Book Review: Responsibility and Atonement by Richard Swinburne

Overview. In Part I of Responsibility and Atonement, Richard Swinburne examines various moral concepts that apply to dealings among humans. In Part II, he uses the conclusions of Part I to determine what versions of central Christian doctrines (e.g., morality, sin, redemption, sanctification and corruption, heaven and hell) are most plausible.

Part I: Responsibility. In Chapter 1, Swinburne examines various kinds of moral goodness. He is primarily concerned with how we determine that an act is good, not so much with which specific acts are good. He examines various proposed definitions of moral goodness, and finds them all to be more or less on the right path but in various ways incomplete. He then proposes his own definition of moral goodness.[1] Swinburne endorses the traditional categorization of good acts as either obligatory (constituting duties) or supererogatory (good acts which go above and beyond duty). We are praiseworthy for doing supererogatory acts but not obligatory acts. Conversely, we are blameworthy when we fail to do obligatory acts but not blameworthy when we fail to do supererogatory acts. There are different kinds of supererogatory acts—favors which benefit groups or creative acts which may only (directly) benefit the agent doing the action. For example, my donating food to a family in need is a favor which directly benefits the family and (perhaps, given pure motives on my part-see more on this below) myself as well. A creative act would be me learning to become a great painter. This would be a supererogatory act which benefits only me directly (although, as we will see, it arguably indirectly benefits my community).

It is not only the goodness of act which is important, but my intentions in performing some act. A man who risks life and limb to save a friend is equally praiseworthy whether or not he succeeds in saving that friend. The counterintuitive implication of this, Swinburne thinks, that this means there is something good about someone doing what is a subjectively good act (one that they perceive to be morally good) even if in fact the act is not objectively good (that is, the act does not correspond with objective, external goodness). Swinburne gives the example of a Stalinist who sincerely believes it a good thing to wipe out the kulak class. The subjective goodness of the act arises from his intentions, even though his action is objectively wrong.[2] We can sincerely hold false beliefs which can impact our moral behavior. Our false beliefs may be either moral beliefs (as in the Stalinist example) or nonmoral beliefs (e.g., I give money to a homeless person thinking he will spend it on food, unaware that I am in fact worsening his addiction to alcohol because he will use it to purchase whiskey).

The upshot of all this is as follows: there are two kinds of good actions, those that are spontaneous (arising out of our natural inclinations) and those that are acts of perseverance (as when we do good in spite of a temptation to not do good). There are also three kinds of good character: being naturally inclined to do the objectively good, having correct moral beliefs, and being naturally inclined to do the subjectively good. So the Stalinist possesses the third kind of good character but not the other two kinds. Swinburne also notes that there is some tension among the types of good actions and good character. If one has a good character that is, through and through, naturally inclined to do the objectively good, then there is no opportunity to commit acts of perseverance (because there is no temptation to not do good).

In Chapter 2, Swinburne examines the concepts of moral responsibility and weakness of will. He argues that we are blameworthy when we fail to do obligatory acts, and praiseworthy when we refrain from bad acts in the face of great temptation. We are not blameworthy when we are subject to conflicting obligations as long as we are not culpable for being in the conflicted situation. Further, we may not be blameworthy when we fail to do the obligatory if we decided instead to do a supererogatory act. He spends a good deal of time focusing on the role of the will in all this.[3] A person is also more blameworthy the longer the intention to do a bad action persists (as in manslaughter versus premeditated murder). He also addresses an issue going back to Plato and Aristotle: how is it possible for a person to do an act which they do not believe to be the best? To do an act intentionally (a restriction necessary because, Swinburne thinks-plausibly enough-that only actions that involve intentions or our free will are those for which we are blameworthy or praiseworthy) is to see it as good—in some sense, Swinburne acknowledges. But this ‘some sense’ does not necessarily mean seeing it as good in a moral sense. Thus, objective goodness and badness of action can be opposed in a way that subjective goodness and badness are not. An act is subjectively good if the agent sees some reason to do it (in either a moral or a non-moral sense). An action is subjectively bad if the agent sees no reason to do it. But that means it is not an intentional action, and thus not a moral action—and hence not under discussion. Thus, Swinburne thinks the most that can be said is that we have subjective goodness of actions and weakness of will. The latter would be situations in which we yield to a non-moral reason to do an action.

In Chapter 3, Swinburne sketches the central role that free will plays in this whole discussion. He defends the coherence of free will and rebuts various arguments for incompatibilism. In Chapter 4, Swinburne turns to a more explicit consideration of the concepts of merit and reward. He revisits his concept of creative acts from Chapter 1 and argues that although the creative act (my becoming a great painter) may only directly benefit me (say I refuse to let others see my paintings, great though they are), nonetheless my community gets some indirect benefit in that their investment in my education and nurturance were not in vain. I did not ‘waste’ their investment in me. Merit involves an agent having worth in light of doing supererogatory acts. Merit may itself be either subjective or objective. Say, for example, I promise to help you out if you ever become poor. You do become poor. I help you out, and think that my helping you out was objectively meritorious. In fact, it is merely subjectively meritorious—that is, I perceived the act to be supererogatory and did it anyway. But my promise to you means (given the plausible existence of an obligation to keep promises) that the act was objectively obligatory. So I am not fully praiseworthy: the action was not both objectively and subjectively meritorious. If I helped you out solely for the praise of others, then the act is no longer meritorious at all. The subjective merit is not present because I did not do it in light of a belief that it was supererogatory, a good thing to do. Nor does it, on balance, have any objective merit—that is counterbalanced by the objective wrongness of my deceiving others as to my motives.[4]

In Chapter 5, Swinburne turns to an analysis of guilt, atonement, and forgiveness. We can be objectively guilty or subjectively guilty (we committed an objectively wrong act, or violated our consciences by committing an act we judge to be morally wrong). Subjective guilt is worse than objective guilt and requires more by way of atonement. Just as creative acts can directly benefit an agent and indirectly benefit society, so too can failure to do obligatory acts directly wrong a victim and indirectly wrong others. Similarly, supererogatory acts—given the co-dependence between human beings—always benefit someone other than the agent. Guilt is to be distinguished from shame, which is an emotional reaction involving remorse. I can be objectively guilty (I punched someone) and subjectively guilty (I believe that punching him was wrong) but not shameful (I feel that punching him was justified, or worse I take sadistic pleasure in it). This is a bad state of affairs-our desires should align with our moral beliefs. Atonement is a complex notion which involves several sub-components. A debt can be paid by a wrongdoer taking certain actions (paying a fine, or returning a stolen watch) or by the victim taking compensation. But for total removal of the debt, some action on the part of the wrongdoer is required. Sufficiently bad actions may require full atonement: repentance (making the ‘present’ me as different in attitudes and actions from the ‘past’ me which performed the wrong act as possible), apology (outward statement of repentance), reparation (some act which restores the status quo) and penance (an additional gift of some sort to the victim). While it may sometimes be good for a victim to forgive without atonement, sometimes it is not. What would we think, Swinburne asks, of Mister X who-despite knowing that Mister Y had murdered X’s wife and made no attempt at atonement-socialized with Mister X? We would judge that X was morally deficient in some way. First because he was not taking his wife seriously and second he was not taking Y seriously—the attitudes expressed in Y’s actions deserve condemnation. Swinburne notes that it may not always be possible to make atonement in our limited earthly life, and without taking a firm stance on the matter floats the possibility of Purgatory making post-mortem possible. Finally, we can sometimes bear indirect responsibility for others. If a parent fails to be a good influence on a child, then they bear some responsibility for that child’s bad actions. We can also help a child atone in some circumstances. If my child breaks a neighbor’s window, I can help ‘foot the bill’-give the child money for the reparations. But repentance must come from the child.

In Chapter 6, Swinburne analyses the concept of punishment. If a wrongdoer does not make reparations to the victim the victim (or the state as his representative) has some right to take reparation, and more than the original loss, as punishment. (Punishment may be seen as the flip side of penance.) Swinburne further argues that it is good that there be further utilitarian justifications for punishment, but that these cannot be the only justifications.[5] Sometimes we may have a duty to punish, and sometimes a duty to refrain from punishing. Swinburne notes that mercy (conceived of as a supererogatory act) does not negate the validity of punishment, but in fact supports it. Earlier Swinburne argued that the obligatory and the supererogatory were different types of good. If something is obligatory it cannot be supererogatory, and vice versa. So if mercy is a supererogatory act, that must mean that in some cases it is no longer obligatory to avoid inflicting pain (punishment) on someone. In Chapter 7, Swinburne briefly examines man’s moral behavior and discusses how some of our desires are given to us by nature (genetics) and others by nurture (society). Society and communities can help or hinder our moral and nonmoral desires.

Part II: Its Theological Consequences. Here Swinburne applies the results of Part I to traditional Christian doctrines. That is, when conjoining the results of Part I to what we might call “The Christian Assumption”[6], what versions of those traditional Christian doctrines emerge as the most plausible?

In Chapter 8, Swinburne argues that God’s existence changes the nature of some moral truths, but not all.[7] Some things would still be obligatory, even if God did not exist (such as showing gratitude to our benefactors). Still, even those duties would be deepened. Our duties to our parents as benefactors, for example, become doubly obligatory in that by failing in our duties to our parents we are also failing in our duty to God, our ultimate benefactor. Some things, however, do become obligatory which would otherwise not be. Worshipping God as our ultimate benefactor becomes a duty only if God in fact exists.[8] God’s perfect goodness means that he would not command us to perform certain acts, and that while our moral status is lowered (God’s existence means that there are duties there would not otherwise be, e.g., worship, and hence we are greater moral failures given God’s existence) our futures are infinitely brighter. (Swinburne notes that we like spending more time with interesting friends. So what could be greater than spending infinite time with the most interesting friend possible?) There are also goods that are necessarily, intrinsically good—but pursuing these goods (like desiring infinite happiness for ourselves) becomes an even greater good when commanded by God.

In Chapter 9, Swinburne turns to the concepts of sin and original sin. All humans commit actual sin. We have a proneness to sin (original sinfulness, one of three cardinal doctrines underlying original sin). Swinburne argues that original sinfulness is transmitted genetically.[9] Swinburne sees original sinfulness as a proneness to sin, not one that necessitates (makes unavoidable) sin. He thinks that the necessitarian view led to Calvin’s doctrine of total depravity, as opposed to a broader Christian tradition of a libertarian view of freedom. (In any case, his earlier arguments shored up the link between libertarian freedom and moral responsibility.) Swinburne also thinks that Augustine and Calvi were led to their deterministic position by the belief that if man had freedom to choose what God did not prefer then God was less than omnipotent. In contrast, Swinburne argues that God’s omnipotence is not impugned if (a) he generously allows creatures to determine their own destiny and (b) it is only by God’s continuing choice that creatures have the power to go against his preferences. Swinburne also discusses Adam, and argues that it is plausible that there was a first ‘man’ (a creature whose biology and psychology—holding moral beliefs, for example—made him a human). Swinburne notes that there have traditionally two views of Adam held by Christian theologians. The first is that Adam was a feeble creature with strong passions and a weak will (proposed by Irenaeus). The second is that Adam was essentially perfect without proneness to sin (proposed by Athanasius and Augustine). Swinburne plumps for the former view, suggesting that a ‘lesser’ Fall is more consistent with the modern account of evolution which entails a very gradual evolution of man from more primitive creatures and an attendant gradual development of his various capacities. Swinburne then turns to the doctrine of original sinfulness, which is the claim that the sin of the first sinner which caused in other min original sinfulness (the proneness to sin) via descent from Adam.[10] RS thinks this is mistaken, given that Lamarckian evolution is false and hence there is no inheritance of acquired characteristics. While it is possible that God could have overridden the normal genetic processes on this unique occasion, the second and stronger consideration against it is that original sinfulness (troublesome desires) exist in apes as well. So Swinburne posits that Adam’s responsibility for our sinfulness is confined to a responsibility for beginning the social transmission of morality which made sin possible—a corrupt morality given his sinful example and perhaps false moral beliefs. Bad, yes-but each of us and so many others have helped it along the way. Finally, Swinburne tackles original sin—the idea that all of Adam’s descendants are guilty for Adam’s original sin. However, he argues that few of the Church fathers showed an awareness of this—it being denied by Clement of Alexandria and Gregory of Nyssa and watered down by Duns Scotus. Earlier Swinburne had argued that we might have some guilt for another person’s bad actions if we failed to prevent them from committing the act when we could have stopped them, or if we could have provided a better example for them than we did. Neither of these, of course, apply to Adam who existed before we did.[11] Swinburne does note his earlier caveat: we are all ‘involved’ in the sins of our community-although in this context the community is the entire human race. We are members of this community given that we owe so much to other humans, especially our ancestors. So we must regard ourselves as involved in their failures—above all that first failure, the original sin. Further, we should want to forward the purposes of our ultimate benefactor (God). one of his purposes is to have humans atone. So we should help them do so. How that can be done is handled in the next chapter.

In Chapter 10, Swinburne focuses on the concept of redemption. He argues that just as it would have been wrong for Mister X to ignore Mister Y’s slaying of X’s wife, so too God simply turning a blind eye to our sin in the absence of atonement would be to take our transgressions and our attitudes (as expressed by our actions—that is, our sin) seriously. Further, a good parent might refuse to accept an apology until atonement had been made, even if the opportunity for atonement came from the parent himself. Swinburne notes that there is less agreement among Christian theologians (both past and present) on models of the atonement than there is on other matters like the Chalcedonian interpretation of the Incarnation (Christ possessing two natures in one person). Swinburne raises various objections to the ‘battle victory’ model of the Atonement[12]. He concludes that the biblical ‘sacrifice’ model is the most satisfactory of the models of the atonement.[13] He further argues that God might help us make atonement is further buttressed by an argument that Christ’s life and death would be adequate for that atonement. The argument that Christ’s life and death would satisfy the demand for atonement is relatively straightforward: we have failed to live perfect (or even good!) lives and so the most satisfying sacrifice would be a perfect life. The greatest sacrifice would be that of a being who owed nothing to us, but whom we owed everything. In any case, fallen man would have made an impure sacrifice. Swinburne rounds out the chapter by discussing how atonement makes sense of other doctrines like the Eucharist.

Chap 11: Sanctification and Corruption. Sanctification is achieved by pleading the atonement made by Christ and by gradually coming to form true moral beliefs and making oneself naturally inclined to conform to them. It is the function of the church to facilitate this process. Total corruption occurs when a man yields so often to bad desires that he becomes their prisoner and loses the moral beliefs that incline him to resist them. We shape our characters. A good act now begins a habit—so that good acts later become easier. Bad acts now lead to (probably) more bad acts later. And so on.[14] Society can shape us for good or ill (often ill)—it is thus the role of the church to be a counterweight for the good, a motivating force which pushes us toward sanctification. He also discusses cognitive dissonance (a perceived contradiction between some moral truth we assent to and actions we take). We strive to reduce cognitive dissonance, either by ceasing the bad acts or suppressing our moral awareness. Self-deception is gradual, or at least total self-deception is—just as sanctification is gradual.

Chapter 12: Heaven and Hell. Swinburne thinks everlasting damnation for finite sins (because committed in a finite life) would be unjust.[15] He thinks it is at least not clear that the Bible teaches eternal, corporeal damnation and thus is open to annihilationism of the damned. If the damned continue to exist, they will not (indeed, cannot, given their voluntary status as totally corrupt) behold the Beatific Vision. [16] Perhaps the most enlightening portion of this chapter is the all-too-brief discussion of God giving us ‘fixed good wills’. Saints, while not yet sanctified, do desire to overcome their non-moral urges. Swinburne speaks movingly of the role that God, as the greatest being, plays in the enjoyment of the afterlife.

And the saints’ desire to know more of God will be fulfilled by God showing them more and more of himself. Since he alone has an infinite depth of wisdom and goodness to reveal, he can go on forever yielding to man’s longings and yet leave him unsated.[17]

Swinburne also holds that ‘good pagans’ will wind up in heaven. He thinks that those of ‘good will’ (a desire to pursue and be good) will be seen as such a precious thing that God, as perfectly good, will give them time in the afterlife to gain the correct moral beliefs and thus become saints. [18] More controversially he also holds that not only good pagans but atheists who are atheists ‘of no fault of their own’.[19] In defense of this, he draws on the doctrine of limbus patrum, which explains how the Old Testament patriarchs came to reside in heven. He also quotes Jesus:

Although it has taken a couple of millennia to make the point explicitly clear, it is, I believe, the point which Jesus is reported as making, that the servant which did not know his master’s will and did things worthy of a beating ‘shall be beaten with few stripes’, in contrast with the servant who knew the master’s will and still did not do it, who ‘shall be beaten with many stripes’.[20] 

Because Swinburne, plausibly enough, sees libertarian freewill as a precondition for moral responsibility, he rejects Calvinistic views of predestination. In discussing the fates of babies and others who passed before they could reasonably held morally accountable for their beliefs and actions, he posits that babies may be accepted as is or maybe (like the ‘good pagans’) given a post-mortem opportunity to shape their own characters so as to gain moral accountability. He ends the book with the following paragraph:

We choose, but God choose which choices to give us. And it is good that our bad choices and good choices should not form our characters with instantaneous effect. We need to show the resoluteness of our commitment over a short period. For a year or two we can change our minds and postpone any irrevocable choice. But sooner or later, whether we like it or not, our refusal to choose the good begins to have a permanent effect on our nature. Our freedom to choose between good and bad has a heavy cost. We may choose the bad, and cause much suffering to our fellow men, and in other ways also wrong them and our Creator. The longer the period of time for which God allows us this kind of freedom, the heavier the suffering to our fellow men may be. There are limits to God’s right to allow others to suffer for the sake of our having this kind of freedom. Further, if God allowed us always to be able to postpone any irrevocable formation of character he would be putting in our way a very strong temptation never to face up to things, to trivialize our lives and forever to waste God’s gift, and to avoid ever making ourselves worthy to enjoy Heaven. It is good that God should allow us to be tempted over a brief period to do evil, for that gives us a real choice. But to allow us to be subjected forever to a temptation to avoid any ultimate commitment would make it very difficult for us ever to make that commitment; and a good God might well avoid putting us in that position. He will do this by giving us a limited period of earthly life in which by our actions we can choose our character; he will leave us free to choose, and he will give to each with his resulting character the kind of life appropriate to such a character. What more could we want?

Assessment. As with any of Swinburne’s works, it is hard to express the subtlety and depth of the original in a (somewhat!) brief review. There are always fascinating side roads to explore. This will challenge the believer to ponder deeply central propositions of the Christian faith. Some arguments will strike a reader as plausible, others less so. However, I think any Christian who wants to seriously grapple with their faith would do well to read this book. It is always a pleasure to ‘meet’ a mind which has spent so much time and effort in exploring the connections between the Faith and other beliefs we hold (like moral intuitions).


Apologetics 315 Book Reviewer Latter Day Inkling is a U.S.-based research psychologist for the military. He is especially interested in epistemology and natural theology.



[1] His definition of goodness is that an act must be one of overriding importance derived from its possession of universal properties. The universal properties must be connected with widely held moral intuitions. Admittedly, this definition is not very enlightening without understanding how the terms are defined and how this definition arises from his consideration of the inadequate (albeit partially correct) definitions of moral goodness offered by other philosophers.


[2] Swinburne ties this to the problem of doxastic voluntarism—the idea that we cannot easily change our beliefs, although we can gradually change them by choosing to place ourselves in specific situations and make ourselves vulnerable to specific influences. Swinburne also notes that many people would suspect that the Stalinist is merely claiming that he is following the dictates of his conscience. This suspicion arises, Swinburne thinks, because we judge it unlikely that an individual could truly be so morally blind.


[3] He cites the Old Testament book of Numbers as an example where a person is more blameworthy if an action was taken in the belief that it was a sin compared with an action taken in the absence of such belief.


[4] Swinburne cites Christ’s admonishment of “They have already received their reward”.

[5] For example, a utilitarian might argue that punishment would deter other people from committing the same bad act. Swinburne points out, however, that mere utilitarian grounds would justify punishing an innocent person, as the deterrence effect would be the same in either case, all other things held equal. 

[6] That there is a perfectly good, omnipotent, omniscient, omnipresent Creator and sustainer of the world who incarnated in Jesus Christ, led a perfect human life, underwent an atoning death, resurrected, and established a church to promote his teachings. Swinburne also explicitly says that by ‘assumption’ he does not mean ‘insufficiently justified’. Rather he means that the defense of the various parts of his assumption take place in other books.


[7] He thus takes a compromise position on the Euthyphro Dilemma.

[8] Swinburne argues that we have a duty to worship God because God has certain properties-i.e., he is our benefactor. But that means that the duty to worship derives from a more general, necessary moral principle that we have duties to benefactors, and so the obligation to benefactors does not in fact ontologically founded in God. This seems to be a minority position among Christian philosophers. See Good God by Baggett and Walls and http://www.reasonablefaith.org/can-god-ground-necessary-moral-truths


[9] Swinburne cites the Council of Trent stance on original sinfulness (‘propagation, not imitation’) in support of this. He also claims that ‘social’ inheritance of sin (as opposed to biological inheritance) was a hallmark of Pelagianism.

[10] Doesn’t think that all men were descended from the first man, but raises the possibility that moral concepts were acquired independently by different hominids in different parts of the globe, who each then did wrong.

[11] Cites some OT passages in this vein, esp. Ezekiel 18:20 “The soul that sinneth, it shall die. The son shall not bear the iniquity of the father, neither shall the father bear the iniquity of the son: the righteousness of the righteous shall be upon him, and the wickedness of the wicked shall be upon him.” 

[12] The ‘victory’ model fails to explain why a victory can remove sin and why the victory required such a costly struggle. The ‘ransom’ model fails to explain why God would pay ransom to the devil, who he could simply annihilate. The penalty (e.g., penal substitution) model fails to explain why man would need to do aught but repent and apologize, etc.

[13] Swinburne cites the book of Hebrews as support for the sacrifice model. He notes in passing the Old Testament tradition of sacrifice as well, and draws parallels between the life and death of Christ and other concepts. E.g., Christ gives something of ultimate value (his life of obedience to God the Father and his own death) as a present to God, whose benefits flow to others through the Resurrection.

[14] See Chap 10 of The Existence of God, especially p. 222.

[15] Some have handled this problem by positing that the damned continue to sin while in Hell, thus incurring more punishment. See the debate between William Lane Craig and Antony Flew.


[16] Swinburne cites in passing Origen as a universalist. My read is that Swinburne does not affirm this. 


[17] P. 191.

[18] He cites the Catholic doctrine of ‘implicit faith’ on this score.

[19] Contrast this with the discussion of sin in Alvin Plantinga’s Warranted Christian Belief.


[20] P. 192

Written by

Brian Auten is the founder emeritus of Apologetics315. He is also director of Reasonable Faith Belfast. Brian holds a Masters degree in Christian Apologetics and has interviewed over 150 Christian apologists. His background is in missions, media direction, graphic design, and administration. Brian started Apologetics315 in 2007 to be an apologetics hub to equip Christians to defend the faith.

Type at least 1 character to search
Catch the AP315 Team Online:

The mission of Apologetics 315 is to provide educational resources for the defense of the Christian faith, with the goal of strengthening the faith of believers and engaging the questions and challenges of other worldviews.

Defenders Media provides media solutions to an alliance of evangelistic ministries that defend the Christian worldview. We do this by elevating the quality of our members’ branding to match the excellence of the content being delivered.