Our books

Become a Fan

« Retracted post | Main | A Campaign for Better Philosophy Citation Practices? »



Feed You can follow this conversation by subscribing to the comment feed for this post.

David Shope

I wonder what Tomasik's position is on quitting video games. It seems like that would be a form of killing as well.

On another note, what about respawning enemies? Thinking about that raises the same personal identity issues relevant to debates about the possibility of resurrection, functional duplicates and so on (and what about saving a game, quitting, and resuming? Have the NPCs been killed? Did time cease for them only to resume?)

Marcus Arvan

David: Nice questions.

In point of fact, when you quit in "Halo" games, your character lets out a groan and falls over dead, as if from a heart attack or something -- so, yes, I think Tomasik would call it killing too.

On your note about personal identity, I think that's exactly right. I think *we* may survive post-bodily death in the same kind of way (viz. it doesn't matter how many trillions of years in the future a functional duplicate of you exists, if it exists, "you" would skip from here to there without missing a beat).


The post is great, and likely to bring interesting issues to the fore.

Some unordered thoughts.

(A) Aren’t there three different cases?

(1) I play a single-player (i.e. me against the computer) shooting game. *I* kill my enemies because I (supposedly) willfully aim my (virtual) weapon at them and click the “shoot” button.

(2) I play against other human players (online FPS). Again, *I* am the killer, but the characters I shoot are not simply the result of a computer program. “Their” actions are actually other players’ actions.

(3) I do not have direct control on the characters’ actions (The Sims). All I can do is “play god”, i.e. set the level of free will, etc. So if a character kills/does something awful to another, it depends on me only insofar as I chose to give them the kinds of attributes that allow them so to act.

(B) This seems to raise an interesting issue for philosophy of language. (Since I am definitely *not* knowledgeable in this area, this should be read as very, very sketchy at best, and pointless gobbledygook at worst.) Videogames characters are, after all, lines of code in a specified language (e.g. C++, Python, etc.). What we see on the screen depends on how the computer reads the code. Now we could imagine that the same language could have different meanings, depending on the reader (i.e. the environment). Imagine that we have two languages with exactly the same items and syntax, so that we could, in principle, write identical codes in these languages. However they differ in that the meanings of the items is not the same in the two languages.

We could thus have two games whose code is strictly identical, but that rely on different readers. In one game (call it The Killer), the line @S{i}{W}@W{3}{D} is read as “If you shoot individual i, then he’s wounded; if one is wounded three times, then he is dead.” In the other game (The Lover), the very same line is read as “If you smack individual i, then his welfare is increased; if one’s welfare is increased three times, he’ll be delighted.” Suppose that only the player knows whether he’s playing The Killer or The Lover (that is, whether he’s using a computer that reads the line as Killer or one that reads as a Lover). A third party could only know that this line of code has been read. So the third party cannot decide whether the action you performed led to a character’s violent death or to her being delighted.

It is plausible that computers are in such a position (that of a third party), and thus that they cannot have access to your actual actions. All they could “know” is that you appealed to this line of code. They then translate it a language you understand (the image on the screen), but which is arbitrary to them (it makes no difference to them whether *you* read the image as The Killer or as The Lover). And thus the computer could not say that you killed a character.

In this sense, a goomba in Super Mario Bros is a Goomba only to the extent that *we* read it as a Goomba. From the computer’s standpoint, it makes no different whatsoever whether it is a goomba or anything else, except that it will combine the language with different reading tools. If this is right (or at any rate intelligible), killing a videogame character makes sense only for us, not for the character or for the computer.

(I hope this was not nonsensical --- I wouldn’t like to have lost your time, and others’ as well.)

Marcus Arvan

Hi Pierre: Thanks for your comment! That's a lot to process, but here goes.

They are different cases, but I'm not sure they're *that* different. When my player takes "damage" (loses health) or dies, the game codes his damage just like the non-player-controlled characters (NPCs). So, I think, if any videogame characters have experience, they probably all do. My character's "goals", however, are more complex than the NPCs, because their NPCs are executed by a much simpler algorithm.

In terms of the "god" case and the Sims, you're right, if you set the characters' "free will", you're not *directly* responsible for action, but you are indirectly responsible for the entire state of affairs (namely, that *these* characters exist, and are doing the awful things they are doing according to the game's algorithm).

Finally, I like your first point about coding. I actually made a similar point about consciousness in an unpublished paper I wrote several years (which I still think is good, but which never found a home in a journal). It would, I think, be *possible* for different ways of coding the same functions to lead to different character qualia.

However, I don't think your final point is right (viz. "killing a videogame character only makes sense for us, not for the character or for the computer"). What do you mean by "makes sense"? If a character dies, there is *some* algorithm representing that character's desires, etc., as well as that character's death. On a panpsychist view, that algorithm will have *intrinsic* experiential content (i.e. content for the character the algorithm comprises). Similarly, on a teleofunctional account, the code that comprises the character *comprises* intentional states of the sort that we ordinarily take to have moral value (i.e. desires, goals, etc.).


Thanks for the reply!

By “make sense”, I mean simply that, from the computer’s standpoint, all the computer does is performing a formal task in response to the player’s input (e.g. click on a mouse button). There is at least one part of the computer that has no access to the dictionary that makes these tasks meaningful for the player or for the game characters. So, even if the computer has experiential content, it might have no access to the player’s and the characters’ experiential content.

In other words: the character’s experiential content is not entirely given by the algorithm that produces this content. It requires a dictionary that makes the character able to have this experiential content. From the character’s standpoint, of course, its own experiential content may be the only it has access to, even though there are many alternative possible worlds, including some in which the same algorithm would produce radically different experiential content.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Your comment could not be posted. Error type:
Your comment has been saved. Comments are moderated and will not appear until approved by the author. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.


Post a comment

Comments are moderated, and will not appear until the author has approved them.

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)

Job-market reporting thread

Current Job-Market Discussion Thread