Oliver Laas
“Father” Zachary Hale Comstock is a self-professed prophet, religious zealot, and racist, who has kept his “heir” under lock and key in the floating city of Columbia. This inheritor is a young woman, Elizabeth, who has the ability to open “tears” or “doorways” into alternate worlds and different times. Comstock subjects her to his will, ultimately leading her to destroy New York in 1984 to fulfill his “prophecy.” Clearly, Comstock is morally responsible for his actions.
Booker DeWitt is a washed-up, disgraced ex-Pinkerton agent haunted by his participation in the Wounded Knee Massacre. He enters Columbia to rescue Elizabeth in exchange for having his gambling debts settled. After much bloodshed, Booker saves Elizabeth and kills Comstock. Certainly, Booker is blameworthy for the carnage, and praiseworthy for saving Elizabeth.
But (spoiler alert) Booker is Comstock. In the past, Booker attended a baptism to assuage his guilt over Wounded Knee. In some worlds he rejected the baptism, got married, had a daughter, Anna, and ended up as an indebted alcoholic ex-Pinkerton agent. In other worlds he accepted the baptism, took the name Zachary Hale Comstock, founded Columbia, and acquired his “heir” by opening a tear to the Booker world and buying Anna in exchange for wiping away Booker’s debts. In the Comstock world, Anna is renamed Elizabeth in order to conceal her true origins. Who is responsible for what? According to common sense, both Booker and Comstock are responsible for their own actions. Yet the game’s ending implies otherwise.
Questions of responsibility are intimately related to the free will problem—a perennial issue in philosophy that has at least two aspects.
The first aspect concerns the kind of freedom that can generally be defined as the ability to act without external or internal constraints. This differs from surface freedoms, such as the freedom to travel anywhere or to buy anything one likes. The relevant deeper sense of freedom is supposedly captured by the notion of “free will.”1
The second aspect involves moral and legal responsibility. Free will is related to notions of accountability, blameworthiness, and praiseworthiness. For example, in BioShock, Jack (or the player) can do one of two things to the Little Sisters: he can either harvest them—this yields more ADAM for enhancing his abilities, but kills the girls—or cure them—this yields less ADAM, but saves the girls. Suppose that Jack harvests one of the Little Sisters. We would say that Jack is blameworthy for what he did because he could have chosen otherwise. However, suppose that someone puts a gun to Jack’s head and orders him to cure the girl or forfeit his life; Jack rationally chooses the former. Despite the good deed, we would not praise him since he could not have reasonably chosen otherwise.
This example rests on two commonsensical intuitions. The first intuition is: “People should be held responsible for their actions unless there are exonerating circumstances.” The second intuition says: “Circumstances in which people are unable to act freely are exonerating.”2 The latter can also be called the Principle of Alternate Possibilities (PAP): a person is morally responsible for what he or she has done only if he or she could have done otherwise.3 PAP is generally accepted in philosophy, everyday deliberation, and legal reasoning.
Determinism has been known in many guises throughout history, but William James (1842–1910) describes the underlying intuition lucidly:
[Determinism] professes that those parts of the universe already laid down absolutely appoint and decree what the other parts shall be. The future has no ambiguous possibilities hidden in its womb: the part we call the present is compatible with only one totality. Any other future complement than the one fixed from eternity is impossible.4
Determinism is a kind of conditional necessity: if the antecedent determining condition occurs, then so does the consequent condition. If some form of determinism is true, then PAP is false. And without PAP there seems to be no moral responsibility either.
The contrary to determinism is indeterminism, which claims that some events simply occur without determining causes or prior conditions. Such happenings are often characterized in terms of chance, uncertainty, or randomness.
In philosophy there are two different standpoints on the free will problem: compatibilism argues that free will and determinism are not mutually exclusive; incompatibilism claims that free will and determinism are mutually exclusive.
Compatibilists rely on two additional commonsense intuitions about determinism and free will. The third intuition is: “Discoveries in physics and the life sciences indicate that determinism applies to people as well as inanimate particles, but this does not undermine ascriptions of moral responsibility.” And the fourth intuition concerns the nature of freedom, namely: “Freedom is the absence of compulsion, coercion, or constraint.”5 Compatibilists wish to maintain all four commonsense intuitions along with the scientific worldview.6 Influential philosophers like Thomas Hobbes (1588–1679), John Locke (1632–1704), David Hume (1711–76), and John Stuart Mill (1806–73) have been compatibilists.
Incompatibilists reject the third intuition while maintaining the others. Their position could be summed up by this maxim: “There is no determinism—man is free, man is freedom.”7
The question of Booker’s responsibility depends on whether the world is deterministic or indeterministic, and whether one is a compatibilist or an incompatibilist.
Classical physics postulated that the universe changes over time according to definite and unequivocal physical laws, so that its state at any time is completely determined by its state at prior times and the physical laws of nature. This leaves little room for incompatibilist free will. Quantum mechanics, on the other hand, seems to imply that the world is indeterministic.8
BioShock Infinite paints a metaphysical picture that treats the many-worlds interpretation of quantum mechanics as reality. This view was originally proposed by the physicist Hugh Everett III (1930–82)9 as an answer to the measurement problem in quantum mechanics, and as a way of avoiding indeterminacy.
The crux of the problem is this.10 According to quantum mechanics, each observable physical system—anything from a particle to a person—has wavelike properties, and their evolution in time is represented formally by a wave function. At one point in the game, it is crucial for Booker and Elizabeth to determine whether the imprisoned Chen Lin is alive. Suppose that in the cell with him is a device consisting of a Geiger counter, a small amount of a radioactive substance, a hammer, and a flask of poison. In the course of one hour, an atom of the substance either decays or not. If it does, the Geiger counter goes off, triggers the hammer, which shatters the flask and kills Chen Lin. The wave function of this system—composed of Chen Lin, the device, and the cell—is interpreted statistically. This means that prior to observation, Booker and Elizabeth only have a certain probability of finding Chen Lin alive. According to the traditional Copenhagen Interpretation of quantum mechanics, one would say that prior to observation it is objectively indeterminate whether Chen Lin is dead or not—he is both dead and alive. Yet when Booker and Elizabeth open the cell door, Chen Lin will be either dead or alive. The discrepancy is explained by the collapse postulate: by observing Chen Lin, Booker and Elizabeth compel the wave function to “collapse” to a determinate state, and assign to Chen Lin either the property of being alive or the property of being dead. It seems to follow that whether Chen Lin is alive or not depends somehow on observation.11
The many-worlds interpretation proposes that the wave function never collapses. Instead, the universe literally “branches” into two or more different, non-interacting but equally real worlds, each of which corresponds to a definite possible measurement. With every observation the world is continually split into an astonishing number of branches.12 The upshot is that “all possible outcomes [of an observed event] occur; each one is the point of departure for other forkings.”13 Hence, the universe is really a multiverse, a set of possible parallel universes or worlds, all of which have more or less different histories14 that together “embrace all possibilities of time.”15 As a result, indeterminacy and observer participation are denied,16 because everything in the multiverse is determined, and all worlds obey deterministic physical laws.17 Since PAP does not hold in BioShock Infinite’s multiverse, Booker is not responsible for his actions, says the incompatibilist. But does responsibility require PAP?
The contemporary philosopher Harry Frankfurt has presented an influential argument for compatibilism by claiming that there is no link between moral responsibility and PAP.18 The illustrations involved in his claims have subsequently become known as “Frankfurt-type examples.” These usually involve some constraint that does not play a role in the agent’s choices or behavior, but nonetheless renders that behavior inevitable.19
As Jack exits the bathysphere, when he first sets foot in Rapture, he is contacted by a man named Atlas (the game’s main antagonist, Frank Fontaine, in disguise), who guides him through the ruins of the city and has him perform various tasks. Atlas’s instructions often involve the phrase “would you kindly” that, unbeknownst to Jack, is a programmed trigger phrase for controlling his actions. Suppose that Jack is faced with the choice of either harvesting or curing a Little Sister. Atlas wants him to harvest the girl and, due to the trigger phrase, he can ensure that Jack complies. However, in order not to tip his hand, Atlas only interferes if Jack shows a prior sign that he is going to choose to cure the girl.20 But suppose that Jack chooses to harvest the Little Sister. He thereby does what Atlas wanted without any need for intervention. It follows that Jack acted without compulsion, and is therefore morally responsible for harvesting the girl. Yet he also could not have done otherwise.
Frankfurt-type examples are designed to show that an agent is morally responsible even if PAP does not hold.
BioShock Infinite can be interpreted as a Frankfurt-style example on the level of both its structure and its narrative content.
The structure of modern first-person shooters (FPS), such as BioShock Infinite, is characterized by constants—narrative points with predetermined outcomes connected by a small number of forking paths—and variables—the players’ restricted freedom in navigating the game’s constants. Player actions are confined to realizing one out of a limited number of prescribed outcomes.21 The game’s structure acts as a constraint for the player, letting him act with little to no meaningful possibility to do otherwise.
BioShock Infinite’s plot is intimately tied to the game’s structure, so much so that it could be treated as a metaphor for the latter. Booker’s relationship with the Comstock world is similar to the player’s relationship with the game’s structure. The plot’s constants, like the coin toss always coming out heads, indicate that the Luteces have brought numerous Bookers into the Comstock world before. The variables that are there—choosing Elizabeth’s necklace or deciding Slate’s fate—have no effect on the game’s outcome. Each Booker introduced into the Comstock world is a foreign element caught in a reality that does not originally involve him, and precludes him from doing otherwise than set by that world’s established history.
Thus, if Frankfurt is right and if compatibilism is correct, Booker is morally responsible. However, the game’s ending seems to imply an even stronger thesis about responsibility.
To redeem himself, Booker must never make the choice of accepting the baptism in the first place. Moreover, every Booker in every sufficiently similar world must either renounce the baptism or never make the choice at all. To ensure that no Booker in the multiverse ever makes the choice, every Elizabeth and every Booker travel back in time to a moment before the baptism, and every Elizabeth drowns every Booker. This can be interpreted in at least two ways: (1) every Elizabeth kills every Booker in the multiverse at baptism, or (2) every Elizabeth kills only those Bookers in the multiverse who accepted the baptism. Let’s concentrate on (1), since it is the more interesting case.
Intuitively, it seems that each individual in the multiverse is responsible for what they have done in their world, since worlds are spatiotemporally and causally isolated. BioShock Infinite suggests that moral responsibility is not world-bound. Why else would a Booker, who renounced the baptism in the past, still be accountable, and end up being drowned? The reason could be that identity is a necessary condition for ascribing responsibility. To hold someone accountable for something after the fact requires that they be the same person they were while committing the act. Booker has multiple counterparts across the multiverse. Which one of them is he? In a sense, he is all of them. As the universe splits at each observation, so do the observers.22 The proper name “Booker DeWitt” when uttered within a world refers to the name’s bearer in that world. But from a hypothetical multiverse perspective, “Booker DeWitt” refers to all individuals bearing that name in all alternate worlds.
BioShock Infinite’s ending implies that responsibility is ascribed to an individual from a multiverse perspective. From this, it follows that there is mutual responsibility between all counterparts in the multiverse, since they are all parts of a transworld individual—each Booker DeWitt in some world is a part of the Booker DeWitt that exists across all worlds and all times. Call this the thesis of transworld moral responsibility. Not only is every Booker responsible for the actions of every other, it is also impossible for any Booker to redeem himself.
The price of clemency in BioShock Infinite is paradoxical, since every Elizabeth travels back in time before she was born to kill her father. This is a variant of the grandfather paradox, which has been taken to show that time travel is impossible, because it implies a break in the causal chain that links the time traveler with his father.
Backward time travel within the many-worlds framework takes the time traveler to a different branch of history than the one he departed from. When he kills his grandfather, the act takes place in, or results in the creation of, an alternate world where the traveler’s counterpart did not exist. The original world, from which the traveler departed, however, remains unaltered.23 The upshot is that restitution for Booker is impossible, since each time an Elizabeth kills a Booker in the past, the universe simply branches, leaving the original world’s timeline intact—nothing changes.
BioShock Infinite has a rather stark stance on moral responsibility. First, one is morally responsible even if one could not have done otherwise. Second, from the fact that responsibility is attributed from a hypothetical multiverse perspective, two things follow: moral principles are not relative to worlds (moral relativism), but are the same in all of them (moral objectivism); and, as a corollary, one is responsible for one’s own deeds as well as those of one’s counterparts in the multiverse. Third, the past cannot be altered, and choices cannot be unmade. The upshot is that we are all inescapably blameworthy for something, because either we or one of our counterparts somewhere in the multiverse have done, are doing, or will do something blameworthy. This can be seen as a secular version of the Christian doctrine of original sin,25 especially when interpreted as resulting in the human race being comprised of a massa damnata,26 a condemned mass. Another shared feature with religious doctrines is BioShock Infinite’s objective view of moral principles as being the same in all worlds.
To my knowledge, the possibility of transworld moral responsibility has not been explicitly proposed in the philosophical literature.27 This is understandable—the thesis seems incredible because it assigns praise or blame unreasonably and unfairly. Suppose that there are two worlds, w1 and w2, identical except that in w1 Jack1 harvests all the Little Sisters he meets, whereas in w2 Jack2 cures all the Little Sisters he runs into. If transworld moral responsibility is true, then Jack1 should receive praise whenever Jack2 saves a Little Sister. But intuitively this seems wrong: Jack1 should not receive praise for something he did not do nor had the chance to influence. Likewise, Jack2 should not be blamed each time Jack1 harvests a Little Sister.
One reason for not taking the idea of transworld responsibility seriously is that it conflicts with the generally accepted maxim of “ought implies can”;28 that is, if an agent ought to do X, then it is possible for him to do X.29 There is a connection between its being the case that an agent ought to do X, and the agent’s being blameworthy for not doing X. But there could be cases in which an agent is blameworthy for X-ing or not X-ing, but in fact cannot X. To expect in such situations that an agent should or should not have done something that he or she really cannot do is to expect him or her to have done the impossible. Because Jack2 cannot influence the actions of any of his counterparts in other worlds, it would be unfair and unreasonable to consider him blameworthy or responsible for their actions. Indeed, what should he have done?30
In response, one could try to challenge the intuition summarized in the “ought implies can” maxim or argue for moral relativism. However, in this world our philosophical adventure has come to an end. We will not pursue the plausibility of transworld responsibility further. My mission has simply been to show that BioShock Infinite raises intriguing questions about free will and offers possibilities for contemplation. The rest is up to you.