The Possible and the Virtual

Tod's picture
Forum: 

I've been studying a lot of Deleuze in the last few weeks. Because of this, I have a giant bundle of tangled thoughts working their way toward design-oriented theories, but they are difficult to elucidate and may take me some time. Today one such idea crystallized, and here it is...

Deleuze made a distinction between the Virtual and the Possible. He said the former was something Real that could become Actualized, while the latter was something Not Real that could be Realized (that is to say, "Real-ized," or "made real").

The distinction can be tricky. I'll try to explain. I'm going to use the words "system" (a word we are all familiar with) and "assemblage" (Deleuze's term) interchangeably here.

The Possible, said Deleuze, is NOT Real, because it hasn't happened AND is not guaranteed to happen, given nothing but the existence of the system itself. However, contingent upon certain events occurring between objects within the system, it might be Real-ized. The Possible, therefore, is a multiplicity of system states, each of which depends upon some event or action in order to become Realized. At the moment that happens - IF it happens - it becomes no longer Possible, but Real.

The Virtual, on the other hand, is always Real, whether or not it is currently manifest in an Actualization. That's because the Virtual is a "field" of effects that emerge via contingencies within the entire assemblage (or system). When the system as a whole enters a certain state, this or that particular Virtuality may become Actualized. It might be said that at that point, the objects within the system are guided by the imperatives of the Virtual, rather than "causing" anything to happen. In addition this new Actuality may affect the Virtual Field, giving rise to new Virtualities.

For Example:

THE POSSIBLE: If you throw a basketball toward the basket, hoping to sink it, you may or may not succeed. The event of sinking the ball is Possible. But before it actually happens - i.e. before sinking it becomes Real-ized - it is Not Real. And if you miss the shot, it never becomes Real.

THE VIRTUAL: If you release a bunch of hydrogen molecules and oxygen molecules into a space with no intensive properties (i.e., no extremes of temperature, weird gravitational fields, or whatever else might cause a different effect), the fact that many of them will form into water molecules is Virtual; that is to say, it is an emergent property of the assemblage (the whole assemblage in this case being hydrogen molecules, oxygen molecules, and the space they meet in). The fact that hydrogen and oxygen bond to form water is Virtual; that is, it was already present in the Virtual Field of the system, even before the first two molecules bonded. We might even say that the Virtual Fact of bonding drives the actions of the individual molecules, for it does not seem that the molecules engaged in any decisions or actions "on their own."

So.
The Virtual is Always Real even when not manifest, but emerges into Actualization in certain system states.
The Possible is Not Real until some action occurs within the system that causes it to become Real-ized.

Let's see if we can put this in game system terms:

THE POSSIBLE: We're playing D&D. Your character has been stung by a giant wasp. You need to make a saving throw. At this moment your character's death is Possible (let's call that State 1), and your character's resistance to the poison is also Possible (State 2). You roll the dice and look at the result. The moment you do that, one of these two states becomes Real-ized. The other simply ceases to be Possible, it was never Real anyway.

THE VIRTUAL: We're playing Fiasco. At a certain point (the mid-point) of the game, we all know that The Tilt is going to happen. We don't know what it will be, of course, but we know that the story will take a drastic and destabilizing turn at that point. As long as we don't stop the game early we can say that the fact of The Tilt (ie, the fact that The Tilt will happen) is Virtual, but depending on what we roll, the precise details of that Tilt - each 1 out of 36 distinct potential results - are only Possible.

It seems to me that almost all of the game mechanics we use almost all of the time involve Realizing the Possible, rather than Actualizing the Virtual.

It also seems to me that this sort of thinking - about Actualizing the Virtual - may be (?) more applicable to genre sims and narrativist forms than simulationist or gamist forms. (Let me know if you concur, I'm still thinking that through.)

In any case, it might behoove us to begin thinking more about Actualizing the Virtual, and finding ways to incorporate these sorts of dynamics into play. We might just turn up a whole new category of design possibilities and playstyles.

Tod's picture

These are just some thoughts I've had while ruminating...

Actualizing the Virtual if performed directly by the GM might have a feeling of "railroading" to it (suddenly the country is at war and the whole game has changed!), but by making it a mechanic rather than fiat, the GM can waive personal responsibility for a change of Game State.

Actualizing the Virtual might be done in the OSR by randomly rolling something extreme (suddenly the country is at war). Lots of GMs do this kind of thing, sure, but then they need to use reverse-reasoning to explain it. But such drastic changes in Game State rarely feel natural; they may break suspension of disbelief. Some of the old Arduin Grimoire tables have this quality. The problem is not that your country might never go to war - of course countries do, and that's probably out there in the Virtual Field at all times - but rather that by rolling it randomly you have no particular context or reason why this particular country is suddenly at war with that one. There's no continuity. It would be like ice suddenly changing to vapor; we don't usually see that happen. Instead we see that as the heat increases this actualizes frozen H2O into first its liquid modality, and then its gaseous one. That increase in temperature must happen for a reason, and that reason should be something in the "Actual" gameworld.

Here's a notion: Imagine a series of Portents (I prefer DW's Portents to AW's Fronts) involving stages of "at war-ness" (picture the Terrorist Threat diagrams, perhaps). But instead of being triggered by MC moves, changes in state are triggered by specific events which are possible or likely to happen in the game world... perhaps even pushed forward directly by action/inaction of the PCs. Or NPCs. Or whatever. The point is: they are pushed forward by potential events in the "Actual" gameworld, they may be known or suspected beforehand, and they may change the entire nature of play.

What I'm picturing is more like what Vincent did with his Principles and Moves: to create formal mechanics for global changes in Game State or even the Game System (!), based on known or knowable types of possible events.

I'll probably think of smaller examples than "WAR!" as time goes by; right now it's easier to see really big ones.

Tod's picture

Talk to me about how Escalation happens in Dogs in the Vineyard. I think it may be an example of Actualizing the Virtual. Another example might be the placement of Crises and Acts in a Trad-GM'd game of DayTrippers. I believe these both satisfy a similar critera: that "Story" elements as they unfold are guided by mechanisms adhering to the designer's (and thus, the GM's) intended Narrative Structure. In such a case, the Narrative Structure may be seen as a Virtual Field, and events within this Field have a sort of "inevitable-yet-conditional immanence" to them which goes beyond the mere existence of (say) the fact that we consult a Wandering Monster table every X minutes.

(This line of reasoning was what led me earlier to opine that mechanics for Actualizing the Virtual might better be suited to "Narrativist" games, although I still withhold judgment on such a blatantly categorical assertion).

Paul T.'s picture

I think I understand what you're talking about now, Tod. Cool!

I'm not sure I see any clear distinction between those things, but as poles in a conceptual model of various "things that haven't happened yet", they are interesting to consider.

Escalation in Dogs is something that any conflict participant has the option to do. If you and I are talking, and it's not going well, you can choose to throw a punch.

We could call that Realizing the Possible ("these are your options; you can do this if you want!") or Actualizing the Virtual ("the threat of violence is always present, whenever we talk, ready to be committed").

Corvinity's picture

I agree with you that actualizing the virtual (as you've explained it) is more common in narrativist and genre-simulating games, but I don't think it's new. I think "actualizing the virtual" is a decent way of describing one of the fundamental shifts that a lot of "narrativist" games made from trad games in the early 2000s. One extremely simplistic way of telling that story would be that people got frustrated with World of Darkness games that promised certain kinds of stories, but didn't have mechanics that made those stories virtual in Deleuze's sense. The mechanics dealt with lots of possibilities (Did you succeed or fail? Did you hit? How much damage did you do? Did you lose a point of humanity?)

By contrast, games like Dogs in the Vineyard and Apocalypse World reliably produce certain kinds of stories (actualize the virtual). What's uncertain is the details within the story. So in a good genre-simulating game, the possibilities that are in play are all always in-genre. What this tells us if we're trying to design such games is that there should never be a roll for which any possible result would contradict the basic intention of the game. Or, stated positively, Every result of every roll should advance the basic intention of the game. PbtA does this explicitly, but there are many other good ways of doing it.

I don't know if everything I just said is obvious, or was already implicit in the OP. If so, just read the subject line.

Emmett's picture

Actualizing the virtual happens all the time in trad games. Feats in D&D are usually a good example. They have the potential to be actualized when the rules say they fit and there's often no test to see if they work. They're virtual until called on and then they're actual.

I kind of see this as "Do you roll for it?" (or use some other randomizer).

Chess moves are all Virtual. Monopoly moves are Possible.

Tod's picture

@Corvinity -- Yep, you're right.

@Emmett -- D&D Feats, given your description, can be seen as Virtual in regards to the character who possesses them. They are always present even if not currently manifesting. But they straddle an ontological line because they need to be activated/called for by the Player; i.e. they are not "automatic" transitions of the character's state. If they were automatic (like an error-free or computer-driven application of an AW Move saying "When you do this, that happens"), they might be considered truly Virtual. I suppose we can say they are Diegetically Virtual but only if activated. On the non-diegetic level - up here in "Reality Prime"(TM) - they are merely Possible, because the Player has to actually decide to use them. Once that decision is made, the Feat (provided it requires no action resolution test) is Actualized in the gameworld.

Which is an interesting distinction, and one of the cool things about our hobby: the multiplicity of reality levels is definitely a factor!

But I think it's too simplistic to make the distinction (Possible vs Virtual) simply a matter of "roll or not." I disagree on chess moves, for instance. Or I can agree on chess moves if we're talking about the state of the entire game, that is to say: given a particular move being played, the state of the game shifts from an Unactualized Virtuality to an Actualized one. The move itself, however, is merely a Possible action in real space. It is this action which brings about the Actualization of the new Game State.

Tod's picture

The Virtual seems to cleave more easily to the level of "rules" - or perhaps we can say "structure" - than the typical outcome of an Action Resolution (which, as I think I've mentioned, is almost always a Possibility Realized). But beyond that there is a qualitative difference, in that the Virtual is always true of the assemblage - whether we're talking about a character or a location or a law of physics or the state of a narrative (for instance, the fact that the narrative will have a "climax" even if we don't know when or what it will be) - whereas something Possible either gets realized or it doesn't, and if it doesn't, it continues to have never existed (like all the numbers you didn't roll when you consulted a table).

The existence of the STR stat in D&D is merely a quality that all characters possess, but your PC's particular score is a Realized Possible.

Character Levels - assuming you don't die before reaching them - might be considered Virtual, along with all the new qualities your character picks up when reaching this or that level. In other words, your thief character having a 47% chance of Lockpicking is Virtual, and will be Actualized the moment you reach Level 6.

"Dead" is a Virtual.
"Stung to death by giant wasps" is only a Possible.

Emmett's picture

I can see the distinction you're making in the original post, I think what tripped me up is that later on you describe moves as virtual. Moves are literally the same concept as a chess move. They both are triggered by choices. So while the rules of the move are virtual the choice that leads to them would be possible. The Tilt in Fiasco is virtual because it's going to happen. Escalation in Dogs In The Vineyard is going to happen so it's virtual.

It's mentioned that a war breaking out being virtual. Then it's mentioned that in an OSR game that virtual war might be rolled for on a table. Anything rolled for on a table is strictly possible. So my earlier statement should be amended, if the state of the game depended on a dice roll it was possible realized.

Going back to the original idea of a GM deciding that the war was going to happen. Is that virtual? Let's say the GM is going to railroad a war. They think it's the best idea ever and they'll shoehorn it in no matter what. What's interesting about that is, it's a choice made by the GM. They could get distracted, they could get overwhelmed by events in the game and never get to their war. They might be dissuaded by player complaints. What happens to it's virtual state then? Sure it's likely to happen, but is it virtual?

The war is a thing in the GM's mind, it only becomes real when they declare that it has. It's not mandated that the war will happen except in the GM's mind. What this comes down to is that if the war is virtual then the GM's mind is part of the system. If the GM's mind is not part of the system then it's possible.

So if the GM's mind is part of the system, are the other player's minds part of the system? At what point does making a choice make something virtual? If the player decides before the start of a game that they'll sacrifice themselves (something in their power to do) is that virtual?

I'm hanging way out on limbs here but that's where the concept takes me.

Tod's picture

I don't think I ever spoke of Moves as Virtualities, but I did use the random war as an example, and it was a poor one. It needs clarification. My thinking at that moment went like this: Every country always has the possibility of being at war, and there are a lot of things we know about the state of war that would immediately come into play in that event: attacks upon the city walls happen, city defense steps up to an extreme level, conscription and volunteerism begins to occur, xenophobia, paranoia and fear of spies sets in, some spies actually do get in, some people become traitors, the economy changes (prices of certain goods go up, other goods become totally unavailable, a black market arises), the king makes sweeping political decisions, curfews are established, encounter tables are replaced by different encounter tables, etc - a kingdom "at war" has shifted into a totally different state that we know lots of things about, even before it ever happens. Point being that the entire state of the "City Assemblage" - and thus the nature of the whole game for anyone in the city - has changed radically. Playing in the city at war is a different game than playing in the city at peace.

Like the Player choosing to activate a Feat (thinking about your example has made this clearer to me), the GM in my "random war" example has rolled something that's a Possible for us in Reality Prime, and that roll determines that a Virtuality (the state of being at war) has become Actualized in the gameworld. Still, it's not the most satisfying example. It becomes much more clearly a Virtuality if the GM has written a module in which war is secretly guaranteed to happen in session 3 (this might be considered "railroading"), or if the module says that war happens ONLY IF certain events transpire in the gameworld (and this would not be considered "railroading" if it depends on something the Players do, or fail to do). Even more clearly if the game rules actually include a section on "things that change when your city is at war." In that last case it's understood by all who read the books that the state of war is a Virtuality that may occur in any city at some point.

So yeah, it can be tricky determining the difference between the Possible and the Virtual, even when dealing with only one "level of reality." It becomes even trickier (and even more interesting for philosophy wonks) when we take into account the fact that we are actually operating in two realities when we play (i.e., "Reality Prime" and "The Gameworld").

Virtualities are always present, hanging overhead like the Sword of Damocles just waiting to fall (given certain conditions), while Possibilities come and go or else never exist at all, like forks in a shifting river.

I feel like the question about "what's part of the system" is kinda orthogonal to this topic, and different GMs/games/playstyles will answer that question differently. Personally I do consider myself part of the system when running, but others do not, so I'll save that question for another thread.

DeReel's picture

From what you describe, the divergence between the 2 is : a possible has competitors, a virtual has none. A possible creates a one way fork (at best, maybe it's a chaotic whirl), a virtual is linear (one can mentally reconstruct the cause from the effect)
The main use of virtual in a game would be to be there in a player's head, as an horizon, informing a strategy. In my game, in Act I you try to take advantage of the discount on introducing facts, because you know it will close soon. Or you close it if you think you have enough material. It's possible that Act I ends at any time. But, virtually, when you begin the game, Act I is closed, the game itself is ended. Mektub?

Thanuir's picture

Virtuality is a matter of degree. There is the chance of the game ending, or having a rules argument, or a player dying of heart attack, and so on, before the virtual thing actualizes. (Maybe the heat death of the universe is guaranteed. Maybe. But in the context of games, everything seems always to have a shade of uncertainty.)
We might want to ignore this and instead treat the, ahem, virtually certain things as virtual. Very convenient for designing games.

Virtuality happens all the time the usual rpg mechanics. Whenever you have committed to a cause of action, rolled the die, and are doing the arithmetic, reading the spell description, and updating the game state, you are turning something virtual into real. Or checking your notes for the next scene or the room description and then narrating.

Tod's picture

The key difference, per DeLanda's readings of Deleuze, is that a Virtuality Actualizes by way of its own internal dynamics. It's not so much a "decision" as a "state." You could say it lies dormant within the assemblage at all times, even when not Actualized, just like the propensity to form ice is present in H2O at all times, even when it is not freezing.

I agree with DeReel's example of a virtuality, it feels similar to my earlier reference to Fiasco.

I agree that the rooms in the GM's notes are Virtualities for the Players, but if we start talking about both layers of reality at once I'd have to agree that everything is a Virtuality in a sense, because we are playing a game designed to do exactly what it's doing. But that's no good, because in order for the term to be useful it needs to be more specific, so I want to think about Diegetic Virtualities supported by Mechanics. That is to say: People, Places, and Things that may be encountered by the PCs in two or more radically different states depending on Condition X, Y or Z. Best if the Condition is something the PCs have been involved in enabling or inhibiting. We avoid railroading, because the GM really doesn't have any control over when the Condition will occur. That control may be ceded to dice, or PC actions, or NPC actions, Fronts/Portents, or the Narrative Arc itself (as in Fiasco).

My purpose here is not to point at any game and say "THIS is Virtualizing the Actual" but rather to prompt discussion about the differences we might tease out of this subtle distinction, in order to see whether it affords us any interesting design possibilities.

DeReel's picture

Now I think I understand Thanuir's idea of degrees : I can either describe a possibility as an act of creation, an event that is yet untamed by reason and judgement : an ontological leap. Or I can say this appeal to metaphysics shows the limitation of my human mind, and yes, the way water freezes is as chaotic and deserving of wonder as the way persons in a game can ping pong a fiction into existence : a question of degrees. Both are true and rational* ways of thinking. (*although not scientific)

Pragmatically, personnally, and for this whole present week at least, Virtuality in design is about Content, the huge keyring of options and material available to players. The actual way they'll use one specific key, is full of "ipseity", it will happen in its own specific way, because it is by definition different from any other, etc. But I can choose to rule these variations as not pertinent. Of course, it's a choice I make of a perspective, for practical reasons.
BTW, w hen y'all talk of "virtuality in narrative games" in the previous posts, I hear Structure.

While on the other hand, I can say that Possibilities deal with "Player interactions". And in saying this, I rule that in some parts of the game, I want to refrain from using any frame or thought pattern, and be as open as I can to events, because, hey, let players be, it's their game after all.

But the concepts introduced by AsIf possibly give me something more : that as soon as I know certain rules about how arguments develop, or techniques to settle them, that becomes part of my "design keyring", and something I can use.
For symetry's sake, I would like to find a mechanical part that would leave room for Possibility, but except the basket ball, or its table top cousin, the rolling dice, I can't see anything obvious. Maybe PbtA moves, because they have so flexible ducts and hoses ? Or maybe it's just how some tables use the Game as a collection of Toys (capitalizing to show these are old terms established in a bunch of old discussions) ? Hm... In which case I'd need comparisons that highlight the differences between the 2 discussions.