Fairness,Hierarchy,and Moral Rationalization,or What's Wrong With Paradise Lost ?

Stories clearly have something to do with emotion. They make us laugh at bumpkins in the big city, cry at the deaths of innocents, dread the dangers faced by the entire planet due to our ecological insensitivity, rage at the injustices perpetrated against “our people” by the heartless enemy, and more. Ethics too is bound up with emotion. If the point was not obvious before (it really should have been), Jonathan Haidt demonstrated it pretty clearly (2012). Just get test subjects to associate disgust with a random word and the appearance of that word will make them react to Dudley Do-Right helping an octogenarian across the road as if he was having sex with a chicken carcass. Well, okay, it's not that extreme, but some of your test subjects will find the whole “helping old ladies” business suspicious. As it turns out, there is also a close relation between stories and morals (and not only through weighty allegorical figures with such implausible names as “Sin” and “Do-Right”). If you are like me, you cringe inwardly every time you hear that x—ethics, politics, medicine, law, you name it—is “narrative.” Such claims often border on vacuity, as the notion of narrative simply expands to encompass, well—ethics, politics, medicine, law, you name it. But the connection won’t be vacuous here. Trust me. I’ll get to it. (You can’t expect everything in the first paragraph.)

Before going on to any of that, though, I need to say a few words about what is perhaps the most famous retelling of one of the most famous stories of the Judeo-Christian world—John Milton's rendering of the Fall of Adam and Eve in Paradise Lost. It's no secret that there is a problem with Milton's epic. Though widely considered the greatest poem in English, it seems to fail in its main purpose, its “great argument” (as Milton puts it [2003/1674]): to “justify the ways of God to men” [1.26]. You would think that a poem treating God and Satan would have an easy time of at least one thing—leading the reader to favor the former over the latter. After all, the former is the source and paradigm of good, the latter of evil. But, as William Empson famously pointed out, this God “kills Adam and Eve and all their descendants for eating an apple,” which is what we might expect from “a merciless tyrant” (in John Leonard's summary of Empson [2003, p. xxv])—or even some sort of sociopath. To put the problem in a perhaps overly banal way, exterminating an entire lineage because two members of the clan ate a piece of fruit doesn’t seem terribly fair. And this wasn’t some hack; the author is almost universally acclaimed as one of the three greatest writers of English (along with Chaucer and Shakespeare). Surely, he must have had something in mind.

Now, gentle reader, permit me to request of you the following indulgence. Take the preceding comments on Paradise Lost as you might take some unexplained scene shown, without context, at the start of a film, a scene to which the film will return only at the end; allow it to hover at the back of your mind, a sort of unresolved quandary. At this point, a title card might appear, reading “Fifty-Nine Years Earlier,” after which we would cut to a scene of, say, a Seventeenth-century childbirth, or “Twenty-Five Years Earlier,” followed by a montage of scenes re-enacted from the English Civil War. Here, in this essay, the title card reads:

Literature and Moral Feeling

This refers to a book I completed in 2021 (Hogan, 2022). That book does not provide a resolution to our dilemma about Paradise Lost. In fact, its relevance is almost the precise opposite. It addresses ethics but manages to defer the entire issue of fairness. As you may have guessed from this (and from my previous reference to fairness), I believe an understanding of fairness will lead to at least a partial explanation of what Milton is doing in his epic. What is more important, I believe that the problem posed by the poem itself contributes valuably to our development of an account of fairness, which (if true) would have implications well beyond Milton criticism, reaching consequential aspects of the relations among emotion, ethics, and stories.

More precisely, in that book on moral feeling, I begin by arguing that, functionally, ethics is a type of constraint on what I call “egocentric self-interest.” ¹ The idea here is that every act we engage in is either motivated or unmotivated. If unmotivated, it may be reflexive, habitual, or random (there may be other possibilities also). If motivated, it involves the pursuit of some goal defined by an emotion system. For example, the activation of my attachment system with regard to my spouse is likely to make me want to be near her. The integrated activation of attachment, sexual desire, and the reward system led Romeo to pursue an enduring union with Juliet. All motivated actions are “hedonic” in the sense that they concern actions that, if achieved, would fulfill the goals set out by the relevant emotion system, which fulfillment should increase happiness or decrease unhappiness. In cases where this hedonic quality is (as far as one knows) consistent with one's long-term well-being, we may refer to the action as not merely hedonic, but also as “self-interested.” Thus, the self-interested options are those which are not only (simulated as) pleasurable, but also prudent.

Self-interest may be understood as egocentric when it is governed by the hedonic value of the goal solely for oneself. What other kind of self-interest is there, you may ask? Well, there is also a class of goals and motivating emotions whose fulfillment is dependent on the emotional states of other people. These states are “allocentric.” Allocentric goals and emotions come in two varieties as well. Specifically, drawing on work by Nico Frijda and Klaus Scherer (2009, p. 10), along with other researchers, I use the phrase interpersonal stance to refer to one person's attitude to the emotions of another person. The emotions are parallel—thus cases of affective empathy—when the two people experience emotions of the same valence (positive or negative). They are non-parallel when the valence differs, as in Schadenfreude. In Literature and Moral Feeling, I take it that moral feeling is an empathic response to the configuration of another person's current emotions and those he or she is likely to feel in a consequence of acting one way or another. That empathic response should then be parallel to the complex, self-interested emotion of the other person as that would be felt by him or her, given his or her goals and circumstances. In other words, it is a prudential, empathic response.

Most people would presumably agree that moral feeling should qualify or even displace a person's own self-interested response in some cases, but not others. But individuals differ on just what cases are of which sort. In part, those differences can be understood by reference to the way we set parameters on a few basic moral principles. For example, some people conceive of moral obligations as in principle uniform across possible targets. This is sometimes called an “equal pull” ethics (see Sommers, 1986). Others argue that some groups have a stronger moral claim on us. This is sometimes called “differential pull” ethics. Often, advocates of the differential pull position set the scope parameter, as I call it, at the minimum distance from ego-centrism, marking our highest moral obligation as that bearing on persons with whom we have a particularly close, affiliative relation, most obviously family members. Giving ethical preference to some identity group, such as one's nation, would fall between the extremes of pure egalitarianism and preference for affiliation groups. In terms of empathic response, both affiliative and identity-group preference would be consistent with our spontaneous empathy (some of the research suggesting the link between empathy and in-group status may be found in Hein et al., 2010; see also Gazzaniga, 2011, p. 164; Hess, 2009, pp. 253–254). As Robert Sapolsky explains, there is a close relation between “in-group loyalty and favoritism” and “an enhanced capacity for empathy” (2017, p. 86). The extension of ethical commitment beyond affiliative relations and in-groups, thus to some form of egalitarianism, requires self-conscious effort at imagining–in their specificity and complexity—the target's emotions (e.g., the emotions of a person affected by war in a part of the world that is unfamiliar to us).

Developing from this general delimitation of ethics, and the broad orientation given in fundamental principles and parameter settings, I argue, we make particular, ethical decisions by categorizing targets. In ordinary life, we usually categorize targets not by the necessary and sufficient conditions, as we do in the sciences, but rather by reference to prototypes, thus (roughly) standard or average cases. ² In addition, the targets here are not static objects, but sequences of actions by persons with emotions and goals, actions which take place in particular circumstances. Thus, the prototypes at issue cannot be prototypes of things; they are, rather, prototypical sequences of actions and events, which is to say stories. The most prominent story prototypes would seem to be those that define the cross-culturally recurring genres, including heroic, romantic, sacrificial, family separation and reunion, seduction, revenge, and criminal investigation. Thus (again, I am summarizing the conclusions of Literature and Moral Feeling), our ethical interpretation of and response to events and circumstances of life are bound up with narrative prototypes, particularly those that are cross-culturally prominent. Of course, this is not a unidirectional process. On the one hand, our thoughts and feelings about a situation activate one or more story prototypes. On the other hand, the activated prototypes lead us to construe, infer, respond emotionally, and act in certain ways.

Just to get a basic sense of what I have in mind here, you might consider my personal inclination to focus on romantic emplotment. I am far more interested in romantic stories than those of any other genre. Moreover, I tend to care more about the separated lovers than about characters of other kinds—though the lovers are rivalled in my affection by abandoned children; the basic idea here is that I am most sensitive to attachment insecurity and attachment loss, which is highlighted in both genres. For this reason, I tend to be deeply opposed to anything that separates lovers. This is presumably the reason why, as a high school student in the 1970s, I had rather banal, mainstream political views, basically taken over from my parents, who were culturally quite conservative. The one main exception to this was my fervent belief that gay or lesbian couples should be allowed to marry—this was at a time when gay marriage was hardly even imaginable for anyone I knew. Moreover, when I thought of other issues, such as racism, they commonly appeared to me in the form of romantic love (e.g., love forbidden by anti-miscegenation laws, in the case of racism). The connections extend to other types of ethical issues as well. For example, romantic stories tend to favor individual choice over the dictates of parents or other social authorities. This is directly contrary to the deference to social authority adjured by the heroic prototype. Here, unsurprisingly, I am on the libertarian side.

In the book, I go on to argue that the various “moral foundations” identified by Haidt (2012) may be understood as abstractions from the ethical ideals implied by the cross-cultural story prototypes and their associated emotions, which define the relevant goals pursued by the characters in those stories and thereby define the genre prototypes. For example, the “moral foundation” of care may be viewed as a particular (though by no means exclusive) ideal in the family separation and reunion story with its sorrow over uncared-for children or children with ruptured attachment bonds.

There was, however, one “moral foundation” that I did not include. That was fairness.

The Trouble About Fairness

So, why did I leave out fairness? Clearly, it is an ethical concern and, as such, a topic that elicits emotions and inspires action. Indeed, when we appeal to ethical principles, fairness—or, rather, unfairness—may be the characteristic property that we are most likely to invoke when damning some act or practice. Moreover, there is an obvious way in which fairness at least appears to be a central ethical concern of the criminal investigation genre in particular. In contrast with the vigilantism of revenge, with its surreptitious scheming and attendant errors that frequently multiply the initial harm, the criminal investigation seems to be a prime case of fairness. That is true. But it is also true that the centrality of fairness to this genre results in part from the very nature of fairness, which enters in some manner into almost any ethical assessment. In a sense, the usurpation of the rightful ruler (in the heroic plot) is unfair to both the ruler and the people. The parental rejection of the hero's beloved (in romantic narratives) due to her social class or race or lineage can easily be framed as a matter of fairness—and so on.

For these and other reasons, I came to view fairness as an ethical meta-principle that bears on each prototype. Or, better, our (idiolectal) ethical norms may be divided into two broad, psychological systems, defined by structures and processes that interact with one another, not always consistently. The first system, which I will call “morals,” is the system I considered in Literature and Moral Feeling. It comprises our personal sensitivities to the feelings of other people as well as our routines for exerting effort to extend our empathic response—both our cognitive understanding (or Theory of Mind processes) and our affective alignment (or parallel interpersonal stance)—when empathy is not spontaneous. This is, fundamentally, a first-person system, a system that treats moral norms from one's own perspective, even insofar as it adjures us to simulate the perspectives of other individuals.

The second system, which I did not explore in that book, is the one that treats fairness. This system addresses, not our individual attitudes toward particular moral decisions, but rather communal principles for cooperation in social practices bearing on or issuing from moral decisions. To understand what this means and how it might operate, we need to step back for a moment and consider a more general feature of mental architecture.

One way in which mental systems may develop involves a sort of perspectival duality, with one system treating a given topic from the subjective or first-person perspective and another treating it, so to speak, objectively or from the third-person perspective. The division is familiar to most people from literary narrative. But this is not a mere projection from narrative analysis. At the very least, it is clear that we have two systems of this sort treating space (see Clark et al., 2005, p. 43). The hippocampal system treats space objectively, determining, for example, that the upstairs porch is about twice the width of the downstairs porch or that the distance between the restaurant and the theater is about five blocks. The parietal system, in contrast, treats space from one's individual, bodily perspective. It tells one, for example, that the pitcher is or is not close enough to reach without standing up, or that the man on the bus has or has not crossed into one's personal space.

I am hypothesizing that there is a similar division between two ethical systems—the system treating fairness and that treating empathic norms, a difference I will refer to by the names justice and morality. Note that I am not speaking here about what is most widely accepted socially or what has been codified and institutionalized. For example, in the case of “justice,” I am not referring to law or the legal system. Like the first-person (moral) system, the third-person (fairness) system is psychological or idiolectal. Indeed, that is why we can feel, for example, that our legal system (of justice) is unfair (or unjust). Both systems—those of justice and morality—may be understood as involving a basic definition (which tells us what systems we are dealing with) along with a set of principles and parameters that generate the fundamental orientations of our idiolectal moralities and senses of justice. In Literature and Moral Feeling, I have discussed these elements in the system treating morality, but not in the system treating justice.

Before continuing, I should forestall one common misunderstanding of my claims here. I am clearly inspired by Chomsky's “Principles and Parameters” account of language (for an excellent overview of the theory, see Baker, 2001). However, the inspiration here does not go beyond the definition of basic types of ethical systems by parameter settings for shared principles. What is most important, I do not believe the principles or the parameters are innate. (It is possible, but seems to me unlikely.) Rather, the principles simply formalize, for purposes of theoretical clarity, key issues one needs to resolve—usually not self-consciously—before one can even begin to have moral preferences. As to the parameters, my inclination is to think that these are for the most part universal, but that this universality arises separately as different groups of people (with innate emotional propensities, and so on) implicitly or sometimes explicitly address these issues. For example, the scope parameter merely states the logical point that, once one is not guided by one's own hedonic preferences, but by empathic responses, there are different ways that such empathic responses could themselves be ordered. As to the specific values of this parameter, we have the capacity to respond empathically to a wide range of human and non-human targets. Moreover, perhaps the most consequential factors limiting or intensifying the degree of one's empathy are (a) our attachment relation to the target and (b) our in-group or out-group categorization of the target (e.g., sharing or not sharing ethnicity). This would seem to be enough to yield the three main values for this parameter. Empathic sensitivity to, for example, facial expression or vocalization, attachment bonding, and social identity-defining categorization are presumably at least in part innate. But there is no reason to think that there is, in addition, a distinct and also innate program for setting the moral scope parameter. ³

So, what is the basic functional role of fairness or justice? It is the social resolution to the (social) problem of allocation. “Allocation” is vague here because it refers to the apportioning of anything that a society wishes or needs to divide among individual group members. What is distributed may be broadly divided into positively- and negatively-valued objects, experiences, or possibilities for action. Specifically, a society must have procedures for allotting such goods as wealth, property, or authority and such burdens as labor or punishment. Allocation, then, combines distributive and retributive justice. In parallel with the valence parameter ⁴ in one's (idiolectal) morality, any individual might stress the unfair allocation of either positive or negative obligations, experiences, and so on, though this division does not appear to be so pronounced as in the case of morality. For example, I suspect that people are largely in agreement that the unfair distribution of criminal punishment (i.e., disproportionate retribution) is more distressing and objectionable than the unfair distribution of wealth, at least until the distribution of wealth becomes more like punishment for some parties, as in starvation. I suspect this because of hedonic asymmetry, the degree to which we habituate to positive experiences being much greater than that to which we habituate to negative experiences (Frijda, 1986, p. 323).

The most significant issues with regard to fairness are, rather, parallel to the scope parameter in ethics. I take it that all societies treat the problems of fairness by appealing in part to merit (given some dominant view of what constitutes merit), in part to a set of conditions that (so to speak) relieve some individuals of merit-based restrictions, and in part to egalitarian principles. As Pascal Boyer explains, as a first approximation, people believe that benefits should match contributions (or, alternatively, effort). In keeping with this, “experiments show that even three-year-olds have the intuition that rewards should be proportional to contributions, in places as different as Japanese cities and the camps of Turkana nomads in Kenya” (2018, p. 175). Generalizing to negative cases as well, there appears to be a common, human tendency to recognize a homology between the contribution of an individual to a (beneficial or harmful) outcome and the degree to which that individual merits reward or punishment for the outcome. For example, suppose ten people are sent out to collect tubers for the village and Nell collects as many tubers as all the others put together. Most people would probably acknowledge the potential relevance of this fact to the social distribution of the tubers. (This is an instance of our wariness of “free riding” or cheating; though I do not necessarily agree with the explanation of this phenomenon by evolutionary psychologists, it seems clear that this wariness is not some peculiarity of a few cultures [see Cosmides & Tooby, 1992].)

And yet, most people would presumably reject the idea that Nell should get half of the tubers. The reasons are obvious enough. Some of Nell's relative success was mere good luck on her part or bad luck on the part of others (e.g., the result of happening upon spots where there were many or few of the items being sought). Moreover, Nell owes something to villagers who engaged in other works (e.g., fishing) from which she will benefit. Finally, some of Nell's superiority may have been due to others suffering disabilities produced by age, health, etc. In being part of a community, we in effect agree to give up advantages at some times to gain security at other times. (This too is the sort of principle that I believe could easily develop as a universal without being somehow innate as such.) For example, young children with little experience or elderly people with weakened eyesight and diminished strength shouldn’t be left to starve because they couldn’t gather as much as someone in the prime of life.

The two main norms here—stressing the individual's specific actions, on the one hand, or the community's purposes and functions, on the other—clearly have been widely accepted. Even so, they are weighted very differently by different individuals (and, derivatively, by different societies). Indeed, even among those who favor communal purposes, the mitigating factors—mere good (or bad) fortune, the communal division of labor along with the associated implication of reciprocity, and the (usually tacit) commitment to communal security for the disabled—are themselves valued quite differently by different people. For example, my inclination is to stress the last (commitment to communal security for the disabled) almost to the exclusion of all the others. By “inclination” here, I mean my spontaneous emotional response, which is something along the lines of empathy with (simulated) persons suffering deprivation due to disability. Indeed, I implicitly absolutize this qualification, seeing everyone as disabled (by age, health, injury, discrimination, and so on), though in different degrees. At the same time, I recognize that the virtual elimination of personal responsibility would almost certainly foster “free riding” and other problems. Still, my vision of justice is fundamentally one of extreme egalitarianism. I start with the principle that everyone is equally deserving of whatever society produces, then (were I, for example, to write a Utopia) I must formulate and justify deviations from that absolute equality.

In keeping with the example of Nell and her tubers, I have been speaking largely of rewards. The same basic points apply to punishment. If Nell steals tubers from others, then most of us are inclined to think that she should receive whatever punishment is to be doled out for that crime. The most obvious way of calculating that punishment is by equivalence, as in “an eye for an eye.” That punishment may be thought to be a matter of (future-oriented) deterrence or of (past-oriented) compensation for a misdeed that has already been committed, or of some combination. ⁵ However, here too we recognize mitigating factors. In principle, these allow the range of orientations for criminal justice to extend all the way from an absolutism of individual guilt to an absolutism of collective responsibility.

The key mitigating factors for individual guilt are parallel to those for individual merit. First, there is moral luck (see chapter two of Williams, 1981 and chapter three of Nagel, 1979). For example, in Ngũgĩ wa Thiong’o's A Grain of Wheat (Ngugi, 2012/1967), Gikonyo intends to murder Mumbi, but trips, hits his head, and is knocked unconscious. As a result, he does not commit murder, though this is not due to any virtue or free (moral) choice on his part. It is just good luck—good moral luck in particular because it is not fortunate otherwise (e.g., physiologically). In contrast, Mugo has no desire to kill anyone. But he is put in a situation where he is in danger of being killed by the British colonial government if he does not betray—and thereby cause the death—of the revolutionary leader, Kihika. Though he chooses to betray Kihika, the extremity of his situation results simply from bad luck.

Second, parallel to the goods one receives through the division of labor, it is clear that we all are subjected to some losses, inhibitions, or disabilities by society, and that some are subjected to these in much greater degrees than others. For example, the murder of (European) settlers by the Kenyan revolutionaries seems to me clearly immoral. Admittedly, depending on where one sets the scope parameter and on other factors, not everyone will agree. But, even in judging it immoral, I have to recognize that it is intertwined with a series of actions by others, including the settlers who have occupied land that they had no right to, and the government, with its decision that the (European) settlers not only owned the land but had the right to evict Kenyan farmers whose forebears may have been working that land since before the development of British imperial control and European settlement (see Anderson, 2005, p. 388).

Finally, just as disability limits one's capacity to act in particular ways, a history of interaction with social institutions and practices shapes one's imagination of possibilities and develops one's skills (or fails to develop them). At the extreme, one may decide that these and other experiences combine with genetic factors to guide one's actions in all cases, thereby eliminating free will along with moral blame (or praise). But even if one does not adopt this deterministic view, one might very well view some sorts of action as more difficult for individuals for reasons that are not entirely dissimilar to being disabled.

Ontological Inequality

But there is a further twist to the issue of fairness. That is, there are rewards and punishments that do not bear on either individual contribution/responsibility or on any of the mitigating factors that surround such contributions as gathering food or such responsibilities as committing robbery. Indeed, they are not tied to actions or events at all, but are a consequence of social structure. The parameter governing this orientation establishes degrees of ethical worthiness before any particular, individual actions. To draw an analogy, I sometimes find myself in the absurd position of defending the difference between truth and falsehood to skeptical humanists. They commonly believe that they can utterly crush my naïve claims (e.g., that there were hundreds of thousands of civilian deaths caused by U.S. bombing in Vietnam or there were not). They believe they can do this with the simple question, “Who decides?” I (unsuccessfully) point out to them that whether a fact is or is not accepted by the general public is certainly affected by who decides what we should accept as true (or reject as false). But no one makes a decision that determines truth or falsity itself. Indeed, it is precisely because of the independent truth or falsity of a given claim that we can dispute official misrepresentations. The fairness parameter I am speaking of here is, in effect, a response to the “Who decides?” question, now applied to social or third-person norms. It concerns the individuals whose voices sound alone or at least more loudly in debates over fairness, individuals whose preferences have greater authority and are more consequential, with regard to distributive and retributive justice. In other words, it concerns the relative ethical hegemony of individuals and groups (with the groups typically being affiliative networks, such as families, or identity-defined in-groups, such as socio-economic classes or ethnicities). This most basic parameter in fairness addresses the issue of whether some groups or individuals have more authority than others in dictating social attitudes and practices regarding fairness and, if so, what groups or individuals those might be.

We may refer to this simply as the hierarchy parameter. As any hierarchy of this sort is before and definitive of fairness, it cannot be justified on the basis of fairness. I will therefore refer to it as “ontological,” as a matter of what or who the individuals or groups at issue are, not what they have done. Put differently, the merit (or demerit) that determines their allotments of rewards or punishments is intrinsic, rather than occasional. This is in part parallel to a possible moral parameter that I introduced in Literature and Moral Feeling, but did not develop there. That is the target parameter, which isolates the primary target of one's moral evaluation. Specifically, that parameter determines whether the target is a particular act or (the character of) the person who commits the act. For example, one might not really find a specific act—such as masturbation with a dead chicken, in Haidt's example (2012, pp. 111–112)—immoral in itself, but one may find a person who engages in such an act to be so disgusting as to be untrustworthy, including in the area of morality. In that case, in claiming that the action is immoral, one might more accurately be said to be claiming that one takes the action as evidence that the person is immoral.

It is also important to note that the central emotional concern in the case of the hierarchy parameter is still a function of interpersonal stance (that is, one's overall response to the emotions of the other person). However, in this case, the stance bears not so much on empathy as on trust. Theories of trust most often treat it as if it were a matter of (cognitive) information processing only. Specifically, trust is widely seen as involving two components (see, for example, Hardin, 2006). One is the belief that the trustee has good will and thus will not be deceptive. The other component is the belief that the trustee is competent and will not misjudge the situation. For example, to say I trust my podiatrist is to say that I believe he will tell me what he thinks is best to heal my broken metatarsal bone and that I believe his thoughts on this topic are the result of genuine knowledge. This all makes sense as the justification for trust or distrust—that is, the (veridical) evaluation of a person's trustworthiness—is a matter of such (cognitive) information processing. Such justification is also what the evolved mechanisms of trust must approximate if trust is to have adaptive value and thus to produce a selective advantage.

However, it seems clear that, on the whole, our trust in another person goes from emotion to (cognitive) appraisal—as Haidt argues generally occurs with ethical evaluation (2012)—not the other way around. In other words, I do not assess my podiatrist's good will and competence, and then come to feel trust. Rather, I feel trust and therefore assume that he has good will and competence. Put in more straightforwardly affective terms, I feel safe with regard to the other person, concluding in consequence that he or she has good will. I also feel respect or admiration for him or her, thus concluding in consequence that he or she is competent. I should note also that we could understand the assumption of good will as a sort of “benevolence attribution bias.” Since out-grouping typically entails a hostility attribution bias (Lukianoff & Haidt, 2018, p. 158), trust is therefore likely to vary with identity categorization, much as does empathy. This is almost certain to be enhanced by our tendency to find members of out-groups less competent than members of in-groups.

Return to Paradise

If what I have just written is at all correct, it may seem that taking up the problem of Paradise Lost is at best anticlimactic. If this is a plausible (descriptive and explanatory) account of the psychology of fairness, then surely that is far more valuable for what it says about the human mind and human society than for what it says about common reactions to a Seventeenth-century poem. However, I want to return to the poem for two reasons. First, the poem itself is important enough in the history of poetry, and the problem with the poem is obtrusive enough, that I believe the literary issue is worth exploring in its own right. Second, and more significantly, I believe that exploring the particulars of the literary issue will help us to understand the psychology of fairness, and thus ethical feeling, in greater particularity and detail, and to further extend the analysis of ethical feeling in general as well.

In some ways, Milton's case in Paradise Lost is very simple. God is, almost definitionally, the ideal trustee. He is absolutely beneficent and perfect in knowledge. But, if that is the case, shouldn’t it be easy to “justify the ways of God” (1.26)? Shouldn’t any reasonable portrait make Him likable? The problem here is that this all tacitly presupposes that we can evaluate God by reference to accessible canons of trustworthiness and fairness. But God cannot be evaluated ethically, either by reference to individual morality or social justice. To “justify” God would seem to require an appeal to some distinct, higher, governing principle, which is in turn known to the (human) judge. In Paradise Lost, Milton tacitly accepts an ontological hierarchy; indeed, he absolutizes such a hierarchy. That is why the transgression seems so trivial and the punishment so monstrous. To assert the hierarchy as ontological, as itself justice rather than being some result of justice, it is crucial that Milton not give the act any properties that readers would judge to be ethically objectionable independent of God's will. If it had any such properties, it would appear to readers that it was these properties that made the act wrong. It is only by making the act—eating a piece of fruit—apparently entirely non-ethical, making its proscription seemingly arbitrary, that the entire ethical quality of the act can operate for the reader as a function of God's moral supremacy. As far as humans (or even angels) are concerned, given their severe limitations in knowledge, there is in practice only one ethical principle here—Do what God orders. In the final book of the poem, Adam draws the ethical conclusion explicitly. “Henceforth I learn, that to obey is best” (12.561), he explains. ⁶

But here another problem arises. Whether it is due to dispositional (e.g., personality) or situational factors, the parameter valuing or rejecting hierarchy is often generalized (see Caprara & Vecchione [2013, pp. 37–39] on theories of authoritarianism and Social Dominance Orientation treating this issue). Someone who affirms an ontological hierarchy in one case, judged relative to someone who affirms equality in that case, appears more likely to see embedded hierarchies at lower levels as well. In and of itself, there is no problem with this, so long as the hierarchies are adequately and consistently hierarchized themselves. Milton presents us with just such an authoritarian generalization, as God Himself apparently establishes the various subordinate hierarchies. On the other hand, when spelled out, it no longer seems entirely clear that Milton's primary aim was to justify the divine hierarchy. Rather, he seems to particularly stress another hierarchy which the divine hierarchy underwrites.

Specifically, Milton claims three subordinate, ontological hierarchies defined by God. They are (a) Human superiority over (non-human) animals. Thus, he explains that God gave humanity “dominion …/Over all other creatures that possess/Earth, air, and sea.” (IV, 430–432). Subsequently, he repeats the idea, writing that humanity holds “dominion … /Over fish of the sea, and fowl of the air,/And every living thing that moves on the earth” (VII, 532–534). (b) Adult or, more exactly, parental superiority over children. Thus, he asks, “what if thy son/Prove disobedient, and reproved, retort,/Wherefore didst thou beget me? I sought it not:/Wouldst thou admit for his contempt of thee/That proud excuse? yet him not thy election,/But natural necessity begot./God made thee of choice his own, and of his own/To serve him; thy reward was of his grace;/Thy punishment then justly is at his will” (X, 760–768). (c) Male superiority over females.

The third point is the most fully elaborated in the poem, supported by statements and, even more, by the implications of the characters’ actions. The obvious action is, of course, the disobedience itself, for it is committed first by Eve, and repeated by Adam only out of (in Milton's view, misguided) devotion to her. Indeed, given the precise way that Milton has developed the story, what he has stressed or elaborated and what he has minimized or left aside, I believe that a quite plausible case could be made that this third point is actually the motive that drives the other three hierarchizations. Even the theologically self-evident claim of God's (purely self-willed) absolute goodness does not appear to drive Milton's fervid imagination in the way it is driven by the thought of Eve's sin. Without ontological superiority, Eve does not appear to be that different from God. God knows that Eve will succumb to the serpent's guile and that Adam will fall when faced with the possible loss of Eve. Thus, by forbidding the fruit of the tree, by allowing Satan into the Garden, and by other acts, God arguably contributed to the fall of humanity's parents at least as much as Eve did in choosing to undertake her gardening tasks alone, giving ear to the serpent, and so on. The difference is due only to the ontological authority of God.

Thus, in book IV, Milton writes that “true authority” lies “in men” (IV.295), as the sexes are “Not equal” (IV.296). Adam, and others of his “sex” were “For contemplation … and valour formed” (IV.297), while Eve and descendants of her kind, “For softness … and sweet attractive grace” (IV.298)—thus intellect or soul and mere mortal body respectively. The culmination of the claim actually sets up a homology between Adam and God, almost making explicit the function of divine hierarchy as a justification for male “Absolute rule” (IV. 301) and female “Subjection” (IV. 308): “He for God only, she for God in him” (IV. 299). The Fall is subsequently traced to over-valuing “worth in women” and allowing a woman to “rule” (IX. 1183, IX. 1184). The Son goes so far as to mock Adam with the question, “Was she thy God” (X. 145). He then chides Adam for “resign[ing] thy manhood,” for that manhood was made by God precisely to “set thee [Adam] above her” (X. 149). This male “perfection far excelled” (X. 150) the qualities of women, making it “Unseemly” for the latter “to bear rule” (X. 155). If this is so, it gives us another possible explanation for the lackluster performance of God, despite His leading role—the purpose of the poem was never to justify Him to begin with (though, like his Satanic partisanship, this may have been unknown to our poet).

This suggests a solution to another quandary in Milton’s scholarship as well, also related to the unappealing quality of God. That is, the seemingly clear parallel between the defeated forces of Satan and the defeated forces of the English revolution. For example, Christopher Hill has pointed out that, “the magnificent Satan of the early books of the epic does convey some of the defiance that Milton himself must have felt tempted to hurl in the face of omnipotence as the republic crashed about his ears” (1979, p. 368). The problem here is not so much that, in Paradise Lost, the revolutionaries are sympathetic; that is explained by Milton's commitment to the revolution. The problem is, rather, that the angelic–demonic insurrectionists are, the poem indicates, ethically wrong; conversely, God (who is likely to strike us as parallel to the English monarch executed by the revolutionaries) is right. That would appear to suggest a repudiation of the English revolution. There is, of course, a straightforward solution to this problem, as Lewis pointed out (1942, p. 73). By supporting one set of revolutionaries against one monarch, Milton hardly commits himself to supporting all revolutionaries against all monarchs. But this does not really solve the emotional problem, for (as just noted) the poem displays some sympathy for the infernals, but little if any for God. ⁷ Moreover, some strands of the English revolution were profoundly anti-hierarchical, which seems to be broadly at odds with the hierarchical character of Milton's vision of justice, as suggested in the poem.

Milton does offer a rather perfunctory explanation for this apparent contradiction (phrased differently, of course). First, he explains his antipathy for monarchs, writing (in Adam's voice) that for any “son” of Adam “to aspire/Above his brethren” and “assum[e] Authority” is “execrable” (XII, 64–65). This is because God “gave us only over beast, fish, fowl/Dominion absolute; that right we hold/By his donation; but man over men/He made not lord; such title to himself/Reserving” (XII, 67–71). Then, he accounts for political and religious hierarchies, stating that man “permits/Within himself unworthy powers to reign/Over free reason” (XII, 90–92). In consequence, “God in judgement just/Subjects him from without to violent lords;/Who oft as undeservedly enthrall/His outward freedom: tyranny must be” (XII, 92–95).

One interesting point about this (rather opaque and indirect) justification of false hierarchies on earth, ranging from monarchy to the papacy, is that this exclusion of a hierarchy for people is clearly confined to men—no one who is a “son” of Adam should aspire “Above his brethren,” violating the proscription against placing “man over men.” This seems to suggest once again that, motivationally, Milton's fundamental orientation on justice was not so much toward some form of egalitarianism, as found in certain revolutionary groups with whom one might wish to connect him. It was, rather, toward whatever configuration of parameter settings would most fully support the hierarchy of patriarchal authority. This in turn suggests a more general conclusion—that, at least in certain cases and to some degree, one's views on fairness may not be the product of ethical thought or even ethical feeling as such. Rather, they may result from one's political preferences and sectional self-interests. In other words, the settings of certain parameters may serve to rationalize motives and actions that might otherwise trouble us, motives and actions that we might intuitively feel to be unfair. This would be in keeping with Steven Pinker's view that, we have “too much morality” (2011, p. 622), too much because it seems that ethical resolutions—especially as they become emotionally consequential—are as likely to excuse (and thus foster) brutality as the reverse.

Conclusion

Literature and Moral Feeling analyzed one large set of ethical concerns, but left aside questions of fairness or justice. The present essay takes up the latter. It argues, first, that we may distinguish two cognitive-affective systems that bear on ethical response. They are systems in the sense that they incorporate different cognitive-affective structures and processes. However, I do not, for the most part, take them to be either “modular” or otherwise innately distinguished. Rather, at each level of explanation, I view them by and large as the product of convergent social developments, which result from groups of people encountering the same sorts of social problems and responding to those problems with the same sort of cognitive and affective capacities and inclinations. There is, however, one important, differentiating factor that is probably innate—that is the difference between first- and third-person point of view. The ethical concerns treated in Literature and Moral Feeling, referred to here as morality, address the former point of view, treating one's personal obligation to act in certain ways, as well as stressing empathy, both being first-person concerns. Fairness or justice addresses the latter point of view, treating the social allocation of reward and punishment.

Both ethical systems may be understood as sets of cognitive-affective processes that fall into a limited range of logical possibilities (sometimes with simplifications, such as the reduction of a continuous scale to a smaller set of alternatives, such as “low/medium/high,” depending on the purposes of one's analysis). These possibilities are well-captured by a few broad principles with specifiable, parametric alternatives. (The principles and parameters appear to be largely universal, but that is probably for reasons of logic and social function; it does not seem necessary to posit innateness here.) The crucial parameters in the case of justice may include valence (whether we view punishment or reward as the primary concern) and method or alternatively ethical function (e.g., in the case of retributive justice, this would address the issue of just what purpose one wishes punishment to serve). In the preceding pages, I concentrated on two parameters. One governs the scope of allocation, specifically the degree to which one's idiolectal principles of merit-based justice privilege a small set of individuals with whom one is affiliated; favor some larger, identity-based group (the majority of whom may be strangers); or define no preference by affiliation or identity.

The second parameter concerns the target of justice. Though closely related to scope, it is valuable to distinguish the two. For the most part, treatments of justice focus on acts, the contribution of individuals or groups to acts, and modifications of reward or punishment by considerations of luck, disability, and other factors. Thus, the scope parameter may be limited in application to windfalls (where there cannot be a calculation of individual contributions because no one contributed) or may modulate allocations based on contribution. In contrast, the target parameter first of all tells us whether any given individual's or group's actions may or may not be subjected to evaluation in terms of justice at all. If there are individuals or groups that are “above the law” in this sense, then we have an ontological hierarchy, ontological because it is not a matter of what the person or persons do, but of what they are. As this tends to go along with further sub-hierarchies, we may refer to this as a hierarchy parameter, rather than a target parameter.

I was led to recognize the hierarchy parameter by considering some notorious problems with Milton's Paradise Lost, such as the unsympathetic nature of God relative to Satan, with which the essay begins. I return to these problems at the end, arguing that some of them are solved by recognizing the nature of an ontological hierarchy in ethics. Indeed, Milton's treatment of God is entirely reasonable, even exemplary, given Milton's views. I conclude by noting that this account of justice does not, by itself, solve all of such problems. Indeed, further examination of Milton's poem suggests consequential complications in our account of the psychology of justice. Specifically, we appear to find Milton using ontological hierarchy to rationalize another form of hierarchy that he might otherwise be hard put to defend, but which was arguably his primary concern in writing Paradise Lost. Presumably, Milton was far from being alone in this.