Introduction

People throughout history have believed that they belong to a special categorical class called persons. One of the special features of this class is moral agency, which may be based on criteria such as being able to make choices, form intentions, and perform actions.¹ However, if the nature of reality is at odds with such criteria, our moral agency could be reduced to a pleasant fiction.

Moral agents have the capacity to make moral judgments and take actions that comport with morality. While challenges to the proposition that persons are moral agents have come from many different disciplines, few have posed a greater challenge to it than metaphysics. Metaphysics is a discipline in philosophy with a very long history. The etymology of metaphysics derives from the Greek words met'a (after) and physik'a (physics), and it simply refers to Aristotle’s work that contiguously followed after his chapter on physics. However, over time the meaning of met'a has been reinterpreted to mean something more substantial, like beyond. Thus, metaphysics is the study of that which is beyond physics. And arguably it is the most fundamental of all disciplines for it studies the categorical structure of reality itself.

There is some disagreement among meta-physicians about this categorical structure. Disagreements are not just over things like the kinds of categories that ought to be recognized, but they are also over things like whether reality even has a categorical structure. That is, do categories represent absolute ontological differences or do they represent the useful ways in which humans organize their experiences? Does reality have an inherent categorical structure (which may someday be discovered) or do minds invent conventional categories to order their experiences?

Moral agency To say that we are moral agents is to say that we can be held morally responsible for our actions, which can run the gamut from good to bad. For example, people who are around a village during a volcanic eruption may demonstrate good moral conduct if they rescue villagers from the destructive power of the volcanic eruption. Alternatively, they may demonstrate bad moral conduct if they do nothing to help the villagers. The volcano eruption is causally responsible for the destruction to the village, but it is not morally responsible; the volcano has no moral agency.

Under what circumstances are things morally responsible? A theory of moral responsibility might touch upon, at the very least, the following subjects:

1. The subject of morality itself.

2. The criteria for being a moral agent.

3. The circumstances under which moral responsibility can be ascribed.

In due course, after expanding upon each of these points, one might arrive at a satisfactory explanation that accounts for moral responsibility. However, the discipline of metaphysics has not only undermined the notion that persons are morally responsible, but also the notion that reality can even support the possibility of moral responsibility.

Morality

“Morality is simply the attitude we adopt towards people whom we personally dislike.” - Oscar Wilde

The etymology of morality derives from the Latin word moralitas (manner, character, proper behavior), and it can be used a) descriptively to refer to a code of conduct promoted by some group, or b) normatively to refer to a code of conduct that, given specified conditions, would be promoted by any moral agent. In what follows, our primary concern will be of the second normative type where the nature of morality can be explored in greater isolation from cultural norms.

What is the basis for morality? From a behavioral point of view, morality is arguably the study of right and wrong conduct. However, what is the basis of right and wrong? One possibility is that its basis is authoritative prescription, in which case morality is a dictum of might makes right– something is morally good or bad because an authority says so with no further explanation required. For instance, a religious person may believe its basis is the authoritative prescription of a deity (divine command theory). A forceful criticism of this type of moral theory is revealed by asking: is an action good because, say, God commands it, or does God command it because it is good? If it is good because God commands it, then morality is the arbitrary preference of God, but if God commands it because it is good, then morality has a basis independent of God.

Many would contend the basis is essentially rational; persons use reason and experience to inform their conduct, e.g., how can a group best realize common interests? In which case, a sense of right and wrong is, for the most part, the habituation of conduct that promotes cultural dominance or health, i.e., the principle of natural selection applied in the context of a social Darwinism.\footnote{Imagine a society that did not inculcate respect for the property among its citizens, e.g., stealing was not taught as wrong.}

While there are perhaps many other moral theories – some of which may not support the premise of moral responsibility – each must fall within one of two general categories: the conventional, and the intrinsic. On the one hand, if morality is conventional, then persons are only morally responsible by convention and moral agents are not ontologically significant. On the other hand, if morality is categorically intrinsic, then arguably moral agents have an inherent duty to conform to it. But what does it mean to say that morality is categorically intrinsic?

Realism

There are many things (particulars) that are red. For instance, there are red cars, red apples, and red signs. But what makes them red? The philosophical discipline of realism submits that all red particulars are red in virtue of the existence of a Universal– a single abstract entity that is a part of all red particulars. More precisely, Universals are repeatable entities that can be exemplified simultaneously by different particulars, where entities are things like properties, like red, relations, like behind, and kinds, like dog.

Realists

Someone who subscribes to the doctrine of Universals.

Realists argue that attribute agreement\footnote{How an entity, like an apple, can have attributes, like the color red} is due to relationships between entities and Universals. Furthermore, they contend that the subject-predicate\nomenclature{subject-predicate}{A complete sentence contains two parts: a subject and a predicate. The subject is what the sentence is about, while the predicate tells something about the subject.} grammar of sentences reinforces this view, e.g., in the sentence “My apple is red,” the predicate “is red” seems to be picking out something (a noun)– call it redness– and linking it to the subject, “my apple.” Let us examine how a realist might account for following two sentences:

1. My apple is red.

2. My friend’s apple is red.

In each sentence, there are arguably two Universals being exemplified. First, there is the Universal property, redness, that is being exemplified by the two particulars, my apple and my friend’s apple. Second, there is the Universal kind, apple-ness, that is also being exemplified by my apple and my friend’s apple. The redness being exemplified by the entity, my apple, is numerically identical² with the redness being exemplified by the entity, my friend’s apple. Likewise, the apple-ness that is being exemplified by my apple is numerically identical with the appleness that is being exemplified by my friend’s apple.

In the sentence, “This apple is red,” a metaphysical realist maintains that this denotes “This apple exemplifies redness,” where redness is a Universal that inheres in this apple. More generally, a realist claims that any subject-predicate sentence of the form “$a$ is $F$” can be paraphrased as “$a$ exemplifies $F$-ness” where $a$ is an entity\footnote{An entity can be a particular or a repeatable} and $F$-ness is a Universal. In the same way, a realist could maintain that morality, like color, is a Universal\nomenclature{Universal}{A type, a property, or a relation that can simutaneously inhere in more than one thing.}. In the same way that an apple can exemplify a color like redness, a person can exemplify a virtue (a moral excellence) like honesty, e.g., “Bob exemplifies honesty.” However, it turns out that the existence of Universals is one of the most important disputes in the discipline of metaphysics.

Paradox If “$a$ is $F$” denotes “$a$ exemplifies $F$-ness,” then we can construct a paradox\nomenclature{paradox}.{A statement that contradicts itself.} Let $F$ mean “does not exemplify itself” and so $F$-ness could be the Universal property, non-selfexemplification.\footnote{Examples: redness is non-selfexemplify, noncorporeal-ness is self-exemplifying.} Thus, for some $a$, the proposition “$a$ exemplifies non-selfexemplification,” should be either \True or \False. But the existence of this property leads immediately to a paradox, for if a thing does exemplify itself, then it does not exemplify itself and if a thing does not exemplify itself, then it does exemplify itself. It is self-contradictory.

On the one hand suppose that an entity, $a$, does exemplify itself. Then, since it is a property an entity exemplifies when it does not exemplify itself, $a$ does not exemplify itself. On the other hand, suppose that $a$ does not exemplify itself. Then, it does exemplify itself. It is similar to the sentence, “This sentence is a lie.” If it is \True, then it is \False; and if it is \False, then it is \True. It is self-contradictory.

To avoid this paradox, restrictions on what constitutes a Universal must be imposed, such as denying that non-selfexemplification is an actual Universal. However, as a result of this, the realist cannot take it for granted that “$a$ is $F$” means that $F$ must be selecting a Universal, $F$-ness. This is not necessarily a problem for morality, but the realist cannot simply claim, without qualification, that if a sentence can take on a subject-predicate form that necessarily means that the existence of Universals can be assumed to account for things like attribute-agreement. For instance, “Socrates is honest” does not necessarily mean that the adjective, honest, is pick out out a Universal entity, honesty.

A vicious regress? There is perhaps a deeper problem with the realist’s account. This problem can be illustrated by pointing out that in the statement, “a exemplifies $F$-ness,” the word exemplifies itself picks out a Universal– a repeatable entity that serves a relational\footnote{For example, next to in “Socrates is next to Plato” is also a Universal relation} role. After all, consider that two entities can enter into a numerically identical asymmetric relationship, exemplification. To explore this further, let us consider the following two sentences:

1. Socrates exemplifies courage.
2. Plato exemplifies honesty.

In each sentence above, there is the Universal entity, exemplification, that is being exemplified by the particulars, “Socrates exemplifies courage” and “Plato exemplifies honesty.” The exemplification exemplified by “Socrates exemplifies courage” is numerically identical with the exemplification exemplified by “Plato exemplifies honesty.”

$$ \operatorname{F}(n) = \text{a set of propositions ${p_1, p_2, \ldots, p_n}$.} $$ where $p_1$ = $a$ exemplifies $!1$ $F$-ness and $p_i = { p{i-1} }$ exemplifies $!_i$ exemplification. More generally, if exemplifies picks out a Universal, then $F(n)$ denotes a set of $n$ propositions for any sentence of the form “$a$ is $F$.”

Example:

$F(3)$ generates three propositions, the third of which is ${$ $a$ exemplifies$_1$ $F$-ness. $}$ exemplifies$_2$ exemplification. $}$ exemplifies$_3$ exemplification.

Moreover, $F(\infty)$ denotes an infinite regress of never-ending propositions and the worry is that this regress is vicious. Whether this is the case is disputable, but one presumes there is a right answer and, if so, Realism’s veracity hinges on this issue.

A vicious infinite regress

An infinite regress is vicious if each proposition $p_i$ in the series requires the support of proposition $p_{n+1}$. If this is the case, then the infinite series of regressions required to provide such support can never be finished\footnote{Unless it provably converges to a final proposition.}, e.g., $p_1$ requires $p_2$, but $p_2$ requires $p_3$, ad infinitum. Consequently, the realist’s solution to problems like attribute agreement ("$a$ is $F$" $\rightarrow$ “a exemplifies *$F$-ness*” where *$F$-ness* $\in$ set of Universal entities) would not be valid.

Nominalism

There is competing philosophical doctrine to realism, called nominalism, which disclaims the existence of Universals. Nominalists believe that realism needlessly\footnote{Occam’s Razor\nomenclature{Occam’s Razor}{Entities should not be multiplied needlessly; the simplest of two competing theories is to be preferred.}} introduces the strange notion of Universals, exhibits incoherency, and is burdened by a busy ontology\footnote{For instance, “$a$ exemplifies $F$-ness” embodies an infinite number of propositions}. In its place they propose a supposedly simpler but sufficient account for the apparent categorical structure of reality.

Nominalists submit that only actual particulars\footnote{Particulars like this book, that apple, and Bob’s car} have independent existence. After all, how could a repeatable entity, like redness, simultaneously inhere in multiple entities, like an apple and a firetruck, such that the redness in the apple is numerically identical to the redness in the firetruck? The Realist is introducing a strange idea that entities with independent existence have non-local, unbounded physical presence. So, realists claim that the sentence, “a is F” denotes “a exemplifies F-ness” where F-ness is a Universal. In response, nominalists counter that “a is F” denotes “a is a member of the set of F things.” For instance, “Socrates is courageous” denotes “Socrates is a member of the set of courageous things.” To try to avoid the strangeness of Universals, Nominalists introduce sets in their place.

Set

An unordered collection of entities considered as a whole. The identity conditions for sets is such that, if $a$ and $b$ are sets, they are numerically identical iff for each entity in $a$ that entity is in b. Using mathematical language, $a \equiv b$.

Abstract reference For simple subject-predicate sentences, the nominalist’s account as already given would seem sufficient. However, how might they interpret a sentence including abstract parts, like “Honesty is a virtue?” The realist would contend that this can be understood to mean, “Honesty exemplifies being a virtue,” honesty and virtue picking out abstract Universals. But what might nominalists say? One possible interpretation might be, “The set of honest acts is a subset of the set of virtued acts.” While this gives nominalists a little trouble, their response might seem sufficient enough. However, their problems do not end here.

Set equivalency How might a nominalist deal with the supposition that all the things that have hearts also have kidneys and vice verse? In other words, the set of things that have hearts and set of things that have kidneys have the same members. But according to the identity conditions for sets, that means that the set of things that have hearts is identical to the set of things that have kidneys. Consequently, having a heart is the same thing as having a kidney (the set of things that have hearts$\equiv$ the set of things that have kidneys), which is plainly absurd.

What is the nominalist’s response to this? Generally, they invoke the terminology of possible entities. Instead of “$a$ is $F$” denoting “$a$ is a member of the set of $F$-things,” it denotes “$a$ is a member of the set of possible $F$-things.” The set of possible things that have hearts and set of possible things that have kidneys do not have the same members, therefore they are not equivalent. Unfortunately, part of the justification for nominalism was that it has a simpler ontology than realism, but the introduction of possible worlds somewhat undermines this effort.

Conventionalism Even if nominalism did offer a satisfactory account of attribute agreement and subject-predicate discourse, it is none-the-less nominative. Accordingly, there is no entity, honesty, there are only sets of resembling things. The nominalist posits that the sentence, \textquotedblleft My friend is honest\textquotedblright{} simply means that my friend resembles other persons that are called honest. This does not appear to give ontological significance to honesty. Thus, we come back to where we started: if right and wrong is conventional, then is being a moral agent ontologically significant?

Criteria for moral agency

“I know only that what is moral is what you feel good after and what is immoral is what you feel bad after.” - Ernest Hemmingway

What is the criteria for being a moral agent?\footnote{And thus be subject to ascriptions of moral responsibility.} By definition, a moral agent is a being who can make moral judgments. Thus, at a minimum, it would seem a moral agent should have the capacity to make rational decisions– that is, reason about the effects their decisions have on reality. So, how do persons measure up to this criterion? At the very least, they seem to have the capacity to reason about a limited domain of reality, and so are perhaps morally responsible within the limits circumscribed by that domain. But putting all of that aside, one can instead ask, “Do moral agents exist at all?” That is, are there subjects making decisions?

Persistence

To consider this question, let us analyze Chisholm’s essay, Identity through Time. In this essay, he distinguishes two ways in which entities persist.

Intactly persisting entities In the first way, entities persist in a numerically identical sense such that an entity, $a$, at time $t_1$ is the same as entity, $b$, at time $t_2$ iff the set of parts for entity $a$ at time $t_1$ is equivalent to the set of parts for entity $b$ at time $t_2$. Therefore, if $a \equiv b$, said entity persisted, at a minimum, $|t_2-t_1|$ units of time.

Non-intactly persisting entities If an entity changes even a single part\footnote{Even a single atom!}, it is not strictly the same. Therefore, since physical entities are in a constant state of flux, when one says physical entity $a$ at time $t_1$ is the same as physical entity $b$ at time $t_2$ (where $t_2 \neq t_1$), we do not mean that entities $a$ and $b$ are numerically identical. Rather, we mean to say that they are conventionally the same.

To clarify, let us consider the thought experiment of the Ship of Theseus, which raises the question of whether a ship which has had all its parts replaced retains the same identity. If the ship at t$_{\text{1}$} has different parts than the ship at t$_{\text{2}$} then they are not numerically identical; so, if the ships are identical, a non-intactly persisting sense of identity is being used. Chisholm contends that the question is not is it the same ship but does it constitute the same ship? And if it constitutes the same ship, that means it satisfies the Theory of Persistence.

Theory of persistence : A composite entity persists if conditions $1$ and $2$ are satisfied:

1. The precedent composite entity evolved from a antecedent composite entity. Note: Evolved denotes that after each successive change, the antecedent composite must have at least one part in common with the precedent composite.
2. The composite entity must satisfy additional quantifiable criteria, e.g., the criteria that it must have the same sailing schedule (in the case of a ship’s persistence).

In practice, how might this theory provide a solution for the Ship of Theseus dilemma? First, we must determine whether it satisfies, from initial state to final state, the evolution criterion. And second, it must satisfy additional criteria, e.g., has the same sailing schedule. For example, if the evolution of the ship undergoes fission\footnote{That is, if it evolves into more than one ship} then the evolution criterion by itself cannot determine which ship constitutes the Ship of Theseus. However, if only one of the ships after the fission has the same sailing schedule as the ship before the fission, then according to the has the same sailing schedule convention, that ship constitutes the Ship of Theseus.

Persons

It is not even clear how persons should be defined, but providing an unambiguous definition for it would help to resolve many controversial. For instance, to resolve the dispute over abortion, one could categorically determine when a bundle of cells constitutes person-hood.\footnote{And therefore qualifies for legal consideration as a person} There appears to be a consensus that motile sperm cells and ova are not persons, but that newborn babies are. What was the critical turning point in which it from from just being a bundle of cells to being a person who has a bundle of cells? Expending considerable time and effort could be avoided if a definitive categorical account of this problem was given, but like with the pile of sand, such a definition seems problematically conventional.

Chisholm on persons In light of the Theory of Persistence, what can be said about persons? As with the Ship of Theseus, are person’s identities also conventional? Or are they like sub-atomic particles which intactly persist? In Which Physical Thing Am I, Chisholm he contends that persons are not conventional. In support of this, he asks you the reader to consider a hypothetical situation in which you are about to undergo an operation. The doctor presents you with two options: you can either choose the expensive option where you are subjected to total anesthesia during the operation, or you can choose the cheaper option. The cheaper option consists of the doctor giving you a pill before the operation to induce complete amnesia so that during the operation there is no memory of you, and giving you a pill after the operation so that everything that transpired during the operation is forgotten.

The question is, would it be reasonable for you to choose the cheap option? Chisholm contends that it would not be reasonable because you would believe that you would be the person enduring the operation. Furthermore, you would not be persuaded that this is not the case by any convention that denies that the person on the table is you. In other words, you would not consider your identity a matter of convention.

Critical response to Chisholm How might Chisholm’s arguments be disputed? In the situation he presented, perhaps you would believe that you and the person enduring the operation in the cheaper option is the same person. Furthermore, perhaps no convention could persuade you from this conviction. However, your conviction could simply be based on a convention that you deeply subscribe to. At this point, it does not seem like Chisholm has established that persons warrant special categorical distinction.

Bundle Theory and Ego Theory In the essay Divided Minds and the Nature of Persons, Parfit contrasts two competing views of persons, Bundle Theory and Ego Theory. Bundle Theory proposes that unity of consciousness cannot be explained by referring to persons; Bundle Theorists maintain that persons only exist as a feature of language. Persons are a convention– they are a label for a bundle of things like thoughts, memories, and sensations that are unified by causal kinds of relations. There are no subjects in which these bundles inhere. But the Ego Theorist proposes that these subjects really do exist– as subjects of experiences– and that they have a bundle of things like thoughts, memories, and sensations.

In support of Bundle Theory, Parfit presents the reader with a hypothetical situation in which a replica of you is created, atom by atom.\footnote{Like the transporter in Star Trek– beam me up Scotty!} At one end of the spectrum, if one-hundred percent of your atoms are replaced, this is clearly a replica of you. At the other end, if only one percent of your atoms are replaced, this is clearly you. Thus, in between these two extremes there must be a critical turning point where it goes from being you to being a replica of you. However, this appears implausible because any point chosen would seem arbitrary. How could a single atom make a difference? It is similar to the question, When does sand become a pile of sand? It is absurd to think that, at one point it is not a pile, and with the addition of a single grain of sand, it becomes a pile.\footnote{Although the definition for “a pile” could be based on structure rather than quantity, e.g., “A pile is a collection of objects laid on top of each other."}

Ego Theorists have difficulty responding to this question because they believe that persons are subjects of experiences, not just bundles of things. As such, there must be some special unknown thing that makes you who you are that may not be included in the replication. Bundle Theorists have little problem answering these questions, however, because they deny that there is a you– only a bundle of things. If an exact copy of the bundle of things that constitutes you is made, what we choose to call this thing, you or a replica of you, is a matter of convention.

On split-brain studies Parfit believes that empirical results of split-brain case studies reinforce Bundle Theory. In this study, patients are shown a placard whose left half is blue and right half is red where each eye is allowed to see only one of the halves. Curiously, the indication is that patients end up expressing two separate streams of consciousness, one in which there is an awareness of seeing only blue and another in which there is an awareness of seeing only red. That seems to indicate that within each of these patients there are not one but two subjects of experience.

Problematically, Ego Theorists define a person as a subject of experience yet patients seem to be expressing two. Consequently, the Ego Theorist must now propose that subjects of experiences need not be persons at all, which seems like a difficult stretch to make. But in Bundle Theory, since persons are just bundles of things it does not seem odd that in this bundle there could be a pair of mental states, one in which there is an awareness of seeing only red, and another in which there is an awareness of seeing only blue. There is no subject in which awareness inheres.

Ascribing moral responsibility

Responsibility: A detachable burden easily shifted to the shoulders of God, Fate, Fortune, Luck or ones neighbor. In the days of astrology it was customary to unload it upon a star. – Ambrose Bierce, The Devil’s Dictionary, 1911

Under what circumstances can a moral agent’s conduct be ascribed moral responsibility? At a minimum, it would seem to require that its conduct be a product of being able to select, in a way that is unconstrained by causal agencies, a specific action among a set of possible actions– that is, a product of exercising a free will. For instance, an agent’s decision to go shopping could be a product of a free will if it was possible for the agent to have decided not to go shopping. Conversely, if an agent stumbles off a building and free falls towards the ground, the act of free falling is certainly not a product of free will. Under the circumstances, the agent had no choice but too fall.

Four-dimensionalism

It is often said that entities\footnote{Such as persons!} have location and extension in a three-dimensional space. That is, physical entities can be specified in a three-dimensional Cartesian coordinate system (see \Cef{fig:3dcart}), $x$, $y$, and $z$, which provides three spatial dimensions– respectively, length, height, and width, e.g., a cubic entity extends $m$ units along the $x$, $y$, and $z$ axes.

\begin{figure} \caption{Three-dimensional Cartesian coordinate system} \label{fig:3dcart} \includegraphics[scale=0.85,bb = 0 0 200 100, draft, type=eps]{3d_cartesian_coordinate_system.png} \end{figure}

However, our coordinate system is missing a crucial detail: at what time, $t$, will a particular entity be at a particular $x$, $y$, and $z$, and how long will it endure? Therefore, a physical entity’s location and extension is more precisely specified in a four-dimensional coordinate system, $x$, $y$, z,and t, which provides three spatial dimensions\footnote{Length is $x$, height is $y$, and width is $z$} and one temporal dimension\footnote{Duration is $t$}, e.g., a hyper-cubic entity extends $m$ units along the $x$, $y$, $z$, and $t$ axes, or a person may extend an average\footnote{Since a person’s dimensions (height, etc.) can change over time, a life-time average is being used} of $\approx 0.6$ meters along the $x$-axis, an average of $\approx 1.8$ meters along the $y$-axis, an average of $\approx 0.3$ meters along the $z$-axis, and $\approx 85.6$ years along the $t$-axis.

Four-dimensional space-time has taken on a new-found importance with the advent of modern science. Under this model, when persons observe an entity at a particular time, they observe a single time-slice of it. More precisely, they observe a three-dimensional spatial “slice” (which is perpendicular to the temporal dimension) of a four-dimensional space-time entity. For example, when they observe a two-dimensional spatial slice of a sphere, they observe a disk; when they observe a three-dimensional spatial slice of a hyper-sphere\footnote{A hypersphere is a four-dimensional sphere– the set of points a distance $r$ away from a center point in four-dimensions}, they observe a sphere.

At any time $t$, persons can only observe a three-dimensional spatial “slice” located at time $t$ of a four-dimensional space-time entity\footnote{Actually, persons only see a two-dimensional plane but can infer a third, depth, by using other clues like shading}. However, if they had access to a fifth-dimension, they could possibly rotate in such a way in five-dimensions to instantly observe the entirety\footnote{The entire time line of entities would be viewable as snap-shots} of four-dimensional entities. From this ``God’s eye'' vantage point, the passage of time would represent a vector through a four-dimensional space-time\footnote{Perhaps some entities experience a different passage of time which could be represented by a different four-dimensional vector}.

Foreknowledge and choice Is this model of reality incompatible with moral agency? If free agents are immutable four-dimensional space-time entities, then would that not denote that their freedom is illusionary? After all, they are unchanging– there does not appear to be an opportunity for choice, there is only the opportunity for existence. But perhaps this notion of timelessness\footnote{The passage of time could possibly be an illusion} is too challenging to our intuitions, so let us instead ask the question, does infallible foreknowledge\footnote{Such as knowledge acquired by seeing a person’s entire as a static higher-dimensional structure} remove the possibility for choice? In other words, given that (1) a free agent, given two options a and b, can freely choose between a and b, and (2) there is infallible foreknowledge pertaining to which option, a or b, the free agent will choose, are (1) and (2) compatible?

On the one hand, it seems infallible foreknowledge forces the free agent’s hand: the free agent must act in accordance with the foreknowledge. It has no choice; it cannot do otherwise. On the other hand, infallible knowledge about the future is arguably no more a limitation on choice than infallible knowledge about the past as long as that foreknowledge is not based on causal determination\footnote{Causal determination: the doctrine that everything which happens is caused by antecedent conditions}. And indeed, a four-dimensional space-time is not necessarily causally determined.

Causal Determinism

Supposing that reality is not an immutable four-dimensional space-time, let us turn our attention to the question of causal determination. Causal determinism is the doctrine that all events are the inevitable result of antecedent conditions, including the actions of persons. The totality of existence is a game of cosmic billiards where, given an initial state as input, the final state could be output by a computable function.

Is causal determinism compatible with moral agency? In causal determinism, there is only one possible future; acts of apparent choice is the mechanical expression of a supposed moral agent’s heredity and past environment. Since its future is determined, that would denote that moral responsibility is not possible, for at the very least a moral agent must have the power to make choices between morally good and bad actions.

An argument from absurdity: the coherency of choice For the sake of argument, assume choice is not an illusion. How then might we offer a rational account of it? But if something is rational, does that not signify that it can be explicated in terms of its antecedent conditions? How else can a physical process be rationally explained if not by cause-and-effect? But if choice can be understood in terms of its antecedent conditions, then it is not free– it was determined by cause-and-effect. This is problematic for those who wish to maintain that choice is a coherent concept.

Concluding thoughts

While metaphysical inquiry has not removed the possibility that people belong to a special categorical class called persons, it has served to undermine the criteria for moral agency, such as being able to make choices. But what would be entailed by the realization that moral agency is not a legitimate concept? Would society collapse in on itself or would we go about our daily lives contented by the fact that, at least it feels like we are free moral agents? That, at the very least, it is a useful fiction?