This is the second post in our series on metaethical issues surrounding the undead in Pathfinder. To start at the beginning, with the introductory post, click here.
The single biggest question in this series is, “Are all undead necessarily evil?” Now, because it’s only five words long, that question seems like a simple one, at least philosophically. Unfortunately, as anyone with experience in analytic philosophy already knows, one of those words is a real humdinger*: that nasty little “necessarily.” The fact is that necessity (and its philosophical converse, “possibility,” or perhaps “contingency”) has been and remains one of the slipperiest and most difficult concepts in analytic philosophy, whether we’re talking metaethics, philosophy of language, or any number of other subdisciplines. Unfortunately, it’s also phenomenally important to this specific issue (we’ll talk more about why shortly), so if we’re going to treat this topic with the care and rigour it deserves, we’ll have to spend some time clearing up what we mean by “necessity,” and by extension what it would mean for undead creatures to be not just evil but necessarily evil. Be careful on this post, folks: it may get a little technical, but more than anything, there is a serious risk of learning something that’s not strictly about Pathfinder this time…
So, a little terminological background: the logic of necessity and possibility is known as “modal logic,” and talk about these concepts is often called talk about “modality”** – if someone is talking about modal logic or using that sort of language, they just mean “logic about what is possible and what is necessary.” This is in contrast to older, more established forms of logic, about what is and is not true (these usually fall under what are called “sentential” and “predicate” logics…but that’s neither here nor there).
Now, for centuries, philosophers had a surprising amount of trouble with modal logic, because no one was really sure exactly how to understand what necessity and possibility mean – in technical language, the semantics of modal logic was ambiguous. This ambiguity essentially occurred as a result of there only being one world: that is, we often don’t have trouble saying whether something is true, but we always have trouble saying whether it could have been false. When the skeleton strikes at your fighter with a rusty sword, did that happen merely because it was possible, and things could have been otherwise…or did it happen necessarily, such that it could not have been otherwise? The structure of our world, and the fact that we can’t just go back in time and see how it runs on a second try, apparently makes this question totally opaque. Even if your GM rerolls the skeleton’s attack because the first result was contested or something (probably the closest possible thing you’re going to get to actually going back in time and checking again), you’re not actually checking whether that first die result was necessary or simply possible; you’re just rolling the dice again and stipulating that it will serve in place of the previous result.
As you can probably tell, this question of distinguishing between events which are necessary or merely possible is sort of tied in with questions about the nature of free will, the truth of determinism, time and space, and so on and so forth – put another way, this seemingly simple question was just thought to be way too complex to be addressed head-on in any meaningful way.
Enter Saul Kripke, who comes onto the philosophical scene in the mid-20th century and is a leading contender for the title of “most important living philosopher” (he’s still alive as of my writing this, anyway – click the link above for updates on that). While he hardly does all this alone, Kripke’s work is sort of the impetus for our coming to understand how to interpret questions of necessity and possibility, and for the resulting development of useful, helpful modal logics.
It’s also a surprisingly accessible kind of solution (though hardly without its complexities), for such a seemingly technical and difficult problem. Essentially, Kripke introduces the notion of “possible worlds semantics:” the idea that we should understand what these confusing modal ideas mean in terms of “possible worlds.” A possible world is just a logically consistent world (i.e., one which doesn’t contain logical contradictions) that is more or less like our own. The actual world is, of course, one possible world…and then there are an infinite number of other possible worlds, more or less different from the actual world, where some things are the same and some things are not. Now, to be clear, Kripke doesn’t think that these worlds exist in a metaphysical sense (although some philosophers, like David Lewis, called “modal realists,” think that other possible worlds actually do exist); they’re just semantic constructs to help us understand modality.
In the context of possible worlds semantics, then, x is possible (or contingently true) if it happens that x is true in at least one possible world, while x is necessarily true if x is true in every possible world; if there is no possible world in which x is true, then x is necessarily false. To return to our skeleton attack example: if there is at least one possible world (i.e., a logically consistent world which we can imagine) in which the skeleton hits you and at least one possible world in which the skeleton misses (which strikes me as plausible), then it is contingent but not necessary that the skeleton hit you. This is the case even if, as a matter of fact, the skeleton did in fact hit you: because there is a possible world in which it did not hit you, the fact of its having hit you is contingent, not necessary. Note that this understanding doesn’t require that we solve any of those difficult issues of determinism, free will, etc.; possible worlds semantics gives us a way of understanding the meaning of contingency (i.e., possibility) and necessity without having to understand all the underlying metaphysical issues. It’s effectively a semantic workaround…and a damn good one!
So that, broadly speaking, is how we’re going to understand what it means for the undead to be necessarily evil, thereby structuring the questions in the rest of this series. It may be that the undead in Golarion are all evil, but that is only one possible world. If there is a possible world in which there are non-evil undead, then the undead of Golarion (and untold other settings of Pathfinder games) are contingently but not necessarily evil…much like people, presumably.
On the other hand, if there is something (probably not something obvious, but something nonetheless) that builds evil into the very fabric of what it means to be undead, then there is no possible world in which there are non-evil undead (because something about the concept “undead” implies the concept “evil”), and the undead are necessarily evil. And that is the very detailed, clearly-laid-out version of our main question, which we’ll be addressing over the course of the rest of this series.
Alright, that got unpleasantly technical at points, I know…but hey, look where we are! Now we have a better understanding of what we’re actually talking about, and any future arguments about whether undead in Pathfinder are necessarily evil can be much more focused and much less subject to equivocation, intentional or otherwise. I hope you saw fit to power through that, and are willing to post any questions or concerns you might have; either way, we’re be onto much more substantive and direct questions on this issue in no time now!
*Actually, at least two are – there’s no question that the concept of “evil” is tremendously complex and difficult no matter what kind of moral philosophy we’re doing. But that’s the kind of topic that’s covered either elsewhere on this blog or nowhere at all – in any case, we don’t have time to go into it here.
**If we’re being super technical (and obviously we are), “necessity” and “possibility” are actually just alethic modalities. There are several other different types of modal logic: deontic modalities (the logic of “required” and “permissible”), temporal modalities (the logic of time: “eventually,” “always,” “until,” and the like), epistemic modalities (the logic of knowledge and the absence thereof), and so on. Confused? Good; that just means you’re still sane. We’ll ignore everything but alethic modal logic (by far the most common anyway) for our purposes here.