What exactly is the agency of the virtual non-human human? This question hit me when I was in the midst of editing what is shaping up to be my first publication, (Re)Building Çatalhöyük: Changing Virtual Reality in Archaeology, a text refashioned from my more stridently titled paper presented at the World Archaeological Congress, Get Real: A Manifesto for Virtual Reality in Archaeology. In this paper I establish Second Life as a pretty good venue for changing the way we reconstruct archaeological sites and examine the use of avatars and non-player characters (NPCs) in these reconstructions. My good friend (and professional copy editor!) Lei-Leen had done me the favor of proofing it and she highlighted my use of “person” while I was discussing these NPCs. Certainly these people-created constructs were not people themselves.
In the paper I argue against the creation and use of NPCs in virtual archaeological reconstructions:
NPCs are “non-player characters,” entities generally created as a kind of set dressing, present to answer scripted questions about the reconstruction, or to provide ambiance. This tends to bring out the worst in reconstructive impulses; in one example, Julius Caesar reclines on his couch, reciting information about his villa to the user. A recent simulation of an African American site in Oakland, California contained prominent members of the community who stood by the side of the road to robotically impart their knowledge when approached by the user’s avatar. Turning people of the past into mere mouthpieces for their architecture diminishes the rich potential of reconstructions to impart information about complex lifeways. Using programmable objects such as the previously mentioned mano and metate allows avatars to act as their own guides to the past, populating the re-created ancient landscape with avatars of people interested in the past, interacting with artifacts and taking on roles suggested by these artifacts. This is simple for archaeologists who are accustomed to telling stories through objects and adds another level of interactivity to the virtual reconstruction.
I still believe this to be true, that fundamentally we are better off wearing Caesar’s crown for ourselves rather than asking a poor simulacrum about the weather in the Republic. Thinking of Caesar as a non-player character in history is a stretch by any means. But game developers (and digital archaeologists) will probably not stop populating virtual worlds with fake people. These NPCs are nonhuman manifestations of a network of agents (polygons, “modern” humans, fiber-optics, and the dead person herself) and the relationships between these agents and as a result should be studied as such. But does this understanding of an NPC as a network make it ethical to take such liberties with the visages of the dead? What does it mean when Joey Ramone comes back from the dead to sell Doc Martins?
Of course, I could just be reading too much Latour. Let’s ask Eliza.
* Hello
> How are you today.. What would you like to discuss?
* Latour
> Tell me more…
* Have I been reading too much Latour?
> Can you elaborate on that?
* He’s a modern social theorist.
> What does that suggest to you?
* That I should be reading him for my dissertation.
> Say, do you have any psychological problems?
* Possibly.
> Tell me more…
Colleen Morgan 
my friend tony and i were talking about making an eliza bot this weekend that would claim to be pyschic and do cold readings and guided questioning.
We figure if someone’s stupid enough to fall for cold reading, they’re probably also stupid enough to fall for an eliza bot.
the transition from NPCs to Eliza was hilarious. Also, something about this post reminded me of late, crazy Robert Heinlein with his anthromorphic ship computers who ended up with bodies according to his imagination.
…Which reminds me of the irony of the Henry Ford Museum, which doesn’t even try to use the employees inhabiting its “historic” buildings and toiling in its reconstructed agricultural sites (which puts “the farm of 1850” right down the path from “the farm of 1750”) as re-enactors. The employees wear the poke bonnets, they work the fields by hand, they attempt to make cakes in 1830s kitchens using 1830s Michigan cookbooks … but they do it all as themselves: 21st century Michiganders who are working a shift job which happens to be at a museum.
If you can’t tell, I really admire the museum curators’ respect for both the reality of their employees’ lives and the uncrossable distance between the people doing the work of sharecroppers and the actual sharecroppers of a century ago.