11
COULD BATMAN HAVE BEEN THE JOKER?
Sam Cowling and Chris Ragg
A Modal Question
Suppose that you and I are citizens of Gotham City, who read the Gotham Gazette with all its Batman-related headlines and have suspicions about the identity of Batman. We’re convinced he’s either billionaire playboy Bruce Wayne or—courtesy of a grandiose conspiracy theory—the criminal mastermind known only as the Joker. One day, the Gazette announces, “Batman Unmasked: Billionaire Wayne Is the Dark Knight!” Our suspicions have been confirmed; we now know Bruce Wayne is Batman. Despite this, it certainly seems to be true that the Joker could have been Batman. But is this really the case?
There are many ways to make sense of the claim “Batman could have been the Joker.” The way that will be relevant to our discussion is as follows: “It is possible that the Joker is identical to Batman.” Claims of this sort—claims about possibility, necessity, and impossibility—are modal claims. Typically, modal claims assert that the universe could have turned out a certain way. For example, “Batman could have two sidekicks” and “Catwoman couldn’t fight crime if her suit were any tighter” are modal claims about what sorts of things are possible for Batman and Catwoman. In ordinary language, these sorts of claims usually include terms like “would” and “might.” The truth or falsity of modal claims depends on facts about what sorts of things are possible and, in the particular case we’re interested in, on whether Batman could bear a certain relation—the identity relation—to the Joker.
In this chapter we’ll tackle a single modal question: Could Batman have been the Joker? Answering this question requires a fair bit of modal investigation and some serious metaphysics. Fortunately, metaphysics—the study of what exists and how it goes about existing—is the realm of philosophy that has the most in common with comics. It often gets complicated, not to mention absolutely bizarre. But before we consider whether the Clown Prince of Crime might have been identical to the Dark Knight, we’ll try to bring together a few key pieces of a very complex modal-metaphysical puzzle. And after introducing some (we hope) plausible metaphysical assumptions, we’ll consider an argument aimed at showing that, perhaps surprisingly (and perhaps not), Batman could not have been the Joker. Finally, we’ll raise some problems for anyone who accepts this argument and suggest why answering modal questions involving fictional characters like Batman and the Joker is a trickier matter than one might expect.
Some Not-So-Secret Things about Identity
Let’s start by introducing the first and most important piece of the metaphysical puzzle: identity. We often say things like “They have identical haircuts” or “Their outfits were identical.” These sentences involve a notion of identity between two separate yet extremely similar things. It’s important to note that the notion of identity we’ll be discussing is different in a very important way—in fact, it’s a relation that things bear only to themselves. The notion of identity we’re interested is expressed when we say things like “Dr. Jekyll is identical to Mr. Hyde” or “Chicago is identical to the Windy City.” So, while a painting and its forgery might be identical, in the sense of being extremely similar, it’s not the case that they’re identical in the way that we’ll be concerned with. After all, you might own one but not the other.
So, when we consider whether Batman could have been identical to the Joker, we’re interested in whether they can be one and the same individual in the same way that you’re identical to yourself and would be distinct from your twin, if you had one. We’re not interested here in whether the Joker might undergo extensive surgery, purchase a new costume, and near-perfectly resemble Batman. Rather, we want to know if Batman and the Joker could have been the very same, identical individual.
Here’s one thing we know for certain about identity: everything is identical to itself (or self-identical) and not identical to anything else. Given that every thing is self-identical, many philosophers endorse a principle called the Indiscernibility of Identicals (let’s call this “IOI” for short). According to this principle, in order for things to be identical, they must share each and every property they have. So, for example, you and yourself have all the same properties: you’re both human, you’re both literate, and you both know who Batman is. Because you and yourself share each and every property, you and yourself are identical. Now suppose that you have a twin sibling and that your twin was born a minute later than you. You and your twin are distinct: you do not share the property of being born at the very same instant.
Similarly, if Bruce Wayne and Batman are identical, we can say that if Bruce Wayne is a billionaire, IOI entails that Batman is a billionaire. While the fact that Batman is Bruce Wayne’s “secret identity” might make this claim seem somewhat counterintuitive, it’s important to realize that the sense in which one might think that Batman is not a billionaire is likely a looser, more metaphorical sense than we’re interested in. Because IOI entails that identical things never differ in what properties they have, it will be helpful to keep IOI in mind when we attempt to determine whether Batman could have been identical to the Joker.
Picking through Possible Worlds
Modal questions are framed in terms of “possible worlds.” Despite the name, these possible worlds—other ways this world could have ended up—are more like alternate universes than alien planets.1 By using possible worlds as a tool, we can distinguish between different ways for sentences to be true. For example, it is true that dinosaurs might not have become extinct, since there is a possible world where we live side-by-side with dinosaurs. Similarly, it is true that the automobile might never have been invented, because there is a possible world where there are no cars. Our world, the actual world, is just one of these many, many worlds.
We can use possible worlds to explain important concepts like necessity. If a sentence is necessarily true, then it is true in every possible world. “2 + 2 = 4” and “Triangles have three sides” are both necessarily true; so, in every possible world, 2 + 2 = 4, and triangles have three sides. If a sentence is only contingently true, then it is true in some, but not all, possible worlds. “Butlers exist” and “Superheroes exist” are both contingently true, since there are some possible worlds where butlers and superheroes exist and other possible worlds where there are no butlers or superheroes. Another category of sentences, including “2 + 2 = 3” and “Triangles have only one side,” are necessarily false; it is impossible for these sorts of sentences to mean what they actually mean and still be true.
Possible worlds represent what we could or couldn’t have done, so when investigating modal questions, we have to look to possible worlds for answers. Curious about whether there could have been invisible planet-sized penguins? Well, if there is a possible world where there are such things, then it is true that there could have been invisible planet-sized penguins. In addition, when someone asks whether you could have been late for a meeting, we can use possible worlds to determine the answer. Very roughly: if there is a possible world (very similar to the actual world) where you were late for the meeting, then it is true that you could have been late for the meeting.
You might wonder how this person who was late for a meeting is indeed identical to you. After all, you have different properties: one of you was late for the meeting, one of you wasn’t. These are thorny issues, but one way to understand this is by thinking of yourself in other possible worlds as you would think about yourself at other times. Five minutes ago, you were standing; now you’re sitting down. Despite this change, however, you’re still identical to yourself now and to yourself five minutes prior. In what follows, we won’t discuss the metaphysical details of how things can be identical over time without violating IOI. Instead, we’ll just assume that individuals can be identical across different worlds just as they are identical across different times.
One further thing to note before continuing is that for most of what follows, we’ll assume that Batman, the Joker, and the rest of DC Universe are merely possible entities. Merely possible entities exist in possible worlds, but not in the actual world. Given this assumption, the world could have turned out in such a way that Batman, Gotham City, and Two-Face would have existed. We’ll also be interested in a particular possible world: the possible world where all the usual truths about Batman and the rest of the DC Universe are true. For short, we’ll call this New Earth, as the current mainstream universe is called in DC. Having introduced notions like necessity, possibility, and possible worlds, we can now put these notions to good use in trying to answer the question at hand: could Batman have been the Joker?
If the correct answer to this question is yes, then there is a possible world where Batman and the Joker are numerically identical. This will be a possible world where Batman and the Joker exist as one and the same object. Put differently, in some possible world, the object picked out by our term “Batman” is identical to the object picked out by our term “The Joker.” This talk of terms “picking out” objects is actually shorthand for some philosophy-speak, and while the philosophy-speak is complicated, it’s important for getting to the bottom of matters modal in nature.
First, terms like “The Joker” and “Batman” are names, and names have a unique feature: when they refer to, or “pick out,” a particular object, they refer to that very object in each and every possible world. This unique feature isn’t shared by descriptions. To see why, consider the description “the police commissioner of Gotham.” On New Earth, the object that this description refers to is Jim Gordon. In other possible worlds, where things went rather differently, Alfred Pennyworth is the police commissioner of Gotham, so in these sorts of worlds, “the police commissioner of Gotham” refers to Alfred. Descriptions are, for this reason, very different from names in the way that they refer to objects. So, while “Harvey Dent” refers to the Harvey Dent in all possible worlds, “the former district attorney of Gotham” does not.
Second, the question we’re investigating isn’t merely about whether Batman could have been called “The Joker” and the Joker could have been called “Batman.” There is good reason to think that the English language could have developed so differently that the word actually used to refer to Batman could, in fact, have been used to refer to the Joker. What we’re interested in are the modal properties of the objects in question: Batman and the Joker. Like you and I, these objects both possess and lack certain modal properties. For example, many philosophers believe that you and I lack the modal property of possibly being a poached egg, but we do possess the modal property of possibly being an inch taller than we actually are. The fact that the terms “Batman” and “The Joker” are names doesn’t change the modal properties of these objects. Rather, the fact that these terms are names helps us better understand what reality must be like for sentences like “Possibly, Batman is identical to the Joker” to be true.
Necessary Secret Identities
Modal logic is a formal language—much like mathematics—used to simplify the way we reason about possibility and necessity. In modal logic, certain logical rules (axioms) are near-universally accepted. Here are few examples: If something is necessarily true, then it is possibly true. If something is true, then it is possibly true. A more complicated example would be the following: if something is possibly true, then it is necessarily possibly true. An absurdly complicated example is this: if something is necessarily possibly necessarily possibly necessarily true, then it is necessarily true!
The ins and outs of modal logic aren’t of crucial importance here, but one thesis of modal logic is important: the Necessity of Identity (hereafter, “NI”). If NI is true, then identity claims—claims that include the identity symbol “=”—are necessarily true, if they’re true at all. Actually, NI is more specific. It applies only to certain identity claims. Specifically, NI says that identity claims that have names on either side of “=” are necessarily true if they’re true, and necessarily false if they’re false. For this reason, NI guarantees that “Harvey Dent = Two-Face” and the equivalent sentence “Harvey Dent is identical to Two-Face” are necessarily true. NI does not, however, guarantee that “Harvey Dent = the former district attorney of Gotham” is necessarily true. NI doesn’t apply to this sentence, since “the former district attorney of Gotham” is a description rather than a name.
Why should we believe NI? Because of IOI and the obvious claim that, necessarily, everything is identical to itself. Imagine that x and y are like variables in algebra; they stand for objects. Now, if everything is necessarily identical to itself, then, in every possible world, x = x. And if x = y in some possible world, then IOI entails that x and y must have all of the same properties. This means that if x = Batman and y = Bruce Wayne and x = y, IOI guarantees that Batman and Bruce Wayne share all the same properties. Well, one property x has is the property of being necessarily identical to x itself; therefore, IOI entails that y must also have the property of being necessarily identical to x. In Bat-terms, if Batman and Bruce Wayne are identical in some possible world, then, since Batman has the property of necessarily being identical to Batman, Bruce Wayne must also have the property of necessarily being identical to Batman. The conclusion of this compact but complicated argument is this: when things like Bruce Wayne and Batman are identical, they are necessarily identical.
If NI is true and “Batman” is a name, then we can make an argument showing that Batman couldn’t have been the Joker: Batman and the Joker aren’t identical on New Earth. Since Batman and the Joker aren’t identical on New Earth, “Batman = The Joker” isn’t necessarily true. But, given NI, if an identity claim is true, it must be necessarily true. Because “Batman = The Joker” isn’t necessarily true, we can conclude that “Batman = The Joker” can’t possibly be true. And, since “Batman = The Joker” can’t possibly be true, Batman couldn’t have been the Joker! (Wasn’t that easy?)
“The Batman” and “The Robin”
Okay, we admit it—we’ve offered up a rather complicated argument for why Batman couldn’t possibly be the Joker. It relies upon some unfamiliar, although plausible, assumptions about identity and necessity. So here’s an objection against the argument we’ve formulated, and thankfully, it relies on vastly less complicated premises. Let’s call it the Robin Argument. We’ve assumed that “Batman” is a name, and we’ve concluded that “Batman” couldn’t have been anyone other than Bruce Wayne, much less the Joker. It seems, however, that “Batman” and “Robin” are the same sorts of terms, so, if one is a name, then they’re both names. If one is a description, then they’re both descriptions. As a matter of superheroic fact, there has been more than one Robin. Dick Grayson, Jason Todd, and Tim Drake (among others) have all been Robin at one time or another, so it isn’t necessarily true that Robin is identical to Tim Drake, or to Jason Todd, or to Dick Grayson. And if identity claims involving “Robin” aren’t necessarily true or necessarily false, then “Robin” can’t be a name, given what we know about names. So “Batman” can’t be a name either. And if “Batman” isn’t a name, then the above argument fails: NI just doesn’t apply to sentences involving “Robin” or “Batman,” since they’re descriptions rather than names.
There is something appealing about this conclusion. It seems that different individuals can take up different secret identities, and one way to explain why this is possible is by holding “Batman” and other “secret identities” to be descriptions. For example, “Robin” might be shorthand for “the guy—or girl, in Stephanie Brown’s case—wearing the Robin outfit.” In fact, one might think that this is fairly obvious. First of all, Bruce Wayne is often called “The Batman” and one might imagine that “Batman” is merely an abbreviation for a certain description. Second, others (such as Jean-Paul Valley and Dick Grayson) have donned the cape and cowl and have “been Batman” in Wayne’s stead. Third, the fact that “Robin” and other terms are so similar to “Batman” in use, and that “Robin” and other terms might not be names, provides reason to think that “Batman” is not a name.
But there are also good reasons to think “Batman” is not a description and is, in fact, a name. First, most descriptions pass a certain linguistic test. As with most linguistic tests, there are exceptions, but this test is nevertheless a generally good one. If certain terms show up in the description, certain inferences are usually good inferences. Consider the description “the ugliest criminal in Gotham.” If this description picks out an individual, we can reasonably draw certain conclusions about the individual in question. In particular, we can infer that that individual is ugly, a criminal, and located somewhere in Gotham. “Batman” seems to fail this test—we cannot reasonably infer from the fact that “Batman” refers to an individual that the individual is a bat. While this consideration isn’t conclusive, it at least gives us good reason to suspect that “Batman” isn’t a “disguised description.” According to some philosophers, disguised descriptions are commonplace. For example, if “Bruce Wayne” were a disguised description, “Bruce Wayne” would really just be a shorthand version of “the son of Martha Wayne and Dr. Thomas Wayne.”
Second, and more important, descriptions refer to objects by virtue of specifying certain properties. “The ugliest criminal in Gotham” picks out Killer Croc because Killer Croc is ugly, a criminal, and located somewhere in—or under—Gotham. But, if some individual wanted to, that person could strive to become uglier than Killer Croc, a criminal, and a citizen of Gotham. If he accomplished this rather strange goal, this description would then refer to him. Notice that “Batman” doesn’t work quite this way: no matter how good a costume you assemble, or what cave you build under your house, “Batman” will refer to the fictional character Batman rather than to anyone else. For this reason, “Batman” doesn’t seem to behave like a description.
It seems, then, that the Robin Argument gives us some reason to believe that “Batman” is not a name, which means that the argument involving NI fails. But there is also good reason, as shown by the two arguments just considered, to believe that “Batman” isn’t a description. For this reason, it seems that “Batman,” whatever it is, is a strange term indeed.
There are a few ways one might go about resolving this problem and determining whether “Batman” is a name or a description. One might argue that despite failing these tests, “Batman” is really a description and, since it is a description, it could refer to the Joker in some world other than the New Earth. One might also argue that although it seems like there have been multiple individuals that “Robin” refers to, there have really been multiple individuals with different names that are always “abbreviated” in some disguised way. If Batman and the Joker are merely possible individuals, then in order to figure out whether Batman could have been the Joker, we would need to resolve this issue. That said, we’ll close by discussing exactly what sorts of entities Batman and the Joker are rather than whether “Batman” is a name or a description. Ultimately, we’ll suggest that certain metaphysical considerations seem to suggest that Batman could indeed have been the Joker.
Fictions and Possible Worlds
Up to this point, we’ve assumed that although New Earth and its inhabitants are not actual, they still could have existed. According to this assumption, “Batman” and “The Joker” are merely possible entities like your merely possible twin brother or a merely possible galaxy-sized piece of French toast. If these entities did exist, they would be a lot like the physical objects that make up our universe. They would be concrete objects like you or me or this book or the Empire State Building. Most of the objects that we’re familiar with are concrete; they’re subject to the laws of physics and they’re located somewhere in time and space.
But fictional characters, like Batman and Robin, aren’t concrete. They’re more like numbers and stories; it just doesn’t make any sense to say that you have the number five in your pocket or that Hamlet is located on Coney Island. For these reasons, fictional characters are abstract objects like numbers rather than concrete objects like this book. There’s good reason to think that concrete entities are very different from abstract entities, but fictional abstract objects, like Batman and the Joker (but not the number two), seem especially different from concrete objects. One way that fictional abstract objects might be particularly unique is in the way that they possess modal properties.
The modal properties of everyday, concrete objects like tables, shoes, and books are determined by objective facts independent of what anybody thinks about them. In our argument, we treated Batman and the Joker like concrete objects, but as we have just indicated, they are quite different; they are abstract fictional entities. Still, if characters like Batman aren’t concrete inhabitants of merely possible worlds, it isn’t clear how we can make sense of the modal claims we make about them. Despite this, we can say that of the modal claims about fictional entities, some are true and some are false. Batman could have killed the Joker. Mr. Freeze couldn’t survive walking on the sun (or even Miami Beach). So how do we make sense of the modal properties of abstract fictional characters? Perhaps fictional characters like Batman have their modal properties in a very special way: they have these properties because they are created by authors and artists like Bob Kane, Grant Morrison, Jim Aparo, and Jim Lee.
More precisely, because abstract fictional characters aren’t merely possible entities, we can’t use possible worlds to make sense of their modal properties. This means that an alternative account of how fictional characters have modal properties is needed. Here’s one account we’re fond of: the modal properties—the properties of contingently or necessarily being a certain way—of fictional characters are stipulated by those characters’ authoritative creators. The modal properties of Batman are unlike the modal properties of giraffes and gazebos: the former are stipulated by authors like Bob Kane, while the latter are fixed independently of anyone’s intentions. If this is true, the things that Batman can and can’t do are determined, not by what possible worlds there are, but by what the authors of Batman comics believe to be possible for him.
Fleshing out this proposal more fully will prove a very complicated affair, but the consequence for the question we’ve been interested in is clear. Whether or not Batman could have been the Joker can’t be determined by our usual methods of modal investigation. To answer this question, we need to know what modal properties the creators of the stories involving Batman believe Batman and the Joker to have. It seems, then, that one way to answer this question is by old-fashioned investigation: reading.
Well, we did the reading—after all, that’s what we do—and the results are in: Batman and the Joker could have been identical. Why? Because in Batman: Two-Faces (1998), Batman and the Joker are one and the same. This Elseworlds tale—a Batman story that seems to takes place outside the typical continuity of the DC Universe, or on an Earth other than New Earth—describes a scenario where Batman and the Joker are, in fact, numerically identical, much like Dr. Jekyll and Mr. Hyde. So this particular case provides us with good reason to think that fictional identities (secret or otherwise) might not be quite so necessary as we would have thought.
All Joking Aside, This Is a Modal Muddle
At the beginning, we set out to answer the question “Could Batman have been the Joker?” We assumed from the get-go that “Batman” and “The Joker” are names of merely possible individuals—things that exist in other possible worlds—and, given these assumptions, we argued that there’s reason to think Batman couldn’t have been the Joker. Despite this, there are some reasons to believe that “Batman” might really be a description and, if it is, then Batman could very well have been the Joker. We then suggested that Batman and the Joker might be strange, abstract, fictional entities rather than merely possible individuals.
Some philosophers might complain about both of these options. These philosophers, sometimes called nominalists, prefer a sparse, desertlike view of reality. They deny that there are merely possible individuals or abstract entities. They take seriously the metaphysical maxim that “less is more” and believe only in actual entities and concrete individuals. While there are attractive features of the nominalist metaphysical picture, the nominalist has to bite at least one unappealing bullet: since the nominalist denies the existence of abstract or merely possible entities, the nominalist must deny that Batman, the Joker, and Robin exist at all. This consequence might make nominalism seem unattractive, but interestingly enough, you might think that Batman himself ought to be a nominalist. After all, if nominalism were true, there would be no crime in Gotham. That said, there would be no Gotham at all.2
NOTES
1 The “alternate Earths” of the DC Universe (before Crisis on Infinite Earths and after 52) are a lot like what philosophers call possible worlds; they are separate, alternate universes where reality took a different turn somewhere in the course of history.
2 Thanks from the authors to Chloe Armstrong, Barak Krakauer, Eitan Manhoff, and Chris Tillman for discussion and helpful comments.