Archaeogaming

December 8, 2019, will mark the one-month anniversary of the release of Death Stranding (Kojima Productions), and to commemorate it I wanted to write about the game’s “rapid archaeology” (my term). The game defies easy description, and I’ll oversimplify it by saying you play as Sam Bridges, postapocalyptic delivery guy who also installs your cable/Internet bundle. Death Stranding is an open world walking simulator, which means you can point yourself (or your vehicle) to the horizon and get there, navigating considerable hazards along the way. Hazards include deep rivers, snowy mountains, “timefall” (rain that ages your equipment and parcels), “MULEs” (delivery-addicted non-player characters), and “BTs” (tar-ghosts, not to be confused with the K-Pop band by the same name). But you are not alone, which is one of the game’s core themes.

As Sam, you have the ability to build things to avoid/conquer hazards. Bridges help you cross rivers and chasms; timefall shelters get you and your gear out of the rain; ladders/ropes help you climb; safehouses give you a place to rest and shower; ziplines let you travel safely; generators recharge your batteries, etc. Players can also erect signs, literal semiotic waypoints indicating danger, encouragement, and even desires. But what you choose to build, you also build for others. One of the ways Death Stranding is unique is the fact that this is a solo-multiplayer game that is based on the premise that “we are all alone together.”

For argument’s sake, let’s say 10,000,000 people bought Death Stranding. That’s 10,000,000 separate Sam Bridges characters saving the United States one package at a time. Like other games such as World of Warcraft, it would take too much computer processing power on both the client and server side to manage all of the stuff built by ten million Sam Bridges. Instead, the game-world is broken into shards, or identical worlds on separate servers shared by a lot of players. Unlike WoW, however, you never see a soul (sort of), but you do see what they have left behind as well as how it has impacted other players.

For example, I built a ladder so I could reach the top of a cliff (ladders can also be used to ford streams, as shown above). Once built, others can use it. After they use the ladder, they have the option of “liking” it, parroting social media such as Twitter and Facebook. The more likes you get, the better your reputation. But it goes both ways. If I use someone else’s ladder, I can like it, too. The more likes you give, the better your reputation as you and others interact with temporary, human-built structures.

The structures are temporary. Timefall decays them to the point of disappearance. They can be damaged through misadventure. And they can be removed from the world. In Death Stranding, we see a human culture of cooperation, of helping people we cannot see. As these structures decay, players can return to them to either reconstruct or upgrade them, which is also rewarded in increased reputation. Building and maintaining roads—the great connecting strands of pavement—yields the most benefits for all players, but these, too, decay and require constant attention and upkeep. Having “finished” the game, I found myself returning to places in the landscape to put up signs and rebuild roads and other structures to help people I don’t know have an easier time navigating a hostile landscape than I did. The game doesn’t require this, but it just feels like the right thing to do. And every time I log back into the game, it tells me how many people I’ve helped, which makes me feel pretty good.

As an archaeologist interested in how flesh-and-blood people interact within synthetic worlds, I was curious to see who was doing what within a predefined space. I wanted to see if other players were continuing to build things in the game’s starting area of Capital Knot City within about a 500m radius of the city’s center. I wanted to check the rate of growth and decay and to test my theory that things change in open worlds like the one in Death Stranding at a fast rate requiring rapid archaeology.

At 1200 on December 1, 2019, I took a screenshot of the map of the area immediately surrounding Capital Knot City. The map showed 31 separate structures (signs, shelters, ladders, etc.). Two of these (an encouragement sign and a watchtower) were created by the non-player character (NPC), “Nick Easton.” The other 29 structures were created by human players sometime between November 8 and December 1, 2019. The structures are not date- or time-stamped. They are, however, labeled with the player’s name (i.e. gamertag), and one can drill down to see more data about the player’s experience in Death Stranding via a remarkable data visualization tool. Instead of seeing how experienced a player was when they placed a structure in the landscape, you can see the player’s current progress and reputation.

Over the course of cataloguing the 31 structures around Capital Knot City, I ran into the problem that some of the things I was recording in the field at 1230, did not correspond with icons on the map as screenshotted at 1200. As it happened, other players were populating the landscape with additional signs and structures faster than I could record them, tying them back to the original map I had annotated.

I decided to return to the same spot the following day, December 2, 2019, and was able to log in at 2100. As before, I screenshot the map and then labeled the icons of human-placed structures. This time there were 24 (plus the same 2 by the NPC). Most of them matched what I had recorded the day before, but there were a few new ones. As for the structures that were now missing, they were likely lost to the decay caused by timefall. Different structures decay at different rates, signs being the most temporary because of their size.

The speed of construction and decay is fast in Death Stranding, and the archaeologist could log in every hour in a single spot and observe change affected by other invisible players as they race against decay in a temporary effort to assist other unseen humans. The futility of the endeavor lends additional positivity to these efforts to assist people we don’t know, that we expend resources knowing full well that the infrastructure will either be used and/or salvaged by players or destroyed by the elements.

There is another kind of destruction. Each of the ten million Sam Bridges can choose to de-clutter their worlds. At first, I thought this would be an ethical nightmare: if a player deconstructed a bridge created by someone else, what would happen to the community and to the relationships built between players in the same shard? After consulting the game’s reddit page, I learned that if I destroy a bridge in “my” world, it still exists for all of the other players in “their” worlds in the shard. By destroying something helpful created by another player, I am actually making things harder on myself. As an archaeologist, though, this feature gave me an excellent resource especially after my ethical question was answered.

I destroyed everything in my survey area. I knocked down every tower and pulled up every sign. I “burned” every bridge, and then I left on a trip to Florida. When I returned on December 5, I logged back into the game at 1700 and brought up the map. What was once virgin landscape was now populated by 13 new structures (plus 2 by the NPC) by 13 different players. I checked the placement of the new signs against the signs recorded on December 1 and 2, and they were grouped into three main clusters along with a few outliers in the countryside. The first cluster was at the top of the ramp leading up from the city’s Distribution Center. The second cluster of signs sat in a transition zone between the distribution center and a plaza giving way to the landscape. The third—and largest—cluster stood at the base of an incline leading from the city into the country. The content of the signs was variable between December 1 and December 5, but the placements were similar, the signs occupying sites of transition from one space to the next. The outliers were also placed generally at riverbanks, although their locations varied because of the number, position, and lengths of the rivers in the delta.

I will come back each week, and then monthly to this survey area to see how things have changed over time, but in doing so will miss the micro-changes that happen minute-to-minute in the landscape. Twenty-first-century digital archaeology, if it wants to keep pace with change in synthetic environments, must operate rapidly in its data collection. The data can then be compiled over time, which can then be analyzed for patterns of behavior. Based on this small test balloon with Death Stranding, I think that digital archaeology must operate at two speeds: 1) rapid for data collection in the field where the human pace and number of participants work at quantum-like speed; and 2) slow for data analysis, interpretation, reflexivity, and publication, which typically works at near-geologic time. Perhaps archaeology shares a kinship with physics then, where we might see a division of labor: quantum-archaeologists working at the speed of light, their agency affecting what happens in the worlds they observe; and the slow, interpretative archaeology that comes after, where theory is built upon data and reflection.

We need to develop a rapid archaeology for studying the temporary, fugitive nature of digital spaces inhabited and modified by humans, and this archaeology must take place alongside these digital events as they happen. This haste is not done in the name of preservation of digital sites, but rather to collect data as soon as they are generated, knowing full well the events and people generating this data will disappear in an eye-blink. But perhaps this is like conducting the archaeology of a river. We see the river as it is. We could conceivably record the river as it flows, trying to keep up with it (“rapids archaeology”). Or we can dip a bucket into the river, sample it, and then return later to the same river to sample it again to see what’s changed. Repeat sampling can then identify patterns of behavior, patterns of change, without the need of being in the river to collect all of its data all of the time. Each bucketful holds a representative sample. Changes and similarities in the samples lead to interpretation. Maybe this is the direction to take instead, knowing that it’s okay to miss everything just so long as we’re sampling something. But maybe I’m wrong.

—Andrew Reinhard, Archaeogaming