Original sin, control, and divine blame: some critical reflections on the moderate doctrine of original sin

Abstract

This article examines a construal of the doctrine of original sin which affirms the cognitive corruption of human faculties but denies that humans carry original guilt for Adam's fall or for cognitive corruption. All humans require Christ's atonement, because they either inevitably commit at least one sin or are rejected by God for other reasons. We go on to identify three problems with this account. The first problem is the ‘inevitability’ of sinning. Here, the defender is forced either to accept a compatibilist analysis of responsibility or provide a libertarian-friendly analysis of ‘inevitability’. The latter option contradicts the Augustinian contention that it is impossible for sinners to lead a life of virtue and holiness. The second problem has to do with the mechanics of the cognitive effects of sin. The cognitive effects of original sin make it very difficult or inevitable for humans to perform meritorious actions and very easy or inevitable for them to commit sinful actions. If the sinner's degree of control over her sinful actions is so low, it seems that she does not deserve divine blame and punishment for failing to commit meritorious actions. Finally, we raised some problems regarding the fate of the non-culpable sinners.

One central function of the doctrine of sin is to form a backdrop for the doctrine of atonement.¹ Why is it the case that every human individual (excluding Jesus), regardless of how moral and holy they appear, needs atonement in order to be united with God? The doctrine of original sin attempts to answer this question by positing the universality of sin among all human individuals. The universality of sin, however, seems to raise difficult questions about moral responsibility.

First, it seems obvious that God holds humans responsible for their sinful actions and wrongdoings. Biblical stories and language are replete with descriptions of God's demands towards us. God issues commandments so that people would not sin, blames sinners, promises to punish them, invites sinners to repent and promises forgiveness. God expresses his anger over human sins and displays forgiveness towards the repentant (e.g. Mk 2:17; Lk 23:47; Act 2:38). All this suggests that human beings are responsible for their sins in the eyes of God. However, it is also clear that sin does not seem to be under the control of humans. Sin is a universal power that afflicts all humans and they are powerless to overcome it by themselves. If some humans were able to overcome sin by themselves, the need for Christ's atonement would no longer be universal. So, sin must have a universal scope. Now, the question is as follows: if it is the case that sin is a power over which no human exercises control, how can it be the case that all humans are, nevertheless, blameworthy for their sins? The universal scope of sin seems to lead to problematic conclusions about human moral responsibility.

In this article, we will focus on a recent, philosophically oriented construal of the doctrine of original sin. This version of original sin accepts that all humans are severely corrupted by the Fall. However, humans do not share any original guilt, they merely share some form of original corruption. This moderate version of the doctrine is supposed to circumvent the problems of the strong version of the doctrine, which entails shared guilt. While we are sympathetic to this project, we raise some issues with the moderate doctrine. We suggest that the moderate view solves the problems related to unjust attribution of guilt, but it creates another problem (or, it just kicks the can down the road), which is so far left undiscussed. We pursue our argument by using contemporary theories of freedom, control, and responsibility. More precisely, we identify three particular problems that arise from this discussion. But first, we need to define the essential concepts and the state of art.

Original sin without original guilt

All Christian churches accept some form of the doctrine of original sin, even if they diverge on some important details. A basic distinction is made between Eastern and Western traditions, according to which Eastern theologians acknowledge the inherited and shared corruption of the human race, whereas since Augustine the Western tradition, or at least some Western theologians, also accepts the idea of shared guilt. Thus, we can distinguish between strong and moderate forms of the doctrine of original sin.

Strong doctrine of original sin

All humans are born with a sinful nature that inevitably leads them to sinful actions.

All humans share the guilt for the primal sin.

All humans are condemned because they are guilty of sinful actions and possess sinful natures.

Moderate doctrine of original sin

All humans are born with a sinful nature that inevitably leads them to sinful actions.

All humans are either (1) blameworthy for their sinful actions or (2) if they do not commit sinful actions; they are, nevertheless, rejected by God, because of their sinful nature.

Some humans are condemned because of their sinful actions and others because they have a sinful nature.

Timpe (2021) uses the terms ‘constitutional fault’ and ‘original guilt’ to distinguish the constituent components of the views. Constitutional fault implies that there is something significantly wrong with the cognitive, moral, and spiritual capacities of all human beings. Humans inherit this fault from their ancestors, who committed the primal sin, and it makes sinful humans such that they are not pleasing to God. The main difference between the strong and moderate doctrine concerns the notion of original guilt. The strong doctrine entails the idea of shared original guilt, namely, that all human persons are blameworthy for the primal sin of the first humans. Not only is every contemporary human afflicted by the constitutional fault, but all humans also carry guilt for the primal sin. These facts together make all human individuals deserving of God's negative judgment and punishment.²

Many contemporary theologians and philosophers find the doctrine of original guilt problematic, because it seems to entail implausible views of justice and responsibility (Swinburne 1989; McFarland 2007, 2010). We will later outline the critiques more carefully, but here is the core argument. The main reason why many thinkers arrive at this conclusion is best explained in terms of the conditions of moral responsibility. Original guilt is unjustified, because it seems to go against our basic intuitions about responsibility and desert. Our intuitive views about just desert require that a person can only be held responsible for those actions, events, and outcomes that are up to the person. This can be formulated in terms of control:

The control condition: a person deserves to be blamed and praised only for actions, events, and outcomes over which she exercises some control.³

Original guilt seems to go against this condition, because it seems obvious that there is nothing contemporary humans could do to change or even influence the actions of their distant ancestors. Because the sinful actions of my ancestors occurred way before I was born and I exercise no control over the past, I exercise no control over their actions either. Thus, it is unfair to blame me for those actions.

A number of contemporary theologians have rejected the notion of original guilt as unfair. Here, we want to address the account of Crisp (2019, 139–156; 2020) in particular. According to the moderate doctrine, original sin is the universal constitutional fault inherent in the human intellect and will. However, humans do not bear original guilt, nor are they blameworthy for the corruption of their mental faculties. Post-Adamic humans should not be blamed for primal sin, even while they inherit the constitutional fault due to their ancestors’ actions. The constitutional fault, then, explains why sin is universal: those individuals who develop moral agency will inevitably act sinfully and those who do not are justifiably rejected by God because of the constitutional fault. According to Crisp (2020, 35), the moderate view consists of the following seven claims:

1. 1. All human beings (barring Christ) possess original sin.

2. 2. Original sin is an inherited corruption of nature, a condition that every fallen human being possesses from the first moment of generation.

3. 3. Fallen humans are not culpable for being generated with this morally vitiated condition.

4. 4. Fallen humans are not culpable for a first, or primal, sin either. That is, they do not bear original guilt (the guilt of the sin of some putative first human pair or human community being imputed to them along with original sin).

5. 5. This morally vitiated condition normally inevitably yields actual sin. That is, a person born with this defect will normally inevitably commit actual sin on at least one occasion provided that person lives long enough to be able to commit such sin. (The caveat ‘normally’ indicates limit cases that are exceptions to this claim, such as infants that die before maturity and the severely mentally impaired.)

6. 6. Fallen human beings are culpable for their actual sins and condemned for them, in the absence of atonement.

7. 7. Possession of original sin leads to death and separation from God irrespective of actual sin.

Crisp and others think that the moderate doctrine is preferable to the strong doctrine for multiple reasons. In what follows, we will take issue with (5) and (7) in particular. Before we go on the discuss these, we will outline Crisp's reasons for (1)–(4). We will focus on the condition of responsibility Crisp employs in defence of these claims. It seems to us that these defences implicitly assume something like the control condition outlined above.

When discussing the notion of imputed original guilt, Crisp (2020, 38) claims that original guilt would be ‘monumentally unjust, for then a sinful condition that I did not choose is immediately transferred to me, rather like having someone else's debt immediately transferred to my bank account so that money is debited from my account as a result’. He further claims that traditional Reformed explanations of the transference of original guilt are ‘convoluted and not terribly intuitive’ (Crisp 2020, 39). While Crisp does not offer a full-blown account of moral responsibility, he seems to assume the control condition. For him, ‘guilt is intimately connected to a person's moral agency’ (Crisp 2020, 44). Blame can only be ascribed to agents, who are capable of acting morally. Moreover, blame can only be attributed to agents for their own actions. For these reasons, ‘guilt is an inalienable property of the person who has sinned’ (Crisp 2020, 44). While Crisp does not explicitly adhere to the aforementioned control condition the truth of the control condition would offer a good explanation for the fact that original guilt appears convoluted and counterintuitive in the light of justice, fairness and desert. Given the control condition, every justification of original guilt involves the transference of blame from the wrongdoer to someone else who had no way of controlling the wrongdoing of the guilty party. Because responsibility entails control and the lack of control excuses the sinner from blame, ‘guilt is nontransferable’ (Crisp 2020, 44). Therefore, original guilt is unjust.

Crisp (2020, 45) also argues that contemporary humans should not be held morally responsible for the inherited constitutional fault. He provides an analogy from slavery. Suppose an ancestor has sold Jane's parents into slavery. Because of that, all progeny of Jane's parents are born slaves. When Jane is born, she inherits the moral status and the debilitating condition that slavery imposes on her. Crisp holds that it would be unjust to blame Jane for the fact that she has the status of being a slave. Applying this to reasoning to the case of original guilt, Crisp (2020, 40) holds that:

the child born with the condition of original sin cannot be culpable for being generated in that condition. How could she be? Culpability presumes some sort of action on the part of the agent to whom it is ascribed, and a child born with original sin cannot have acted in a way that ascribes culpability.

Again, this argument makes sense, if we explain the slavery case by invoking the control condition. Given the fact that Jane can do nothing to control the conditions of his birth, it is unjust to blame Jane for it.

The defenders of the moderate view hold that it is more plausible than the strong view for a number of reasons. For our purposes, the most relevant reason is that it supposedly manages to avoid the counterintuitive moral assumptions entailed by the notion of original guilt. However, what we want to suggest is that if the defender of the moderate account now accepts something like the control condition, she will encounter a number of new problems to which no easy solution is in sight. In what follows, we will identify three such problems: (1) the problem of the inevitability of sin, (2) the problem of constitutional fault, and (3) the problem of the ‘stain of original sin’. In what follows, we will outline each problem and provide a critical look at solutions offered so far.

Two interpretations of inevitability

Crisp (2020, 45) writes that ‘all those born with original sin will normally inevitably commit actual sin for which they do bear responsibility’. By ‘normally’, Crisp refers to human beings who develop the level of moral agency required for moral responsibility. However, a problem emerges: What exactly is ‘inevitability’? The everyday meaning of inevitability is that a particular state of affairs is certain to happen and cannot be prevented or avoided. In the discussion so far, two main interpretations of ‘inevitability’ have arisen. According to the first, ‘inevitability of sin’ refers to the fact that sin is necessary. According to the second, ‘inevitability of sin’ means that given the constitutional fault sin is almost impossible to avoid. This latter is what we will call the probabilistic interpretation of inevitability. We will take these in turn.

Let us now assume that the inevitability of sin means the necessity of sin. Inevitable sins are sins that are determined to happen. If the defender of the moderate doctrine adopts this interpretation of inevitability, serious consequences will follow. Indeed, one significant consequence will be that libertarians about free will and moral responsibility can no longer accept the moderate account of original sin, because it undermines the moral responsibility of all human beings.

Incompatibilists (including libertarians) about determinism and moral responsibility hold that the notion of control at stake in basic desert moral responsibility is incompatible with determinism. Many Christian philosophers and theologians accept a libertarian account of moral responsibility (Timpe and Speak 2016). They make sense of the control condition for moral responsibility in terms of the agent's decisions and choices. If the agent's action is causally (or logically) necessitated by prior conditions, the agent cannot be said to control that action in such a way as to deserve blame or praise for it. Morally responsible control over action is, on libertarianism, incompatible with necessity.

Timpe (2014) usefully distinguishes two groups of libertarians: leeway libertarians and source libertarians. According to leeway libertarians, the clash between necessity and control is due to the fact that control requires that the agent has the ability to choose between at least two metaphysically alternative possibilities at every given occasion. So, a necessary condition for responsibility endowing control is that at least two courses of action are open to the agent.⁴ It seems obvious that if an agent's action is a product of prior necessity, then she does not have two courses of action available to her, but only one. This is the reason why necessity excludes moral responsibility. For the second group, source libertarians,⁵ necessity and control do not necessarily contradict one another. Some morally responsible actions might be such that an agent has no alternative courses of action available to her. However, those actions must be outcomes of the agent's character and nature with respect to which the agent has had some control in the past. In other words, the responsible agent can be said to control those actions that flow from her character, when the agent has been able to exercise control over that character in the past. If all the agent's actions were determined by prior necessities, she would have lacked the possibility to shape her character and, thus, not qualify as blameworthy or praiseworthy for her actions now.

Consider now a simple argument against libertarianism from the inevitability of sin formulated by Vicens (2023). First, according to Paul, sin is a kind of power that holds people in their grasp. Human beings are described as slaves to sin and, thus, having as much freedom to choose their actions as slaves have:

So, on Paul's view, it is not up to us that we sin, any more than slaves’ actions are up to them; and it is not up to us that we sin, not simply in the sense we cannot be perfect, but in the sense we cannot avoid doing grave evil – evil that makes us liable to divine wrath. (Vicens 2022, 152)

Despite the fact that sinning is inevitable for humans, it is clearly culpable. If it were not so, divine wrath would be unjust as a divine response to human sin.

Now, the libertarian defender of the inevitability of sin has only one option here: she must present an analysis of ‘inevitability’ that avoids collapsing inevitability into necessity. There are a number of candidates on the table. Copan (2003, 531), for instance, argues that:

though we do not sin necessarily (that is, it is not assured that we must commit this or that particular sin), we sin inevitably (that is, in addition to our propensity to sin, given the vast array of opportunities to sin, we eventually do sin at some point).

Copan attempts to demonstrate that sinning in general could be inevitable for sinners without any particular token sinful action being necessary. If successful, this would alleviate the tension between inevitability and libertarian control over action. Even if necessity were incompatible with responsible action, inevitability might not be.

Copan's solution is countered by Franks (2012), who has argued that it entails a contradiction. The moderate doctrine claims that (a) all moral agents inevitably sin at least on one occasion and (b) no moral agent inevitably sins on some particular occasion. Franks takes it to be obvious that our actual world has moral agents who have only one occasion for moral action. Consider one such moral agent. If (a) is true, this moral agent inevitably acts sinfully on this one occasion. However, if (b) is true, the moral agent with only one occasion to sin does not sin inevitably on this one occasion. So, the conjunction of (a) and (b) leads to a situation where there are moral agents who do sin inevitably and do not sin inevitably. The existence of moral agents, who have only one occasion for moral action combined with (a) and (b), leads to a logical contradiction. Something must give.

Rejecting (b) would be an option for those philosophers and theologians who hold that an agent can deserve blame or praise for necessary actions. Many compatibilists are control theorists and they argue that we can make sense of the control condition for moral responsibility in determinist-friendly terms. In other words, moral responsibility conferring action control over some action is possible even if that action is necessary for the agent (e.g. Fischer and Ravizza 1998). If this is correct, then (b) is false and the compatibilist can escape Franks's dilemma. For a traditional leeway libertarian, however, this move is not possible, for the reasons explained above. So, only one option remains: rejecting (a). This will result in rejecting the very notion that the constitutional fault necessarily leads to sinful action. In other words, the ‘inevitability of sin’ is interpreted not in terms of necessary sin, but rather in terms of sinful actions being highly probable. For Franks (2012), for instance, the constitutional fault only strongly inclines or disposes all sinful human beings to commit sinful actions, rather than making sinful actions necessary.

Franks is not the only one understanding the doctrine of original sin and freedom in these terms. Swinburne (1989) reasons as follows. First, he contends that scripture teaches the culpability for sin. Humans, for the most part, deserve blame or praise for the evil and good actions they commit. Second, he defends a libertarian account of free will and responsibility by invoking both everyday reason and scripture. For Swinburne, no culpable sin can be necessary, and he interprets inevitability in terms of necessity. From these two premises a conclusion flows: no culpable sinful action can be inevitable. Human moral agents, who have many opportunities to commit sins, will eventually commit one or more sins. On this view, it would not be strictly true that a person is unable to avoid sinning, but rather not sinning is very difficult for her – so difficult that he will most likely sin at least once.

By interpreting the inevitability of sin probabilistically, the defender of the moderate doctrine might be able to keep leeway libertarians in the fold. However, the price for that move will be rather costly – so costly, that Reformed theologians like Crisp are, very likely, unwilling to pay it. For it follows that there could be sinful agents who succeed in living a holy and virtuous life, namely, succeed in avoiding sinful actions altogether. This would, in turn, prove false the notion that it is impossible for sinful moral agents to enter the beatific vision without Christ's atonement. The probabilistic version of the inevitability of sin would fail to yield the result that Crisp and others want, that is, the universal need for atonement. Crisp (2019, 150, 154) states that Augustinians, like himself, are committed to the claim that no sinful moral agent can lead a life of virtue such that she would thereby merit salvation without atonement. In other words, it is impossible for any sinful normal human to live a life without actual sins.

So far we have assumed that there are two options: inevitability is interpreted in terms of either necessity or probability. However, there might be a third option, which has recently been explored by some libertarian defenders of the inevitability of sin. We want to highlight this possibility here, while acknowledging that we do not have the space the discuss it properly.⁶ This defence strategy will admit that sinful actions might be inevitable in the sense that they must happen, but it holds that they are not necessary. Traditionally, Molinists have attempted to offer this kind of solution. Given that many libertarians have been critical towards Molinism, there is a need for a non-Molinist libertarian strategy.⁷ In a recent paper, Timpe (2022) has argued that such a strategy could be logically possible. We have no space here to address Timpe's argument in any detail. So, as far as we see it, this is one possible way out of Vicens's anti-libertarian argument. However, we want to highlight the demands that such a strategy should meet in order to be successful. The main requirement is that it should explain the inevitability of sin by invoking the constitutional fault as its cause. Furthermore, this explanation should yield the result that sinful actions are unavoidable (so a sinner must commit them), but they are, nevertheless, not necessary. Even Timpe has yet to offer an account along these lines. He merely argues that such an account is logically possible.

So, to sum up the discussion above, the defender of the inevitability of sin faces a difficult choice. The first option is to interpret inevitability in terms of necessity, which rules out libertarian accounts of free will and moral responsibility. The second option is to adopt the probabilistic notion of inevitability: sin is not inevitable, but ‘near-inevitable’. The problem here, however, is the possibility of a fallen human being leading a life of virtue and holiness that would merit salvation without Christ's atonement, which would go against the universal need for atonement. Finally, we mentioned a possible avenue of retaining the libertarian analysis by distinguishing inevitability from necessity. In what follows, we will focus on the first two interpretations and highlight further problems for them.

The constitutional fault and control over sinful actions

The second problem of the moderate doctrine we want to examine is a follow-up to the first one. The problem has to do with the mechanics of inevitable sins. We will examine two variations of the problem. The first variation interprets the inevitability of sin probabilistically, as outlined above. The problem here is that the more probable sinning becomes, the less degree of control the agent exercises over sinful actions. The second problem has to do with the ‘necessitarian’ reading of inevitability. Even a compatibilist account of control has a number of conditions, which might be threatened by the constitutional fault, especially if it is given a very strong reading. Before we explain these problems in more detail, a few words about the constitutional fault and the control condition are needed.

We can, very roughly, divide the effects of the constitutional fault as follows. The corruption of our faculties pertains to both the human intellect and the human will (Couenhoven 2013, 30). With respect to the human intellect, the defect involves a serious failure to identify moral and spiritual reasons, which results in a failure to grasp truths about right and wrong (the moral law) and truths about what God is and what he wants. Following some Reformed theologians, we can call this the noetic effects of sin. The sinful moral cognition is, to a high degree, ‘blind’ or ‘dead’, that is, not responsive to moral and spiritual facts. The second component of the constitutional fault has to do with the will of a sinful human being. The human will is corrupted so that it often fails to implement the best judgment of the intellect. Even if the intellect manages to discern what the best course of action is, the will suffers from conflicting desires and motivations that impair its implementation. On Augustine's account, the disorders of the will are due to the evil concupiscent desires and emotions (Nisula 2012).

We can interpret these effects of the constitutional fault in the light of the two previous readings of the inevitability of sin. According to the first, the noetic effects of sin and concupiscent desires make it very improbable that the sinful agent will succeed in living a moral and holy life. In other words, acting morally and living a holy life are very difficult, but not impossible. According to the second reading, the constitutional fault has so thoroughly disintegrated the human cognitive system that it determines (that is, makes it necessary) that I will fail at least once (or every time, according to most Reformed and Lutheran theologians) in discerning moral and spiritual truths. The result is that sinners are determined to commit one or more actual sins.

What we will suggest next is that both the probabilistic and the necessitarian reading of the inevitability of sin have difficult problems with moral responsibility. This is because both kinds of inevitability seem to lead to a sinner having no significant control over sinful actions and thereby becoming excused from wrongdoings and undeserving of God's negative moral attitudes. In order to demonstrate this, we must look at the notion of control and its connection to moral responsibility a bit more carefully.

We work with a rather simple notion of control: the more successfully agent A is able to bring X (an action, outcome, etc.) in line with his intention, the more A controls X. Alternatively, we could say that the more successfully A is able to make a difference with respect to X, the more she is in control of X.⁸ Control is a success notion, so it comes in degrees. Of course, in the case of limited and constrained humans, there can never be perfect control, since our actions are always dependent on multiple environmental, social, and psychological conditions with respect to which we can do nothing. We are not gods. Nevertheless, by looking at various counterfactual scenarios we might be able to determine a reasonable amount of control that would allow us to justify our moral responsibility practices.

The assumption we are making here is that our attributions of blame (and many other responsibility attitudes) are responses to the intentions and judgments of others. Strawson (2008) calls these reactive attitudes. Reactive attitudes consist of judgments, practices like punishment and forgiveness, and moral and social emotions like resentment, pride, and hate. Reactive attitudes emerge as responses to the quality of will that others express towards us through their actions and attitudes. We react to others (and ourselves) with anger, disappointment for failing to follow norms, failing to adopt attitudes of respect towards us, and disregarding our interests, for instance. This reveals the reason why a moderate degree of control is a core condition for the appropriateness of moral responsibility attributions. If the moral agent exercises no control or a low degree of control over some action or an outcome, he no longer expresses the ‘will’ (intentions, attitudes, and judgments) of the agent, thereby making the blame inappropriate. Furthermore, given the fact that the agent's intentions, judgments, and attitudes can be more or less expressed in some action, it makes sense to adjust blameworthiness accordingly. The more control the agent exercises over some action or an outcome, the more it expresses her intentions, judgments, and attitudes towards others, making the agent more liable to blame and praise.

Crisp (2020, 41) explicitly states that God's moral demands towards us should be conceived in ways that are analogous to our standards of the appropriateness of moral responsibility.⁹ For Crisp, it is natural to read Christian scripture as confirming this. God holds humans responsible by expressing his judgments and attitudes towards humans. Not only does God not want humans to perform sinful actions or be sinful, God expresses this by voicing his judgment over sin and evil, blaming humans for our sinful actions, as well as expressing his negative emotions, like his wrath. The wrath of God is his response to human sin. God's negative reactive attitudes towards sin and evil are also expressed in his ultimate and final judgment, which he promises to deliver in the future. If we assume, like Crisp and others do, that divine reactive attitudes require sufficient grounding and we have some epistemic access to the appropriate conditions (like the control condition), we can, to some extent, evaluate whether God's reactive attitudes are appropriate or not. Indeed, we might even say that since God is the perfect being and God is perfectly just, God's reactive attitudes will, necessarily, be perfectly just and fair. God cannot, according to perfect being theology, be unjust or unfair in his blaming and punishing human beings. Given that we know some of the correct conditions for moral responsibility, we are in a position to, modestly, evaluate God's moral response to our sin.

We now return to the issue of constitutional fault and responsibility for sins. The question at hand is this: given that we are not to blame for the constitutional fault, is it appropriate for God to blame sinners for their sinful actions? To put the question another way, if the sinful actions of the sinner are either highly probable or necessary, can they express the sinner's intentions, judgments, and attitudes to such an extent as to ground God's wrath or even eternal punishment of sinners?

We should note that according to Augustinian theologians, God expects a very high standard of conduct from sinful humans. Let us call this the Perfection requirement. Based on Matthew 5:8 it states that in order for humans to fulfil the moral and spiritual demands of God, they must be morally and spiritually perfect. Humans can only ever merit positive reactive attitudes from God, if they fulfil a high standard of moral and spiritual conduct, and if they fail that standard, they deserve the full might of God's punishment. The bar for meritorious actions is set high in two ways. First, the Perfection requirement imposes a high demand on the nature of a meritorious human action – traditionally referred to as ‘good works’. A moral agent performs a good work only if (1) she knows the relevant moral and spiritual facts, (2) those moral and spiritual facts actually motivate the actions, and (3) her motives are purely selfless. Thus, if a person acts on the basis of false or unjustified beliefs about moral and spiritual matters or if some of her motives are selfish, her action fails to merit positive reactive attitudes from God.¹⁰ Furthermore, the Perfection requirement also states that if a sinner performs even one sinful act, she deserves to be the target of a harsh set of divine reactive attitudes, including blame and eternal punishment.

We mentioned above that we should not expect an extremely high degree of control from any human individual, because many features that shape human moral environments and moral agency are outside the control of individual moral agents.¹¹ Similarly, our cognitive systems are limited, and the operations of our will are constrained by many psychological and environmental factors. Given this, the human ability to control actions is constrained, so it makes sense for us not to demand a high degree of control from each other. Against this, the perfection requirement demands a very high degree of control from humans. Not only does it demand that the moral agent succeeds avoiding sin, say, on 100 out of 100 occasions, but it also demands that the agent manages to perform extremely well every time. If the will and the intellect of sinners are severely compromised due to the constitutional fault, it is far beyond the ability of sinners to fulfil the demands of the perfection requirement.

This leads to a tension in the moderate doctrine of original sin. On the one hand, the perfection requirement sets the bar for meritorious conduct very high and, subsequently, the bar for sinful actions very low. It is rather easy to commit sinful actions. On the other hand, Augustinian theologians usually have a very strong view of the effects of original sin on human moral and spiritual abilities. The constitutional fault makes it either very difficult or impossible for sinful humans to succeed in performing meritorious actions and very easy or inevitable to perform one or many sinful actions. Given the low degree of control that sinners exercise over their actions and such a high demand, divine blame and punishment do not seem like a fair response to human failure.

Let us consider the case where the constitutional fault makes virtuous and holy life very improbable. Given the considerations above, it is easy to see why a moral agent should be attributed a low degree of culpability for those actions that she will commit with a high degree of probability. If sinful actions are ‘almost inevitable’, like, for instance, Swinburne (1989) insists, this means that the agent plagued with constitutional faults of original sin exercises very little control over whether she lives a virtuous and holy life pleasing to God. This might be enough to maintain the basic condition of leeway control, since the sinner always has the possibility to not sin. However, the lower the degree of control, the less blameworthy for moral failures the agent should be. In other words, the more the effects of the constitutional fault impair the sinner's ability to act in a virtuous and holy way, the less culpable she becomes of her moral and spiritual failures.¹²

If the improbabilist reading of the constitutional fault is correct, the moral and spiritual situation where moral agents find themselves is rather dire. While the sinner might be able to do something to alleviate the effects of the constitutional fault, she exercises only a very low degree of control over it. The sinner cannot fundamentally change her misfiring cognitive system and develop skills that lead to a life of virtue and holiness. Due to the noetic effects of sin, sinners often fail to discern the moral and spiritual good towards which they should be striving. Even when they sometimes succeed in discerning the moral and spiritual good, their will is plagued with conflicting desires and emotions that make it difficult for them to act accordingly. All in all, sinners, due to the constitutional fault, are rather bad at acting in ways God expects them to behave and they have a very low degree of control over this fact. So, when God notices that the sinner exhibits disregard of or disrespect towards the moral law and God himself, it is difficult to see why God would respond to this in the harshest possible way, by means of eternal punishment. If sinners often fail to understand the moral law and what God expects from humans, their actions do not appropriately express their lack of disrespect or ill-will towards God and the moral law. So, it would not make sense to unleash the full spectrum of God's negative attitudes against the sinful humans.

Finally, let us briefly consider the case where we interpret the inevitability of sin in a way that makes some sinful actions necessary. We argued above that this interpretation forces its adherents to adopt a compatibilist-friendly account of control. While this kind of control is supposed to be compatible with the fact that the sinner sins necessarily, we have so far said nothing else about the nature of such control. The problem for the moderate doctrine is this: there might be some effects of the constitutional fault that undermine the necessary conditions of compatibilist control, thus making the sinner less culpable for her sinful actions.

Let us now suppose that the noetic effects of sin are such that the sinner is completely blind to the moral law and God on many occasions. In other words, the constitutional fault makes it such that on multiple occasions the cognition of the sinner fails to identify and respond to facts about right moral conduct as well as many spiritual facts. Let us further suppose that the sinner's action control system suffers from failures as well: there are many occasions where she fails to implement her intentions through action.¹³ Does it follow from these facts that our imagined sinner lacks control over her sinful actions or that her degree of control is so low as to significantly mitigate her blameworthiness for her sinful actions? If the defender of the moderate doctrine wants to give a negative answer to this question, she should provide a compatibilist account of control, one that incorporates the corrupting effects of the constitutional fault. Crisp and others have not provided an account like this. As far as we know, only Preciado (2019) has attempted to do this. Given the constraints of this article, we have no space to offer a detailed analysis of this account. We will simply conclude by highlighting some of the difficulties it must overcome.

The most prominent compatibilist account of control is that of Fischer and Ravizza's (1998) semicompatibilism. This is also the account that Preciado uses and develops in the context of Reformed theology. The core idea is that a moral agent can be blamed or praised for actions that are inevitable, if (and only if) those actions are products of reasons-responsive mechanisms that belong to the agent. Our cognitive system might be more or less sensitive to moral reasons in various counterfactual conditions. If the system producing a particular action is sensitive to a very small range of reasons, the agent should no longer be held responsible for her actions. Now, consider a case where the noetic effects of sin lead to significant cognitive failures in identifying and processing moral reasons and spiritual knowledge. The sinner is either completely insensitive to moral and spiritual facts or she only seldom manages to discern those facts and implement them in action. Similarly, we might ask whether the moral agent's degree of control drops so low as to disqualify her from blame and praise altogether. On the Fischer and Ravizza theory, it is difficult to see how a sinful agent with a comprehensive constitutional fault could pass the minimum threshold for morally responsible agency. However, we do not want to argue that providing a control-based, compatibilist account of moral responsibility for sinners is impossible. We only point out that it might turn out to be a difficult task – a task which the defenders of the moderate doctrine of original sin should undertake, if they want to present a plausible case for their view. Moreover, if the compatibilist wants to succeed in this task, he must also assume that the constitutional fault does not completely destroy the sinner's degree of control over her actions. In other words, the fault must not lead to complete mental blindness to moral and spiritual facts and complete corruption of human volition.

Beatific vision and the stain of original sin

We will now briefly discuss the third problem of the moderate doctrine we identified above. The moderate doctrine admits that there are human individuals who inevitably fail to pass the bar for moral agency and individuals whose moral agency is permanently impaired so that they are unable to carry responsibility for their actions. Since blameworthiness for sinful action requires moral agency, these individuals either never commit a single sinful action or they are not blameworthy for their wrongdoings. While these individuals suffer from the constitutional fault, they have done nothing to deserve God's punishment and, therefore, do not seem to be under any kind of threat from eternal damnation. The defenders of the moderate doctrine argue that such individuals are, nevertheless, barred from entering the beatific vision, because their constitutional fault alone (without guilt). This is where Crisp switches from the language of justice, desert, and fairness to that of holiness and uncleanness. He presents a story of a man plagued with leprosy granted an audience with the king. Once it is realized that the man has leprosy, he is disbarred from meeting the king, because he is unfit to be in the presence of the king. ‘Similarly, . . . fallen human beings are in a morally corrupted state irrespective of actual sin. Although they have inherited this corruption through no fault of their own, the presence of such corruption renders them unfit for the presence of God’ (Crisp 2020, 47). Here, Crisp and his Reformed sources depict original sin as a kind of disease, infection, and uncleanness. While Crisp (2020, 48) admits that, for instance, Calvin and Zwingli are content for God to allow the damnation of such individuals, he himself holds that given God's overflowing goodness, God very well might count the merits of Christ's atonement in favour of these individuals as well. So, the moderate doctrine, minimally, allows for the salvation of sinners, who are incapable of moral agency (and having faith).

We want to highlight a problem in this account. The problem has to do with the universal need for atonement. The strong doctrine of original sin has a straightforward and simple explanation for the universal need of atonement. All human beings are culpable of either sinful actions or the state of sin or both. Without atonement, they merit eternal punishment.

In the language of reactive attitudes, atonement is needed to overcome God's negative reactive attitudes towards culpable sinners. As we have already seen, the defender of the moderate doctrine rejects original guilt, so she cannot accept this solution. In the case of a non-culpable constitutional fault, there are no divine negative reactive attitudes to overcome. The sinner has expressed no ill-will towards God or the moral law. She has not failed to meet a justified moral or spiritual standard. Indeed, such standards do not apply to her, since she is not an agent who is capable of incurring any kind of negative reactive attitude from God. So, for such individuals there is no debt to be paid to God.

Now, there might be all sorts of reasons why individuals suffering from non-culpable constitutional fault might still need atonement.¹⁴ Atonement might be needed to remove some other barrier (than guilt and God's wrath) between the individual and the beatific vision.¹⁵ So, if God is abhorred by the uncleanness of non-culpable sinners, God could put in motion a process to deal with the uncleanness via Christ's atonement, for instance. However, what these doctrines do not explain very well is the reason why non-culpable sinners are under the threat of eternal damnation in the first place. They have done nothing wrong and therefore do not deserve to be barred from the beatific vision.

The problem becomes more pressing if we consider what being left outside the beatific vision might actually mean. We take it that barring an individual from the beatific vision amounts to allowing her to exist and suffer the consequences of being deprived of connection to God. Traditionally, this is how hell has been conceived: a state where God is absent. Moreover, the traditional account of hell entails that the absence of God is a state of great suffering (Timpe 2014, 69). So, the claim is that God would be justified in allowing a group of individuals to end up in a state of great suffering through no fault of their own. This claim, we submit, goes against a very Augustinian notion of God's justice. Why does God allow evil and suffering of created persons in the first place? Augustine himself felt very strongly that the problem of evil is a significant challenge to Christian faith. If God is perfectly loving and just, God will not allow undeserved human suffering. So, if God actually allows a person to suffer, it must be because the person deserves it (Couenhoven 2013, 49–50). This is one of the motivations for Augustine to posit the strong doctrine of original sin in the first place. God's justice demands that evil and suffering are deserved punishments for the primal sin. The constitutional fault and all the evil it creates are just punishments for the primal sin. Now, the claim that God is justified in barring an innocent moral agent from the beatific vision and thereby allowing that agent to suffer enormously from the lack of connection to God goes against this basic Augustinian motive.

In sum, while the moderate doctrine solves some of the counterintuitive moral problems inherent in the strong doctrine, it creates additional problems in their place. The problem we outlined in this chapter is that the strong doctrine has an elegant and simple explanation for why all created individuals deserve eternal punishment, namely, culpable sin. The moderate doctrine, however, has no access to such an explanation. The defender of the moderate doctrine would have to offer some reasons why non-culpable sinners should suffer eternal punishment without any recourse to guilt and culpability. Intuitively, it seems to us that justifying eternal damnation by invoking guilt is much easier than justifying it by invoking non-moral features of the damned individual. We are not saying this is impossible, of course, but simply more difficult.

Conclusions

We have suggested that the control condition for moral responsibility is a good explanation of the moral intuitions that drive the rejection of original guilt. However, we argued that the control condition generates multiple problems for the defender of the moderate doctrine of original sin. These problems have to do with the inevitability of sins and other commitments of Augustinian theologians. First, if the defender of the moderate doctrine wants to hold that all moral agents commit at least one actual sin inevitably, she will be forced to adopt a compatibilist account of control and moral responsibility. The inevitability of sin could also be given a probabilistic, libertarian-friendly interpretation, but this will not justify the Augustinian contention that it is impossible for sinners to lead a life of virtue and holiness. The second problem we identified has to do with the mechanics of the constitutional fault. The effects of original sin on the human faculties of intellect and will make it very difficult or impossible to perform meritorious actions and very easy or inevitable to commit sinful actions. Given that the sinner's degree of control over her sinful actions is so low, it does not appear just for God to demand perfection from the sinner and blame her for failing to achieve it. Finally, we raised a problem regarding why non-culpable sinners would be under the threat of eternal punishment in the first place, if they have done nothing wrong.

Given the problems that the control condition creates, one option is to reject the problematic condition wholesale. There are two ways to go here. The first one is to replace the control condition with something else that will perform the same function, namely, the justification of the moral intuitions behind the rejection of original guilt. Augustinians, like Couenhoven (2013), have attempted to do this by defending what are now called attributionist accounts of moral responsibility.¹⁶ These proposals are interesting and definitely have merit, but we will not address them here.

The second option is to adopt a certain kind of scepticism. Some Augustinian theologians hold that while the control condition might be a plausible principle in the secular ethical domain, we have no reason to think that it would apply in the theological or religious domain. In other words, the norms that apply to the appropriateness of humans holding each other responsible might not apply to God. So, even if it would be inappropriate for a human being to blame another human for some wrongdoing, it might be appropriate for God to do so. Moreover, given the noetic effects of sin, we have no access (or very low degree of access) to the conditions God must fulfil in order to be just and fair. Since we know very little about God and are often mistaken about the moral law, including the correct conditions for moral responsibility, we are in no epistemic position to evaluate God's fairness. Augustinians, who defend the strong doctrine of original sin, could hold that sinful human beings are for this reason unable to evaluate correctly whether God's actions and attitudes are fair and just from our perspective. As a consequence, we cannot conclude that God is unjust in his treatment of sinners even if it looks to us that God places unreasonable demands on sinners.

While replacing the control condition with attribution or rejecting all attempts to find moral conditions that apply to God might have venerable Augustinian roots, they are not attractive options for the defender of the moderate doctrine. For rejecting or replacing the control condition would undermine many of the moral intuitions that drive the moderate doctrine in the first place, especially the ones having to do with guilt over the constitutional fault. If a sinner can be justifiably blamed for the constitutional fault (might be possible given the attributionist theory) or if we cannot know whether God's demands are unfair, the list of reasons to prefer the moderate doctrine to the strong doctrine grows shorter and shorter. There seem to be very few reasons to prefer the moderate doctrine over the strong doctrine, if one is willing to allow for moral responsibility over non-controllable actions or character traits.