaeology in that some sites are known, and have a well documented history of excavation and interpretation, while other sites are only just-discovered, and require critical thinking both prior to and during fieldwork as to what questions to ask, what tools to use, and what interpretations to make “at the mouse’s edge.” The outcomes of all three case studies will then be compared side-by-side with an analysis of what worked, what didn’t, and how to move forward. An analysis of the data recovered in all three case studies will also follow, with the possibility of applying rules learned in the virtual spaces to real-world excavations.
In my January 20, 2017, conversation with Dr. Julian Richards, head of the Archaeological Data Service (ADS), he is willing to let my work be a test case in best-practices for virtuality, documenting games-as-sites, but doing so with traditional publication, data, and archive. The ADS will host the work. It may be possible to include game content as part of the archive, along with multimedia, but that depends on whether or not British intellectual property law permits the archive and access of video game content for public research purposes.
This thesis is divided into seven chapters, which gradually make the case for an archaeology of interactive digital built environments, and how to do fieldwork in them.
Chapter One will conduct a needs assessment for the archaeology of video games. What questions should be asked? What hypotheses do archaeologists have for studying games? Why should games be considered as built environments, and what do archaeologists hope to answer through their study? As Lucas wrote in 2013 in “Archaeology and the Science of New Objects,” “What new entities can archaeology propose? What does archaeology show us that we did not know already?” (Lucas 2013: 374). He continues:
“It is one thing to point out that archaeologists might discover new types of pottery . . . but what the debates over scientific realism involve is not simply new types or new species, but new kinds of objects altogether. Is this possible in archaeology? Is there an archaeological equivalent of an electron? . . . . What possibilities are we not realizing . . ?” (Lucas 2013: 370).
With video game archaeology, we are seeing something new.
Chapter Two will consider games-as-archaeology, discussing further the nature of digital material and materiality, and the false dichotomy between “real” and “virtual.” Video games embody the relatively recent anthropological phenomenon of the “ontological turn.” The archaeologist is situated in the middle, observing the real and virtual melting together. Boellstorff writes: “This is an understanding of the gap between virtual and actual as real precisely because it is a space of connection and alteration—a space of mutual possession—rather than a teleology from online to offline, as if from spirit to flesh. It is not that virtual worlds are potentially real, but that they are additional realities” (Boellstorff 2016: 395).
Meades writes in Understanding Counterplay in Video Games, “we have seen an ascetic approach to games, where the methodical, archaeological scrutiny, and deep reading of simulation and code expose a profound and unique understanding and enable the creation of identity” (Meades 2015: 183). It would seem that understanding identity (player, developer, or other) is possible regardless of the material being examined, pixel or stone.
Chapter Three will explore what it means to have material culture within a game-space. It will address the human notion of what it means to create material culture for a game and how games themselves are examples of material culture. Machine-created culture (MCC) will be more fully introduced. To better understand it, archaeologists will need to be able to know how to identify game-imposed rules as well as how to interpret those underlying instructions in “code palaeography.” It is a nature-or-nurture kind of fundamental question: is MCC really an interpretation by machine of code, or is it instead a product of execution of code that when combined with other routines yields emergent behavior? In the end, are we seeing culture, or merely cultural noise? Is the game producing an actual city, a simulation of a city, or an emotional messaging of what a city “ought” to look like?
Because games are interactive and are built for human habitation (if only for a few moments or hours), one must consider human needs vs. code needs within the overarching context of the game itself. The archaeology of video games is the understanding of the needs of the code to serve the narrative environment of the game-space. In the real world, objects happen because of human necessity. Games, however, isolate a handful of human needs in order to fulfill them for the player. It is a potent distillation.
Chapter Four will contain the first, “controlled” case study, which will apply existing/established archaeological tools and methods to a game previously played by the author. This includes following the lead set by Catalhoyuk, which includes reflexive methodology in support of efficient workflow and record-keeping including both text and visualization data (Berggren 2015: 443), and ensuring that all data are made available via the project database as quickly as possible: contextual metadata, photos, diaries, and videos from the excavations are made available (Farid 2015: 74). The case study will also attempt to apply GIS/mapping software over the top of the game so as to record where events happen, as well as the absolute location of features within the game-space. An attempt will also be made at in-game photogrammetry.
Chapter Five will detail the application of adapted/modified archaeological tools and methods to two case studies of games new to the author, one of which contains “traditional” material culture, and the other of which contains machine-created culture. Both games will utilize AI, and the author will note emergent behaviors derived from complexity. If possible, the nature of the algorithms in these games will be documented, creating a base set of rules that can be followed to repeat in-game behavior(s).
The author will also look actively for the occurrence of glitches, software aberrations in games caused by unintended agent-behavior. Glitches form their own kind of special artifacts within game-spaces, have their own established typologies (Meades 2015), and require special documentation as “significant finds.” The archaeology of digital games is not just about what happens within the game-as-written, but also about what happens when the game breaks. This is part of the life of the game-as-site, and forms an event within the game’s biography.
Chapter Six will analyze the tools, methods, and data from all three case studies together, determining successes and failures as well as a look at the data recovered from the three “expeditions.” Following Bayliss and Whittle’s lead, Bayesian statistics will be used: “We analyze the new data we have collected about a problem in the context of our existing experience and knowledge about the problem” (Bayliss and Whittle 2015: 217). Based on the results, what new tools need to be developed and what new questions need to be asked of interactive digital built environments?
Chapter Seven will consider real virtual fieldwork, making an attempt towards establishing best-practices for conducting archaeological fieldwork in virtual spaces. It will conclude with lessons learned (and potentially rules/algorithms) that can be applied to real-world fieldwork after being field-tested in games-as-simulations or as agent-based models. In 1999, Puyol-Gruart hinted at this future in “Computer Science, Artificial Intelligence, and Archaeology”: “Simulation, in archaeology, is devoted to the simulation of the objects, of archaeological study, the people and their societies, their relationships with the environment and other people, commerce, hunting, etc. . . . A multiagent program consists of the programming of agents, with their particularities—proles—in the society, the communication capabilities with other agents, the perception and reaction behaviour to the environment, etc. We can consider defining this behaviour, by means of the rule formalism” (Puyol-Gruart 1999: 26–27).
Basing analysis of data created by rules is called “agent-based” modeling, and will be especially important when documenting machine-based culture. The agent (which can be a player, but is more often a software construct created to react to the rules established in the environment into which it is placed) can interact with its environment as well as with other agents. The agent might have goals to achieve as well as the ability to learn and adapt based on its experiences (see Macal and North 2009 for a readable summary of conducting agent-based modeling and simulations). Archaeologists can use agent-based modeling systems (ABMS) in order to test video game environments, and actions done within them in order to induce emergent behavior.
Video game archaeology perhaps is the first example of a new New Archaeology, one that is post-material and post-human, one that not only intersects past and present, but that also uses the screen as the sole method of accessing new archaeological spaces. These spaces are made by people (or by machine) for other people to use, and are invested with creativity and examples of material culture. They are kinetic and also kinesthetic. They contain their own space-time. Each game is its own discrete entity, its own site. At the same time, each game exists in multiple, identical copies, circumventing the problem of the “unrepeatable experiment” of total excavation. They pose both classic and new questions to the archaeologist who operates in both the real and the virtual simultaneously using very real archaeological craft. “[The archaeologist] has no direct contact with that [archaeological] evidence, which cannot be physically touched. Yet in another sense she displays all the attributes of a craft practitioner, demonstrating embodied skills of computer use alongside intellectual reasoning in the ongoing investigation….” (Edgeworth 2014: 54). The outcome of this thesis will be ultimately the creation of a field manual for conducting archaeological survey and “excavation” craft in digital games.
The precipitate of this work within digital built environments is the creation of what Champion calls “virtual heritage”: “the attempt to convey not just the appearance but also the meaning and significance of cultural artefacts and the associated social agency that designed and used them, through the use of interactive and immersive digital media” (Champion 2016: 64). In his article, Champion is speaking about this interactive and immersive digital media intersecting with real-world cultural artifacts. I would argue that the same media will be used at some point in the not too distant future for conveying the meaning and significance of purely virtual artifacts, events, and sites. Already there are both real and virtual video game museums, and in-game memorials. It is not a question of what but of when. The potential for artificial life to emerge from new and future games is the stuff of science fiction, yet has already been seen on a smaller, cellular scale (e.g., Conway’s Game of Life). Machine-created material culture, however, is already here, and is becoming more complex and harder to distinguish from those objects, artworks, and buildings purpose-built by designers. At the same time, it is possible that MCC will generate something completely alien based on the rules used to program the procedural generation of new, digital artifacts. Will people be able to recognize that new material culture for what it is, something that resonates with machine-logic far removed from human pots and pans.
[OPTIONAL: Feel free to leave constructive comments to this post. This will help me with the revisions/rewrites and will improve the overall quality of the work and hopefully of archaeogaming generally.]
—Andrew Reinhard, Archaeogaming