Part III - Rights and Moral Cognition

6 Which Kind of Mind, Which Kind of Morals, Which Kind of Rights?

Was ist der Mensch? konnt’ ich beginnen; wie kommt es, daß so etwas in der Welt ist, das, wie ein Chaos, gärt, oder modert, wie ein fauler Baum, und nie zu einer Reife gedeiht? Wie duldet diesen Heerling die Natur bei ihren süßen Trauben?¹

Friedrich Hölderlin, Hyperion oder der Eremit in Griechenland

6.1 Ethics and the Theory of the Human Mind

The content of justice and the good forms one prime focus of human reflection about practical matters. Another question that has occupied practical philosophy ever since its beginnings concerns the mental means by which human beings achieve moral cognition. The answers to this question are manifold, mirroring the changing conceptions of what human beings are like. One constant of this debate is the notion that human beings possess a particular faculty through which they are able to gain insight into true normative propositions. One famous example of this is Socrates’ daimonion, his inner voice that guided him, taking his decisions about what it was right and wrong to do.² The Stoics reflected on a human faculty of moral understanding – an idea that influenced the natural law tradition they so profoundly inspired in other regards, too.³ In Christianity, Paul’s idea of a law written in the human heart already was very influential in Patristic thought: Moral precepts seemed somehow ingrained in human nature.⁴ In later natural law theory, the question persisted. Thomas Aquinas, for instance, believed that synderesis was the means of understanding natural law.⁵ Others assumed a natural light or a recta ratione.⁶ In modern times, the thesis of innate ideas gained prominence, not only concerning the foundations of mathematics, scientific insight and logic, but also with regard to practical questions. Descartes did not say much about ethics or natural law, though there are some interesting remarks.⁷ Leibniz, however, developed quite a differentiated theory of moral judgment in which innate moral ideas played an important part.⁸

One influential idea advanced by the Scottish Enlightenment was that human beings have a moral sense that allows them to perceive the moral status of things.⁹ Hume argued that morality is part of humans’ mode of thinking because of “some internal sense or feeling which nature made universal in the whole species.”¹⁰ Kant argued with practical reason. In his view, the moral law is a “fact of reason,” a given property of human understanding.¹¹ Other psychological assumptions supplement this view: For Kant, one central idea is “respect for the law,” the ultimate reason why moral precepts matter.¹² According to this doctrine, the human psyche is structured in such a way that the cognition of the moral law gives rise to respect for this law. This respect for the moral law is the reason why morality matters in practical terms and influences both human motivation and ultimately – if other inclinations do not gain the upper hand – human action.¹³

The same cognitive interest motivated authors such as Locke who considered determining the nuts and bolts of the machinery of human thought a central task, even though they were critical of psychological theories that assumed the existence of innate ideas.

In contemporary thought, moral psychology continues to play a role in influential theories of morality. Habermas, for instance, took Piaget’s and Kohlberg’s theory of moral development as the basis for his understanding of the ontogenesis of moral thought.¹⁴ A further example is Rawls, who considered the idea of a parallel between Chomsky’s concept of universal grammar and the sense of justice, although he did not develop this systematically after encountering substantial critique.¹⁵

This rich tradition of thought about the characteristics and the functionings of the mind and moral understanding comes as no surprise. The question of how human beings acquire moral knowledge and which cognitive mechanisms are used to put this knowledge into practice suggests itself too evidently to have been missed in the past.

Throughout much of this tradition, moral psychology was understood as equally important for the understanding and creation of law as for morality. The faculty of understanding that enabled human beings to cognize natural law or the law of reason was seen either as dealing with norms regarded as directly valid law or as constituting the foundation for the determination of the legitimate content of the law. The inclusion of moral psychology in theoretical reflections about the law thus also comes as no surprise. The discussion that follows consequently represents yet another small piece of reflection on a topic embedded in a rich history of creative thought. Despite its brevity, it hopefully will shed light on some problems of human moral understanding that hitherto have not been fully illuminated.

Some theories assume that human beings have a general learning ability that both allows them and at the same time makes it necessary to learn everything anew, including moral concepts. Some of these theories understand humans as protean beings who are infinitely malleable and whose mental makeup is determined by their social surroundings, as we have seen. It should be noted again that all such theories, despite their constructivist bent, formulate a substantial, empirical psychological hypothesis including assumptions about human nature, assumptions that are true or not, depending on evidence: Their thesis is that the human mind is structured in such a way that it can and indeed must learn everything anew. In this sense, consequently, even radically culturalist or constructivist theories are naturalistic or nativist because they presuppose an empirical thesis about the cognitive foundations of the development of morality in human beings. There is no escape from psychology and theories of the structure of human cognition.

The view that there are species-specific properties of the human mind other than general learning abilities that constitute the preconditions of the development of morality has gained considerable influence, although perspectives often differ radically about what this means in detail and whether this is good news for an egalitarian ethics of human justice and solidarity and for the political and cultural project of human rights in particular, or whether it rather calls these projects into question.

It is easy to discern the reason for the view that such species-specific properties of human beings exist. Everybody agrees that human beings acquire a vast number of norms as they grow up, through their upbringing, education and socialization. These norms are products of their culture. At the same time, it is not really plausible that the complex cognitive capacities forming the foundation of morality are not species-specific: The cultural influences are obvious, “[b]ut the household pets growing up in the same cultural and religious contexts do not thereby become moral beings.”¹⁶ Therefore, it seems quite reasonable to assert that “there is absolutely no question that human children are biologically prepared for the process as well.”¹⁷

The inquiry into the cognitive preconditions of morality is important, regardless of whether we believe that any findings in this area have normative consequences in addition to explaining some of the cognitive mechanisms involved in moral understanding, or whether we believe that this would mean committing a naturalistic fallacy or neglecting the is/ought dichotomy.¹⁸ There are various reasons for this importance. First, the question of whether facts of moral psychology have normative consequences or not must be informed by what these facts actually are, and so we need to inquire into these facts. Second, the comprehension of moral cognition and its relation to the law is itself a valuable, even indispensable scientific goal if one wants to gain a profound understanding of the nature of morality and law. We clarified this in the Introduction of the present study. Third, the empirical theories of moral cognition very often do have a direct or indirect normative edge, because they influence the way human beings think about the mechanism that directs their volitions and actions. Just because there are good reasons to take the is/ought dichotomy seriously (as will be argued here) does not mean that everybody in fact agrees with or, even if they avowedly agree, adheres to this stance. Misconceived theories may have significant practical impact. What has been called “evoconservatism” – the justification of reactionary political visions by means of evolutionary theory – is just one such example we will discuss below.¹⁹ Finally, we have to answer the question of the relation between what seems right in normative theory and the facts about human moral cognition. To put it in concrete terms: Do human rights demand the impossible of human beings, as some influential voices argue? Thus, as things stand, the critical assessment of theories of human morality and the law by psychology and neuroscience is of great significance. A lack of interest in the cognitive preconditions of human moral understanding is not something moral and legal theorists can afford.

As indicated above, these questions are not of minor concern. As Hannah Arendt correctly observed, a core problem brought to the fore by the horrors of the Holocaust is the question of whether human beings have a faculty of moral understanding that enables them to distinguish between right and wrong, even if this means setting themselves in radical opposition to their environment.²⁰ The assumption that everybody was able to know that Auschwitz was a crime, irrespective of the drums of Nazi propaganda, presupposes at least two things: firstly, that there are standards of right and wrong; and secondly, that people have the ability to understand them. The first question already occupied us in our discussion of the justification of human rights, with encouraging results. The second must be informed by what we know about human moral cognition, not least because some theories maintain that moral cognition does not facilitate but rather obstructs doing the morally right thing.

6.2 The Epistemology of Moral Cognition

There are at least two acts of cognition relevant to this inquiry. First, there are concrete assertions of rights along the line of “X has a right to Y.” These can take the form of assertions in concrete circumstances relating to specific people, such as Creusa’s complaint about the violation of her rights by Apollo. They can also take the form of more abstract and general assertions – for example, about the rights of women to sexual self-determination.

Second, there are acts of cognition that concern the justification of such rights assertions. The cognition of the justification of a right is achieved by a series of mental acts, by a set of cognitive judgments. The final judgment has the propositional content “right X is justified” – for example, expressions such as “freedom of speech is worth defending.” This proposition in fact relies on a number of complex reasons. It implies, as explained above, assumptions about the usefulness of a right such as free speech in human communities (for the expression of humans’ personality, the pursuit of truth, the protection of democracy and the like) and – although these are not always made explicit – anthropological claims about the importance of freedom for human beings. In the case of legal rights, further complications arise from their nature as positive law – for instance, as to their justiciability.

One further element of this bundle of predications is of special concern for normative theory: the predication of the moral justification of a right. Complex systems of moral legal rights, many details of which are historically contingent, cannot be derived directly from basic moral intuitions. As we have seen, substantial steps are necessary to move from concrete moral judgments to the formulation of explicit moral human rights. Basic intuitions of justice or impermissible harm are one thing, the formulation and full justification of a norm in the form of an explicit moral human right, let alone in the technical form of a legal fundamental right and its regime of limitations, quite another.

This notwithstanding, making the moral case for human rights forms a necessary element of any justification of human rights.²¹ There are many considerations that speak for human rights, but these rights cannot be legitimate if they are not justified from a moral point of view. Nobody would have taken the UN General Assembly seriously if it had stated that adopting the Universal Declaration made good sense even though it was quite an unjust catalogue of rights. A human rights catalogue that does not claim to lay down a just and morally appropriate set of norms is not a proper human rights catalogue.

Legal human rights sometimes appear in very abstract form. More often than not, however, the respective norms are stated in more differentiated terms, including a regime of justified interferences. This is of crucial importance, because this regime of justified interferences determines the true content of the right, as we have seen. The final judgment “right X is justified” in this case thus encompasses intricate arguments not only about the prima facie scope of a right, but also about other values that justify restrictions of this right and the degree to which this is possible. In this context, principles of proportionality have become a core element of modern human rights catalogues,²² sometimes buttressed by the protection of the essence of a right.²³ The determination of a human right’s concrete content thus involves the recursive application of normative principles to limit the prima facie scope of the right and determine the meaning of the clauses of limitations, not least by weighing and balancing different rights. As explained above, similar considerations apply for moral rights.

The predication of the normative justification of a right in this sense by a moral judgment (as one element of a bundle of highly complex justificatory arguments about human rights) must be a principled mental act. Caprice cannot provide justification. Rather, the application of normative principles to the case evaluated – the human right under consideration – gives rise to the judgment that this right is morally justified (or not). But what are these principles? And how are these principles, used to justify norms such as human rights, justified themselves? These two major questions of practical philosophy lead us straight to the heart of the normative component of the theory of human rights. Accordingly, over the course of the history of ideas, many principles have been formulated that seek to pin down at least some core elements of morality, from Socrates’ belief that it is better to suffer injustice than to do injustice to Kant’s categorical imperative and beyond. Given what has been said above about the justification of rights, it seems plausible that principles of equality and thus of justice, duties of human care and respect for the worth of human beings play a crucial role in this respect.

It seems likely that these principles also guide moral judgments about concrete claims, the first category of cognitive acts mentioned above, and not only the normative justification of these intuitions following their transformation into an explicit concept of human rights. Such concrete claims are regularly based on intuitions about justice and the moral obligations of others in concrete circumstances that give rise to specific rights and duties. For example, Creusa’s rebellion against her abuse implies that what Apollo has done to her is unjust. These kinds of claims, too, originate in the justificatory principles of justice, benevolence and respect.

The problems of the content and justification of the normative principles guiding moral judgment in the context of concrete claims and the justification of rights lead to the core question of interest here. Is the content of these justificatory principles perhaps dependent on the structure of human thought? Are these principles possibly even in one way or another determined by the properties of the human mind? Is the nature and structure of human thought a factor that defines the results of reflection – in the moral domain for moral judgments? This is, as indicated above, the Cartesian, Lockean, Humean and Kantian question about the preconditions of the possibility of human moral cognition. Are there such Verstandesbegriffe, such “concepts of understanding” in the moral domain, to use a Kantian term? If so – what is their effect? Are the particular nature and structure of human thought crucial to objective, foundational moral insight? Or are they indispensable elements of moral cognition that, however, do not reveal “morality in itself” (to modify another Kantian idea), do not reveal the true, objective morality? Is there thus a parallel to theoretical cognition, in the same way that human thought, in Kant’s view, never grasps the nature of the “Ding an sich,” the thing in itself?²⁴ Or – a third possibility – do such structures of the mind exist and decisively influence human moral thought but create nothing but cognitive illusions, the moral equivalent to the Müller-Lyer visual illusion mentioned in the Introduction?

These questions lead us to another important challenge to human rights to be considered in this study. A particular line of cognitive psychology and neuroscience has formulated exactly the latter thesis. It holds that the nature and structure of the human mind is indeed decisive for human moral reasoning, but that the moral judgment thus determined yields not insight, but error. Accordingly, the idea of human rights is part of this erroneous reasoning, and thus is not qualified to decidedly influence human affairs. This challenge will be considered over the next sections of this chapter, before we turn our attention to the fertile research in behavioral law and economics and then to a related challenge from evolutionary psychology in Chapter 7. Finally, we will consider other potentially very fertile approaches to the relation between mind and rights in the last chapter of this book.

6.3 The Neuroscientific Attack on Human Rights: Human Rights and the Mental Gizmo Thesis

The thesis to be examined here accepts as its general framework the dual-process model of the mind. This model holds that there are two kinds of mental processes: “thinking fast” and “thinking slow,” to use the popularized terms.²⁵ Thinking fast means using heuristics, framing operations or biases to solve everyday problems. These mechanisms are hardwired into the human brain. Humans cannot but use them and do so intuitively and unconsciously.²⁶ Thinking slow means using reflective rationality that abides by standards of logic.²⁷ Thinking fast works well for many aspects of everyday life but is skewed in important regards. Human decision-making is thus not fully rational. It is possible to become aware of the factors that skew human rationality, such as heuristics, framing effects and biases, and overcome their influence through slow thinking. However, this may not necessarily happen, because the control system of slow thinking may fail in its task.²⁸ This dual-process model of the mind has become highly influential beyond psychology, inspiring research in many other fields, not least behavioral law and economics. Its pioneers have also explored the place of moral reasoning within this model.²⁹ The mental gizmo thesis, however, aims to provide more substantial guidance. Most importantly for our inquiry, human rights are prominent topic of the discussion.

The mental gizmo thesis runs as follows: Moral cognition is part of the dual-process mind.³⁰ Deontological judgments are part of fast thinking. There is a mental “gizmo” that yields such judgments involuntarily, unconsciously, as a product of the fast, automatic and emotional operations of the human mind. These deontological judgments are like heuristics or biases: They are useful in certain respects but should be disregarded as general guides for moral judgments because they systematically skew human moral rationality.³¹ The mental gizmo thus causes “moral illusions”³² (in the same way that the Müller-Lyer illusion causes visual illusions), and one prime example of a product of its operations is Kant’s principle of humanity that one should never use other human beings merely as means and not as ends.³³ This in itself is already important for the topic of human rights, because this principle is widely regarded as an important element of the concretization of guarantees of human dignity in various national, international and supranational legal systems.³⁴ Human dignity, in turn, is a constitutive part of the whole architecture of human rights. If human dignity is a moral illusion, then this architecture starts to totter. However, the mental gizmo thesis goes even further than this. Human rights as such are themselves seen as products of the mental gizmo, useful as rhetorical devices and exploitable for good causes but without any claim to rationality as such and often quite harmful in their effects.³⁵ Instead, for truly rational moral thinking one needs to resort to utilitarianism. Utilitarianism constitutes slow thinking, which is what should govern human moral reasoning in the last instance.³⁶

What is the evidence for the mental gizmo thesis? Some of many variants of the familiar trolley problem form the starting point of the analysis.³⁷ In the so-called bystander case, a bystander can turn a switch so that a runaway trolley is redirected with the consequence that it kills not five persons on one track but one person on another track. In the variant called the “footbridge case,” a person is thrown down onto a track from a bridge to stop a runaway trolley, thus saving the five people working on the track. Proponents of the mental gizmo thesis argue that a proper analysis of cases where deontological judgments seem to be at play, because the observers judge it not to be permissible to sacrifice the life of one person to save five (footbridge case), shows that this judgment is in fact determined by hardwired emotional reactions.³⁸ This analysis is supported by neuroimaging studies. These studies, it is argued, show that when deciding on these cases, the ventromedial prefrontal cortex (VMPFC) is active, an area of the brain associated with the production of emotion.³⁹ The VMPFC is active because these cases involve the agent directly; they are “personal” and thus trigger emotional reactions.⁴⁰

In other cases that are “impersonal,” where the judgment reached is different and participants consider it permissible that one person dies and five persons are saved (bystander case),⁴¹ the dorsolateral prefrontal cortex (DLPFC), the center of cognitive control in the brain, is active, showing that this utilitarian judgment is a rational, not an emotional one.⁴² Further evidence is provided, it is argued, by studies showing that if the VMPFC is damaged, utilitarian judgments are made by the participants in both the bystander and footbridge cases.⁴³

From this perspective, deontological arguments are nothing but the post-hoc rationalization of hardwired emotional reactions.⁴⁴ This is not only the “secret joke of Kant’s soul,”⁴⁵ but, one might add, the secret joke of the souls of a great many thinkers on moral issues from antiquity to the present. Thus, say, Plato’s dialogues, the Critique of Practical Reason and A Theory of Justice are all exercises in self-delusion on the part of their authors. Plato’s defense of a nonconsequentialist concept of justice, the formulation of the categorical imperative in its formal and material versions and John Rawls’ nonutilitarian principles of justice are ultimately expressions of the secret workings of the emotional mental gizmo, rationalized post hoc and writ large.⁴⁶ Human rights are part of these post-hoc rationalizations: “‘Rights’ are nothing short of brilliant. They allow us to rationalize our gut feelings without doing any additional work.”⁴⁷ The many people concerned with human rights, such as lawyers, judges, activists, politicians and – most importantly – the people claiming human rights, fighting for these rights to be respected, hoping, sometimes desperately, for their protection, are all under the spell of a “moral illusion.” Given the importance of human rights in practice, this “moral illusion” has massive consequences that dwarf any practical effects that other elements of a skewed rationality, such as framing effects, may possibly have. These consequences are quite harmful: “Rationalization is the great enemy of moral progress, and thus of deep pragmatism.”⁴⁸

The mental gizmo thesis is part of a wider trend in contemporary neuroscience that can be called neuroscientific emotivism. This trend uses the means of cognitive psychology to defend traditional emotivism’s idea that human morality is no more than the expression of certain emotions of appraisal and disgust. It represents a very radical critique of human rights: Its punchline is not to deny the cognitive reality and impact of deontological judgments and the idea of human rights, but radically to reinterpret their status and meaning. In this view, deontological judgments and human rights are not manifestations of practical reason, but on the contrary are some of the causes of human moral irrationality. This irrationality has such far-reaching detrimental consequences that it needs to be overcome for the sake of the survival of the species. Only utilitarianism, only slow thinking, it is maintained, is able to solve the great problems of humanity and transcend the parochial moralities of the human tribes created by the mental gizmo,⁴⁹ “to free philosophers from the ups and downs of their automatic settings,”⁵⁰ and to free all the rest of us, of course, who also often suffer from the “moral illusion” of human rights.

How to answer this interesting challenge, which presumably will be paradigmatic for quite a few discussions to come?

6.4 The Mental Gizmo Thesis Reconsidered

One fundamental problem is that the analysis of the trolley problems (and its predecessors, such as those formulated by the German criminal lawyer Hans Welzel) underlying the mental gizmo thesis is deficient.⁵¹ The cases that are taken to prove the workings of emotional gut reactions (instances of the footbridge case) in fact show something quite different, namely the relevance of the means/side effect distinction in explaining the empirical patterns of moral evaluation observed and, more precisely, the prohibition of the instrumentalization of human beings.⁵² A proper analysis of the trolley problems therefore in fact vindicates the relevance of Kant’s famous formula of the principle of humanity and thus the relevance of a crucial element of the idea of human rights. In addition, it is far too rash to conclude that a utilitarian reasoning is at play in the other cases where choosing the death of one person is taken to be permissible if the alternative is the death of five (or more victims, as in Welzel’s initial train case). Thinking that it is permissible to choose the lesser of two unavoidable evils, though perhaps feeling at the same time that this is a tragic choice, is one thing, endorsing utilitarianism in the sense that it is always permissible simply to count lives quite another. The analysis of the trolley problems needs to be much more sophisticated and transcend such simple dichotomies.⁵³

Another problem of the mental gizmo thesis is that it is self-refuting. The reasons for this are as follows. Utilitarianism in all its classical and contemporary rule- or action-based variants is guided by the principle of utility: Any rule or action that creates the greatest happiness of the greatest number is normatively justified.⁵⁴ The foundation of this principle is the idea that each person’s happiness counts equally because the persons who are happy count equally. This egalitarian thrust of utilitarianism explains its persistent attraction and forms the core of what is truly admirable in this line of thought.⁵⁵

This maxim rests on two pillars: the equality of persons and the prescriptive principle that equal persons ought to be treated equally. Only given these two propositions does it follow that everyone’s happiness shall count equally, as presupposed by the principle of utility. Consequently, the obligation to respect the equality of equal persons by counting their happiness equally in the utilitarian calculus is not the consequence, but the precondition of utilitarianism. The obligation to respect the equality of persons is not and indeed cannot be derived from the application of the principle of utility: Being its foundation, the obligation to treat equals equally cannot be the consequence of that latter principle. Rather, the obligation to respect the equality of persons is a principle that forms the nonconsequentialist normative precondition of consequentialism. There is thus a deontological residue at the core of utilitarianism, because the principle of the obligatory equal treatment of equals (the reason for being obliged to value everybody’s happiness equally, as presupposed by the principle of utility) is the foundation of any utilitarian argument.⁵⁶

This analysis of utilitarianism leads us to the central catch of the mental gizmo thesis. The catch is this: Either it is true that utilitarianism is slow thinking (as utilitarianism presupposes deontological principles of equality, it follows in that case that such deontological principles are slow thinking, too, because these deontological principles of equality are the normative core of what is regarded as slow thinking) or it is true that such deontological principles are fast thinking (as utilitarianism presupposes these deontological principles of equality, which is fast thinking, utilitarianism must constitute fast thinking, too). In either case, the mental gizmo thesis is refuted by internal contradictions.⁵⁷

When thinking about these kinds of claims advanced by certain psychological theories, we should bear one thing in mind: While neuroimaging studies have provided many fascinating results in recent years, as yet there are limits to these insights. These limits are not only the product of the constraints imposed and problems implied in the methods of neuroimaging, such as the level of the spatial and temporal resolution of the methods applied, “voodoo correlations,”⁵⁸ circular analysis,⁵⁹ the effects of the statistical “smoothing” of results and so on.⁶⁰ For example, the way that complex mental phenomena are realized in the human brain still has not been clarified. One question to ask in this context is whether the long-dominant focus on the localization of functions in the brain is fruitful or whether the research on patterns of activation is not more promising.⁶¹ This has consequences for the interpretation of patterns of neuronal activity that form the basis for the hypotheses of neuroimaging studies, because it is far from clear what a given observed neuronal activity really means. This is due not least to a classical problem of neuroimaging studies: reverse inference. One cannot conclude from the fact that a brain region is active when performing a certain task that whenever this brain region is active, this cognitive task is being performed. This is because a single brain region may perform many tasks – interacting with other areas of the brain, for instance. Reverse inference can produce research hypotheses but cannot provide conclusive evidence of how a mental function maps onto the brain.⁶² The fact that a brain region – say, the DLPFC – is active when performing certain tasks involving cognitive control thus does not entail that whenever the DLPFC is active, cognitive control tasks are being performed. The same is true of the VMPFC and emotional reactions. Consequently, if they stand the test of further research, the results of neuroimaging studies such as the ones referred to are in no way proof that deontological morality is the expression of emotion. In addition, there are many competing empirical findings,⁶³ interestingly – if critically interpreted – also in studies coauthored by Greene himself.⁶⁴ In light of what has been said, there is thus ample reason to reinterpret such findings and what they tell us about the workings of the mind, considering closely the analysis above of cases such as trolley problems and what they mean for plausible theories of moral judgment.

In this context, we should stress that no one doubts that emotions are a central part of moral evaluation. The question is, however, whether such emotions constitute moral (deontological) evaluation, as emotivists contend, or play a different role. Again, the theoretical power of imagination evident in emotivist accounts seems far too limited to explain the complexity of the human moral world.⁶⁵ One important point is these accounts’ analytical failure to distinguish emotions that are the consequences of moral judgment from emotions that constitute moral judgment. Consider the case of outrage after witnessing a grave injustice. Here, the cognition of injustice is the precondition of and thus not identical to the feeling of moral outrage. The fact that moral sentiments arise is predicated upon certain preconditions, such as the indignation about an injustice upon the unequal treatment of two persons who are equal in the respects that are relevant in the situation at hand. Only if the agent thinks that these elements of an immoral act exist will the respective feeling ensue. This is why such feelings disappear when one understands that the facts actually were different – for example, that there was indeed a relevant difference between the persons being unequally treated.

For the study of the neurological basis of morality, this means that it would be very surprising if brain functions relevant for human sentiment (whatever they turn out to be) were not engaged when an agent evaluates a situation in moral terms. However, these moral sentiments do not constitute moral judgment, but are the consequence of an analysis of structural elements of the action (agency, patients of action, intentions and their kind and object, relations of equality, etc.).

In addition, as our analysis of the concept of fundamental or human rights and the history and justificatory theory of human rights have shown, the preconditions of the predication of rights are highly complex both in form and in substance. Equating such complex judgments with an emotional gut reaction does not seem a very promising approach to the matter.

All of this is nothing more than a reminder that the interpretation of empirical data is dependent upon theory: Data only have meaning within a theoretical framework. In our concrete case, the value of neuroimaging studies of the neurophysiological basis of moral judgment is dependent on the merits of the theoretical framework in which these studies are developed. If this framework is deficient, the interpretation of the data will be insufficient, too.⁶⁶

In addition, identifying rationality with utilitarianism seems somewhat naive.⁶⁷ The question is: Why should the scope of practical reason, to use a traditional term, not be wider? In the history of thought, it was the default assumption that human thought is made up not only of some kind of instrumental rationality, but that there are other, qualitatively different yardsticks, most importantly those of justice and moral goodness. This is the common denominator of much of the greatest thought on these matters. Why should this be wrong? Why are deontological principles a priori not rational, or reasonable, if you prefer? What is intrinsically better about the principle of utility (forgetting for a moment its deontological foundations) as compared to the prohibition of instrumentalization, the principle of the justice of the equal treatment of equals or the obligation to care for others?⁶⁸ A strange impoverishment of the richness of human thought is at play in such theories that fails to live up to the insights of practical philosophy and legal theory.

It is important to emphasize that the mental gizmo thesis is not a necessary consequence of the dual-process model of the mind as such. It is possible to believe that this model describes an important aspect of human cognition without finding the mental gizmo thesis convincing. Deontology could be part of slow thinking; there is no a priori reason why this could not be the case. Consequently, not being convinced by the mental gizmo thesis does not say anything about the value of the dual-process model of the mind. The mental gizmo thesis is just an implausible thesis developed within this model of the human mind.

Nor do neuroscientific or psychological approaches to ethical and legal issues as such wed us to a certain perspective on the cognitive origin of ethics and law. In particular, nothing in the theory of mind, in neuroscientific research or in psychology forces us to develop an impoverished account of human practical thought. The answer to psychological skepticism is consequently neither to ignore neuroscience and psychology, nor to escape into a normative theory where psychology, whatever it says, simply does not count. Rather, the answer is to develop a substantial concept of human moral cognition as an element of a wider theory of human rights that is more plausible than its skeptical alternatives.

6.5 Rights and Behavioral Science

One large and creative area of research concerns the analysis of law on the basis of behavioral science. Classical law and economics operating within the parameters of rational choice theory assume that human beings are rational maximizers of their expected utility in absolute terms. This assumption is not only meant to be descriptive, but often forms part of a normative theory: Justified decisions must be based on such rational choices.

Behavioral law and economics take their impetus from systematic differences between the assumptions of rational choice theory and actual human decision-making.⁶⁹ One influential element is prospect theory.⁷⁰ The central thesis of prospect theory is that people make choices not on the basis of utility determined by a final state, but rather in the light of changes relative to their specific situation or reference point. Losses and gains are central elements of perceived utility. People are loss averse – the utility of gains is perceived to be smaller than the disutility of losses, even if gains and losses are equal in absolute terms. When making decisions, people use heuristics, general decision principles that may work well in many cases but may yield results that are irrational – for example, the representational heuristics that ascribe properties, sometimes falsely, to individuals because of the class to which they belong. Judgments are also influenced by biases, such as the omission bias – all things equal, people prefer not to act. Framing effects are a third example: The way a decision situation is framed – for instance, in terms of losses and not of gains – influences the decision, even though the outcome is the same in absolute terms.

Within this conception of the mind, theories of rights are developed. One such theory explores the possible effects of loss aversion on conceptions of rights. The distinction between civil and political rights on the one hand and social and economic rights on the other, or between so-called first- and second-generation rights, is of interest in this context. A key question is whether or not there are reasons to protect both kinds of rights equally or whether only the former qualify as true human rights. A possible approach to this question is not to engage in the debate about the justification of these different rights, but instead to try to explain the reasons for the widespread perception (be it justified or not) that there is such a difference.⁷¹ One way to do so is to refer to loss aversion as a psychological mechanism. Taking this approach, “the crucial distinction is not necessarily between governmental acts and omissions, but rather between government giving and not taking.”⁷² Civil and political rights are perceived as being about the government not taking something, such as the unrestrained possibility of free expression. Social and economic rights, by contrast, are about giving something to the rights-holders. As losses loom larger than gains, the former kind of rights enjoys greater plausibility than the latter. Reference points are one major factor in analyzing this problem. The fact that many legal systems demand that if social or economic benefits are provided then this must be done without discrimination is based on such a reference point: Not receiving the benefit is experienced as a loss, because those similarly situated persons who do receive the benefit serve as reference points.⁷³

This view is of considerable interest when attempting to explain why civil and political rights are perceived as being different from social and economic rights, at least in modern debates. Whether this difference also exists in this form from a historical perspective is far from clear, given the prominent place of rights to material goods in older reflections.⁷⁴

This approach leads to the following question, however: Why does loss aversion (assuming that it plays a role in this context, if only for the sake of the argument) not settle the case? Why do questions about social rights not only arise at all but do so very powerfully, and arguably since the beginnings of the discourse on rights? This seems to indicate that certain normative principles have an influence on human perceptions of the justification of human rights beyond loss aversion, including principles of justice, which are particularly relevant for social and economic rights.

This does not speak against an impact of loss aversion on such debates, but rather against assuming that this is the decisive or even only psychological influence on the perception of whether these rights are justified.

Other examples seem to confirm this kind of analysis. One such example is affirmative action, which is a controversial issue in the interpretation of equality guarantees and thus in the interpretation of an important element of human rights. Legal systems apply affirmative action to benefits that people do not yet possess.⁷⁵ Accordingly, affirmative action programs concern, for instance, access to university or employment for people who do not enjoy this access yet, but do not demand that others relinquish places at universities or jobs they already hold. This can be explained by loss aversion: The losses of losing a benefit (admission to university, employment, etc.) loom larger than the gains for those who are not enjoying this benefit yet.⁷⁶

There are interesting questions to be asked beyond loss aversion in this context, including the legitimate protection of trust and the reliance on certain decisions for allocating the goods of students admitted to university programs. Revisiting decisions on university admission should only be possible under very restrictive conditions, such as applicants intentionally providing wrong information, because they will have based many subsequent decisions on their admission. Such considerations are mirrored in some administrative law.

But even if one disregards this for the sake of the argument, an answer still is needed to the question of why the desire for affirmative action arises in the first place. This desire is connected with the idea of justice as equality. The ultimate aim is, after all, to achieve equal access to the benefits of society independently of characteristics that are irrelevant for the allocation of such goods, as skin color is for admission to university. For this reason, minorities who were excluded previously are treated preferentially in order to overcome traditional patterns of exclusion. The question is not whether this connection between affirmative action and justice exists, but rather whether considerations of this sort can trump the principle of equal treatment for a transitional period in order to achieve this aim.⁷⁷

It is argued that loss aversion is compatible with common-sense morality – which adheres to threshold deontology.⁷⁸ Classical distinctions in ethics between doing and allowing or between intending (in the sense of purposeful action) and foreseeing mirror loss aversion, it is argued: The duty to avoid doing harm is stricter than the duty (if it exists at all) to help others (or to not allow harm to happen to them). The two distinctions are not the same, as illustrated by cases in which one intentionally allows harm or harms somebody by an action, but only as a side effect.⁷⁹

The problem with this account is that loss aversion may be compatible with a version of threshold deontology but clearly fails to adequately specify the relevant moral principles that are the real core of the matter. These principles are more complex than the doing/allowing or intending/foreseeing distinctions suggest.

Take the trolley problem: Mere loss aversion gives no reason to judge flipping the switch in the bystander case to be permissible but throwing the person off the bridge in the footbridge case to be impermissible. In both cases, the losses are the same – five people killed, one person not killed, or vice versa, depending on whether or not action is taken. Even loss aversion supplemented with the distinction between doing and allowing is not enough. Consider a variant of the footbridge case: The person on the bridge is a toddler, Hannah, on her tricycle who will fall on the track as the railing is damaged, stopping the train, if the other person on the bridge who is entrusted to care for the toddler does not prevent the child’s fall, which the other person is easily capable of doing. It seems impermissible to let the toddler fall in order to stop the train, even though one is not doing anything, only allowing something to happen. The reason has already been pointed out: More complex principles than loss aversion or the distinction between doing/allowing determine the evaluation of such cases. What is key here is an ends–means distinction and the prohibition of using human beings (merely) as means for other ends, be it by doing or allowing.

Loss aversion consequently has limited power to explain the problems investigated here. However, studies in experimental psychology and behavioral economics deal directly with normative principles relevant for our argument, looking at intuitions about justice and benevolence – the next topic to which we will turn.

6.6 Justice and Benevolence

Research on justice has a substantial tradition in social psychology, encompassing research on relative deprivation, distributive justice and the fairness of outcome distributions, procedural justice and corrective justice.⁸⁰ One prominent example from the current debate is the extensive discussion of fairness or, to use another term, inequality aversion. Yet another example is altruism, which can be seen, for example, in the case of contributions to social goods or in the (related) form of altruistic punishment. Importantly, altruism is understood in the sense of strong reciprocity; that is, behavior that does not lead to individual economic benefit of the agent.⁸¹ There is cross-cultural research on such attitudes.⁸² Another aspect of this research is the development of such patterns of behavior in children.⁸³

From the perspective of our inquiry, one problem of these studies is that they are predominantly concerned with patterns of behavior (e.g. distributive acts, punishments, rewards) and not with the internal mental states of the agents. In particular, they are not concerned with the reflective evaluation of actions by an agent or observer that has deontic, prescriptive consequences either from the first-person perspective of the agent or the third-person perspective of the observer. Reflective evaluation is crucially important for human morality, however, as moral judgment involving a prescriptive dimension is the core of the matter, as already indicated. Morality is not concerned simply with acting in conformity with certain (other-regarding) standards; it is concerned with a moral evaluation that yields prescriptive propositions such as “You should not bombard hospitals in civil wars” and possibly gives rise to action on the grounds of and motivated by such prescriptive propositions.

It is important to remain aware of a traditional insight of moral philosophy in this respect: There is no necessary or deterministic connection between moral evaluation and moral behavior. An agent may very well perceive the morality or immorality of an action but nevertheless fail to act accordingly due to intervening interests, weakness of moral will and so forth. Agents’ failure to display altruistic or just behavior does not allow direct conclusions to be drawn as to the principles governing their moral judgment. They may simply not be acting on the basis of their moral judgment.

There is another problem concerning the theoretical and conceptual framework of such studies and thus the determining framework for the interpretation of data. Concepts such as “inequality aversion” or “other-regarding preferences” play a central role in some of these studies. These concepts do not fully fathom the intricacies of moral judgments, however. Aversions and preferences describe inclinations to act. An aversion to asparagus means that one has no wish to eat this vegetable if it can be avoided. Moral judgment concerns something qualitatively different, as we just have seen, namely the reflective evaluation of an intention or action (e.g. based on such inclinations). This evaluation has deontic, prescriptive consequences, a moral ought. This “ought” is categorically different from a mere aversion or preference. It does not incline – it obligates persons, whatever their inclinations may be. The possible conflict between prescriptions that obligate and inclinations to act, such as preferences, or not to act, such as aversions, reveals the difference between the two categories, showing that we should not mistake the one for the other. Moreover, there is the basic deontic category of permissions, which is important for the concept of rights and cannot be translated into preferences or aversions either. Whether you have an aversion or preference to do something is irrelevant for the question of whether you are permitted to do it.

These methodological and theoretical problems notwithstanding, this area of research offers important findings. There are a large number of studies that provide empirical evidence about the egalitarian intuitions of human beings, prominently in the ultimatum game, for instance, which involves the following: A player receives a sum of money and distributes it between themself and another player. If the recipient accepts the distribution, both keep the amount distributed; if not, nobody receives anything. There are many variants of this game – for instance, the dictator game. As a baseline, the results show that proposers offer an almost equal share and that responders will not accept just any distribution but reject very low shares.⁸⁴ We should note that neither an unequal proposal nor the acceptance of an unequal distribution means that the offer is considered just. Selfish impulses evidently have a strong hold on human beings (whether the proposer has a persistently good conscience is, however, a different question), and the responder may consider receiving a smaller, unequal share to be a lesser evil than receiving nothing at all.

The most plausible interpretation of these results is that human beings are not (only) maximizers of utility, but that their evaluation of distributions is based on moral principles – for instance, of (proportional) equality.⁸⁵ If they were simple utility maximizers, they would accept any distribution that improves their situation, however small it might be. Importantly, the preservation of relations of equality seems to be a value in itself, and greater than at least some material benefits.

The results of studies on altruistic punishment point in the same direction: Human beings value certain standards and act to enforce them, even if this comes at a certain cost to themselves. Whether the reason for this kind of behavior is that they expect to benefit from the maintenance of such structures themselves at some point or that they consider defending certain normative principles to be of intrinsic value is quite another question.

There are other patterns of behavior that are discussed in connection with questions of social norms and morality. To take some examples:⁸⁶ Communication has the effect of increasing cooperative behavior. Many people are conditional cooperators – they cooperate in proportion to the cooperation of others. In finitely repeated public good games, levels of cooperation deteriorate over time, despite high initial cooperation rates. Stable group composition leads to higher cooperation rates. Framing a game as a community game has positive effects on cooperation in comparison to framing the same game as a stock market game. If the games are played sequentially, however, the effect of this framing disappears. Peer punishment leads to greater levels of cooperation, but its effects can be undermined if the punishment indicates selfish intentions. There may even be forms of punishment of cooperators. Punishment means a certain investment. Nevertheless, people prefer an environment with peer punishment. Rewarding cooperators increases cooperation. Bringing cooperative individuals together also augments cooperation.

How to explain such patterns of behavior? One way to approach the problem is to account for these patterns by the effects of social norms. This is how Ernst Fehr and Ivo Schurtenberger proceed, for instance. In their influential analysis, they supplement a direct social norm approach with the key idea of conditional cooperation and a set of particular psychological mechanisms. The direct social norm approach calculates the utility and disutility of following a social norm. The (dis)utility depends on an intrinsic desire to comply with norms.⁸⁷ The conditional cooperation approach postulates that people will cooperate dependent on others’ level of cooperation. Because of the mechanism of conditional cooperation, the mentioned patterns of behavior arise, Fehr and Schurtenberger argue, with the exception of: peer punishment; punishment of cooperators that threatens the positive effects on cooperation; and the preference for environments with peer punishment. Here, Fehr and Schurtenberger’s theoretical account relies on additional psychological mechanisms, such as social preferences for fairness/equity, reciprocity, a prosocial self-image or an aversion against guilt.⁸⁸ The social preferences for fairness and equity are themselves determined by social norms.⁸⁹ From these authors’ perspective, institutional structures play an important role: Punishment is necessary to maintain cooperation and institutions that guarantee norm conformity.⁹⁰

It is true that “unconditional normative prescriptions like ‘be selfless’” alone cannot account for such patterns of behavior.⁹¹ But this does not mean that the normative principles that we have discussed in the present inquiry are irrelevant or nonexistent and that the moral world is limited to conditioned cooperation and the particular understanding of fairness and equity considered by this experimental work. This is because a moral obligation provides a sometimes-powerful motivation but is always just one of other impulses that influence human behavior. According to the view defended here, human beings are not selfless beings, but beings who have the faculty to limit their many selfish impulses because of moral judgments and the volitional consequences of these judgments. Moreover, cooperation is a complex affair and evidently not just based on moral impulses of selflessness, justice and altruism.

Therefore, the patterns of behavior recalled as examples of this strand of research come as no particular surprise: Communication can help with cooperation for many reasons – for instance, defining one’s mutual advantage, clarifying common interests or strengthening the will to adhere to moral intuitions. That levels of cooperation can be influenced by the declining cooperative behavior of others is easily reconcilable with a moral orientation, too, as the latter does not imply a readiness to be exploited. Even if we feel a moral obligation to act in certain ways, our preparedness to do so may diminish if we see others pursuing egoistic goals. The sobering effects of repeatedly played games with declining levels of cooperation may have related reasons but do not say anything about the reality of more exacting standards of behavior. In the same vein, partner matching or the framing of a game as community-oriented action can help agents to adhere more faithfully to some moral standards and prudential principles that are also important for cooperation. The sequential playing of these games may have the mentioned sobering effect.

Altruistic and peer punishment has much to do with principles of justice. Distinguishing just punishment from unprincipled, potentially boundless revenge and determining the principles that make punishment just have been permanent themes of the theory of justice, dealt with in many contributions, including Aristotle’s classic treatment of the matter. A preference for entrenched systems of punishment is a possible product of cool-headed assessments of the pacifying, conflict-reducing effects of such arrangements. That rewards may help to motivate people to do anything, including to cooperate, needs as little proof as the positive effects of working with like-minded cooperators.

The reference to fairness, equity and moral emotions like guilt takes us straight to the question of the structure underlying human moral cognition. Here, one has to be analytically precise and try to determine more concretely what a moral judgment is about, avoiding in particular the category error of mistaking a preference or aversion for an obligation or prohibition. We already tried to give some indications of what such an analytical theory of morality could look like, and we will proceed further on this path in the last chapter. Moreover, this body of empirical evidence does not conclusively answer the question of the origins of foundational moral intuitions; in particular, it cannot tell us whether they are culturally induced or based on the innate structure of the human mind. We will also ask, therefore, whether it is really true that ideas of fairness and equity are wholly the offspring of social norms or whether these social norms are rather in at least some part the expression of basic moral intuitions, not least in the light of child psychology.

The behavioral studies we have reviewed contain no findings that have a direct and robust bearing on the question of human rights. We identified some problems of the theoretical framework of the interpretation of the empirical findings. Nevertheless, the thrust of this important research underlines the fact that intuitions about justice such as equality and altruism are not mere theoretical chimeras, but psychological realities of substantial importance for human beings, their decision-making, the explicit norms they develop and their subsequent actions.

7 Where Did It All Come From? Morality and the Evolution of the Mind

And this tattooing, had been the work of a departed prophet and seer of his island, who, by those hieroglyphic marks, had written out on his body a complete theory of the heavens and the earth, and a mystical treatise on the art of attaining truth; so that Queequeg in his own proper person was a riddle to unfold; a wondrous work in one volume; but whose mysteries not even himself could read, though his own life heart beat against them; and these mysteries were therefore destined in the end to moulder away with the living parchment whereon they were inscribed, and so be unsolved to the last. And this thought it must have been which suggested to Ahab that wild exclamation of his, when one morning turning away from surveying poor Queequeg – “Oh, devilish tantalization of the gods!”

Herman Melville, Moby Dick

7.1 Morality and Evolution

Human beings are part of the natural world and thus owe their makeup, including their cognitive capacities and limitations, to the same forces as any other organism. Therefore, it is natural, at least for the purposes of a scientific theory-building whose methods are not concerned with religious interpretations of human existence, to ask what the theory of evolution can tell us about the nature and content of human morality. As the law is a central element of human beings’ normative world, the answer to this question potentially has important consequences for the understanding of the law as well.

Evolutionary thinking has permeated many disciplines. Theorems of evolutionary biology in particular have been put to use – not always felicitously – in many other contexts, from cultural development based on war¹ on to social theory.²

The key question for the concerns of our inquiry is whether there are hardwired, genetically inherited mental structures that influence or even determine human moral evaluations, intentions and concrete actions. As these moral evaluations clearly impact upon both the making and (according to the most plausible theories of interpretation) the application of law, the influence of such hardwired mental structures on the law may be substantial.

There is a huge variety of research in this area. Some investigations emphasize the importance of the biological foundations of human morality, some the evolution of culture after modern human beings started populating the Earth some 80,000 years ago. Some research challenges traditional assumptions about the content of morality, explicitly including the idea of human rights or the idea of a universalist moral outlook. One example of the former, the mental gizmo thesis, was already discussed in the Chapter 6, although we did not go into detail with respect to the underlying evolutionary theory. From this point of view, evolutionary mechanisms constrain the moralities possible in a way that rules out any natural morality consisting of generous, solely other-regarding altruism or egalitarian justice, because such a morality could not form a behavioral evolutionarily stable strategy for an organism. By contrast, other approaches confirm notions of altruism, fairness and mutual respect not limited to small groups, albeit to varying degrees, in the framework of a naturalistic theory of morality.³

As far as the topic of human rights is concerned, there are two dimensions to the problem. The first concerns human rights themselves, whose evolutionary origin forms the object of some such theories, including the mental gizmo thesis. The second dimension concerns the normative principles that we have identified, which underpin intuitions about individual claims and the justification of human rights, including justice, obligatory concern for the well-being of others and respect for their worth. Our review of evolutionary theories needs to deal with both of these dimensions.

7.2 Puzzling Altruism

Evolutionary biology has long been concerned not just with the physical traits of organisms, but also with behavior and the mental structures and inner states that underpin behavior.⁴ Evolutionary explanations form a central part of ethology. Sets of behavioral patterns include social behavior, which occurs in many different forms. Some species, like bees and ants, live in highly organized, complex communities. How do such behavioral patterns and the underlying cognitive structures arise? What is their evolutionary origin?

In this context, behavioral traits that benefit organisms other than the actor are of particular interest. Altruistic behavior is a special object of scrutiny for some approaches to evolutionary theory, because it does not seem to fit well into what is regarded as the basic mechanisms of evolution, and into natural selection in particular. How can evolution select for traits that are beneficial not for the bearer of these traits, but for others?

These questions also arise for highly cooperative organisms such as human beings. Human morality is crucial to understanding this cooperative behavior. The mental world of humans includes prescriptive moral rules that affect both the formation of intentions to act and actual action. For humans at least, the evolutionary point of altruism is not the only problem to arise: Justice is another equally puzzling issue. Its principles limit the gains people can reap for themselves. How does this constraint fit into the picture of evolutionary theory? An equivalent question can be asked of the idea of obligatory respect for others. What are the evolutionary roots of this phenomenon?

Many approaches to the issue observe that the gains an organism reaps by cooperating are a central factor in the evolutionary process. It is argued that other-regarding behavior is only an evolutionarily stable strategy if it increases the chances of reproduction by augmenting the reproductive fitness of the cooperating organism. Moreover, every behavioral trait needs to yield such reproductive benefits if it is to continue to form part of the evolutionary process. Otherwise it will be weeded out by the forces of selection.

7.3 Various Forms of Cooperation and the Problem of What Morality Is

Before embarking on a review of the main findings of evolutionary theory that have a bearing on the theory of human rights, it is worth mentioning a few points that are sometimes overlooked in discussions of this matter and that will inform our analysis.

First, sociality and cooperation are not necessary Forms of existence for organisms – solitary life forms can be very successful at assuring the reproduction of a species. There is no evolutionary necessity to be social or cooperative; it is simply one of many possible ways of living. Arguments claiming that humanity’s specific form of life had to develop – a kind of natural teleology in the form of evolutionary theory – are therefore untenable.

Second, cooperation comes in many shapes and sizes. This is a simple but absolutely crucial observation. A herd of animals staying together to create protection in numbers; active group defense; temporary intragroup coalitions to outcompete other group members or to kill their offspring;⁵ the complex functional differentiation of ant colonies or a beehive, including the infertility of some organisms for functional purposes; the mechanisms of cooperation in a human society of hunter-gatherers in 20,000 bce; or human society in 2023, organized through the institutions of a constitutional state based on human rights, part of the global community under international law – these are all very different forms of cooperation.

Third, human forms of cooperation face the normative standard of legitimacy – bees have no other option than their hives, but how humans should build societies is far from clear. In human history and thought, very different forms of cooperation were used for long stretches of time: A stratified slaveholder society under an authoritarian ruler cooperates, but not in terms of justice, beneficence and mutual respect. In Politeia, Plato envisaged a highly cooperative society, where all members devoted their best abilities to the good of the polis – a vision that 2,500 years later was denounced as the seed of totalitarianism.⁶

Fourth, the diverse forms of cooperation presuppose very different cognitive abilities on the part of the organisms cooperating. The mechanisms used by antelopes to cooperate in herds, by bees to assemble in a hive, by ants to form a colony and by human beings to form a constitutional state within the UN system are not the same. These forms of cooperation consequently also come about in very different ways. Herds of antelopes or ant colonies are not created by centuries of reflection, reform and political revolution, while human social institutions are. These developments and the often-radical reorganization of human societies are of major interest, even though humans were living in hunter-gatherer societies, not constitutional states when the survival of the human species was decided, because they may offer clues to human beings’ potential, including cognitive capacities that were put to use only in later stages of their history. The fact that human beings are able to form highly complex mathematical theories about the world with explanatory power is certainly of interest for understanding the cognitive machinery of human beings, even though differential equations were (most probably) only rarely discussed in the average cave 60,000 years ago.

Fifth, the mechanisms underlying cooperation are not immediately obvious and are certainly manifold. This is true for human cooperation as well. The way that human societies are integrated into something like an organized structure is a classic and highly contested sociological question. Moral precepts are only one of the mechanisms that may be relevant here. Strategic interaction based on instrumental rationality, as in trade and other tit-for-tat relations, is another. Further elements include force or the threat of violence to command obedience, or the belief in an authority’s legitimacy because of its divine origin or because of the leader’s particular charisma. Force and belief in the legitimacy of some sort of authority are two very significant factors assuring cooperation in human history, used, for example, to build the pyramids, to construct the Palace of Versailles and to lead millions of Germans to become agents of bloodshed during World War II, often sacrificing their own lives on the way into the moral abyss.

Sixth, it is not obvious that human morality is only or even primarily a tool for cooperation. If one is to even start considering this problem, one needs to know what kind of cooperation is at issue, as we have seen. The effects of morality are not limited to what is functionally necessary for any form of cooperation. Morality can even become a lasting impediment to particular forms of cooperation: A sense of justice is a substantial obstacle to certain highly efficient forms of cooperation that would be useful if only reproductive fitness were at issue – say, an authoritarian command structure. Morality often demands the reorganization of society, thus challenging existing forms of cooperation, not on the terms of what is functionally needed to enhance reproductive fitness, but for the sake of the intrinsic value of justice. The ensuing conflicts have often torn human societies apart. Morality thus does not serve simply any form of cooperation; rather, it defines legitimate forms of cooperation – for instance, a just political order. In addition, it provides the tools to make such an order a reality.

The oft-encountered references in evolutionary theory simply to “cooperation” without any further qualifications are therefore inadequate. The explanatory task at stake is much more demanding: What is needed is an explanation of human sociality’s very particular forms of cooperation and – as the foundation of the specific forms of human cooperation – of the emergence of the peculiar phenomenon of human morality in the natural history of cognition.

Seventh, care must be taken when determining the explanandum, which in our case is human moral precepts and their workings in the human mind. The determination of the explanandum is particularly important in the case of morality because what this explanandum is is far from clear. There is not only no consensus about how to explain an explanandum whose properties are generally accepted (say, apples move perpendicularly to the ground), but the explanandum itself is controversial (say, whether humans are psychological egoists or not). Moreover, one encounters another problem in this context: The preferred evolutionary explanatory theory sometimes fallaciously determines what is assumed to be the content of morality. The following line of argument may serve as an example: Because it is assumed that the mechanisms of evolution can only produce psychological egoists, it is held that human beings must in fact be psychological egoists, because humans are ultimately the product of these evolutionary mechanisms and accordingly can only have the properties that these mechanisms can create.

It is important to avoid this fallacy, which ascertains that what cannot be explained from the point of view of some evolutionary theories cannot in fact exist. This would mean committing a fallacy analogous to the argument that gravitational force cannot exist because gravity cannot be explained within the framework of contact mechanics. The world is full of surprises, and we must remain open to the possibility that human moral cognition contains some of them. Therefore, a proper analysis of the phenomenon to be explained is the precondition of any attempt to formulate an explanatory theory of this phenomenon.

Eighth, it is important to be mindful of the fact that there are quite substantial controversies about the fundamental aspects of evolution – there is not one, let alone “The One” uncontested evolutionary theory, but a plurality of evolutionary approaches that merit serious consideration.

7.4 Animal Morality?

A first step towards getting a grip on these matters is to ask whether there is anything species-specific about morality. This is not an easy question to answer. However, there is a substantial body of work on animal cognition that can help to formulate some plausible theses.

The starting points of this research are the great apes, particularly chimpanzees and bonobos, which are human beings’ closest relatives. This does not mean that there is not considerable evolutionary distance between us and them – in fact, the evolutionary development of human beings branched off from that of the great apes some 7 million years ago. Human beings are thus separated from great apes by 14 million years of development.

Studies on great apes have produced evidence of differentiated cognitive abilities. To name just some of these abilities that are important for moral psychology:⁷ There is reason to believe that great apes act with instrumental rationality, at least in some instances. Under certain circumstances, they comprehend intentional states such as the goals and desires of others, particularly in the context of competition. They are able to maintain social relationships over long periods among subordinates, dominants and friends within their group. They are able to identify such relations between familiar third parties. They have and express emotions and are emotionally empathetic. They communicate intentionally. Within this framework, they engage in prosocial behavior, including instrumental helping: “When costs are small, and food competition is absent, great apes help others.”⁸ They are able to control some of their impulses for immediate self-gratification for prudential reasons, such as avoiding conflict. They are able to inhibit the impulse to forage individually and collaborate with others to obtain resources not available without such cooperation.

As far as morality proper is concerned, some have argued that great apes have some kind of proto-morality, others have denied this and others again have proposed a middle ground.⁹ There seems to be evidence that great apes exhibit signs of what is called sympathy towards their kin and friends, including acts of instrumental helping.¹⁰ But various experiments, including adapted forms of the ultimatum game, for instance, produced no evidence that great apes have a sense of fairness.¹¹

What seems clear from all of this is that the human moral world is structured on different terms. Humans certainly entertain many warm feelings towards others, from friendly concern and sympathy to love, feelings that motivate them to act in ways beneficial to others. In addition, however, human moral cognition implies principles of altruism and justice that are prescriptive in nature, implying a moral ought, and that are the product of engaging in different forms of reflection on the object of evaluation, sometimes limited to a spontaneous intuition about the justness or moral wrongness of an act, sometimes the result of painstaking soul-searching. Not only acts, but also internal mental states such as intentions form possible objects of such evaluations. These intentions come in more than one shape – direct and oblique intentions, for example. Accordingly, not only does the evaluation of these intentions presuppose an intricate cognitive machinery, but so do these mental states themselves and observers’ ability to identify them.

Things become even more complex when we come to further key notions of human moral cognition, such as the concept of “ought” and moral responsibility. The moral ought cannot be conceptualized without a notion of freedom, because ought is not a determining “must,” but leaves the agent the freedom to act otherwise. It is related to the notion of responsibility, another central category of the moral world of human beings, including their psychology, which presupposes that human beings can understand what is right, that they can make this insight the motivation for action and that they therefore can be held accountable for said action. All of this obviously presupposes a particular cognitive makeup, and this is true whatever one’s position may be on the question of free will, be it indeterminist, determinist or compatibilistic.

A further point: Morality does not equal sympathy, as illustrated by the moral obligations one has towards those people one does not sympathize with and the shame one may feel if one does not live up to these obligations, despite disliking the patient of the moral action. The same holds for empathy (if distinguished from sympathy). One may understand very well how it feels to be incarcerated and still think that it is not only the legally, but also the morally right thing to do in some cases of crime.

Finally, human moral judgment is connected to a set of particular sentiments that are the consequences of certain moral evaluations. These sentiments include shame, indignation and resentment.

The differences between the cognitive and thus moral makeup of human beings and great apes manifest themselves in the way they are treated in morally relevant contexts:

Despite the fact that they live in cooperative social groups and behave pro-socially towards kin and friends – and are of course worthy targets of our moral concern – chimpanzees themselves are not moral agents. We do not allow them to roam freely in our midst for fear that they will attack our children, steal our food, destroy our property, and generally wreak havoc without regard for anyone else. And if they did all these antisocial things, no one would blame them or hold them responsible.¹²

This kind of differential treatment implies assumptions about the differences between chimpanzees’ and human beings’ structure of cognition that are the reasons why such differential treatment is justified.

Given these observations, it comes as no surprise that any nonhuman animal can be raised with humans without this animal developing the cognitive abilities of human beings, whereas every human infant develops these abilities, whatever human environment they grow up in, drastic cases of deprivation aside.

There is simply no way around the fact that human cognition and nonhuman animal cognition are different and that there is some kind of biological basis for this difference. The question is only what this biological basis is and how to account for its development in evolutionary terms. This is a first important result.

7.5 Evolutionary Psychology

7.5.1 The Morality of Selfish Genes

Evolutionary psychology represents one very influential account of the origin and nature of human moral psychology. It draws on certain well-known ultimate evolutionary mechanisms – kin selection, reciprocity and group selection – to explain the evolution of cooperation.¹³

Inclusive fitness and kin selection are seen to be the solutions to the conundrum of other-benefiting behavior. From this perspective, single “selfish” genes are the objects of the evolutionary process.¹⁴ Their reproduction is assured not only if the individual who is acting survives, but also if other individuals who likewise carry these genes have a sufficiently high chance of reproducing. This is the case if these individuals share a sufficiently great amount of the same genetic material: The behavior that benefits other organisms and that seems to be implausibly altruistic from an evolutionary point of view is in fact a tool to increase the chance of reproduction of the gene responsible for this behavior and will therefore be selected for. The proximate psychological mechanism is other-benefiting; the ultimate evolutionary cause for this mechanism is a higher chance of reproduction for the selfish gene.

7.5.2 Reciprocal Altruism

Another central element of evolutionary psychology is reciprocal altruism, which consists of tit-for-tat relations: An individual does something beneficial for another organism in order to be paid back, either now or at a later point. One variant of this is indirect reciprocal altruism, which factors reputation into the equation: Actors do something to increase their good reputation in order to reap benefits from others through their standing, albeit perhaps not in the concrete relation in which they are acting, but in other contexts. A good reputation may pay off in the long run.¹⁵

Evolutionary psychologists argue that reciprocal altruism is both a proximate psychological mechanism and operates on an evolutionary level. As a proximate psychological mechanism, it inclines an organism to other-benefiting behavior if there is a payback, whether direct or indirect: Organisms help others because they, too, will benefit from offering help – not necessarily now, but perhaps later, when they themselves are in need. On the evolutionary level, reciprocal altruism is a possible evolutionarily stable strategy because any seemingly other-benefiting act ultimately increases the reproductive fitness of the actor through the benefits it produces and thus is advantageous from an evolutionary perspective.

There is more than one possible proximate psychological mechanism to make an organism behave according to the principles of reciprocal altruism. One is conscious reciprocal altruism – the actor thinks strategically and offers the patient an advantage in order to get something in exchange. Another is – to use de Waal’s terminology – not calculated, but attitudinal reciprocal altruism: The actor is not consciously acting in a strategic manner, but has developed an other-regarding emotional attitude.¹⁶ Some forms of social relationships of primates or other mammals exemplify this – for example, the grooming of kin and friends.

Supporting mechanisms include partner control through sanctioning, partner choice and social selection: Actors choose those partners who are good collaborators, shunning cheaters. Social selection is the product of partner control and partner choice: It promotes the selection of certain individuals because of the attractiveness of having them as a partner.¹⁷

There is also the idea of multilevel selective processes: An individual’s adaptiveness depends on the adaptiveness of the way of living of the group to which that individual belongs.¹⁸

7.5.3 The Morality of Tribes

At the same time, these mechanisms restrict the possible outcome of evolution: Only a small-group morality makes evolutionary sense because of the “environment of evolutionary adaptation” of modern humans, it is often argued. This is the core claim of what has been called “evoconservatism.”¹⁹ Only such a morality would have been selected for because concern for others must be limited to those who plausibly share a sufficiently large number of genes. Given the living conditions of early human beings, who associated in small groups and were faced by out-group competition for resources, concern for a small, limited group is the only possible product of evolution. By evolutionary necessity, there is a natural tribalism, based on an in- and out-group morality.

The logic of this argument leads to the conclusion that the idea of rights that all human beings, not just group members, enjoy, with correlative duties of all human beings in turn, is not supported by evolutionary psychology. The same follows for the moral principles that give rise to this idea. As discussed, justice, solidarity and respect are not limited to some in-group, but apply to all human beings. Human rights and principles of justice, altruism and respect in this sense have no ally in human nature from this perspective. They are at best justified cultural artifacts created to tame human nature – or, if one takes a more critical view of their ultimate justification, not even that. Instead, they are a kind of heuristic applied only to in-group members that has a certain rhetorical effect strong enough to make them empirically important factors in human behavior, but that has no substantial legitimacy and does not extend beyond the limits of the group.²⁰

Such accounts are important for the theory of human rights, even if one is as firmly committed to the is/ought distinction as the present inquiry is. Even if one rightly holds that human beings are capable of defying their natural inclinations, whatever they may be, and could and ought to do so in this area, too, by following the commands of a universalistic morality, the question of whether universal human rights are principles and institutions that are contrary to the nature of human beings requires serious consideration. The significance of this point is underlined by various voices that consider the claim that people are in fact not (and cannot be) naturally motivated to respect the rights of every human being to be an important argument against human rights.²¹

7.5.4 Explanatory Problems

The basic mechanisms of evolution mentioned above are of more limited explanatory power than is often assumed, however: As discussed above, there is substantial evidence that the basic assumption underlying many accounts of the evolutionary origin of morality that human beings lived in small groups of hunter-gatherers competing with other such groups is not on the right track. The respective cultures often spread over geographically huge spaces, individuals travelling large distances, creating groups that were not based on biological kinship relations in any discernable sense, but rather on common cultural bonds.²² The perception that anyone outside a particular group was regarded as an enemy is not tenable either, given the evidence for complex social ties beyond small groups.²³

Moreover, kin selection is not a mechanism that could explain the emergence of the particular properties of human morality.²⁴ The architecture of human morality includes such complex phenomena as duties, rights, specific moral emotions and responsibilities. Humans have the ability to create, apply and enforce moral norms. Kin selection is not sufficient to explain this sophisticated cognitive structure. To serve the ends of kin selection, much simpler psychological proxy mechanisms would suffice – say, some straightforward prereflective attraction to kin, without any intricate moral machinery in place. Furthermore, the commands of morality are not limited to kin, neither as an intuition nor in reflective evaluation. Principles of justice can and often do limit the benefits that kin receive, because others can claim an equal share to kin. Morality and justice thus constitute checks on kin favoritism and are not proximate mechanisms to serve it.

Reciprocal altruism raises similar questions. In particular, first, there is the problem of first altruistic acts: What motivates the initial altruistic act, given that no tit-for-tat pattern has been established yet and thus no rational expectation exists that the beneficent of the altruistic act will reciprocate – “blind optimism or accident”?²⁵ The second problem is the free-rider problem, the “powerful incentive to defect.”²⁶ What reason is there not to follow this incentive? The same problems arise for indirect reciprocity, which takes reputation into account.²⁷ Why should there be no free-riding here?²⁸ What is needed to break this deadlock is either an agreement to adhere to certain norms, as has been rightly underlined,²⁹ which is beyond the cognitive capacities of nonhuman animals, or, it is important to add, a form of altruism that is not conditioned upon reciprocal advantages. Similarly, in the case of emotional, attitudinal reciprocal altruism one can ask: Why do individuals form friendly social relationships with nonkin in the first place? Why do they help these friends?³⁰

This aside, it is decisive that human morality is limited neither to calculated nor to attitudinal reciprocal altruism in descriptive terms. As already underlined above and as will be fleshed out in greater detail below, it is a complex reflective capacity to judge corresponding principles of altruism and justice with motivational and emotional consequences, and thus it is neither a calculation of the benefits of nor simply an emotional attitude fostering tit-for-tat advantages.

The paradigmatic case of moral acting is sacrificing some of one’s own goods for the benefit of others – agents perhaps even giving their life to save others. The evaluation of such an act as morally good is not restricted to saving kin and friends. On the contrary, agents who save only their kin and friends when they are able to save others as well, or saving others only because they are expecting advantages as a result of their deed, would not count as performing a particularly impressive example of moral action. Similar observations hold for justice: Distributing something justly to receive benefits oneself is not what the idea of justice is about.

It is not a good counterargument to assert that early modern humans’ morality was different from what we find today. As we will see, we have only indirect clues to the cognitive abilities of early modern humans. None of them conclusively shows that moral cognition was limited to a small-group morality, speculative just-so stories aside, which are, moreover, based on the most probably wrong assumption, as discussed above, that hunter-gatherers did not live in larger cultural spaces.

Pointing to the distinction between proximate mechanisms and ultimate evolutionary causation does not help to counter this critique either: The nonreciprocal altruism of human beings and their sense of justice, among other principles, do not with sufficient probability produce paybacks for those adhering to them. The fact that being moral regularly does not pay constitutes a central conundrum of moral philosophy: For Kant, for instance, the fact that acting morally often does not lead to particular happiness and not seldom leads to the dungeon and the stake was so obvious that it served as an argument for the existence of God and eternal life, so that the scales of justice could be balanced at least in the afterlife.³¹ Consequently, human morality cannot be understood merely as a proximate psychological mechanism for reciprocal altruism as an evolutionarily stable strategy. It does not deliver the advantages that reciprocal altruism presupposes.

A further point of criticism homes in on the assumption that forms the foundation of the evolutionary approach, namely that all of an organism’s traits have an adaptive function. This poses overly narrow constraints on the process of evolution, as will be illustrated below.

There are, however, other accounts that paint a different picture of the evolutionary origin of morality. Arguably, the currently most sophisticated theory of the evolution of morality describes a process driven forward by mutualism and interdependence, enabled, however, by a wide-ranging and complex, inborn, species-unique cognitive endowment of human beings. Because of its paradigmatic importance for current debates, a closer look is necessary to see whether this approach by Michael Tomasello solves the riddles of the evolution of human morality.

7.6 The Power of Joint Intentionality: Interdependence and Cooperation

7.6.1 Some Specifics of Human Cooperation

Tomasello’s theory, too, is guided by the idea of cooperation: Its central thesis is that human moral psychology is a phenomenon emerging from the increasingly sophisticated forms of collaboration of hominids and finally human beings. Cooperation is also key to moral ontogeny: Moral development does not depend on general intelligence, but on a set of inborn, species-specific cognitive capacities universally possessed by all human beings independent of cultural background, which enable human beings to develop their rich moral world through cooperative constructive interaction similarly to the way they acquire other parts of their cognitive and social world:³² “It is precisely this biological preparation – in the form of maturationally expressed capacities – that makes uniquely human sociocultural activities and experiences possible in the first place.”³³ It is not just “general biological preparedness for learning and inference, but also specific biological preparedness for uniquely human cooperative interaction and mental coordination with others.”³⁴ The capacity for joint intentionality is key:³⁵ It is the “ultimate source of human uniqueness.”³⁶

The main criteria for identifying capacities as innate are a lack of cultural diversity of outcome in experimental intercultural designs with small children and very similar maturational age trends across cultures.³⁷ All children, for example, learn the complex activity of upright walking irrespective of cultural background in a “quite predictable developmental period.”³⁸ It is thus safe to assume that the capacity of upright walking is innate.

Cooperation between human beings extends to the rules they apply to their own cooperative activities: The “‘cooperativization of self-regulation’ is of the essence of normative sociality and morality.”³⁹

The theory (quite correctly) emphasizes how fundamental human cooperation is for the human way of living. While cooperative cultural institutions in particular create a framework for competition, competition is not the true driver of human development: “It is only if one neglects the cultural-institutional context of human behaviour that one can hallucinate the competitive cart as leading the cooperative horse.”⁴⁰

As a first very important step, this approach goes to greater effort than others properly to account for the specifics of forms of human cooperation. This includes identifying the various layers of the cognitive dimensions and preconditions of cooperation. This proves very fruitful because cooperation is a highly underdetermined concept – as illustrated above, cooperation can take very different forms and does not necessarily imply moral norms – it could be based, for instance, solely on force and power.

A battery of experiments provides some clues about the differences between the cooperative behavior of humans and that of great apes. More precisely, this research concerns the behavior of small children and thus does not even focus on the mature forms of cooperation between adults because Tomasello argues that the cognitive capacities of children are similiar to the cognitive capacities of early humans. These findings provide at least some good indications of distinctively different cognitive dimensions of forms of cooperation, spelled out in respective behavior. These include the following patterns, for instance:⁴¹

Children cooperate with joint intentionality: They pursue joint goals with others, and with joint attention. They reverse roles in collaborative activities and communicate cooperatively to coordinate collaboration. Children divide goods in particular ways: They share goods even if these goods are readily monopolized by one partner. They display a tendency to share goods more equally if these goods are the result of collaborative efforts. They have a preference for collaboration even if the payoffs are identical to those of solo activity. In relation to their partners in the respective collaborative activity, children share less with a free rider. They help others in various forms. They modify their behavior when peers are watching. They show a particular quality of commitment to the collaborative activity: They continue to collaborate until the end, even if they have already received the advantages offered by the collaboration. Unlike great apes, children from the age of three sanction others for violating social norms, and they do so on behalf of others. This indicates that their attitude towards social norms is not just prudential.⁴² They are as prepared to help when they are alone as when they are with their mother. They help persons who do not know that they are being helped. Such research bolsters the view that small children are intrinsically motivated to foster the well-being of others. None of the above examples forms part of the cognitive repertoire of great apes:⁴³ “[G]roup actions of apes are all about individuals achieving their individual ends in group contexts – they are using one another as social tools.”⁴⁴

This provides some hints on how different the cognitive apparatus of great apes and of human beings already is at a relatively early age. It illustrates vividly the point made above: One needs to spell out what kind of cooperation one is talking about and what the cognitive preconditions are that make it possible before one even can start thinking about an evolutionary theory of this form of cooperation.

7.6.2 Sympathy and Fairness Develop in Small Steps

The evolutionary theory of morality that tries to explain these findings is a theory of the incremental development of morality. This incremental development theory offers an alternative to standard accounts based on kin selection and inclusive fitness, which are regarded as insufficient.⁴⁵

Tomasello’s approach starts by identifying two forms of cooperation: altruistic concern for others and mutualistic collaboration and, correspondingly, a morality of sympathy and of fairness. The former is freely performed, while the latter is accompanied by a sense of obligation and deservingness and by punitive moral attitudes such as resentment or indignation towards others who are unfair. The former is pure cooperation, the latter the “cooperativization” of competition to find balanced solutions for conflicting demands. Morality in this sense is unique to human beings:⁴⁶

We proceed from the assumption that human morality is a form of cooperation, specifically, the form that has emerged as humans have adapted to new and species-unique forms of social interaction and organization. Because Homo sapiens is an ultracooperative primate, and presumably the only moral one, we further assume that human morality comprises the key set of species-unique proximate mechanisms – psychological processes of cognition, social interaction and self-regulation – that enable human individuals to survive and thrive in their especially cooperative social arrangements.⁴⁷

Note that this passage raises an important problem that will occupy us in more than one context: According to this formulation, there are “new and species-unique forms of social interaction” prior to human morality, as morality is a means to adapt to these forms of interaction. The problem is: Do not these “new and species-unique forms of social interaction” presuppose the existence of the cognitive mechanisms necessary for morality? Is a certain cognitive apparatus not the precondition for certain forms of living, including forms of collaboration, and consequently cannot be interpreted as an adaptation to these forms? If the cognitive abilities are prior to the “new and species-unique forms of social interaction” they enable, the evolutionary theory needs to account for the emergence of these cognitive abilities independently of the way of living that was only possible after these cognitive abilities were acquired.⁴⁸

How, then, did this “key set of species-unique proximate mechanisms,” the “psychological processes of cognition, social interaction and self-regulation” that are the natural mental foundations of human morality, evolve?

Tomasello argues that all began far back in time, with hominid species: First, there were dyadic relations of cooperation in hominids that led via role formation to normative standards. These cooperative relations were established to aid survival by foraging together, which led to sympathy beyond kin and friends, including collaborative partners.⁴⁹ In order to enable them to collaborate, hominids evolved joint intentionality – that is, the capacity to form a common goal with a partner and to know things together.⁵⁰ At the same time, Tomasello asserts that a context of collaboration including “obligate collaborative foraging with various and robust means of partner choice and partner control” is necessary for the development of joint intentionality.⁵¹ This is a concrete example of the problem just identified: According to this account, are forms of collaboration the precondition for the evolution of cognitive abilities such as joint intentionality or, vice versa, are cognitive abilities the preconditions for the possibility of certain forms of collaboration?

This collaboration shaped role models of what it meant to be a good partner, which were the first step to impartial normative prescriptions. This had a major consequence: “Recognizing the impartiality of role standards meant recognizing that self and other were of equivalent status and importance in the collaborative enterprise.”⁵² The self–other equivalence led to mutual respect among partners. Each could hold the other accountable for living up to this role ideal. A self-regulating “we” emerged that was considered legitimate because it derived from a joint commitment that the partners had created themselves and because both regarded the other as deserving of respect. This was the birth of an “evolutionarily novel form of psychology.”⁵³ The ensuing behavior is motivated by genuine moral motivation, not just by the wish to avoid punishment or reputational attacks: “And so was born a normatively constituted social order in which cooperatively rational agents focused not just on how individuals do act, or on how I want them to act, but, rather, on how they ought to act if they are to be one of ‘us’.”⁵⁴ Tomasello hypothesizes that this natural, second-personal morality emerged 400,000 years ago.⁵⁵

7.6.3 The Path to Objective Group Morality

A second evolutionary step was taken with the emergence of Homo sapiens. The reasons for this step lay in demographic development: Groups of modern human beings formed tribes that built up a culture, “one big interdependent ‘we’” that secured role conformity through identification with the group and sympathy of the group members towards each other in order to ensure survival. The individual learned to use the perspective of “any rational person,” based on the self–other equivalence, and thus “served as interchangeable (agent-independent) cogs in the conventional cultural practices that kept the culture going.”⁵⁶ Humans become persons through social recognition.⁵⁷ In order to enable the cognitive coordination of their activities, group members evolved “new cognitive skills and motivations of collective intentionality – enabling the creation of cultural conventions, norms and institutions.”⁵⁸ Group members knew about similarities between them and knew that the others knew, too.⁵⁹ Conventional cultural practices were turned into social norms, representing the right and wrong thing to do, and substituting partner control – that is, the sanctioning of cheating – by social control with the aim of creating conformity.⁶⁰ Institutions enshrined social norms and created institutional facts.⁶¹ New forms of cultural agency with respect to the group’s conventions, norms and institutions emerged.⁶² Humans found new forms of self-regulating self-governance based on collective commitments that created obligations and led to moral judgments about others’ moral judgments and to specific moral emotions, such as guilt.⁶³

Group identity was based on similarity; solidarity was therefore felt with in-group members who resembled the actor in behavior and appearance, not with “out-group barbarians” in the vicinity.⁶⁴ The fact that the preferential treatment of in-group members as compared to out-group members already is evident in children is taken to confirm this point: “In-group favouritism accompanied by outgroup prejudice is one of the best-documented phenomena in all of contemporary social psychology …, and it emerges in young children during the late preschool and especially during the school-age period.”⁶⁵

Moral norms were considered legitimate because the members of groups identified with their respective culture, formed a moral identity, regarded themselves as coauthors and felt respect towards the other group members.⁶⁶ Deviation could only be justified by recourse to the shared values of the moral community. These processes led to the second novel form of moral psychology: “It was a kind of scaled-up version of early humans’ second-personal morality in that the normative standards were fully ‘objective’, the collective commitments were by and for all in the group, and the sense of obligation was group mindedly rational in that it flowed from one’s moral identity and the felt need to justify one’s moral decisions to the moral community.”⁶⁷ For modern human beings, the result of this objectification of conventional norms was a “kind of cultural and group-minded ‘objective’ morality.”⁶⁸ As a consequence, human beings harbor three kinds of morality in their breast: the “cooperative proclivities of great apes, the ‘joint morality of collaboration’ and an ‘impersonal collective morality’ of cultural norms and institutions in which all members of the cultural group are equally valuable.”⁶⁹ These kinds of moralities may conflict in dilemmas such as: “Shall I steal for a friend?”

Social norms as objectified conventional norms are moral insofar as they conform with the second-personal morality of sympathy and fairness⁷⁰ – this “natural morality” forms a baseline of human morality, not only in developmental but also in normative terms. Interestingly, Tomasello’s theory incorporates studies that suggest that young children regard norms frequently connected with sympathy and fairness as applicable to all humans, not just in-group members. This does not seem to be easy to reconcile with a simple in-group/out-group dichotomy and the supposedly dyadic nature of second-personal morality.

Cases giving rise to competing moral claims need to be solved on an individual basis using “creativity.” Moral debates are crucially shaped by the second-personal moral principles of sympathy and fairness, what these principles mean and the answer to the question of who is part of the moral community.⁷¹ Exclusion from the moral community was, it is argued, the reason for the justification of slavery or apartheid. The standards of inclusion and exclusion can change.⁷² “Moral entrepreneurs” like Martin Luther King Jr. and Mahatma Gandhi were able to move people by putting suffering and injustice in front of people’s eyes: a “second-person protest writ large.”⁷³

In any case, there is a genuine, not just strategic concern for morality,⁷⁴ although the ultimate evolutionary cause is to increase reproductive fitness – morality ultimately emerged because “early humans who were concerned for the welfare of others and who treated others fairly had the most offspring.”⁷⁵ Moral humans simply had “more babies.”⁷⁶

Children are socialized in the moral community. Second-personal morality is culturally universal, Tomasello asserts, objective morality is not. This can be explained by reference to the ontogenesis of morality: Children first develop a second-personal morality and later the culturally relative social norms.⁷⁷ Humans in general have some “built-in responses to morally relevant situations based on intuition and emotion that have evolved to deal with evolutionarily important situations, especially those in which there is no time for a considered decision.”⁷⁸

Besides the evolution of the principles of morality that are foundational for human rights, rights as a normative category (albeit not human rights) Tomasello explicitly discusses as part of human morality.⁷⁹ In addition, the reflection about fairness and “deservingness” as other considerations seems to imply the idea of the rights of moral patients. Law in general is conceptualized as an institutionalized framework of social integration.⁸⁰ The idea of equal respect, which is foundational for human rights, is also important, albeit somewhat ambiguous: It is limited to a cultural group but can be extended by moral entrepreneurs to all human beings.

Concern for the well-being of others does not mean reciprocal altruism: More demanding principles are involved. The two problems mentioned above – how to explain the first altruistic act and the free-rider problem – indicate the explanatory limits of an account based on reciprocal altruism. The element crucial to overcoming this difficulty, especially the “undermining effect of cheating,” is interdependence.⁸¹

Mutualism (as reciprocity) works at the level of the individual. A stakeholder model is endorsed: Mutualism provides benefits for others, depending on the stake the agent has in the survival (because of the benefits the other provides) of their partner in mutualism.⁸² Group selection is not held to be a major influence, because the gene flow between different groups prevents the development of a genetic makeup that is sufficiently differentiated between groups.⁸³

Tomasello argues, however, for cultural group selection: Those groups with the “conventions, norms and institutions that best promoted cooperation and group cohesion won out, by assimilating or eliminating competitors from other groups.”⁸⁴ Such groups formed collective agents. With the sedentary lifestyles established by the Agricultural Revolution, intergroup competition shifted towards a reconciliation of the moralities of different subgroups.⁸⁵ Gene–culture coevolution can be a factor in this process.⁸⁶ The preconditions for cultural group selection are “species-universal skills and motivations for creating social norms and institutions in the first place,”⁸⁷ as general learning mechanisms are not sufficient to create moral mental categories.⁸⁸ The evolutionary process of moral development was initiated by the shortage of individually available food, a growing population and group competition.⁸⁹

Interestingly, Tomasello posits that similar mechanisms are involved in this phylogenetic trajectory as are at work in the ontogeny of human beings.⁹⁰

7.6.4 Self–Other Equivalence as a Spandrel

Tomasello’s argument contains an important pillar: Self–other equivalence is crucial to the development of morality.⁹¹ It is argued that, at the beginning, this was a purely cognitive judgment that paved the way for the normative “deservingness” of human beings to be treated equally.⁹² This is important, as there is a gap between the judgment that two agents fulfill equivalent roles and the judgment that they deserve equal respect, the former a factual, the latter a normative statement.⁹³

The cognitive judgment of self–other equivalence thus forms the basis for the emergence of new normative principles. It is “moral-structural” in the sense that it is a cognitive structure that could be recruited for moral functions. It is thus, in Gould and Lewontin’s famous terms, a “spandrel”⁹⁴ – an architectural element used for other purposes than the one for which it was built. This is a crucial observation – it opens the door to additional dimensions in evolutionary accounts of cognitive abilities and invites us to take a fresh look at morality.

The theory of interdependence takes the argument very seriously that the evolution of a species motivated to subordinate or equate their own interests to those of others cannot be explained by natural selection. The reference to spandrels sends this “spectre,”⁹⁵ Tomasello highlights, to its final resting place.

7.6.5 Paradigmatic Incrementalism

This theory is arguably currently the most detailed attempt to explain the evolution of moral cognition, relating its findings to current moral philosophy and social science. It stands for a certain research paradigm in evolutionary theory, one that can be called functional incrementalism: Its leading hypothesis is that all cognitive capacities have evolved ultimately to increase reproductive fitness through an incremental process of small evolutionary steps. From this perspective, morality is a functional tool narrowly tailored by natural selection to enable cooperation, albeit enabled by the “spandrel” of self–other equivalence.

The questions that this account raises vividly illustrate some of the challenges encountered by an evolutionary explanation of morality. These questions are thus likewise of paradigmatic importance.

7.6.6 The Analysis of Morality and the Evidence for Evolutionary Incrementalism

A first problem to be discussed is the precise analysis of human morality: As already explained, in morality, there are not only sympathy and fairness, sympathy understood as an inclination to benefit others, but there is also (under qualified circumstances) an obligation to foster the well-being of others. It is not only the commands of justice that have obligatory force. Certain commands of human benevolence do as well: There is an obligation not to let somebody die in front of your porch if you can pick up the phone and call an ambulance. Such actions are evaluated as being morally good. Therefore, they possess a specific deontic status. Moreover, morality concerns not just the behavior, but also specific kinds of intentions of agents. Sympathy and fairness thus do not exhaust the content of morality.

A second problem is that there is no evidence confirming that human moral evolution did in fact take the steps described, that there were first dyadic forms of interaction in which role models emerged, that cooperative forms of foraging lead to sympathy and equal respect, to moral self-regulation and a moral ought, that hominids had a sense of legitimacy and that anything like a second-personal morality existed among them. The same holds for the second step, the development of the objective morality of groups, the exclusion of outsiders based on “similarity,” the individual turning into a “cog in the cultural machine” or the group-mindedness of moral orientation. There is no evidence either that early human beings were like young children in terms of their cognitive capacities. There may be a huge, qualitative difference between the maturing cognitive capacities of the young children of modern humans and the mature capacities of hominids.

Tomasello is entirely honest about the complete lack of evidence that the described stages did in fact exist. The methodological problem of sparse evidence⁹⁶ is acknowledged, and Tomasello highlights that what is presented is not an account of human cognitive development sufficiently based on paleoanthropological evidence, but a “speculative evolutionary narrative.”⁹⁷ It is an “imaginative reconstruction of historical events with little in the way of artefact or paleoanthropological fact to help.”⁹⁸

This comes as no surprise: There are not many clues generally that help us to understand what kind of cognitive skills hominids possessed. Artifacts such as tools are key, as are some indications of behavioral patterns, like signs of cannibalism or burial rites (or the lack thereof). Many cognitive activities simply leave no traces behind. We know from striking pieces of art that early modern human beings evidently had the cognitive capacity to produce art. We do not know, however, whether they engaged in mathematical thinking, in whatever form, because such thinking leaves no traces in the paleoanthropological record, though some artifacts like the Ishango bone raise intriguing questions. This is also true for moral precepts. Whether group life in some hominids included ideas of “self–other equivalence” or not, whether “legitimacy” played any role or not or whether ideas of a moral “ought” structured their thought along the lines of a second-personal morality or not is anybody’s guess, because there are simply no clues that would help to decide these questions. One cannot brush over this state of affairs. It is very important – not to mention an issue of fundamental scientific ethics – to be prepared to leave questions open if there is no way of answering them given the evidence available.

7.6.7 Stumbling Blocks on the Way to Second-Personal Morality

Further important lessons for the understanding of the evolution of morality can be drawn from the details of the “speculative evolutionary narrative.” These concern in particular the conceivable argument that the lack of evidence does not matter as the evolutionary trajectory had to be as described, because this and only this trajectory makes evolutionary sense.

A key question in this respect concerns the transition from one stage of the hypothesized incremental process of cognitive development to the next. There are two dimensions to this process – the development of new skills acquired by individuals and the development of new genetically inherited skills of the species. The first must be based on some kind of learning mechanism and the cognitive ability to create cultural traditions if these skills are passed on to the next generation to learn anew. The second must be based upon genetic change – that is, mutations that inscribe certain cognitive capacities in the human genome.

According to this theory, there is in certain respects a parallel between phylogeny and ontogeny – similar mechanisms are held to drive the development of the species and of the individual forward, such as joint intentionality, role-taking or self–other equivalence.⁹⁹ According to the incremental account of evolution, some of the steps described seem to be taken by individuals, although it is not quite clear how many. In any case, hominids acquire increasing cognitive capacities during the development of the species until the species-specific competencies that Tomasello identifies as the basis of modern humans’ moral ontogeny have evolved.¹⁰⁰

These different competencies are responses to new ecological challenges. Those individuals possessing the properties making them able to meet these challenges have an adaptive advantage that sets the stage for further developments.¹⁰¹ The question is, then, whether one of the stages described leads necessarily to the next. If not, if there are other developmental options, what the evolutionary trajectory really looked like and to what ecological challenges hominids and modern humans adapted remain open questions, given that, as Tomasello underlines, no conclusive paleoanthropological evidence bearing on these questions exists. That any such evolutionary necessity exists is far from clear, however. On the contrary, the apparently continuous development from the dyadic interactions of foragers to human moral codes contains yawning gaps. Crucially, in order to bridge these gaps, certain cognitive resources are necessary that are qualitatively different from those required to engage in other forms of interaction assumed to be prior, without there being any discernible evolutionary necessity that these capacities and no others had to evolve.¹⁰²

To illustrate what is at issue: The development starts, it is argued, in dyadic relations of cooperation. If one assumes for the sake of the argument that there are role models for cooperation (although, as explained above, no evidence is presented either for this or for the consecutive steps), these may give rise to prudential rules (how to be a good forager), but not to moral rules (help other group members in need). Role models derived from certain tasks like foraging are functionally circumscribed, moral rules are not. There is no bridge from one to the other. This problem arises in the next stage of the development as well: Cooperative foraging does not necessarily lead to sympathy for one’s partner, an instrumental relation to them is also possible – and the reality is that there are many forms of human collaboration, from soccer to lawyering, where one can effectively cooperate (play the decisive pass on the field or in the courtroom) without sympathy (because one wants to win, not because one likes the striker or one’s legal teammate).

This problem continues to plague this incremental account: Equal importance in cooperative foraging may not lead to equal mutual respect either. First of all, the roles in cooperative foraging need not be equal but may be very different, according to a division of labor. In modern humans, collaboration does not necessarily lead to equal respect between cooperators – say, the bank teller and the bank’s CEO. The “impartiality of role standards” (if there are any) has the effect that there is an objective yardstick for measuring whether a role-defining task is performed well, but it does not necessarily have the effect that everyone is regarded as being of “equivalent status and importance in collaborative enterprises,” as actors may fail to fulfill these role standards or may simply fulfill a functionally marginal role according to an impartial role standard, leading to a low status – as the low social respect for cleaners in human societies exemplifies (unfortunately, of course). In addition, any such notions of equality could be limited to the collaborative effort and not extend what is supposed to develop, namely equal respect for all. Self–other equivalence in role fulfillment does not help either. Role equivalence (if given) does not entail mutual respect, because the latter encompasses all of a person’s characteristics, not just their ability to fulfill roles. (You can respect somebody as being as good a defender as yourself on your football team and disrespect them in other contexts.)

The assumed legitimacy of such role-derived norms of an “us” indicates another problem. It presupposes (without further argument) that legitimacy was an important concept for hominids. In addition, a kind of social contract, a “joint commitment,” and the idea of the equal status of others are invoked to provide legitimacy – substantial and controversial assumptions about normative reasons. It is implied, not explained, why these reasons were necessarily valid reasons for hominids more than 400,000 years ago – and not just concepts that were alien to the cognitive capacities of hominids or, if comprehensible at all, were as unconvincing as critics of contractarian theory consider them to be today.

Moral “self-regulation” is a demanding cognitive task, including the idea of moral obligations to do or not to do something and the idea of respect for obligations that motivate the agent to perform the obligatory action. The origin of such an “ought” as a central element of self-regulation creates specific riddles. The need to fulfill a role may lead to an instrumental ought (if you want to catch this prey, you ought to approach it without making any noise), but not to a moral ought. Holding others “accountable” – with sanctions, one has to assume – can lead to strategic behavior, conforming to norms when sanctions are likely and not conforming to norms when the chance of being sanctioned is remote, but it will not necessarily lead to moral self-regulation.

How this development leads to joint intentionality is unclear, too. This is no minor point, because joint intentionality is the very precondition for certain forms of collaboration, as the research on the difference between great apes and children shows, not its consequence – another example of the problem flagged above. Again, there are explanatory gaps between the different stages of development. Thus, phylogenetically, there is no evolutionary necessity for one stage to follow the other as described – many other developmental pathways are open, including not developing any of the species-specific capacities of modern humans discussed by Tomasello, as indeed many species, including great apes, have failed to do.¹⁰³ Ontogenetically, similar conclusions must be drawn, because the incremental steps described are, for the reasons just discussed, also not sufficient to explain how individuals ultimately develop the cognitive capacities that are the basis for their normative thinking and culture.

7.6.8 The Objective Morality of Cogs in the Machine

Similar problems are encountered in the second stage, the development from the natural, second-personal morality to the cultural and group-minded “objective” morality, “scaling up” sympathy and justice. To illustrate this with some examples: The perspective of “any rational person” now taken entails many things, not least a concept of rationality. Where does this concept come from? How role conformity leads to a strong idea of universalization and even rationality remains unexplained. Self–other equivalence presupposes that the other is regarded as a person. This seems irreconcilable with Tomasello’s assertion that personhood depends on recognition by others, not recognition of others as equal on the prior understanding of the shared personhood of human beings.

The problems of the relevance of the concept of legitimacy and the reasons providing legitimacy are pertinent here, too. Morality is not just based on a group-minded rationality. According to some elements of the discussed theory, morality is content neutral in a certain respect: If something turns into a group morality, it is obligatory, regardless of its content. Morality, however, is not content neutral: It is about justice and what we owe to others. Accordingly, these norms are not simply obligatory because they are the norms of “us,” but because of their specific content. There are some good reasons to see the set of possible moralities as being substantially constrained. According to the analysis above, this is particularly plausible if one distinguishes between a conventional morality and a reflective morality that is the product of critical scrutiny of moral precepts, ideologies and prejudices inherited from the past. The understanding of the second-personal morality of sympathy and fairness as a baseline of moral systems actually seems to point in the same direction, as it forms a critical yardstick for justified moral norms. However, second-personal morality is conceptualized in too constrained a manner, as we have seen, and its emergence is not sufficiently accounted for.

Then there is the related problem of moral innovation: Many deviations from traditional morality came from individuals who relied on sources other than group morality. Socrates is a leading example of this, as are Jesus and Gautama Buddha, not to mention the many crucially important but forgotten moral innovators in human history. In the light of this, there are compelling reasons to conclude that the “objectivity” of morality is not derived from conformity with a group morality, but is a matter of critical thought and insight, often contrary to group perceptions. Moreover, the history of human morality is not one of intragroup harmony and intergroup strife: The moral battles within what may count as such a group (blurry as this concept is), which were often deadly, teach a different lesson. It is not enough to explaining this strife by referring to subgroups. What is central is a person’s capacity for autonomous moral reasoning that makes them more than a “cog in the conventional cultural practices,” empowering them to become the subject of change. The reference to the need for creative solutions to norm conflicts underlines this point but cannot easily be reconciled with the thrust of the argument emphasizing that objective morality is a group-minded system of rules. Where does moral creativity come from if morality is group based? What are the mental resources of critical moral thought?

The reflective morality that is important in this context is not just that of dyadic second-personal relations. Altruism and justice are not limited to such relations but include every moral patient. This leads us to the next point: The reference to in-group/out-group differentiations certainly captures one feature of human judgment and behavior. But it does not exhaust the principles of human moral judgment, because there are other patterns of judgment and behavior that clearly transcend and subvert this distinction. Again, the theory concedes this point – for example, by admitting that transcending the in-group/out-group distinction happens in dramatic cases such as slavery or apartheid.

It is important to ask: Why do arguments about the inclusion of human beings in the moral community work, as highlighted in the fight against slavery or apartheid? If there truly is a hardwired in-group morality, this success seems hard to understand. If “similarity” of appearance and manners counts, then how does this account for the power of arguments for including people in the group of those who deserve respect, arguments that appeal not to such conventional similarities but to the common humanity of all people? Why is it not only possible but morally compelling to look at others as more than “out-group barbarians”? The task is to explain the reasons for accepting the equal status of human beings as moral patients independently of in-group/out-group considerations and to understand what this tells us about the moral cognition of human beings.

Evolutionary mechanisms affect an individual’s phenotype. Arguments for group selection therefore need to be taken with great caution. Something like cultural group selection is even less plausible. The eradication of the Maya culture by the conquistadores is not an example of cultural group selection for better cooperators, but of the victory of violence and greed using superior technological means over human beings who shared the same properties and possibilities for cooperation (on both moral and other terms) as the conquerors. The (at least temporary) influence of the human rights idea on human affairs is the product not of mechanisms discussed by adaptationist theories of cultural evolution,¹⁰⁴ but of dire historical experience and a normative idea with the power to convince.

A last exemplary point: As in the case of joint intentionality, collective intentionality seems to be the precondition of certain forms of culture, not their result.

7.6.9 Is There an Alternative to Incrementalism?

These criticisms show that the argument stalls on both the phylogenetic and the ontogenetic level: The mechanisms described by the incremental evolution theory cannot plausibly be taken to necessarily produce the cognitive structures that are taken to have developed in the species, and the ontogenetic account appears not to be sufficient to explain the emergence of the cognitive capacities acquired by individuals. The theory therefore in fact formulates a hypothesis about a successive number of mutations that lead to an incremental change in the cognitive abilities of hominids and modern humans, including the acquisition of joint intentionality, self-regulation, obligations mediated through an ought, self–other equivalence, group-mindedness, collective intentionality and so on that are the basis for human ontogeny. It implies that the proximate psychological mechanisms underlying these capabilities became biologically fixed. The ultimate mechanism underpinning the development is that hominids and later humans who cooperate in these increasingly advanced ways have “more babies” and consequently other, differently endowed individuals are “weeded out”¹⁰⁵ by natural selection. The fact that individuals with specific cooperative skills have more babies would not matter if certain proximate cognitive mechanisms were not fixed in the parent generation and bequeathed to their offspring. Otherwise, the next generation would have to learn these cooperative skills anew. The result would be a cultural, not a biological evolution. However, the incremental theory assumes precisely the latter. The account posits that the genetically inherited traits developed in this incremental fashion and proved to be of adaptive value (or to be at least neutral or not sufficiently maladaptive to reduce reproductive fitness), one after the other, and thus persisted in the evolutionary process. Therefore, the newly acquired capabilities are passed on to the next generations. These next generations then move up the ladder, acquiring new, enhanced cognitive skills due to more mutations and passing them on to their offspring, until the level of proximate cognitive mechanisms enabling the formation of an “objective” morality of humans is reached. There is an important qualification: Some of the proximate cognitive mechanisms are understood as “spandrels,” cognitive structures that evolved because of certain nonmoral functions they serve but recruited to new tasks with potentially no (immediate) adaptive value.

All of this means that the baseline of the account is the assumption of consecutive accidental mutations that lead, step by step, to the set of cognitive capacities that ultimately formed the psychological mechanisms that enable human beings to judge and act morally.¹⁰⁶ Some crucial aspects of these capacities are understood as “spandrels,” the primary function of which was not moral. Joint intentionality, for instance, opened up a whole range of new possibilities of human life.¹⁰⁷

As we have seen, the described development does not make unique evolutionary sense on the level of the evolution of the species, so the problem of lacking evidence for this particular trajectory cannot be circumvented.

Moreover, the assumption of a gradual, incremental evolution in small steps in hominids and Homo sapiens is not the only possibility. On the contrary, as we will see in greater detail when reviewing other evolutionary accounts, it is rather implausible given the evidence of the development of complex traits in various organisms and the paleoanthropological evidence about human development. It is in fact widely accepted that after very long periods of comparatively little cognitive development, as evidenced by the lack of major change in tools used, a qualitatively different kind of intelligence emerged with the cognitively modern Homo sapiens. Rather than incremental steps, this indicates a major evolutionary reorganization of the cognitive capacities of modern human beings. How and why this evidence fits into current evolutionary theory is something the following section will explore.

We can already learn something important from these interesting and differentiated arguments about the incremental evolutionary growth of human morality and their critique, however: The incremental succession of mutations, including the creation of spandrels, is not more probable than other accounts allowing for more far-reaching change, let alone the only evolutionary account that makes sense. This is even more so if one stays mindful of the fact that – as the theory underlines itself – there is no evidence that the described stages were those through which hominids in fact passed. It is possible that they are pure imagination.

The cognitive mechanisms enabling human morality, created by mutations that reorganized human cognitive abilities, were the preconditions (not the consequences) of qualitatively new forms of collaboration (not of collaboration as such) and other forms of human thought and action – for example, based on joint or collective intentionality, ideas of self–other equivalence or respect and the morality of altruism and justice that have structured human ways of living ever since. Given the “sparse evidence” of what happened in the history of the human species, the question of the evolution of these cognitive capacities underlying morality thus continues to be an open one. Acknowledging these limits of our current understanding is of fundamental importance for any evidence-based scientific endeavor.

These sobering findings have an important consequence: So far, there is no reason whatsoever to assert that human beings, given what we know about the evolution of human moral cognition, can only have a small-group morality irreconcilable with the ethical principles embodied in human rights. The simple reason is: We know so little about this process that there are no sufficient grounds to come to any such far-reaching conclusions.

The summarized theory therefore does not provide a fully convincing account of the evolution of morality. But what it does do, which is of substantial interest, is help to determine important elements of the set of psychological mechanisms that allow (among other things) for the specific forms of human cooperation, in particular on moral grounds. It helps to distinguish the cognitive apparatus of the nearest relatives of human beings from humans’ mental capacities. It has substantial things to say about the content of morality. There is no explicit reference to the problem of human rights, but it seems clear that rights play a role in the context of the discussion of notions such as “deservingness” or fairness. Morality’s contents of sympathy, fairness and – importantly – equal respect for everybody are in line with the human rights idea. One should take this seriously, at least as showing that contemporary evolutionary theory does not necessarily mean human rights skepticism. These findings are further underlined by other approaches of contemporary evolutionary theory.¹⁰⁸

The theory of the incremental growth of human morality through interdependence makes another important point that shows us which direction to pursue further, given the preliminary results of our reflection: The reference to “spandrels” opens the door to evolutionary theories that take mechanisms of evolution operating in addition to natural selection seriously, among which nonadaptive structures such as spandrels play an important role, albeit not the only one. Such approaches help to paint a much more differentiated picture of the mechanisms of evolution than those which have been reviewed by us so far. These approaches form some of the most influential thought on evolution today and therefore merit closer investigation.

7.7 Evolutionary Pluralism

7.7.1 The Contested Scope of Evolutionary Theory

The review of some influential and paradigmatic approaches to the evolution of morality (which have some direct consequences for the understanding of the law) renders it crucial to answer the following questions: How much scope do evolutionary possibilities offer for the development of human beings’ cognitive capacities? Can any well-defined constraints on the mental mechanisms possible (whatever they may be) be derived from evolutionary theory?

Thus far, our discussion of attempts at an evolutionary explanation of the idea of human rights, including the normative principles that underpin these rights, has provided no reason to assume that this idea is formed contrary to humans’ cognitive nature – because the cognitive faculties underlying human morality are geared to produce a narrow tribal code of proper behavior, for instance. Nor are there reasons to assume that the idea of human rights is a misleading cognitive illusion produced by a hardwired cognitive gizmo. How do these results fit into the framework of the current understanding of evolution?

Evolutionary biology has developed since Darwin, and the transformations of evolutionary thinking in the Modern Synthesis have become a controversial field.¹⁰⁹ Nevertheless, current evolutionary biology appears to provide strong support for our analysis so far. To begin with, it is worth noting that in a considerable number of accounts of the evolution of morality, a set of assumptions play a guiding role: An organism is understood as an ensemble of singular traits. These traits are designed by natural selection for a specific fitness-enhancing function. Given the obvious fact of the interaction of traits, trade-offs are allowed, which are, however, understood as the best possible solutions in the face of the different competing functional demands expressed by natural selection. These are the tenets of what Stephen J. Gould and Richard Lewontin famously called the adaptationist program or the Panglossian paradigm.¹¹⁰

However, some very good arguments suggest that this picture of evolution is too simple. Evolution involves more than just gradual, simple lineage modification through natural selection. First of all, it is not so easy to say what a “trait” actually is, because organisms are “integrated entities, not collections of discrete objects” that allow for an atomization of properties.¹¹¹ In addition, the function that a trait possesses is not necessarily obvious. Take a well-known example: “Bones serve the function of providing rigidity to the body and attachments for muscles. But they also are the sites for the storage of calcium, and the bone marrow is the tissue within which new red blood cells are produced. Depending on the causal pathway of interest, ‘bones’ are either macroscopic structural elements or collections of cells that secrete calcium or embryonic tissue of the circulatory system.”¹¹² Often, there is no evidence available that would allow us to choose between the possible options.¹¹³

Traits that have developed themselves in turn define the conditions of evolution: Without hands that are anatomically able to make tools, tool-making remains beyond an organism’s reach. In order to cultivate fire, an organism needs sufficient body mass to collect enough wood. This is also important for cognitive traits: “[T]he evolutionary questions about cognition are questions both of the evolution of cognition and the effects of cognition on evolution.”¹¹⁴

Furthermore, for theory-building it is important that – as Darwin himself underlined¹¹⁵ – natural selection is certainly a central, but not the only causal factor of evolution. Many other factors need to be considered, as the example of spandrels already indicates. Evolution is, for instance, a stochastic, not a deterministic process. Very different factors that are unrelated to natural selection can influence the evolutionary pathway. Evolution plays out, for example, in finite, not infinite populations. Therefore, sampling effects in small populations – genetic drift – can shape the traits of a given population. Even without the influence of natural selection, because of stochastic inheritance patterns, some traits (including adaptive ones) may become extinct while others are passed on.¹¹⁶

Furthermore, the problem of the stochastic “gravity well” has to be overcome.¹¹⁷ Newly evolved adaptive traits may disappear again because there are only a small number of carriers of the trait that all become extinct before becoming sufficiently numerous to reproduce successfully – for example, because they all accidentally fall prey to some predator:

It is, for example, of little value to be the smartest member of your species, if, in an environment crawling with predators, you are also the slowest – or even just the most unfortunate. What’s more, in an indifferent world your reproductive success may not in the end have much to do with how magnificently you are adapted to any one thing. Whether or not that predator gets you, or whether or not you get the girl, may simply be a function of blind luck and circumstance.¹¹⁸

There are stochastic migration patterns that can lead to hybridization with existing organisms, with evolutionary effects.¹¹⁹

In addition, genes interact in complex ways – as epitomized by Darwin’s observation that blue-eyed cats are deaf. Genetic change may therefore have effects unrelated to natural selection. Natural selection operates on the whole organism – mutations leading to the genotype with the alleles XX may have different effects on fitness if combined with the alleles YY as compared to a genotype XX, ZZ.¹²⁰ Looking at the fitness of single traits therefore possibly misses important factors of development. Other conditions of development have to be considered, too: Fitness levels, for instance, can fluctuate if the frequency of the trait increases, as in the case of overpopulation.¹²¹

Nonadaptive mutations can persevere in the evolutionary process because of such influences. There are also nonadaptive side effects of adaptive properties that are established by natural selection. Architectural constraints are an important limit on the acquisition of new traits – not all developmental paths are even theoretically possible: “[T]here are no animals with wheels presumably because there is no way to make an appendage that rotates on an axle and still can be supplied with blood and nerves.”¹²² Moreover, only a limited amount of what is theoretically possible has been explored by evolution.¹²³ Consequently, the traits of an organism can be determined by path-dependent phylogenetic trajectories that circumscribe the possible future development independently of natural selection and create the oft-observed “inertia” of evolution: “There appear, then, to be basic body plans that are maintained through immensely long evolutionary periods despite dramatic changes in the life activity patterns of organisms and the functions of their parts.”¹²⁴ There are, for instance, no vertebrates with six limbs or insects with eight, not six.¹²⁵ Or, to take another example: “[T]he contingent fact that we have five fingers and five toes may be better explained by an appeal to how toes and fingers develop than that five is optimal for their function.”¹²⁶ There is substantial debate about the reasons for the uniformity of certain common structures of organisms.¹²⁷ In addition, evidently, any development can only unfold within the framework of natural laws (physics, chemistry, etc.) and the physicochemical constraints that they impose.

Some of an organism’s adaptive, nonadaptive or neutral traits can be coopted for new functions – “exaptation” is another factor to be aware of in evolutionary theory.¹²⁸ This term clarifies an ambiguity of the term “adaptation,” which refers either to the genesis of a trait and its reason (features developed by natural selection for their present role) or to current features with fitness-enhancing utility irrespective of how they arose.¹²⁹ Darwin himself noted the existence of traits developed because of “laws of growth” but coopted for other (adaptive) purposes.¹³⁰ Mixing both dimensions might lead to “a common flaw in much evolutionary reasoning – the inference of historical genesis from current utility.”¹³¹ One example of an exaptation that opened up striking new functional possibilities is the evolution of feathers. Finds such as gigantic feathered dinosaurs support the theory that feathers initially served thermoregulation and were only later coopted for flight. Other examples include the acquisition of limbs in a marine environment that later were used for movement on land, or the wings of insects.¹³² After a trait is recruited for a new function, it may evolve further, becoming more adapted to this new function.¹³³ The cooptation of existing traits is of great importance for the evolutionary process: “[T]he enormous pool of nonadaptations must be the wellspring and reservoir of most evolutionary flexibility. We need to recognize the central role of ‘cooptability for fitness’ as the primary evolutionary significance of ubiquitous nonadaptation in organisms. In this sense, and at its level of the phenotype, this nona[da]ptive pool is an analog of mutation – a source of raw material for further selection.”¹³⁴

We already encountered the idea of spandrels, which are unavoidable byproducts of architectural choices – for example, to construct a dome on rounded arches.¹³⁵ Given that they exist, they are put to good use for ornamental purposes and often are so accomplished that it seems as if they are built precisely for this decorative reason. Analogously, traits of an organism can serve a function even though they may have evolved not because of this function, but because of such architectural constraints: “One must not confuse the fact that a structure is used in some way … with the primary evolutionary reason for its existence and conformation.”¹³⁶

It is even less admissible to go a step further and conclude from the assumption that selection could produce only one kind of trait in an organism that this organism in fact has this and no other trait. As explained above, one also encounters this functional fallacy in accounts about the evolution of morality that maintain that human beings have a naturally tribal, small-group morality – not because there is strong evidence that human beings indeed have this kind of morality, but because only such a morality is held to make evolutionary sense.

As indicated earlier, in Tomasello’s account of the evolution of morality we encountered a reference to spandrels at a crucial point of the argument: Proxy psychological mechanisms producing cognitive judgments of self–other equivalence are assumed to have been recruited for moral purposes – which only underlines the importance of these broader evolutionary perspectives.

The evolutionary processes captured by the idea of exaptation are of particular importance when considering the evolution of human beings’ higher mental faculties: As Gould and Vrba note on the debate between A. R. Wallace and Darwin on the evolutionary explanation of the development of the brain, Darwin recognized

that the brain, though undoubtedly built by selection for some complex set of functions, can, as a result of its intricate structure, work in an unlimited number of ways quite unrelated to the selective pressure that constructed it. Many of these ways might become important, if not indispensable, for future survival in later social contexts …. But current utility carries no automatic implication about historical origin. Most of what the brain now does to enhance our survival lies in the domain of exaptation – and does not allow us to make hypotheses about the selective paths of human history.¹³⁷

7.7.2 Nature Does Not Make Leaps, Does It?

A classic topic that already occupied early evolutionary theory is whether evolution is exclusively based on micromutations, as Darwin for instance assumed,¹³⁸ echoing the traditional metaphysical idea that nature does not make leaps (natura non facit saltum). There is conclusive evidence to show that this is an overly narrow conception of what is evolutionarily possible. There are examples demonstrating that rapid change is possible, leading to a very fast divergence of closely related forms, arguably based on small genetic changes – in particular in regulatory genes – that have far-reaching effects, including the classic case of the stickleback, a small fish with spines, or the evolution of the basic elements of eyes (light-sensitive cells, pigment cells), among others.¹³⁹ Thus, the possible evolutionary trajectory of a species needs to include more than just evolutionary accounts of incremental change.

Furthermore, the evolutionary theory of human cognition faces particular challenges. One such challenge we have already encountered: the lack of evidence for the existence of particular cognitive capabilities of early human beings. If certain kinds of artifacts are found, say a tool or a flute, we can be sure that the cognitive abilities necessary for making this kind of tool or this musical instrument must have been in place. So, too, must other capacities – for example, those making it possible to identify the need for tools, or the capacity for purposive functional inventions, for using tools or for producing and enjoying music. But how are we to know whether a group of modern humans sang a song while making the tools or the flute? What were their social relations like? How did they communicate? Did the members of this species tell tales while sitting at campfires? Did they use arithmetic to keep track of the quantity of their provisions? Did they possess a sense of justice that guided them when dividing up prey? Such cognitive activities do not leave any traces behind. So how are we to know?

Giving an evolutionary description of the developmental process of human cognition is equally difficult. It is far from clear who the predecessors of modern humans are, as the paleoanthropological record contains insufficient evidence to determine the lines of relationship between species and distinguish ancestral lines from those forms that are not ancestors.¹⁴⁰ It is equally unclear what cognitive capabilities any of the species that may be part of the lineage of modern human beings possessed, as here, too, the problem of evidence for such capabilities rears its head. The split between Neanderthals and modern humans happened 400,000–600,000 years ago. Tool-making was typical of Neanderthals.¹⁴¹ Recent research on Neanderthal artwork indicates symbolic thought on the part of this certainly cognitively gifted species, but the evidence is controversial and very sparse. And, whatever the results of future research may be on Neanderthal symbolic behavior, there is no evidence of any systematic use of symbolic thought – in contrast to modern humans.¹⁴² We have already observed that the closest living relatives of modern humans are primates, chimpanzees and bonobos, about 14 million years of evolutionary time away from modern humans – a substantial amount of time in which to develop differently. As animals adapted in crucial aspects to life in forests, and in some cases adapted behaviorally to life in the savannah,¹⁴³ there already is a difference from early hominids, who – enabled by their new body form – moved out of forests and into other environments and modes of life.¹⁴⁴ The number of existing “close” relatives is sparse, which makes it difficult to determine a trait’s successive changes: “The evolutionary space is too sparsely populated to be able to connect the points sensibly.”¹⁴⁵ In addition, the paleoanthropological evidence suggests that early hominids developed cognitive skills that are not available to primates living today: “The evidence of tool use, and yet more of tool-making, tells us that the bipedal apes had graduated – perhaps as much as 3.4 million years ago, and at least before 2.6 million years ago – to a cognitive state that lay beyond anything we can infer for the apes as we know them today.”¹⁴⁶ Tattersall argues that early hominids’ “entirely new and radical way of interacting with the world around them” had its roots in an early “leap of nature” and later exaptation of cognitive skills. The best explanation is that “the cognitive potential to make stone tools was born in the large genetic alteration that must have been involved in the acquisition of the new and radically different bipedal body form; and that this potential lay dormant for some time before being expressed in the invention of stone tool making.”¹⁴⁷ The fact that evolution is not a continuous, linear, incremental process creates specific problems for the understanding of the evolution of cognition, because it is possible for closely related forms to diverge very rapidly.¹⁴⁸ The similarity of traits can vary from trait to trait – an organism may be similar in some respects but quite different in others.¹⁴⁹ And there is another crucially important possibility, namely that the trait “simply does not exist in some or all related lines, that it is a novelty, and so has no observable evolutionary history.”¹⁵⁰

Furthermore, there is the functional change of traits – less closely related organisms may share more properties with certain organisms than with others that are more closely related, as illustrated by an evolutionarily convergent system such as vocal learning in songbirds and humans, which are separated by millions of years of evolutionary time.¹⁵¹ There is a need to differentiate between homologous and analogous structures and to identify homologies, especially because homologous structures may change functionally and because of the abovementioned possibility of novelties without homologies in other species.¹⁵² In any case, as in other organisms, it is not singular traits that determined the survival of modern humans but the human organism as a whole, with its many specific adaptive and not-so-adaptive features and their complex interactions,¹⁵³ which renders an account of the evolutionary history of any one element of this integrated organism even more difficult.

All of this arguably leads to the conclusion that there is simply not enough evidence to ever decisively settle the question of how exactly modern humans’ cognitive capacities came into being over the course of evolution.¹⁵⁴

It goes without saying that any evolutionary theory of the development of human cognition has to take into account the existing evidence of human cognitive development. There is widespread consent, given the current stage of knowledge, that around 200,000 years ago anatomically modern humans were living in Africa.¹⁵⁵ About 80,000 years before the present day, there is uncontested evidence of symbolic behavior such as beads, shell ornaments and geometric engravings, though there are some earlier artifacts.¹⁵⁶ This indicates that at least from this time onwards, modern humans had the cognitive capacity for symbolic thought and used it systematically. Other early indicators for new cognitive abilities are bladelets (stone flakes sunk into handles), making tools from artificially hardened materials, pressure flaking, barbed harpoons and, on the social level, the functional division of living spaces.¹⁵⁷

The capacity for symbolic thought is plausibly taken as an indicator that cognitively modern humans had arrived on the scene: With modern humans, an unprecedented and rapid development set in, with cultural and technological innovations that transformed the world – and by now are threatening the continued existence of this simultaneously creative and destructive species. This development is dramatic proof that these beings possessed some revolutionary new creative abilities in both thought and action. Before that, over the millions of years of development there had been many striking innovations like tool-making, the use of fire, spears, composite tools, “prepared core” implements and blades.¹⁵⁸ However, there was a disconnect between technological innovation and the development of a new species of the Homo genus.¹⁵⁹ Technological innovations persisted unchanged for hundreds of thousands of years and were of very limited scope in comparison to what happened later, although they were very remarkable innovations in themselves, beyond the cognitive capacities and other abilities of great apes:¹⁶⁰ Mode 1/Oldowan tools (sharp stone flakes) were invented 2.5 million years before the present and mode 2/Acheulean hand axes 1.5 million years before the present, although they were widely applied only hundreds of thousands of years later. These tools continued to be used even though new species like Homo ergaster developed during this time.¹⁶¹ There is consequently no necessary connection between the appearance of a new kind of hominid and technological innovation, nor is there evidence of any gradualism in the appearance of technologies for long stretches of time.

The cognitive abilities evident in Acheulean hand axes are already remarkable – foresight and the planning of complex sequences of actions, the knowledge of materials, anticipating need by keeping stocks of suitable stone, among others. Hand axes even indicate economic and social specialization.¹⁶² Later developments furnish further examples of the advanced cognitive skills of the various branches of hominids. But with Homo sapiens, change starts to take place at breathtaking speed. In just 80,000 years the species traveled from hand axes that had already been used in different forms for over 2 million years to the technical civilization of today. This change cannot but be based on cognitive abilities of modern human beings that are qualitatively new in comparison to those of their predecessors, abilities that have increasingly been put to use.¹⁶³ Symbolic thinking and action is one such ability, in itself of far-reaching importance for culture-building and science. Technical skills also point clearly to certain abilities – for example, the analysis of complex causal chains as a precondition for developing new technologies in the making of tools and weapons. The existence of art is an indicator not only of symbolic thought, but also of a very rich inner life of early humans, including the search for meaning and beauty – and this tells us a lot about the species’ cognitive capacities. In the case of other cognitive abilities, there is only little evidence until much later in history. Theory-building therefore turns to the second-best option: interpreting the emergence of symbolic thought and action as a proxy for the existence of other cognitive abilities that are crucial to being human but leave no material evidence behind.

Language is at the forefront of many debates, as an instrument of complex thought and communication,¹⁶⁴ but any other aspect of the human mental world qualifies as well, of course. One example is human moral understanding and judgment. There are some indications of forms of care for others in hominid history: One example is a 1.8-million-year-old skull of an elderly male who already lost his teeth during his lifetime but survived nevertheless. This seems to indicate help from members of the group to which he belonged, although other possible explanations also exist. If help was indeed provided, the skull “furnishes us with the first putative instance of social concern in the hominid record.”¹⁶⁵ Other traces are (different forms of) cannibalism or burial practices that indicate concern (or lack thereof) for species members.¹⁶⁶

Beyond this, as for other possible mental abilities, there is no direct evidence of the exact kind of cognitive capacities existing in hominid history that constituted the foundation for the moral world in which modern humans lived, answering the question of which concepts and principles they applied, what moral feelings they entertained and how these were spelled out in social practices and institutions. Did modern humans have a sense of fairness? Did they experience shame? Did they feel bound by obligations?

If one accepts symbolic thought as a proxy for a fully developed modern human cognition because of the reasons outlined, one has to conclude that early modern humans possessed the same cognitive abilities to form moral ideas as contemporary humans (whatever these abilities may turn out to be and to whatever use they may have been put).

When considering these problems, it is useful to distinguish between the evolution of cognitive capacities (human beings have the cognitive capacity to produce symbolic artifacts, primates do not) and the development of skills and insights within the framework of these evolved cognitive abilities. This is particularly important when considering the human species. Evidently, the great historical change that gave rise to the transformative creative innovations highlighted above is a central element of the history of modern humans. There is, however, no evidence of any evolution of modern humans’ cognitive capacities themselves that underpins this striking history of change and innovation. Therefore, it comes as no surprise that one could transfer a child born in a contemporary tribal society whose way of life is similar to a Stone Age culture to Zürich, New York or Beijing and know that it will develop in the same way as any other kid in the world, becoming unhealthily attached to its smartphone at far too early an age.

It is consequently reasonable to conclude that modern humans’ basic abilities in the cognitive domain have stayed unchanged since their appearance at least 80,000 years ago. If this is indeed the case, we can project backwards whatever we know about the structure of the mind of humans living today: If human beings today are distinguished by a general learning ability, then it is plausible to assume that this formed the basis of their early development as well. If they possess the ability for shared intentionality, then plausibly early modern humans had this capacity, too. If there is evidence for other additional or alternative cognitive abilities – say, a language faculty – then most probably language was among the cognitive tools with which modern humans started their journey. Accordingly, the current cognitive capacities of modern humans can be taken as a clue to the mental parameters of life in the remote past.

One should note that making this assumption is justified because it is the most plausible interpretation of what we know at present. Nevertheless, within this general framework, much is unclear. Regarding language, for instance, one prominent account in fact speculates about an evolutionary pathway along these lines: First the semantic–conceptual system of language evolved, then the systems of the externalization of language through speech and other means such as signs.¹⁶⁷ This is relevant for the assessment of the cognitive abilities of all hominids, including Neanderthals, for instance: “It is merely assertion that complex stone working, fire control, clothing, ochre, and the like require language. We may have them all, but that does not mean Neanderthals had to have all the features that co-occur in us just because they had some of them.”¹⁶⁸ This cautionary observation underlines that confident assertions about the precise path of the evolution of human cognition are simply not possible.

There is another problem: For many such cognitive traits, we do not know what kind of heritable variations existed, and it is difficult to determine their reproductive function for the first individuals possessing the trait within a species, which is decisive for these traits to become established.¹⁶⁹ Nor do we know whether and how these cognitive traits led to the replacement of other species. Take language, a central topic of evolutionary research. Despite the large number of contributions to the theory of the evolution of language, the function of language is less clear than it may appear.¹⁷⁰ A cofounder of evolutionary theory, Wallace, famously wondered what function human language may serve, because he saw no biological function that could not be met by other, simpler means than human language¹⁷¹ – a problem still relevant today (although Wallace’s proposed solution, divine intervention, may now be less accepted).¹⁷² The proposals entertained today range widely, including language as a tool of successful cheating, of sexual selection, of communication for the purpose of cooperation or of structured thought, in the latter case a tool that evidently is still very useful.¹⁷³ Moreover, it is impossible to determine whether the function of such cognitive traits had any substantial effect on reproductive success: “[W]e cannot measure the survival advantage, if any, in our remote ancestors of the ability to do arithmetic.”¹⁷⁴ Thus, one cannot determine whether any cognitive trait that might have had an effect on reproductive rates did in fact have an effect on reproductive rates.¹⁷⁵

Such problems need to be considered for morality, too. Experiencing altruistic obligations in a group with members devoid of such inclinations may simply make the respective individual easy prey. The functional value of morality for collaboration is not obvious either, as, for instance, it precludes effective forms of collaboration such as anthill- or beehive-like human societies. Collaboration for survival is not only possible on moral terms.

In this context, too, we should stay mindful of the observation that just because a trait is used in certain ways does not mean that it evolved because of this function. The fact that morality makes possible certain sophisticated forms of collaboration based on solidarity and respect does not mean that it evolved because of this function. The mental abilities creating the possibility of moral thought might have any of the evolutionary origins mentioned and form, for instance, a side effect of an (overall) adaptive cognitive reorganization or an example of exaptation, a spandrel perhaps, which turned out to be a major influence on the modern human life form.

There is no reason to assume that any account of human moral cognition has to satisfy the constraints of the adaptationist paradigm, namely to be favored by natural selection as an isolated trait within the framework of functional trade-offs between different traits. It is entirely possible from the point of view of evolutionary theory that various elements of human cognition are structures that evolved without any clear adaptive advantage but still define the kind of creatures we are.

On all accounts, the paleoanthropological timeline quite clearly indicates a case of rapid evolutionary development that led to a fundamental cognitive reorganization of human beings – albeit perhaps caused by minor genetic change, probably of regulatory genes, which had far-reaching consequences. In comparison to the 1 million years that passed between the invention of Oldowan stone tools and the designing of Acheulean hand axes as a major technological innovation, the development of our current civilization in a mere 80,000 years has been fast indeed. As this rapid change did happen, there is a pretty clear answer to the question of whether it could have happened.¹⁷⁶ With modern human beings, nature simply did make a leap.¹⁷⁷

The conclusion to be drawn regarding the evolution of whatever psychological capacities enable human morality is, then: As things stand, there are good reasons to believe that these cognitive capacities were part of the set of faculties with which modern humans embarked on their journey at least 80,000 years ago – faculties that define modern humans and have made their journey since then human history.

7.8 The Evolutionary Possibility of Human Goodness

This leads to the crucial lesson for the cognitive interests of this inquiry: Evolutionary theory provides no discernible argument as to why the cognitive faculties yielding both the idea of human rights and the identified justificatory principles of this idea could not be the products of evolutionary processes. It may have a “wonderful reductionist appeal” to think that the cognitive abilities of modern humans are determined by the living conditions of early modern humans¹⁷⁸ – for example, yielding a “small-group morality.” But this reductionism does not do justice to an evolutionary theory aware of the complexity of the human mind. Striking things evolved, such as the faculty to create a phenomenon as breathtaking as art,¹⁷⁹ and there is no reason to think that the ability for moral reasoning, motivation and emotion may not be a further example of an equally striking quality. There are no compelling grounds to restrict what is evolutionarily possible to such a “small-group morality” or any other form of psychological mechanisms that make the idea of human rights alien to human cognition.¹⁸⁰ Nothing in what we know about the history of the evolution of cognition, nor anything in evolutionary theory speaks against the possibility that human beings possess a sense of justice, evaluate genuine care and respect for others as morally good, regard just and morally good acts as obligatory and embed moral judgments in a rich, sometimes painful, sometimes ravishing world of moral emotions.

The answer to our initial question on the origin of the cognitive capacities enabling the moral world of human beings is thus: There are many riddles hidden under the human skin, as Ahab despairingly realizes, and the attempt to hunt down the elusive essence of human existence can even lead to a self-destructive chase. But this much is clear: The scope of evolution is certainly wide enough to include cognitive mechanisms (whatever they turn out to be) that enable human beings to form notions of justice, concern and respect for others and, after a long development, ultimately articulate the remarkable idea of human rights.

8 The Mentalist Theory of Ethics and Law

When a man denominates another his enemy, his rival, his antagonist, his adversary, he is understood to speak the language of self-love, and to express sentiments, peculiar to himself, and arising from his particular circumstances and situation. But when he bestows on any man the epithets of vicious or odious or depraved, he then speaks another language, and expresses sentiments, in which he expects all his audience are to concur with him. He must here, therefore, depart from his private and particular situation, and must choose a point of view, common to him with others; he must move some universal principle of the human frame, and touch a string to which all mankind have an accord and symphony.

David Hume, An Enquiry Concerning the Principles of Morals

No doubt wickedness hath its rewards to distribute; but whoso wins in this devil’s game must needs be baser, more cruel, more brutal than the order of this planet will allow for the multitude born of woman, the most of these carrying a form of conscience – a fear which is the shadow of justice, a pity which is the shadow of love – that hindereth from the prize of serene wickedness, itself difficult of maintenance in our composite flesh.

George Eliot, Daniel Deronda

8.1 A Fresh Look at Frameworks of Morality

Our discussion thus far has shown that a solid analytical theory of morality is crucial as a starting point for further theory-building about the nature and origin of moral cognition. Once this analytical theory of morality has been achieved, we can attempt to reconstruct the psychological mechanisms underpinning human moral judgment and investigate how they are acquired. There are no a priori constraints on the ways that human morality could be structured, and in particular no constraints that can be derived from the theory of evolution. What plausibly can be taken as constitutive of human morality is simply a matter of the evidence given. Therefore, any account of morality’s constitutive elements is well-advised to take seriously the rich debates of moral philosophy, whose insights must inform plausible theories of morality.

In order to develop an analytical theory of morality, one needs to identify the building blocks of the human moral world. Such a theory must be based on an analysis of the practice and phenomenology of moral judgment. It needs to tell us what humans actually do when they exercise their moral understanding. One complication of such a study – and a major one at that – is the fact that moral judgments are intrinsically contested. Which moral judgments form the basis of theory-building? The views of a misogynist racist or those of a female Black Lives Matter activist?

One plausible way of proceeding is to look at qualified moral judgments. These judgments have to be “considered judgments,” to borrow a useful term, in the sense that they are reflective, dispassionate judgments that are not skewed by interest, passion, errors of fact and so forth.¹ This methodological move has as its background the distinction between competence and performance, the faculty to perform a certain cognitive task and the actual performance of this task.² Humans have the competence to construct an image of the external world using the specific structures of their visual cognition. This does not mean that the effects of imbibing a certain amount of alcohol may not affect the functioning of this competence and make what is perceived appear strangely blurred. As this example shows, only indirect conclusions about an agent’s competence can be drawn from their performance, because the performance is influenced by many other factors than just the structure of their competence. There is no doubt that the loud sound of techno music will influence the mathematical problem-solving capacity of people exposed to it. However, nobody would ever entertain the idea that techno music is of great relevance to studying the cognitive apparatus enabling humans to do math. The capacity for moral evaluation is another such competence usually possessed by human beings. Nevertheless, the performance of this capacity, the final evaluation of an action can be biased – for example, by the interests of the evaluating person. Consequently, such influences need to be factored out of the analysis if we are to properly study the cognitive competence in question, which is not an easy thing, particularly in empirical work. This crucial issue is sometimes overlooked in recent moral psychology studies, which claim to be studying human moral competence, but in fact to a surprisingly large degree are concerned with performance problems, such as the skewing of moral judgment by nonmoral factors, from smells³ to the feeling of being controlled.⁴ Another methodological approach to dealing with the contested nature of moral judgment is to look at highly idealized and often artificial cases that appear to be as little politically and culturally loaded as possible.⁵ By contrast, studying the human moral faculty by looking at opinions about issues as contested and ideologically charged as abortion, for example, is a methodological nonstarter.

One preliminary result of our discussions so far has been the observation that some concepts of morality in contemporary debates are too narrow, both as to the substantive material principles of morality and as to the subjects of moral concern. In particular, morality is neither an ultimately selfish enterprise seeking to reap the profits of in-group cooperation, nor is it simply a set of preferences or aversions. The analysis below will flesh this out in more detail.

One theoretical approach that explains how these findings may fit into a theory of moral cognition is the so-called mentalist approach to ethics and law. A mentalist model of moral cognition investigates the question of whether it is possible to identify generative principles of moral judgment specific to human moral cognition that are universal and uniform across the species – a universal moral grammar, if you will, to use a metaphor sometimes employed to capture the basic intuition of this approach.⁶ The mentalist model has been a very influential research paradigm in other domains of the theory of mind – for example, in the study of language – stirring many intense controversies and debates.⁷ The thrust of the mentalist argument echoes a long tradition of thought on moral understanding – after all, the belief that human beings are endowed with a particular faculty of moral evaluation has been one of the thoughts guiding moral philosophy ever since antiquity.

8.2 Some Properties of Moral Cognition

8.2.1 The Cognitive Space of Morality

A close and careful look at the phenomenology of morality gives rise to some important observations.⁸ One is that something like a moral experience exists at all. Humans operate naturally within a mental space that has a normative dimension. There is a specific mental domain of morality, a qualitatively distinguished element of conscious thought, an introspectively accessible, distinctive, intuited, subjectively experienced aspect of our mental life, a qualia, as it is often said, or – to use standard understandings of this term – a certain phenomenal character of certain forms of experience. The availability of such a cognitive domain is not self-evident; rather, this domain represents an empirical property of the human mind that not all organisms share. Human beings do not perceive ultrasound, but bats do. Human beings see the world in the distinct colors of morality, but bats do not. This cognitive domain does not concern a side issue but defines nothing less than an element central to the identity of the human species: the moral dimension of human lives. Its existence consequently merits close attention.

A further interesting observation concerns the fact that there is a highly and intricately qualified limited set of possible objects of moral evaluation. This set already restricts the kinds of morality that are possible. The dropping of an apple from a tree into the hands of a hungry person is not a virtuous action on the tree’s part. Or, as Hume observed: “A young tree, which over-tops and destroys its parent, stands in all the same relations with Nero, when he murdered Agrippina,” but does not commit matricide.⁹ This is because agency is a precondition for the moral evaluation of certain events in the world. If these events cannot be attributed to agents, questions of moral evaluation do not arise.¹⁰

Moreover, placing your pen gracefully on your desk cannot be the possible object of moral evaluation either (except under very particular circumstances), even though an agent performs this act. However, this act is a possible object of aesthetic evaluation – another distinctive element of human experience.¹¹ Kicking a ball so as to feel like Dzsenifer Marozsán for a few precious seconds is morally very different from kicking a defender or a dog that is in your way. A precondition of moral evaluation is thus – to put it very roughly – something like volitionally controlled or controllable bodily actions or omissions of agents with consequences for the well-being of sentient beings and other qualified objects of moral concern; intentions concerning such actions or omissions; states of affairs resulting from such intentions, actions and omissions; or qualified emotions directed at the well-being of others.¹²

8.2.2 Principles of Morality

As we already have seen, in analyses of morality it is particularly important to distinguish behavior and motivational inclinations to behavior from the moral evaluation of this behavior and the intentions underlying it. Concrete behavior, including what is called prosocial behavior, and the inclination or preference for such behavior are one thing, the reflexive appraisal of this behavior, inclination or preference with deontic dimensions quite another. Only the latter falls within the proper realm of morality and ethics.¹³

If we turn to the content of morality and carefully analyze some qualified (only seemingly simply structured) considered judgments of the kind described above, we see that these judgments seem to be guided by principles of egalitarian justice and altruism across a wide range of cases. That it is just to distribute party favors equally to the young guests of a child’s birthday party seems as uncontroversial as that it is a morally good deed to help prevent starvation in Yemen.¹⁴ Respect for others is a further important principle. This state of affairs is not particularly surprising, as all of these principles run through the history of ideas, too, as the core of morality, across cultures and millennia, albeit accompanied throughout by influential skeptical voices from Thrasymachus to Nietzsche and beyond, who have argued (with greater or lesser philosophical sophistication) that morality has no content that lends itself to rational reconstruction and that apparently core normative concepts do not mean anything at all.¹⁵

Substantial empirical work carried out in recent years points in the same direction, supporting the observation that certain identifiable elements – more precisely, egalitarian and altruistic principles – seem to play an important role in the evaluation of the justness and goodness of actions, despite some theoretical limits of parts of this research that already have been reviewed above.¹⁶ Other frontiers of research include the prohibition of the instrumentalization of human beings that plausibly lies at the heart of a proper analysis of the extensively empirically researched trolley problems and distinctions between different subjective cognitive and volitional attitudes towards potential actions, including direct and oblique intentions and their relevance for human moral evaluation.¹⁷

If we analyze the empirical work on justice and basic uncontroversial judgments about the justice and injustice of intentions, actions and states of affairs and do not forget what the struggles of social history seem to suggest about ideas of justice, then differentiated principles of equality as already discussed above¹⁸ seem to have considerable explanatory power to account for many patterns of moral evaluation. We have said that a just distribution demands proportional equality between the value of the specific criterion of distribution reasonably related to a particular sphere of distribution on the one hand and the amount of the good distributed on the other. Other normative demands concern the equal treatment of persons, restitution and respect for the equal worth of each human being – as indicated, the ultimate yardstick of just treatment and a just state of affairs. Such egalitarian principles match the empirically identified patterns of moral evaluation recalled in this study, including standard examples such as the ultimatum or dictator game, the many debates about the proper interpretation of these findings notwithstanding.¹⁹ The same also holds for other clues to the moral world in which human beings live beyond experimental data, not least social history and its many egalitarian struggles, some of which we have recalled. It should be noted that a main bone of contention in these political struggles was not the principle that equals ought to be treated equally, but the criteria of distribution and who and what actually fulfill these criteria. Regarding rights, for instance, an important question in history was whether fundamental rights are distributed in a society on the basis of the bearer’s humanity or some other criterion (say, aristocratic birth) and who fulfills this criterion (for instance, whether women are fully human or some kind of deficient being and thus not entitled to a full set of fundamental rights). If we pay due attention to such factors, the core of the human quest for justice becomes considerably more transparent.

To achieve sufficient explanatory depth, however, the analysis of these empirical observations needs to remain aware of the distinction between a moral competence and its actual use, the plurality of motivational factors influencing human action, including nonmoral interests and the complex structure of moral judgments, with their cognitive, volitional and emotional components (to be discussed in more detail below), which are not reducible, for example, to preferences or aversions.

Whether there is such a thing as genuinely other-regarding altruistic behavior or whether any action beneficial to others is ultimately motivated by some (albeit perhaps refined and hidden) self-interest of the agent is one of the traditional questions of practical philosophy.²⁰ As in the case of justice, this is a huge debate that today is enriched by interesting empirical work.²¹ In this context, it is important, too, to rely on a sufficiently complex theory of morality, in particular to distinguish between actual behavior and considered evaluation and thus distinguish the question of whether people are in fact acting because of a genuinely altruistic motivation from the question of whether genuine altruism is the precondition for evaluating something as morally good. There is not much reason to believe that people generally excel in altruistic behavior. However, this observation tells us nothing about the principles that guide moral judgment – for example, when evaluating the selfish behavior prevalent around us (including our own).²² Concerning these principles of evaluation, there are reasons to think that such a genuinely altruistic motivation is in fact a core element of moral evaluation. More precisely, it seems plausible to assume that an action is morally good if it is performed with the direct intention (or purpose), not only the oblique intention (or knowledge), to foster the well-being of the patient. If this is so, it is irrelevant for the moral evaluation whether or not the fostering of the interests of the agent is – at the same time – a directly intended or foreseen (obliquely intended) consequence of the action and forms a second reason for action in a bundle of motives. The direct intention to foster the well-being of the patient of the action appears to be a necessary condition of morally good action.²³

To illustrate the meaning of this principle, it is useful to look at one of the most refined versions of ethical egoism. This form of egoism holds that altruistic behavior is ultimately motivated by the desire to experience the satisfaction of having acted in a morally appropriate manner. This argument makes an important point, namely that moral action does indeed provide some particular form of satisfaction for the agent and that agents are certainly often aware of this. In addition, acting immorally can have unpleasant effects, too, such as feelings of shame. These observations do not settle the issue, however. Consider the following case: Pawel helps Mio, thinking: “I do not care for this person and her well-being at all (what a silly person she is, in fact!), it just happens (unfortunately) that I have to do something for her in order to reap the sweet fruit I really desire, namely to feel the satisfaction of being a truly nice person!” Is this really a morally laudable deed? If doubts arise about the moral praiseworthiness of an action with such an intention, it seems to confirm the analysis above. This is because the Pawel has only an oblique intention to help the other person and not the direct intention to be beneficial to her: His direct intention is to satisfy one of his own personal desires, and helping the other person is only a (perhaps even unwelcome) means to achieve that end.²⁴

Another point is perhaps worth noting: Justice seems to be something like a limiting condition of morally good action. There are reasons to believe that there is no morally good intention that violates principles of justice. If Pawel, for instance, helps three out of four people in need, not because he cannot help them all, but just because he feels like excluding one person on a whim, this is not a morally good action, despite his direct intention to help the other three, because it violates principles of equal treatment.

A third principle is respect for human beings. Respect in a moral sense must be distinguished from admiration for achievements of other kinds – for instance, for a skillful free kick right into the corner of the goal. Unlike admiration of this kind, respect in a moral sense has prescriptive consequences: One ought to behave towards other human beings in certain ways. One important element is to treat them as ends-in-themselves and not only as means to achieve other purposes, to once again use the Kantian version of an ancient idea. Evidently, reducing somebody to a means of some ulterior purpose degrades this person. As we saw when discussing the trolley problem, substantial empirical data suggest that people do in fact apply this principle when evaluating certain morally salient situations: The recorded moral evaluations of the footbridge scenario and its most promising analysis show that these evaluations are best explained by an operative principle demanding that people not be instrumentalized based on a structural means/ends distinction.²⁵

Furthermore, there are other forms of disrespect for human beings that do not instrumentalize them: Regularly flooding a prison cell with feces, not to torture the inmate but simply due to negligence concerning the sanitary conditions of inmates, is an illuminating example of this from constitutional case law.²⁶

The importance of recognition as a being of equal worth has rightly been flagged as a key element of social struggles:²⁷ These struggles are not only about bread-and-butter issues, but also about exploited and degraded groups of people’s demand to be respected as human beings. This is a central ethical and political dimension in the fight against slavery, old and new, and in the fight against the subjugation and exploitation of women. The same holds for struggles for the emancipation of the working classes, which likewise were not only about material concerns like decent wages, but also about respect for the humanity of workers, who demanded (and indeed still demand) that they not be reduced to beasts of burden.

These findings help us to tackle some of the problems that the supervenience of moral judgments over facts poses.²⁸ They help us to identify the principles that determine, first, the moral evaluation and normative consequences of moral judgments that are triggered by certain facts (e.g. by the fact of a direct intention to harm somebody) and thus to determine how moral judgments supervene over facts (not just that they do) and, second, which facts are morally relevant in the first place (e.g. that the agents’ intentions to act matter morally but not their haircuts). These tentatively outlined principles, which are obviously in need of much refinement, are abstract but not without meaningful content, as can be exemplified by the justification of human rights, as we have seen.²⁹ To recall: The principles underlying the attribution of rights to persons have to be equal for all potential rights-bearers. It would, for example, be unjust to let some people enjoy fundamental rights because of their personhood and deny these rights to others because for them the color of their skin (and not their personhood) is taken to be relevant. In addition, the reasonable – more precisely, the only reasonable – criterion for the attribution of rights is a person’s humanity. As a just system of rights has to preserve a relation of equality between the value of this criterion and the distribution of rights, and as all humans are equal in their humanity, only a system of equal rights is consequently a just system of rights. The theories of human rights that imply a diminishing humanity of older people and, correspondingly, a diminishing set of protected rights, which may seem to be at odds with this principle, actually confirm it: These theories simply entertain the (implausible) idea that the humanity of elderly people diminishes, even though the elderly, of course, enjoy the same humanity and consequently the same rights as every other human being – no minor point, as the Covid-19 pandemic has reminded us.

Moreover, fostering the enjoyment of rights is morally good. Accordingly, as has been said before, given the importance of the goods that rights protect and the significance of the rights themselves, the promotion of rights is a prima facie obligation of human solidarity. The theory of the justification of human rights has shown, too, that one of its building blocks is respect for human beings, today widely understood as grounded on their dignity. In sum, justice, the altruism of solidarity and respect for the worth of others take us a long way when trying to identify the normative principles that are key to the justification of human rights.

8.2.3 Basic Harms, Human Rights and the “Seeds of a Collective Conscience”

Human rights play an important role in John Mikhail’s pioneering work on the mentalist framework of ethics and law.³⁰ His reconstruction of a set of principles of a “universal jurisprudence” based on common moral precepts that are “the seeds of a collective conscience”³¹ includes reflections on human rights that constitute some of the most advanced thought on the topic of our inquiry. His main thesis is that the prohibition of basic wrongs that is an element of the inborn structures of human moral cognition is mirrored in fundamental human rights norms.³² The universal moral grammar “implies that human beings possess tacit or implicit knowledge of specific, human rights-related norms.”³³ The small set of principles that guide human moral judgments include the prohibition of intentional battery, the prohibition of intentional homicide, the rescue principle and the principle of double effect.³⁴ Of particular importance for human rights are the prohibitions of intentional battery and homicide, which are “lesser included offenses of a wide range of human rights abuses, including murder, extermination, deportation, torture, rape, genocide, and other crimes against humanity.”³⁵ Human rights protect against a set of “core human wrongs” like battery and other torts, “although they by no means exhaust these violations.”³⁶ A “clear conceptual and empirical bridge between moral grammar and human rights can be built” because “[t]hese basic wrongs and, in particular, the tort of harmful battery likewise supply the basic perceptual and cognitive tasks of many influential research programs in the cognitive science of morality.”³⁷ Mikhail’s work on the trolley problems is itself an outstanding example of this.³⁸

These remarks lead to three questions that can be addressed fruitfully in the light of the results of our analysis.

The first is the question of foundational principles. Our analysis suggests that the principles of justice, altruism and respect that are the normative core of our normative theory of human rights are consistent with Mikhail’s analysis but seem to have additional explanatory power. As indicated, the obligation to foster the well-being of others is the flipside of the coin of the prohibition to inflict harm – the minimum you can do for others is not to harm them, we said. It is the normative core of demands for mutual support and solidarity as well. Moreover, human rights are not just about the prohibition of specific wrongs, but also about the allocation of goods, such as normatively protected spheres of liberty. Such an order of freedom presupposes yardsticks for the justified allocation of such goods – in our analysis, for instance, principles of justice. This is important not least for ascertaining the justified limitations of human rights – for instance, when rights of different persons collide. Moreover, human rights only make sense, as we have seen, if there are reasons to believe that the beings who enjoy them have some kind of worth that justifies protecting these beings’ goods. Principles of human worth, today spelled out in human rights ethics and law as human dignity, are therefore likewise key to the understanding of human rights. These principles and their normative consequences, such as the principle of noninstrumentalization, are arguably also key to the solution of the trolley problems.

The second question asks why goods are guaranteed by rights and not some other means in the first place. If normative means to protect human goods are employed (and not just sheer force), why are duties and prohibitions not enough? After all, influential theories argue for the importance of systems of duty and their superiority at least in some respects to systems of rights.³⁹ Our analysis has suggested (apart from the analytical connection of duties and rights) that normative principles like justice and altruism provide the answer, as these principles give rise to the normative phenomenon of rights that have the important function of normatively empowering people, turning them from patients of others’ obligations to subjects of justified claims.

The third issue is the question of construction. Our analysis has suggested that a process of complex construction is necessary to transform concrete, principled moral judgments about moral claims into explicit, critically reflected human rights in ethics and law, a process that spanned thousands of years. This process includes only seemingly straightforward tasks like the generalization, universalization and objectification of the abstract core content of particular moral judgments. Moreover, human rights systems are highly selective. To account for this selectivity and its justification, among other things, a theory of human goods is essential. The importance of a political theory of human rights has been underlined in our inquiry as well. It is essential to include these complex issues in a theoretical account of the link between human moral cognition and the idea of explicit human rights in ethics and law.⁴⁰

8.2.4 Volitional Consequences of Moral Judgment

A further important element of an analysis of morality is that human moral judgment has volitional consequences: A moral evaluation does not yield information about a fact of the world like a descriptive proposition, but has prescriptive content.⁴¹ Saying “X is just” is different from saying “X is a table,” and an important element of this difference is that a moral judgment tells us how we ought to act. This can have direct volitional consequences: An ought creates a motive for action for agents who are evaluating their options for acting.⁴² Their will is bound to form certain intentions to act, to use a metaphor for a common introspectively accessible internal state of an agent. Everybody knows how it feels when you come to the conclusion that you ought to hand in the smartphone you have found lying on the ground.

If an observer evaluates not his own options for action but the intention or action of another agent, moral judgments still provide information on how one ought to act. For example, they tell the observer how the observed agent ought to act if the observed agent is in a position to act as a person is obligated to act in the particular situation in question – for instance, to hand in the smartphone found, even though the observer himself is not in this position.

A moral judgment can also remain abstract in the sense that there is nobody currently in a position to act accordingly. “One should hand in the valuables of others one has found” is a meaningful statement even if all valuables on the planet are safely in their owners’ pockets. Moral judgment then tells us what one ought to do if the conditions for this obligation obtain.

The fact that a normative statement is about an obligation, a permission or a prescription – in short, about a moral ought that provides a motivation to act – does not mean that the obligation necessarily forms the predominant, decisive, let alone only motivation to act. To be sure, human motivation encompasses many other inclinations that have great power and have nothing to do with moral considerations. Human history is to a large extent the history of greed and the pursuit of power, not the history of moral niceties. The claim is thus that only an element – and perhaps a precious element – of human moral motivation derives from moral insight.⁴³

The prescriptive content of a moral judgment can – and this is important for our particular topic – constitute a right.⁴⁴ If an act is just, the agent has an obligation to act justly and the patient has a right to that action: In the limited time available to comment and set things in intellectual order after an abysmal lecture on the foundation of human rights by a legal theoretician from Zürich, every discussant has the right to the same amount of time because this is a just distribution of this scarce good. The chair of the discussion has the obligation to ensure this fair distribution of time – for example, by restraining the loquacious Swiss theoretician’s vacuous responses. The connection between obligation and right holds for an obligation stemming from a duty to benefit somebody, too, unless it is a supererogatory action: Not only is there an obligation to pick up your phone to call an ambulance if somebody in front of you collapses, but the person who has collapsed has a right that you do (at least) this. The obligation to respect others is the correlative of others’ right to be respected: Respect is not an act of grace, but means acting in a way to which others have a claim.

8.2.5 Questions of Metaethics and the World of Moral Emotions

The principles of justice and altruism that guide reflexive evaluation have cognitive content. Whether or not there is (for example) a relation of equality between patients of actions or between a criterion of distribution and the good distributed in the sense explained above is not felt physically in the same way that cold or heat are, for example, but stems from a complex structural analysis⁴⁵ of the evaluated act that eventually predicates a relation of equality or its absence and thus constitutes a judgment with cognitive content. The same is true for other structural elements of the action that play a role for moral evaluation, such as agency, the properties of patients of the action (e.g. whether they are sentient or not) and intentions and their kind (direct or oblique) and object. The presence of these elements is not physically felt either, but is ascertained by a judgment with cognitive content – the obtaining of a direct intention to benefit somebody, for instance.

As we already have seen, this does not mean that the relevance of moral sentiments is diminished. They are evidently crucial to the impact of morality on a human life. There may even be emotions that are “geological upheavals” of moral thought.⁴⁶ Nor is the importance denied that certain emotions have for the design of law, not least for the skeptical project of constitutionalism.⁴⁷ However, insofar as they are moral emotions, these emotions are not simply emotions relevant for ethics (like the fear of tyrants) but are the consequence of moral judgment with the cognitive content just identified, and thus do not constitute moral judgment. The moral evaluation based on the obtaining of the necessary elements of a good or just intention or action is the precondition for the experience of moral emotions – for instance, the unequal treatment of a person who is equal in the relevant respects in question is the precondition for the feeling of indignation about injustice. If there is no unequal treatment of equals, we will not feel moral indignation because of an injustice.

In light of this observation, traditional arguments about the constitutive role played by human moral sentiment in moral evaluation fail to convince because they are not precise enough. Hume wrote:

If any material circumstance be yet unknown or doubtful, we must first employ our inquiry or intellectual faculties to assure us of it; and must suspend for a time all moral decision or sentiment. While we are ignorant whether a man were aggressor or not, how can we determine whether the person who killed him be criminal or innocent? But after every circumstance, every relation is known, the understanding has no further room to operate, nor any object on which it could employ itself. The approbation or blame, which then ensues, cannot be the work of the judgement, but of the heart; and is not a speculative proposition or affirmation, but an active feeling or sentiment.⁴⁸

Our argument so far indicates that two elements of the structure of moral evaluation are missing from Hume’s account. Hume rightly emphasizes the importance of ascertaining the facts of a case to be evaluated. What he does not address, however, is the structural analysis of the intention or action that determines the outcome of the evaluation – for example, the analysis of whether Norbert killed another person with the direct intention to do so for personal gain or whether he did so to defend himself against an aggressor. We have already clarified that this analysis is not a sentiment, but a judgment with cognitive content. The second missing element is the evaluation that ensues on the basis of this analysis – for example, that killing a person for personal gain is (deeply) immoral. This evaluation assigns a deontic status (good/bad, just/unjust) to an intention or action. This is the propositional content of statements such as “Murder is a heinous crime.” The moral judgment causes and is accompanied by a sentiment – for instance, the feeling of abhorrence towards murder for gain. The feeling of abhorrence does not constitute the entirety of the evaluation of murder for gain, however.

Interestingly, Hume rightly draws our attention to agency as a structural precondition of evaluating something as morally right or wrong. In Hume’s example of the young tree outstripping and eventually overwhelming and killing its parent, the tree’s growth does not elicit any moral feeling because a precondition of moral evaluation, agency, is not fulfilled.⁴⁹ Once again, whether or not a tree is an agent is not felt in the same way as our skin feels a cold breeze. Rather, any such conclusion is a statement about a complex state of affairs constituting agency (responsible, spontaneous beginning of new chains of causation, authorship of intentions to act and of actions, etc.). This cannot be reconciled with simply identifying moral judgment with sentiment. There certainly are moral sentiments, but they depend on such a prior structural analysis of the eliciting situations, including of criteria such as agency, and a consequent moral evaluation.

A central concern for Hume and the many thinkers following in his footsteps is to explain moral motivation. How could a judgment of reason cause a motivation to act? A proposition stating a matter of fact has no motivational impact. “There is a table” is, even if true, motivationally neutral, as we clarified in the preceding section. Only sentiment, it thus seems, can cause people to act. There is, however, a third path beyond the traditional but not exhaustive dichotomy of moral judgment as an act of reason and as sentiment. This third path offers the key to the problem: The motivational effects of moral judgments we highlighted above are crucial in this respect. Moral evaluation has a direct motivational effect, we said, because of its prescriptive content. Moral evaluation is not limited to stating the deontic status of an intention or action. It does not merely inform us about this status like a proposition about a fact of the world. Agents do not react to moral judgments in the same way they react to factual propositions. They do not say after a moral judgment, “Oh, this is unjust, how interesting!” in the same way they may observe, following some scrutiny, “Oh, this stone is in fact blue, how interesting!” Moral evaluation gives rises to a moral ought, and this ought alone can motivate people to act: It binds the human will, to use this basic metaphor. This is very much an everyday experience. Take the lost object example: You see a wallet that somebody has lost lying in the street. You ask yourself whether you should pocket the money in it. You look around – nobody is watching you. Do you do it? Perhaps you do not, because you have come to the conclusion that taking the money would harm the owner and thus would constitute an immoral act. This moral evaluation does not leave you untouched, because it creates an obligation not to do what is immoral that affects your will and may ultimately make you hand in the wallet. Perhaps you overcome this moral impulse by thinking of something nice you want to buy with the money, but even in this case the moral impulse is not nothing. It was simply not strong enough to outweigh the attraction of the thing you want to buy. Unsurprisingly, the direct motivational force of the moral ought has not escaped the attention of important contributions to moral philosophy.

In discussions about the nature of morality, we sometimes find the idea that a cognitivist approach to moral judgment is wedded to conscious, explicit reasoning based on deontological moral rules. As there is ample evidence for spontaneous moral judgments, such intuitions are understood as showing that moral emotivism is correct: These intuitions are emotions, not acts of reason-based cognition. This assumption, too, fails to fathom the nature of morality. Evidently, a structural analysis of the kind discussed – for example, as regards agency – is a quick, largely unconscious process that is not necessarily transparent to the agents themselves. However, this does not mean that the structural analysis is based on a sentiment – it clearly is not, as we just have seen. One thus has to broaden the analysis of moral judgment and include these structural analysis mechanisms, which, while they are largely intuitive, are not simply emotive. Such mechanisms can be made explicit, as Hume did in the case of agency, a process very important for the understanding and practice of morality.

Thus, it turns out that the simple dichotomy between moral judgments as cognitive acts of reason or as expressions of sentiment does not adequately capture the intricacies and rich content of moral judgments, their cognitive, volitional and emotional dimensions that need to be differentiated and accounted for in a theory of human moral cognition.

To repeat: None of this doubts the importance of moral feeling. It is no less than one of the most important elements of human identity. It makes morality the powerful force that it sometimes is in human life. The rich colors of moral emotions are sources of both the beauty and the profound sorrows of human life. These remarks simply intend to clarify what the role of these emotions in moral thought actually is.

A further central function of emotion in moral judgment is to fathom what an action means for the patient of the action. It is a piece of moral heuristics: Without empathy for the experience of victims of racial discrimination, for instance, without the emotional understanding of how it feels to be degraded and humiliated, nobody will be able properly to evaluate the significance of this injustice. If one thinks that being relegated to the back of a bus is a minor issue, the issue that the Freedom Riders in the USA were fighting for will remain inexplicable.

8.3 Explanatory Limits of Emotivism

8.3.1 Ruled by Moral Taste Buds?

Contemporary contributions to moral philosophy and psychology with a – broadly understood – emotivist background do not call these findings into question. We have already investigated the explanatory power of the mental gizmo thesis. A further influential perspective, the social intuitionist model or moral foundations theory by Jonathan Haidt, does not change this perception either. This theory argues along emotivist lines that emotional intuitions form the basis for moral judgments. Arguments that appear to be rational are used to justify such emotion-based moral attitudes post hoc. Moral reasoning is not engaged in to critically improve one’s own moral point of view.⁵⁰ Rather, moral argument is directed at manipulating others – in fact, reason is a “public relation firm” of emotional intuitions for “strategic purposes such as managing reputation, building alliances, and recruiting bystanders to support your side.”⁵¹ At the same time, however, rational argument is supposed to have an influence on these emotional reactions, although the theory does not clarify exactly how this influence is to be understood, in particular whether reasoning causes and changes moral judgments or overrides moral judgment.⁵²

Jonathan Haidt claims that there are six “moral taste buds of the righteous mind.”⁵³ Haidt intends to expand the ethics of autonomy of WEIRD (Western, educated, industrial, rich and democratic) people (who form the sample of many psychological studies) to include moral systems that are “hierarchical, punitive, and religious” because they include ideas of loyalty, authority and sanctity.⁵⁴ The six foundations of morality are the pairs care/harm, fairness/cheating, loyalty/betrayal, authority/subversion, sanctity/degradation and liberty/oppression.⁵⁵ Fairness is about the allocation of goods proportionally to merit; demands for equality are matters of nondomination and therefore are based on the liberty taste bud.⁵⁶ These pairs are moral modules that form the basis and common ground upon which all of the different human moralities develop. They define the possible contents of any given moral code. It is a misconception, Haidt argues, to focus on just some of these foundations as Western individualism does – for example, overlooking other moralities where loyalty and sanctity are of great importance.

The thrust of the argument is illustrated well by the following example:

[W]ithin any given culture many moral controversies turn out to involve competing ways to link a behavior to a moral module. Should parents and teachers be allowed to spank children for disobedience? On the left side of the political spectrum, spanking typically triggers judgments of cruelty and oppression. On the right, it is sometimes linked to judgments about proper enforcement of rules, particularly rules about respect for parents and teachers. So even if we all share the same small set of cognitive modules, we can hook actions up to modules in so many ways that we can build conflicting moral matrices on the same small set of foundations.⁵⁷

This, then, explains why “good people” can disagree about right and wrong (for instance, the right and wrong of spanking) – they are simply operating in different moral matrices.⁵⁸

Haidt uses these moral matrices to explain US politics. If care, fairness and liberty are dominant, one becomes a liberal, if loyalty, authority and sanctity are guiding, a conservative.⁵⁹ The dominant influence of these taste buds is genetically fixed.⁶⁰ These moral taste buds are explained by their evolutionary functions. The moral foundations have evolved, it is argued, in response to the adaptive challenges of “caring for vulnerable children” (care/harm); of “reaping the rewards of cooperation without getting exploited” (fairness/cheating); of “forming and maintaining coalitions” (loyalty/betrayal); of “forging relationships that will benefit us within social hierarchies” (authority/subversion); and of “the omnivore’s dilemma, and then to the broader challenge of living in a world of pathogens and parasites” (sanctity/degradation).⁶¹ There is no clear account of why the liberty/oppression foundation evolved.

Group selection and gene–culture coevolution may have played a role in this evolutionary development, it is argued.⁶² The point of morality is to enable cooperation, albeit limited to the groups of which the agent is a member. Parochial altruism is the ultimate frontier of ethics. Oxytocin is important for explaining this within the framework of the theory of group selection: “Oxytocin should bond us to our partners and our groups, so that we can more effectively compete with other groups. It should not bond us to humanity in general. Several studies have validated this prediction.”⁶³ The benefits of cooperation also are why groups evolved that entertain religious beliefs: These beliefs make people better cooperators.⁶⁴

Human rights are explained in this framework: They are post-hoc rationalizations of the intuitions of people with particularly receptive liberty taste buds.⁶⁵ The different moral matrices complement each other – liberals are, for instance, right about the necessity of some regulation, while conservatives are right that markets are “miraculous.”⁶⁶

This psychological theory is descriptive, it is underlined. Haidt notes that if taken as a normative definition of morality, the simple acceptance of loyalty and authority could lead to a defense of political orders such as fascism, which would be given “high marks” as long as they produced high levels of cooperation.⁶⁷ For questions of normative theory, therefore, other arguments are necessary – more precisely, what is needed is rule utilitarianism, which should be decisive and guide public policy decisions that take into account the importance of loyalty, authority and sanctity.⁶⁸ For the individual, virtue ethics is tentatively endorsed.⁶⁹

The overall vision of morality presented is a narrow one: “Parochial love – love within groups – amplified by similarity, a sense of shared fate, and the suppression of free riders, may be the most we can accomplish.”⁷⁰

8.3.2 A Testing Case: Corporal Punishment – A Question of Taste?

Spanking is an interesting example for assessing moral foundations theory in the context of human rights, because the corporal punishment of children is a standard human rights issue and in a leading decision of the ECtHR, for instance, was declared to violate Art. 3 ECHR on inhuman or degrading treatment or punishment.⁷¹

Was this decision just an emotional gut reaction on the part of the majority of judges of the ECtHR, who were led by their particular moral taste buds in their interpretation of Art. 3 ECHR, while the dissenting judge arguing for the permissibility of this kind of sanction simply used other taste buds?

This example reveals a major shortcoming of moral foundations theory quite clearly: It does not distinguish between factual, traditional morality and critically reflected ethics. One major point of moral philosophy is to critically investigate certain moral issues that may turn out not to stand the test of such critical scrutiny, such as the morality of loyalty to fellow slaveholders or respect for the sanctity of moral rules subjugating women – or the fairness of the physical punishment of children. The effects of this critical scrutiny can make history – as the abolition of slavery and the women’s liberation movement illustrate. Only critical reflection on morality, based on considered judgments, will tell us what human morality ultimately is about, while factual, traditional morality, by contrast, will tell us about precisely those influences that skew moral judgment – say, racist ideas, misogyny or an authoritarian tradition of child-rearing. Studying a traditional morality that sanctions violence against children as a good starting point for understanding moral cognition is comparable (with a dose of exaggeration) to studying mathematical cognition by looking at the belief that 79 + 86 = 164. Such studies may tell us a lot about the factors influencing counting, such as a lack of attention, but they are not key to the psychological foundations of mathematics.

Let us take a closer look at the example of corporal punishment to better understand what the further problems of the moral foundations theory may be. There is much to be said about this example, particularly in relation to children. Thinking about the effects of such punishment on children, its educational merit, brutalizing consequences and so on, is an important exercise if one wants to do justice to children. None of this means simply to rationalize emotional intuitive reactions. Rather, this exercise clarifies the issues – in particular, whether corporal punishment does any good or simply harms children, as very many people (rightly) assume today, and whether it should come with corresponding legal consequences. Beyond these important arguments, the ultimate moral evaluation relies on the elements of the action identified above – for instance, the kind of intentions of the agent of corporal punishment. If the intention is to gratify the cruel instincts of the punishing adult, the evaluation will differ from the evaluation of the ill-advised but well-meaning intention of a loving parent who uses corporal punishment as means of education, even though in the latter case, too, it is ultimately an unjustified harm done to children.

This search for moral understanding is the daily business of any responsible human being. It is thus analytically unconvincing to regard morality simply as a manipulation device to strategically influence others, not as a guide to one’s own actions. One distinguishing feature of moral judgment is that is aspires to correctness based on reasons – and there are certainly good reasons for certain forms of action. This is what motivates the painstaking efforts of many human beings who honestly want to get it right. Interestingly, the social intuitionist model underlines this itself because it includes reflection in its reconstruction of morality, albeit ambivalently and without clarifying its precise role. The fact that it refers to rule utilitarianism as the ultimate yardstick for the evaluation of social policy and as a bar to “high marks” being awarded to fascism bears witness to this. By trying to limit the harmful conclusions that may be drawn from its arguments by reference to the principles of an ethical maxim such as rule utilitarianism, moral foundations theory denies intuitions generated by moral taste buds the ultimate rule that it appears to defend. In this context, we can note once again that utilitarianism itself is based on notions of equality that are not the consequence, but the normative precondition of utilitarian evaluation. Utilitarianism takes us right back to moral principles of justice and offers no escape route from them, as we have seen. Moral foundations theory’s inconsistent impression is only deepened by making virtue ethics a yardstick for individual action. Virtue ethics is, like all ethical systems, contested. However, it does not endorse the rule of emotional intuitions as the core of ethics.⁷²

These observations are important for assessing the merits of the six foundations of morality. They point to further deficiencies in the analysis of morality: Care and harm, fairness and betrayal point to notions of justice and altruism, albeit not very precisely, as betrayal, for instance, is simply one form of unfairness. The relationship of proportionality and equality is analytically misunderstood, too: If the criterion of distribution is equally fulfilled in two agents, equal treatment is the consequence of principles of proportional equality, as we have seen. Therefore, for example, the equality of rights is a consequence of proportional equality: The humanity of humans as a criterion for the enjoyment of rights is simply the same for each person.

Liberty is a central good for human beings, to be sure. Moral foundations theory, however, does not discuss the normative principles that are relevant in balancing the liberty of one with the liberty of all and that may even create rights. The reference to a “liberty taste bud” can be seen as a metaphor that can be reconciled with (but is no substitute for) the theory of liberty as a human good sketched above. However, this reference does not answer the problem faced by any normative theory of liberty that has been at the heart of many debates since antiquity, namely the question of who justifiably enjoys which liberty and to which degree in relation to others.

Another example: Loyalty, authority and sanctity are only secondary virtues. Loyalty and respect for authority are morally justifiable if they serve the good of people; loyalty to dictators and reference of their authority is not a virtue but a vice, as Haidt himself admits, realizing that its content may cause “high marks” to be awarded to fascism if it manages to achieve social cooperation. Sanctity is a difficult concept, too. Respect for the sanctity of a certain conception of marriage may be bad news for gay couples. Here, too, other moral principles, such as respect for the autonomy of other human beings and their dignity, need to be considered.

Moreover, the theory does not engage in any detail with what basic notions of justice or altruism actually entail – for example, as to the intentions of the agents, the effects of actions, means/ends distinctions and so forth, as roughly outlined above.⁷³

Finally, as a last point: The examples of political differences explained by the different operation of the moral taste buds are loaded with prudential arguments, such as arguments about the question of when regulation actually works or whether and under which conditions markets deliver good results. Regulatory choices or the design of markets evidently have ethical implications. But discussing such prudential arguments as part of the core normative machinery of morality does not help to clarify what morality is about.

This notwithstanding, these analytical shortcomings teach us a constructive lesson: They show that the six categories of moral foundations theory do not adequately capture the normative principles that are central for ethical systems. In light of its critical discussion, the principles of justice, altruism and respect continue to seem valuable candidates for the foundational elements of ethics that a psychological theory needs to account for.

The evolutionary framework of moral foundations theory suffers from a failure to present any evidence whatsoever that the evolutionary story it recounts is actually true. It remains one of those “just-so” stories that are falsely taken as valid evolutionary theory-building.⁷⁴ Moreover, as indicated above, the first step for any convincing evolutionary theory of morality is to construct a theory of morality that possesses sufficient analytical precision. As this is lacking, Haidt’s evolutionary theory is unable to get off the ground.

Moreover, as we have seen, the idea of human rights has highly complex roots in history and human normative reflection. To reduce it to the effects of a liberty taste bud and the post-hoc rationalization of its operations seems not to do full justice to these findings.

The social intuitionist model illustrates a danger already identified previously: When the idea that corporal punishment is admissible, for instance, is interpreted as the expression of certain moral taste buds that simply are different from the taste buds used to criticize this form of punishment, the criticism of corporal punishment loses its ground because there is no reason to prefer the one taste over the other. The theory tries to deal with this problem by introducing rule utilitarianism as a normative yardstick for public policy and virtue ethics for individual acting, contradicting its emotivist message in doing so. Nevertheless, ideas of loyalty to in-groups, respect for authority without questioning its legitimacy and reverence for sanctity without inquiring into its origin and content are understood as expressions of moral judgment on par with any other moral code, in particular one based on human autonomy. In this way, psychological theory may serve to shield unjustified moral precepts against criticism by presenting them as products of a genetically fixed and thus unchangeable natural morality.

In line with various theorists of morality, moral foundations theory argues that only a morality of “love within groups,” a “parochial altruism” is the horizon of human moral possibilities. Concern for all humanity is simply too much for human beings. The human rights project, which is a real practice, not an ephemeral dream, indicates that these theses do not fathom what morality is really about.

8.3.3 Sentimental Rules

In another very interesting approach, Shaun Nichols argues for a sentimentalist picture of morality. It takes as its starting point the body of data suggesting that children from an early age distinguish between conventional and nonconventional moral rules.⁷⁵ The former can be changed by authorities, the latter not. The former deal with prudential arrangements, the latter are centered on the avoidance of harm. The latter are more serious than the former. The latter can be generalized over situations and contexts, the former less so.⁷⁶ Moreover, moral objectivism is the psychological default position, Nichols argues.⁷⁷

Nichols claims that moral rules are based on moral judgment, which he argues has two components: information about normative violations and a noncognitive response. The former is provided by rules proscribing certain behavior, while the latter consists of specific emotions. Only those rules relating to behavior that triggers reactive concern and distress cues are nonconventional moral rules.⁷⁸ It is the particular emotional engagement of the evaluating person that turns them into moral rules. These emotions are also the reason why these rules are able to survive historical change and development.⁷⁹

As we saw above, the mental gizmo theory already claimed that emotions explain moral judgment such as that reached in the footbridge case, namely that it is impermissible to kill the bystander to save five lives. The sentimental rules approach goes further than the mental gizmo theory, however, in claiming that this is not only because of the emotionally salient features of the act. Rules on the prohibition to perform this act are also important. However, these rules need to be backed by emotions in order to yield the impermissibility judgments observed in the footbridge case.⁸⁰ A third factor influencing moral judgment consists of cost–benefit assessments along utilitarian lines.⁸¹ These factors interact in a complex way, without there being a unified normative theory that captures all moral intuitions.

Metaethically, morality therefore is not objective, but bound to feelings relative to particular groups of people. There is no argument why one should prefer one set of emotions to another. It is not possible to evaluate the emotions determining the content of morality with these very emotions.⁸² The illusion of objectivity hides the respondent-dependent relativity of emotionally based sentimental rules. This notwithstanding, the objectivist intuition persists and guides human lives.⁸³

This theory faces the same problems already identified as decisive for emotivist accounts of moral judgment: The emotional reaction to some harm is not all there is to moral judgment. A harm may cause considerable unease without it being immoral – to use Nichols’ own example, a medically justified operation, for instance. It is only immoral if the intention and the action have certain properties – for instance, the intention to torture the patient and not to heal them. These properties need to be determined based on an analysis of the intention and action, as outlined above in some detail. This analysis then elicits the moral evaluation. This, in turn, gives rise to moral feelings, which are different and need to be distinguished from feelings such as unease at witnessing an operation. Feeling nauseous at the sight of blood during such an operation is not already a moral feeling, while feeling abhorrence because of the wrongness of an attempt to torture a person is. The evaluation of an action as morally wrong is, however, the precondition and cause of these moral feelings. Like an Escher cube, the same action can change its moral nature, depending on what evaluation the analysis of the action’s intentions and effects yields.

Nichols developed a theory of moral learning that adds a further qualification, underlining another influence that reason has on moral judgment. It is best considered in the framework of the discussion in Section 8.5 on the acquisition of moral knowledge.

8.4 Explaining Moral Disagreement

Given the great variety of moral opinions visible both today and in history, any theory of moral cognition will need to formulate a theory of moral disagreement as part of its explanatory enterprise. Moral disagreement is yet another traditional and vast topic of practical philosophy, and empirical work has been carried out in this field, too.⁸⁴ It is sometimes argued that the mere existence of moral disagreement already proves moral relativism.⁸⁵ In this respect, however, conclusions should not be drawn too rashly: “Establishing the best explanation of stubborn ethical disagreement requires understanding all the possible origins of these conflicting beliefs and all the possible resources that might resolve the conflict – no quick or easy job.”⁸⁶ There may be ways to account for moral disagreement, even radical disagreement, with substantial explanatory power without any implications about irreconcilable foundational moral principles.

A first task, thus, is to determine precisely what kinds of moral disagreement actually exist. This is far from obvious. Some studies on the supposedly different moral orientations of different cultures, for instance, under critical scrutiny turn out to have overlooked important communalities.⁸⁷ The second task is to see what the causal factors for apparently different moral beliefs are and what they teach us about the structure of moral judgment. Not just law, but morality, too, is often based on express moral rules. These rules can embody conceptions of what is good and just that do not withstand critical scrutiny. It is therefore necessary to distinguish between traditional, prereflective social moralities and a critical, reflective ethics in the context of analytical theories of moral disagreement, too. The conditions for the successful criticism of traditional moral practices are key to understanding the principles underlying human moral deliberation and judgment that may in the end lead to the abandonment of such practices.

One factor of considerable importance in explaining the existence of moral disagreement is disagreement about the nonmoral preconditions of moral judgment. This includes understanding what an action, practice or institution means for other persons. These assumptions may even lead to denying certain persons moral status: As we have seen, a central question during the conquest of the Americas in the sixteenth century was whether the indigenous people of the Americas were actually fully human or some other kind of creature. The identification of the proper objects of evaluation likewise shows the relevance that these factual questions hold for moral judgment: For instance, the institutions of the state have to be regarded as products of human volition and action and not as unchangeable elements of the makeup of the world to be evaluated on the basis of principles of justice.

Other factors that rightly play a prominent role in this debate are the influence of interests and the impact of ideological constructions, both the source of extremely powerful emotions and motives of action. An entirely fantastic ideology like National Socialism, for instance, was capable to motivate people to commit mass murder before leading them to their own deaths and the destruction of their country. Taking these factors into account may already reduce the cases of real moral disagreement about basic principles of morality considerably. Take the (important) example of the rights of women: The denial of equal rights of women was (and is) partly based on wrong factual anthropological assumptions, such as the idea that women lack the capacity for autonomous self-determination and rationality and therefore have to be guided by men.⁸⁸ The suffering of women who were denied an equal part in social life because of these assumptions about their interests and capabilities needed to be understood. This required massive cultural and political efforts, sustained over generations and still ongoing today. The interests of men in comfortable structures of domination were an evident further factor: “When, however, we ask why the existence of one-half the species should be merely ancillary to that of the other – why each woman should be a mere appendage to a man, allowed to have no interest of her own, that there may be nothing to compete in her mind with his interests and his pleasure; the only reason which can be given is, that men like it.”⁸⁹ Ideological constructs, partly in religious garb, about the place of women in the world buttressed these social structures, too. Such factors continue to play an important political role today, despite the progress made. If these false assumptions about the nature of women, their experience of repressive patriarchic structures and the power of interests and ideologies lose their influence, apparently irreconcilable moral disagreements can disappear quickly and the equal rights of women appear as an evident truism even across cultural boundaries (as indeed they should).

Another important issue is the process of clarifying ethical concepts – for instance, that responsibility for actions depends on the internal state of the agent, an intention to act, an insight of major importance for the development of criminal law. Some forms of human behavior may be wrongly moralized or wrongly demoralized, often as a consequence of false factual assumptions, interests and ideologies. Take the example of LGBTIQ* rights: If one understands that the only consequence of preventing consensual intimate relations between same-sex partners is to cause suffering while not promoting any discernable good enjoyed by human beings, traditional moralities about the wickedness of same-sex partnerships quickly crumble, as we have witnessed in the last thirty years or so. If one understands that harassing people because of their sexual orientation means doing an unjustified harm to them, then refraining from mistreating them in this way becomes a moral imperative and may even lead to a prohibition by law of certain qualified forms of this kind of behavior. Reflective ethics helps us to achieve consistency – if autonomy, for instance, is a key value, autonomous decisions about the persons one loves should be protected robustly. Ethical reflection of this kind can expand the circle of persons included in moral deliberation: If all persons count equally, the interests of future generations of human beings and thus questions of intergenerational justice should be factored into our debates about the appropriate measures to combat climate change, as courts around the world have now started to do explicitly.⁹⁰ If sentient beings are objects of moral concern, then ethical and legal protections for nonhuman animals are required. Moral values and standards of virtuous acting may not stand the test of critical thought – that it is imperative to fight a duel with somebody who has insulted one has ceased to be a convincing interpretation of proper, self-respecting behavior. A further example illustrating the importance of critical normative thinking is the scrutiny of theories of morality themselves: The critique, for instance, of the (as we have seen) false idea that human rights are products of the post-hoc rationalization of emotional reactions to certain stimuli will prevent the conclusion that human rights should be made irrelevant.

Finally, there are real moral dilemmas to which no easy solution is available – for instance, in the case of conflicting duties, say, the duty to save lives in a pandemic and the duty to prevent the domestic violence increased by lockdowns. In sum, there are a plethora of factors helping to explain different moral points of view without taking recourse to substantially and unchangeably different underlying moral conceptions.⁹¹ There is thus reason to believe that under the surface of apparently insurmountable moral disagreement, there may be a deep, shared structure of common moral principles.⁹²

8.5 The Development of Moral Cognition

8.5.1 How Do We Learn to Be Moral?

One question that remains to be explored is the ontogenetic origin of moral cognition. The development of moral cognition forms the subject of landmark debates in moral psychology.⁹³ Our analysis thus far suggests that central issues here include – among other factors – the acquisition of the cognitive domain of morality, the restrictive principles that determine the possible objects of evaluation, the material principles of morality discussed, the prescriptive, volitional effects of moral judgments, the moral ought, including necessary connections between duties and rights, and the emotional consequences of moral experience. Such phenomena could be constructed and acquired by secondary learning processes (instruction, repetition, role-taking, peer pressure, sanctioning, etc.), as many normative principles are (e.g. the intricacies of Swiss law on unjust enrichment), or alternatively they could be at least in part the product of the unfolding of innate cognitive structures triggered by experience. The latter is the way a mentalist approach to ethics would approach the issue.

It is important to emphasize that the importance of social and cultural influences on ethics and law is not denied in the latter case: As noted, the freedom of the press, for instance, as a legal norm and the underlying principles of political morality presuppose the cultural achievement (a late achievement with plenty of preconditions) of the press and, in addition, the experience of its suppression even by democratic governments, among many other things. “The freedom of the press is protected” is certainly not an inborn principle of human moral cognition. As we saw in our survey of the history and the theory of justification of human rights, fundamental moral judgments about the wrongness and justice of certain intentions and actions are the seeds of the idea of human rights. The real question is therefore: What kind of mind do you need to possess in order to develop such an idea? What is so special about the human mind that only humans and no other organism developed a concept like human rights? What are the cognitive preconditions for starting the long cultural process leading, after many thousands of years of human cultural and social development, to the idea that one can not only wish for or have an interest in, but in fact enjoy a right to a free press? None? Is simply being smart enough? Or does one need to have certain specific conceptual tools with which to build a system of rights?

The simple observation already recalled above that any child acquires a differentiated set of moral concepts and categories whereas no nonhuman animal acquires any of the same, even if it is raised in the same environment, seems to indicate strongly that very different cognitive abilities are in place. Some of the empirical work on this difference between humans and nonhuman animals has been discussed in Chapter 7. To repeat: The question therefore is not whether there are species-specific inborn cognitive structures enabling the formation of moral precepts, but what exactly these cognitive structures are.

8.5.2 The Moral World of Infants and Toddlers

Jean Piaget pioneered the idea that children apply complex moral ideas and concepts, not least when they play, as illustrated by his fascinating analysis of children playing with marbles.⁹⁴ Recent years have produced many more highly creative studies on the moral psychology of children, increasingly focusing on young children. We have already encountered some important examples in our inquiry. As we have seen, these studies operate with different theoretical background assumptions and sometimes yield contradictory results. Some research is constrained by theoretical and conceptual problems, limits of the design of experiments and contestable interpretations of their results. One important problem is the lack of a sufficiently fine-grained analytical theory of moral judgment that does not commit, for instance, the category error of conflating moral judgment with simple preferences and aversions. The thrust of this research is, however, that children operate in a richly textured moral world from very early on. This fits well with research into other cognitive domains that also indicates that human beings are not born blank slates.⁹⁵

A classical set of experiments concerns the moral–conventional distinction, for instance: As mentioned above, children distinguish between conventional norms and moral, nonconventional norms.⁹⁶ Other experiments suggest that toddlers by the age of eighteen months and even earlier exhibit spontaneous helping behavior towards others.⁹⁷ They engage in so-called paternalistic helping: Children want to accommodate the well-being of others, not just their wishes.⁹⁸

Children act with an intrinsic moral motivation: They also help when nobody is watching or when others do not know that they are being helped.⁹⁹ External rewards even seem to undermine their intrinsic motivation.¹⁰⁰ Helping somebody themselves and seeing somebody being helped render them equally content.¹⁰¹ There are studies on extended sympathy – for instance, for the victims of some harmful action after the deed.¹⁰²

Studies indicate that – unlike great apes – small children act according to principles of distributive egalitarian justice, particularly in collaborative contexts.¹⁰³ While there are studies suggesting that children limit fairness to in-group members,¹⁰⁴ there is also substantial evidence that three-year-olds apply egalitarian principles universally.¹⁰⁵ Such intuitions extend to procedures of distribution.¹⁰⁶ Three- to five-year-olds take need or merit as criteria of distributive justice. They protect the entitlements of others, create social norms, employ normative categories and engage in third-party punishment.¹⁰⁷ Intentions are a central factor in this punishment.¹⁰⁸ There are studies on the specific moral feelings of guilt and shame – for instance, arguing that three-year-old children feel guilt over the harm they inflict, not just sympathy for the victims of this harm.¹⁰⁹

According to various studies, it is plausible to assume that infants already operate with at least some kind of proto-morality. Twelve-month-olds categorize actions on the basis of their social valency.¹¹⁰ Studies on social evaluation indicate that six- to ten-month-old infants base their evaluation of another individual as appealing or aversive on this individual’s actions towards others: For instance, they like individuals better who help others and who act more cooperatively to facilitate the achievement of these others’ aims than those who hinder the achievement of others’ goals. They also like helpers better than neutral individuals, and the latter better than hinderers.¹¹¹ These experiments concern actors who are unknown to the infants and actions that have no effect on them as observers. This rightly has been identified as an important element of moral judgment: It is not personal experience with the agent that causes the agent to be evaluated a certain way, but the latter’s action affecting unrelated others. Eight-month-olds showed a preference for individuals with good intentions, basing their evaluation on mental states of the agent, not outcome.¹¹² Other research indicates that actions towards inanimate entities do not enter into infants’ evaluation of intentional agents.¹¹³ Knowledge of an agent’s goals is relevant for ten-month-olds’ evaluations: Only an agent who knowingly helps another is preferred to other agents. It is not far-fetched to draw the conclusion from this research on infants that the capacity to evaluate others based on their behavior towards third parties is universal and unlearned, given their early age. It should be noted once again, however, that a social preference is not the same as a moral evaluation of an intention or action. The former is at best indirect evidence for the latter.

Many of these studies understand these mental capacities as biological adaptations for cooperation and interpret them as tools to identify free riders, cooperators and reciprocators. As we have seen already, this biological interpretation is too narrow and overlooks the many forms of cooperation that are both possible and entirely reconcilable with the constraints of a plausible evolutionary trajectory.

8.6 Poverty of Stimulus and the Development of a Moral Point of View

These studies on child psychology tell us something interesting about the early cognitive stages of human mental development. The younger the children are for whom plausible evidence shows that they operate within a moral cognitive domain, the less plausible it becomes that these children were born as blank slates in moral terms. There are some studies on the lack of cultural variation in crucial areas of the development of moral cognition that point in the same direction.¹¹⁴ This kind of evidence supports the hypothesis that foundational elements of morality are part of the natural cognitive endowment of human beings that matures during childhood, just like other cognitive faculties. But even if clear traces of a mature system of morality were not in place at an early age, the possibility of such a natural endowment could not be excluded, for it is entirely conceivable that a specific mental faculty matures only later in childhood. The fact that puberty (including its cognitive components) occurs in the second decade of human beings’ lives does not speak against it being based on inborn properties of the human species.

The decisive argument is therefore the poverty of stimulus argument: If the input of an organism’s experience is not sufficient to generate a certain cognitive ability, at least some of the cognitive structures underlying this ability must be inborn.¹¹⁵ This argument holds for any cognitive ability of any organism. To take a fascinating example that illustrates the structure of this argument:¹¹⁶ Honeybees communicate the location of food through their dance in the hive. They determine the position of food by the position of the sun relative to a known terrain. They can do this even if the sky is overcast, because their circadian clock and their orientation systems enable them to determine the solar ephemeris – that is, the position of the sun relative to the time of the day. Their dance can thus refer to this position even if they were unable to see the sun itself. How do bees acquire this striking faculty? The following experiment gives the answer: Bees were raised in an incubator and foraged only in the late afternoon and had thus no direct experience of the sun’s morning position. One day, they were let out to forage in the morning under an overcast sky. They still were able to determine and communicate the position of food relative to the position of the sun they had never observed: They had determined the solar ephemeris and with it the position of the food. As they had no experience of the sun’s morning position but were still able to determine it, this ability cannot be learned. The only explanation for bees’ acquisition of this ability is an innate cognitive mechanism of orientation operating in conjunction with the circadian clock.

This example illustrates the point of the poverty of stimulus argument and shows what intricate cognitive mechanisms are in place in organisms like bees. It may encourage us to ask seriously what structures the vastly more complex apparatus of human cognition may contain.

If the poverty of stimulus argument is considered in detail in the context of morality, many questions arise. Given the sheer scope of the issue, some hints at the gist of the argument must suffice here. Hume’s implicit use of it avant la lettre provides a good illustration of its structure:

This principle, indeed, of precept and education, must so far be owned to have a powerful influence, that it may frequently increase or diminish, beyond their natural standard, the sentiments of approbation or dislike; and may even, in particular instances, create, without any natural principle, a new sentiment of this kind; as is evident in all superstitious practices and observance: But that all moral affection or dislike arises from this origin, will never surely be allowed by any judicious enquirer. Had nature made no such distinction, founded on the original constitution of the mind, the words, honourable and shameful, lovely and odious, noble and despicable, had never had any place in any language; nor could politicians, had they invented these terms, ever have been able to render them intelligible, or make them convey any idea to the audience.¹¹⁷

Hume’s point is that basic moral concepts are not learned from scratch, they are the precondition for moral learning. If one takes as examples a constitutive element of human morality like the moral cognitive space, some aspects of the principles governing moral judgment, the concept of ought and moral emotions like guilt or shame, the argument can be fleshed out somewhat more: As to the moral cognitive space, the poverty of stimulus argument asks whether a child can acquire this particular dimension of its perception of the world de novo, through the example of peers, instruction, imitation and so forth. In order to answer this question, one has to imagine a child who has no idea whatsoever of the qualitative content of the moral perspective: It sees the world only as a world of facts and events, including actions of other persons, but not in the very different colors of moral evaluation. How could a child possibly step from this perception of the world into the very different kind of world possessing a moral dimension? Quite aside from the fact that there is no such thing as instruction about this matter in real educational settings, even direct instruction would not help: How could a child understand what is meant by explanations of the moral dimension of human actions if it did not have access to this category of thinking? It would be like explaining the scent of chocolate to somebody who cannot smell.

The same is true for the concept of ought or moral emotions. What is a child to do with the explanation that an ought binds the will? How can the child be enlightened about what is meant by this if it has no access to this experience? Moreover, how is such a verbal definition to be turned into the actual subjective experience of being under an obligation?

Other mechanisms often used to explain the development of a moral orientation do not help either. Sanctions as such do not create the experience of an inner ought, only the experience of outward compulsion. A child with no access to the experience of moral obligation will feel a sanction as a harm inflicted upon it and a prudential reason not to show certain behaviors in order to avoid this harm, but will not feel any inner obligation not to do certain things. Sanctions enforce obligations but cannot trigger the subjective experience of a moral obligation.

Another example is role-taking. Sometimes it is argued that learning to take the perspective of another person leads to the understanding of the idea of a moral ought.¹¹⁸ Role-taking, however, informs us about the perspective of another person only in a qualified sense: We understand how we with our capacities would perceive the world if we were in the other person’s place. A person who cannot smell cannot understand what scent is just by taking the role of somebody who can smell. The same is true for moral concepts: If, as imagined, a child is a moral blank slate, nothing about this state of affairs will change by taking the role of another person. The child will only imagine what a world without access to the idea of moral obligation looks like from the point of view of the other person.¹¹⁹

The case of moral emotions seems even more obvious. How are we to teach a child what the feeling of guilt or shame is like if this child is assumed to have no access to this emotion? Note that this is not about teaching the reasons for feeling guilty or ashamed, but about the emotion itself and its quality. Invoking internalization does not help either. One cannot internalize something to which one has no cognitive access. What the hypothesis of internalization tries to capture in these contexts is in fact the process of some experience triggering the maturation of cognitive capacities that make some cognitive or emotional phenomena accessible to the person in question.¹²⁰

Finally, it is not particularly plausible that a child is ever instructed about the differentiation between direct and oblique intentions for the moral evaluations of seemingly altruistic acts. The same seems true for the prohibition of instrumentalization or the intricacies of proportional justice.

The same kinds of question need to be asked about other elements of an analytical theory of morality, in particular the dependency of moral judgment on agency, the limited class of possible objects of moral evaluation and the foundational relation between rights and the principles of justice and altruism. The concept and normative category of “right” is itself of great interest in this respect, as is what this category entails, namely the intricate web of normative positions sketched above, the necessary connections between claims and duties, privileges and negations of duties, the intentional content of these deontic categories, the semantics of obligation, permission and prohibition and the necessary volitional and emotional consequences of moral judgment’s implications for rights. In all these cases, one needs to consider carefully what the poverty of stimulus argument may teach us about the acquisition of this complex set of normative positions.

8.7 Sentimental Rules, Rational Rules?

8.7.1 The Power of Statistical Learning

It is instructive to look at an account of moral learning developed by Shaun Nichols that takes seriously the argument made thus far, namely that there is no direct input from parents and peers that explains the development of moral cognition.¹²¹ He underlines that there is a consensus that for some capacities an empiricist account and for other capacities a nativist account is more plausible.¹²² His account, already referred to briefly above, complements his moral sentimentalism with a rationalist element¹²³ and presents an alternative to assuming that concrete moral rules are based on certain inborn structures, claiming that statistical Bayesian learning abilities suffice to infer the content of the existing rules of a community from the given evidence of identified violations of rules.¹²⁴ This is how deontological rules arise: “Our hypothesis is that non-utilitarian judgement derives from learning narrow-scope rules, i.e. rules that prohibit intentionally producing an outcome, in a way that approximates Bayesian learning.”¹²⁵ Nevertheless, there is a nativist element in his theory, or his theory is at least consistent with a nativist approach: Learners have an aptitude for concepts like agent, intention and cause, and the capacity for acquiring rules.¹²⁶

More concretely, Nichols’ argument takes its start from the observation that a distinct evaluation of doing on the one hand and allowing on the other is a reality of human moral psychology. This distinction presupposes that humans apply act-based rules, not consequence-based rules. Moral rules are about prohibitions of or prescriptions for action and not about minimizing unwelcome outcomes or maximizing welcome consequences.

Nichols goes on to ask how children can learn that the moral rules that they apply are act-based, not consequence-based.¹²⁷ His argument is that they are exposed to act-based input by adults, such as “Don’t do X!” or “Do Y!” Through mechanisms of statistical learning, the children acquire act-based rules, rather than consequence-based rules, as experimental evidence confirms in his view. The core statistical mechanism is the size principle: When a learner has to choose between two hypotheses, one of which is a nested subset in the other, it is rational on probabilistic grounds to choose the hypothesis with the smaller scope if all of the evidence available is consistent with this smaller hypothesis. This is because there is a higher likelihood that the smaller hypothesis is in fact correct. It would form a “suspicious coincidence” if all of the evidence falls in the smaller hypothesis while the larger hypothesis is true.¹²⁸ Humans are able to perform this statistical operation because they possess substantial statistical learning abilities, Nichols holds.¹²⁹

As the input that learners are exposed to is based on the application of act-based rules (for instance, the command “Don’t do X!”), this evidence is consistent with both the hypothesis that act-based rules are applied and the hypothesis that consequence-based rules are applied, the latter hypothesis also including act-based rules. As the first hypothesis is a nested subset of the latter and the evidence that the learners is exposed to is act-based, it is rational to conclude that act-based rules are applied. Act-based rules are thus acquired by the learners. Nichols concludes, albeit somewhat hesitatingly, that this learning process could explain why people apply the principle of double effect – for instance, in the trolley cases.¹³⁰ It can also account for the acquisition of parochial moral codes.¹³¹

Furthermore, there is evidence that humans expect new rules to be act-based, showing a “pronounced prior” in this respect.¹³² The reasons for this are overhypotheses in the sense defined by Nelson Goodman – because the experience of rules consists of act-based rules, an overhypothesis is formed that rules tend to be act-based.¹³³

Normative systems need a default principle of the normative status of those intentions and acts that are not explicitly prohibited or permitted. Everything that is not prohibited is allowed or everything that is not permitted is prohibited are such closure principles, for instance.

Nichols argues that learners acquire the one or the other depending on the input received: If they encounter permissions, they conclude that what is not permitted is prohibited. If, however, they encounter prohibitions, they conclude that what is not prohibited is permitted. This conclusion is based on pedagogical sampling: Learners assume that their teacher is using methods of rational and efficient instruction. Based on this principle, it is rational for learners to conclude that if they are only presented with prohibitions of certain acts by the instructor, then other acts are permitted, and vice versa.¹³⁴

Nichols also explains perceptions about the universal or relative validity of norms along these lines. If learners encounter widespread consensus, they acquire the idea that the respective norms are universally valid, while if they encounter disagreement, they conclude that these rules are of relative validity. The statistical principle at play is the trade-off between fit and flexibility: The hypothesis has to fit the data but must remain flexible so as to accommodate other factors producing consensus or variety of opinion (for instance, unreliability of the persons evaluating the topic), without becoming too flexible and thus empty, accommodating any data.¹³⁵ These mechanisms also explain the distinction between moral and conventional rules. If people (on statistical grounds) judge an action as universally right or wrong, they should regard it as a wrong in itself, independent of authority, too.¹³⁶

8.7.2 The Limits of Statistical Learning

Importantly, this approach rightly underlines the significance of a descriptively adequate account of morality. It acknowledges the necessity for theory-building to determine properly the “acquirendum,” the mental structure that human beings actually acquire in the moral domain.¹³⁷ It convincingly refutes various theories that identify morality as the expression of a small set of primitive emotions, seeking to deny that the operations of moral cognition are based on a rich systems of structures, rules, principles and representations.¹³⁸ It applies the poverty of stimulus test but arrives at the result that there is in fact no such poverty: The input available to children and statistical principles suffice to acquire the basic elements of the moral world of human beings it investigates. The theory is, therefore, an important constructive contribution to the understanding of human moral ontogeny.

However, several problems arise with regard to the reach of statistical learning. As just discussed, cognitive access to basic elements of the moral world is the precondition for understanding what moral judgment and explicit norms are about in the first place. You need to know what ought means, for instance, before you can grasp what you ought to do, as Hume already argued. Statistical learning is unable to bridge this gap. No quantity of references to ought by others will tell you what ought means from a first-person perspective, just as no quantity of references by others to the pleasant smell of coffee in the morning can teach you the nature of this odor from a first-person perspective if you cannot smell. The same holds for the other elements of human beings’ moral world, including moral emotions like shame.

Another problem stemming from an insufficient determination of the acquirendum consists in the following: The moral principles human beings acquire are not captured with sufficient precision if one looks only at the doing/allowing distinction. The distinction between (intentionally) doing something and allowing X to happen when you have a duty to act to prevent X from happening on the one hand and (intentionally) allowing X to happen when there is no such duty on the other comes closer to a central element of human moral cognition.¹³⁹ The prohibition against a girl hitting her brother as punishment for taking her ball is morally equivalent to the prohibition against the child not preventing her younger brother from falling from a swing as punishment for taking her ball when she is playing with her sibling at a playground and has a duty to see that her younger brother does not hurt himself. In contrast, there is no equivalent moral duty to prevent all possible harm to all other children on the playground. Unsurprisingly, the relevance of this distinction is mirrored in basic provisions on criminal omissions in penal law around the world.

Moreover, children do not just learn that rules are act-based. They acquire the principle that internal states like direct intention (purpose) or oblique intention (knowing) matter when evaluating an action. Moreover, these internal states are relevant in complex ways for moral assessment: If one commits armed robbery of a bank with the purpose of getting money and knowingly creates a risk that a guard will be killed, and this indeed happens, in many legal systems this will count as intentional (knowing) homicide. The foreseen but not intended negative side effect of a purposive morally wrong act only makes it worse. In other constellations, for instance, when one intentionally acts to prevent harm but foresees harm to third parties as a side effect of the benevolent action, the act may be justified, as in the standard trolley bystander case. Thus, the acquirendum seems to be of a different and much more complex nature than is allowed for in the doing/allowing distinction that Nichols addresses. That any of these acquired complex structures can be learned through the statistical operations Nichols employs is far from clear, given our findings thus far. Take his example of intentional act-based rules. The evidence he considers (“Do not do X!” etc.) allows for the conclusion that somebody wants the agent to refrain from doing X, perhaps backing the prohibition to do X with the threat of sanctions. It does not allow for the conclusion that one is to refrain from acting with a specific internal state – for instance, an intention – or that one is to see intentions and actions causing harmful foreseen consequences as sometimes prohibited, sometimes justified, let alone that one is to acquire the capabilities to experience in oneself a moral ought and the other cognitive, volitional and emotional elements of the subjective world of morality already discussed above.¹⁴⁰

The discussion of closure principles is also of interest. It tells us something relevant about the reactions to explicit instruction by prohibitions or permissions concerning the specific type of action (in Nichols’ study, for instance, mice entering a barn).¹⁴¹ For moral theory-building, however, the problem that closure principles deal with is a different one. It concerns the question of how people assess the permissibility of action even without explicit permissions or prohibitions concerning a specific type of action: Do you need an explicit prohibition to take an action as prohibited or do you need an explicit permission to regard it as permitted? What is the default principle (if there is indeed one) if there are no such explicit prohibitions or permissions? Nichols does not investigate this question because he is concerned with reactions after explicit instruction, though he indicates that there may indeed be a default principle of liberty. Only if one addresses this question, however, can one reasonably discuss the problem of how such a principle – for instance, of residual liberty – becomes part of the moral and (in liberal orders) legal world of human beings.¹⁴²

As to the discussion of universalism and relativism, it is important that the validity of a norm is not dependent on the quantity of assent it finds, but on material validity conditions. Thus, the norm that Jews are just as entitled to life as other human beings was valid even in German extermination camps. Statistical learning thus allows for the hypothesis that many people agree about the validity of a certain norm, not that it is (in fact) valid. So-called authority independence is not related to consensus either. The point is rather that such universally valid, authority-independent norms are the basis to challenge even majority opinions. The phenomenon of critical reflection creates the possibility of changing a consensus using reasons – a process in which the universality of some norms is a possible argument. In Nichols’ account, any norm could be understood as universal and authority-independent, as its universality depends solely on factually existing consent patterns. If the input that a child is exposed to consists of a consensus that it is justified to kill Jews, this norm enjoys universal validity, for instance. Nichols is not very clear on when and under which conditions the acquired rules change, but it seems that this change is dependent merely on new experiences of consensus or varieties of opinion. This account appears to miss the point that content matters for the question of universality, and that only certain content is a serious candidate for universally valid norms – for instance, norms prohibiting harm, but not norms prescribing gratuitous cruelty. This is also observed in the child psychology studies on the moral/conventional divide that Nichols accepts – not just any norms are at issue, but in particular prohibitions of harm.¹⁴³ That any of these norms constantly change with the shifting of opinions is hard to reconcile both with experience and with evidence.

As to some concrete rules that we discussed, it is not clear what kind of statistical evidence there is, for instance, for the prohibition of the instrumentalization of people. Do children observe a lot of this? How come human beings are able to critique and transcend norms that were robustly enforced in their environment – for example, on the permissible instrumentalization of women in patriarchal societies?

Nichols rightly underlines the difference between possible learning processes and actual learning and emphasizes that his account is only concerned with the former, not the latter.¹⁴⁴ The reviewed literature on the ontogeny of moral cognition seems to underline the relevance of the identified problems of his account. Moral development in fact seems to take not the course that statistical learning foresees, but a rather different one, including differentiated moral concepts in young children, cross-cultural equal age trends of developments of moral cognition, intention-based moral evaluation and so forth.

An interesting point that Nichols raises is the origin of the limited hypothesis space with which child learners operate. The theory offers no explanation (as it underlines itself) of how learners arrive at the hypothesis space. This hypothesis space is thus understood as possibly innate,¹⁴⁵ though there is some counterevidence against even such constraints, it is argued.¹⁴⁶

The problem of constraints on hypotheses of possible morality is very important: What if children were exposed to the punishment of unintended actions? Some kind of strict liability morality? Would they then acquire a morality in which intention does not count in the moral evaluation of actions? Is there any evidence of this happening? Or is rather the opposite indicated not only by experimental work, but also by the passionate protests of children around the world when they are punished for something that they did not intend to do? The answer must be based on the poverty of stimulus argument: If the learning child does not receive sufficient input to form the hypothesis space – including the constraint that intentions count in a morality, for instance – then the constraints must be innate.¹⁴⁷

In sum, the reference to statistical learning thus does not seem sufficient to account for the acquisition of the basic elements of morality.¹⁴⁸ Exploring the problem of the origin of the initial conception of morality that frames the acquisition of moral knowledge, the hypothesis space and its possible nativist explanation is an important point of the theory and highlights the need for open-minded research in this area. In any case, nothing in this research rules out that the Bayesian learning mechanism would lead to the acquisition of a morality incompatible with the content of human rights – on the contrary, this theory argues that human beings can learn any rule.¹⁴⁹ They are in no way naturally limited to narrow parochial tribalism. Statistical learning is open to acquiring inclusive norms,¹⁵⁰ such as – one may add – human rights.

8.8 Theories of Mind and Human Moral Progress

There is thus a prima facie case for seriously considering the possibility that the moral space of human cognition, the concept of ought, moral emotions or certain elements of the principles of morals such as altruism, justice and respect for others form part of human beings’ natural cognitive endowment that matures throughout childhood.

As illustrated by the mental gizmo thesis, the moral foundations theory, various approaches of evolutionary psychology, behavioral economics and the joint intentionality theory of cognitive development among others, the time has passed when studying the structures of the mind was considered unimportant because only one kind of theory of mind was believed to be plausible, namely a theory that assumed that the only inborn property of the human mind is that it is an unspecified learning device.¹⁵¹ The mental gizmo thesis and moral foundations theory, for instance, are substantial empirical theses about the structure of the human mind, as is any assumption about the importance of shared intentionality, heuristics, framing effects and biases. Some of these theories have triggered a vast amount of research across the globe. Their claims may be right or wrong, but they certainly are serious scientific efforts that need to be evaluated based on their explanatory merits, and the same is true of the mentalist approach to ethics and law.

These remarks show that it is possible to frame an empirically minded theory of moral psychology that understands deontological principles not as cognitive illusions, but as part of the makeup of the human mind that may be the precondition enabling the cultural development of moral systems and the law.

Substantial empirical evidence suggests that there is a faculty of language with highly restrictive principles in which natural languages unfold, and there are many theoretical reasons for assuming its existence. This is a highly contested area of research, but even if this hypothesis is true, the theory of a faculty of language is still a long way from explaining the origin of the verses of King Lear. However, the language faculty is a precondition for humans’ ability to produce and enjoy something like King Lear, whatever the hidden secrets of human creativity ultimately may turn out to be that put the language faculty to such thrilling use. Similarly, what has been said about the history, justification and psychological theory of human rights perhaps renders plausible the idea that a theoretical account of the human moral faculty is a long way from an understanding of the sources and justification of the Universal Declaration or other concrete, historically shaped catalogues of rights in constitutions and international bills of rights. But this moral faculty could – in the same way as the faculty of language for the verses of King Lear – turn out to be the cognitive precondition for the possibility of ultimately producing something like the Universal Declaration and the aspirations it implies.

8.9 Critique and Construction: Explanatory Theory and Normative Arguments

Let us assume that a mentalist account of morality and law has some merits and is preferable, for example, to the mental gizmo thesis or the moral foundations theory. This would be a very substantial insight for an explanatory theory of human moral cognition and for the theory of mind in general. But what normative significance would this have if one wanted to avoid a naturalistic fallacy? This is the next question our inquiry will address.

The discussion thus far has already made clear a first function of the analysis of the relation between mind and rights: identifying unwarranted human rights criticism based on unconvincing theories of the mind – for instance, on unsatisfactory neuroscientific studies or insufficient evolutionary theory. This obviously already is a very important function.

Our findings so far refute, for instance, the idea that deontology is no more than a cognitive or moral illusion from a hard-headed, non-armchair, scientific point of view. Deontological arguments are not discredited in any way by the theory of mind, psychology or neuroscience. Of course, critiquing implausible theories of moral cognition does not in itself justify the normative principles that enter into a theory of justification of human rights. But defending cognitive principles outlined as reasonable is quite a different matter from defending the normative value of principles that are products of the post-hoc rationalization of hardwired emotional gut reactions. Moral psychology cannot substitute normative theory-building in ethics and law. But it is indispensable to show that normative theory is not just the illusionary offspring of hidden mechanisms of the mind and thus can be reconciled with what is known about the mind’s structure and workings.

The same holds for forms of evolutionary tribalism: Our discussion showed that a kind of universalist ethics is at least not an evolutionary impossibility. While this does not make the normative case for such an ethics, it is a different justificatory task to argue the legitimacy of rights that are contrary to human beings’ cognitive nature, assuming, for instance, that “parochial love – love within groups … may be the most we can accomplish,”¹⁵² and where it remains a complete riddle as to how the cognitive machinery that produced them could have evolved, than to argue for a set of rights that is entirely reconcilable with cognitive mechanisms that easily fit into a plausible theory of the evolution of the human mind.

This critique, is, if you will, preparatory work that lays the ground for normative arguments by showing what kind of counterarguments are insufficient to discredit a justificatory theory of human rights.

This critical function of an inquiry into mind and rights already justifies the efforts made thus far. It is quite clear from our review of current thinking about moral psychology that the is/ought distinction does not prevent the theories scrutinized from influencing how human rights are conceptualized. They are among the influential sources that feed into human rights skepticism today.

But can such a theory of human moral cognition provide more than this crucial criticism? Can it perhaps even provide some kind of additional normative justification of human rights?

Of particular interest in this respect is the question of whether a universalist justification of human rights exists or whether human rights are culturally relative. In our survey of human rights history and normative theories of human rights, we encountered no compelling evidence for the latter.

The question of universalism is of substantial interest because it concerns a central claim of human rights – to be valid for everybody, everywhere. This question is important in both practical and political terms: The answer decides, for instance, whether Uighurs have a claim to religious freedom and nondiscrimination against China, or whether the cultural difference between Zürich and Beijing acts as a bar to such claims.

Our discussion of the justification of human rights has helped us to clarify the issue. As we have seen, a theory of the justification of human rights contains three elements: first, a theory of goods and, as part of it, anthropological assumptions; second, a political theory of the role of rights in society; and third, normative principles of justice, solidarity and respect. Anthropological assumptions are a matter of empirical knowledge about human beings and have no direct connection to moral cognition. The political theory of rights has normative elements but concerns other questions, too, such as the factual effects of a political order of rights on people’s well-being. The question of universalism thus mainly concerns the normative element of the justificatory theory of human rights. Are the guiding principles and the normative tenets of such a theory universally valid? This is the core question at issue in debates about the universalism of human rights.

What do our findings so far tell us about the justification of normative universalism? Is the idea of a universal and uniform human moral faculty, a universal moral grammar, perhaps the high road to universalism? Is this the claim made and the ultimate point of the argument? If one takes the is/ought distinction seriously, however, it seems that this cannot be true. Whatever the factual makeup of the human mind may be, it has no bearing on the question of normative justification because no ought follows from it.

On the other hand, there are theories that doubt the relevance of the is/ought distinction.¹⁵³ Does the discussion thus far offer new support for these theories?

Two steps are necessary to answer this question. First, we will clarify in greater detail what normative universalism is about and what its justification might be. In light of this, we will then ask what kinds of consequences a plausible account of human moral cognition has for the understanding of the universalism of human rights.

8.10 The Epistemology of Human Rights Universalism

Are human rights only legitimate for some groups of human beings – say, Europeans, North Americans, Christians and whites? Is the global appeal of human rights based on an epistemological error, namely on the flawed idea that the core normative ideas of human rights can be justified across the borders of different communities? Or are these principles potentially of universal validity, and thus the project of human rights enjoys universal validity, too?

These are standard questions of any discourse on human rights, and ones that we hope to answer, if we are able, to clarify the content of universalism and determine at least the rough contours of its epistemological merits.

Universalism in the sense relevant for ethics and law is an epistemological stance.¹⁵⁴ It holds that the truth conditions of normative claims are the same for all human beings. These truth conditions are thus not relative to certain contingent properties of human beings, such as the groups that they belong to or their social, cultural, religious or other background. This epistemological doctrine applies to any normative propositions, including those concerning the moral rightness and justness of intentions, actions and states of affairs brought about by such intentions and actions. It also applies to the obligations people have, to what they are forbidden and permitted to do and to what rights they enjoy. Such universal standards are valid even if a given subject thinks otherwise. From such a point of view, a man is obligated to respect a girl’s right to education, even if he thinks that this is contrary to important norms of his patriarchal customary morality.

A defense of universalism may be based on two lines of argument: The first is an indirect argument for the plausibility of universalism derived from the implausibility of the opposite of universalism, which is relativism.¹⁵⁵ One argument commonly cited in support of relativism that we have already encountered is moral disagreement. The great diversity of moral opinions is taken to be an argument for the relativity of moral propositions. This is a fallacious argument, however, and for the following reasons.

To start with, universalism is a theory about the justification of normative propositions. It does not imply that universally valid principles are in fact universally accepted. The existence of even deep moral disagreement does not contradict universalist perspectives. Parallel observations are possible in the realm of science. For example, there are quite good reasons to assume that the Earth is not flat. These reasons are in no discernible sense relative to the contingent properties of a thinking subject making this proposition about the shape of the Earth. This does not mean, however, that there were not times in history when the very opposite of this insight was the common wisdom of the age. And even today, flat-earthers courageously try to make their case. However, their disagreement does not imply that there is no justification for assuming that the Earth is round.

Moreover, as we have already seen, there are factors entirely reconcilable with universalism that account for this – evident – diversity, such as nonmoral preconditions of moral evaluation, competing interests and passions or problems of ethical reflection – for instance, concerning the proper conceptualization of morality and its content or the consistency and coherence of normative reasoning.

Another important point is that relativism suffers from an insufficient determination of the factors to which moral principles are supposed to be relative. The reference to cultures or religions is a good example of this. Cultures and religions are not monolithic wholes but encompass a whole variety of ideas. Western culture (whatever the exact boundaries of this entity may be) was characterized for centuries by religious intolerance leading to bloody wars, authoritarian regimes and dogmatic systems of thought, contempt for human beings, various forms of racism, slavery, colonialism and imperialism, to name just some features that may spring to mind. At the same time, great ideas of human benevolence, freedom, autonomy and justice were outlined in forms of which some made history. Who is the true European – Kant or Friedrich Wilhelm III? Or – more precisely – the Kant of the principle of humanity and dignity,¹⁵⁶ the Kant of the disenfranchisement of women¹⁵⁷ or his Prussian king?

Another problem is the following: How do background factors such as culture or religion determine moral perceptions exactly? Cultural determinism is a highly implausible theory given the fact of constant change in the normative sphere. How can normative principles be determined by the cultures that these principles ultimately transform?

This points to the central problem of relativism that we already encountered in our survey of human rights history: It overlooks the importance of human subjectivity and autonomous reasoning for the development of ethics and the legitimation of law. Any cultural influence is mediated by human reflective subjectivity and reasoning. There are many cultural influences on human beings, but no person’s ethical identity is necessarily merely the product of the lullabies sung by the hand that rocked the cradle. On the contrary: Human reasoning is the ultimate source of ethical beliefs. Critical reflection can transcend the given parameters of culture and history and ultimately change their course. This does not mean that such autonomous thinking is what always or even predominantly determines the actual moral and political path of the human species. The reality is that ethics and law often fall prey to outlived customs, the self-righteous perpetuation of principles without thought, reverence to social authorities, powerful and harmful political emotions such as the hatred of minorities, fear and other such influences. But these need not be the last word. People are not just the malleable victims of such factors. As thinking subjects, they are in a position where they are responsible for reducing the importance of such driving forces for the course of history, and sometimes do so successfully – as the many steps taken towards a more humane world indicate, from the partial vindications of the rights of women to the defenses of democracy and human rights.

The second way of arguing for the plausibility of universalism is to defend certain normative propositions as universally justifiable. But is this possible? Is this not epistemologically naive? To answer this question, some remarks on the epistemology and ontology of morals are necessary.

8.11 The Epistemology and Ontology of Morals

The justification of moral judgments is crucial.¹⁵⁸ There is no a priori metacriterion that does not require scrutiny and defense. While ethical reflection is not about the revelation of higher truths, it is not about whimsical skepticism either. One needs reasons to justify any normative proposition, but one also needs reasons to doubt it. Simply maintaining that a certain normative position still could be doubted is not good enough. One can doubt any proposition, including the idea that the Earth is not flat, as flat-earthers happily do. No bolt of lightning will punish such a doubt, no voice from heaven will confirm that the Earth is in fact not flat. Nothing renders doubt about anything impossible. Nothing in the best argument imaginable compels a thinking subject to get its point. Demanding that a good argument be literally indubitable is a flawed demand because it is too exacting. Not even the best argument about the shape of the Earth irresistibly commands assent. This is a necessary consequence of the fact that the assessment of the truth value of any proposition is an act of human mental cognition. There is no epistemic authority over and beyond such acts of cognition. There is no more stable ground of human insight. Such acts of human cognition are the stuff of which the perception of truth is made.

This analysis is confirmed by a plausible ontological theory of morality: The problem of whether moral judgments refer to objective, mind-independent moral facts in the world (as moral realists assert) or not is a classic epistemological question. In the former case, a truth condition of moral propositions consists in correspondence with moral facts, in the latter case other truth conditions are key. As far as the ontology of morality is concerned, it is plausible to take moral cognition as being nonreferential: There are no objective moral facts in the world, the correspondence with which is the truth condition of moral predicates, as moral realists maintain.¹⁵⁹ Nevertheless, moral judgments are not merely subjective in the sense of them being idiosyncratic and relative to the outlook of a specific agent.¹⁶⁰ This is because their truth is authenticated by internal mental yardsticks for justified propositions, as in other areas of thought.¹⁶¹

It is useful to note that not only from the point of view of a nonreferential moral ontology, but also from a moral realist point of view, there is no escaping from acts of cognition that rely on internal standards of human understanding for their truth. This is because even a moral realist ontology ultimately is based on the assumed truth of such nonreferential statements: The correspondence of a moral proposition with a moral fact can only be ascertained by such an act of cognition. Moreover, the moral realist thesis that a truth condition of moral judgments is that moral predicates correspond to objective moral facts in the world does not itself correspond to an objective epistemic fact in the world. Its truth thus depends on other sources of epistemic justification.¹⁶²

Given this epistemic state of affairs, it is always possible to ask: Is this right? The possibility of doubt is the necessary consequence of the human capacity for unfettered, free thought. It is a fallacy, however, to mistake this possibility of doubting any proposition for the proof that no proposition is better justified than any other. It is necessary to have arguments for doubt that are more convincing than the arguments that (for the time being) confirm a certain normative position, at least if the argument is supposed to be more than a superfluous game.

Let us take the example of the principles important for the project of human rights, the idea that every human being has the right to protected dignity, autonomy and equality. There are many thoughtful and rich theories of human rights, as we have seen. And there are rather good reasons to assume that these rights are legitimate not only in Zürich, but also in Laos, Bogotá and Beijing, because the goods they protect, if ever enjoyed, turn out to be of extremely high value. This is shown, for example, by the longing for equal dignity and freedom of the victims of the Tiananmen Square protests, a powerful example of important “Asian values,” as is the struggle for freedom in Hong Kong, the fight against the military dictatorship in Myanmar or the rebellion driven by Iranian women against the regime, although in all these cases expressed from below, not through government announcements.¹⁶³ That there is a political case for human rights to protect such goods has been argued extensively above. That this basic good should be distributed equally among human beings, that it is not justifiable for some groups – say, party functionaries and their partners in the economy – to be entitled to such freedoms while peasants are not, seems equally plausible in light of the basic principles of equal treatment that are constitutive of justice. If these arguments and their foundations are doubted, reasons need to be given for this – for example, that serfdom is in fact good for some people (say, people of color) and that justice can as plausibly mean unequal treatment as it does equal treatment, at least for some (e.g. in the Global South). While the preceding remarks on the justification of rights surely have many flaws, they at least indicate that shouldering this burden is not an easy task.

8.12 Epistemological Resilience

There has been a long debate about foundational theories in ethics and law.¹⁶⁴ What are the ultimate principles of justification of normative precepts and what is their status? How can one argue for these principles without becoming entangled in infinite regress, dogmatism or tautology?¹⁶⁵ The line of the argument seems to point in the following direction: The normative justification of human rights is ultimately based on a fallible account of certain foundational moral principles that have proven to be resilient to systematic, conscious theoretical doubt. This account is informed by critical thinking that is committed to the decisiveness of reasons and open to the possibility that sometimes such reasons are good enough to rest a case. This stance implies a certain amount of trust that human understanding does not lead us entirely astray, and that consequently what seems well justified to us may, all things considered, in fact be true – an epistemological stance that can be defended against skeptical challenges.

This reflexive resilience of fallible normative principles when scrutinized by critical, in principle reliable human thinking that is respectful of reasons is the epistemological alternative to the trilemma of infinite regress, the dogmatic termination of the justificatory argument and tautology. The principles of justice, altruism and respect play an important role because there is good reason to believe that they (or some variant of them) are foundational not only for human rights, but for morality as such. They are not self-authenticating truths, but improvable, preliminary approximations to those principles that constitute morality.¹⁶⁶ Other principles may play an important role, too, such as the principle of the noninstrumentalization of human persons as a concretization of obligatory respect for others or those specifying the permissibility of otherwise-prohibited acts.¹⁶⁷

On this basis, further, more concrete questions can be asked – for example, whether the ethical thought formulated in the principle of humanity and the idea of human dignity can be put to ethical and legal use to guide the evaluation of specific problems, from assisted suicide to abortion to “dwarf-tossing”.¹⁶⁸ In this way, we may be able to approach a more comprehensive theory of human rights at a much-needed level of detail.

8.13 Universalism without Dogmatism and Human Rights Pluralism

There is thus a case for the universal justification of human rights. There is, however, another problem: The ethical ideas about human rights and the legal systems enshrining them differ, and sometimes more than just in detail. How are we to deal with this fact of ethical and legal human rights pluralism?

Can human rights pluralism in the real world be reconciled with normative universalism? This may seem implausible if the assumption is that universalism must demand normative uniformity around the world. But this is by no means a necessary conclusion, and for various reasons.

The normative questions human rights are concerned with are difficult and complex. Even if we assume that normative universalism makes good sense as an epistemological theory, this does not imply that it is reasonable to expect everybody to agree on all normative matters after a little deliberation. Moral and legal problems are more complicated than that. There is no reason to believe that the tortuous path of cognition is easier to walk in practical reflection than in other fields of human knowledge, including natural science. Modern theory of science is very varied, and the historicization of scientific paradigms is an important element of this reflection on the nature of science.¹⁶⁹ However, these qualifications do not change the epistemic claim of science that whatever seems to be the best theory available at a given moment enjoys this status universally. Higgs particles are supposed to make as much sense at CERN in Geneva as in Saigon. Science’s history of getting where it has got so far has not been a smooth ride to insight. On the contrary, its path is so fascinating and rich precisely because of the many difficulties encountered, dead-end roads pursued and breakthroughs achieved. The expectation that it might be any easier to gain normative insight through practical, ethical and legal thought seems somewhat outlandish.

Universalism as an epistemological stance gives no reason whatsoever for anybody to be sure that one has in fact reached normative insights that hold universal validity. This is because human thought, not least practical thought, is irredeemably fallible. It is important both for individuals and their sense of epistemic modesty and for collectives to remain conscious of this fact, so as to avoid unwarranted self-righteousness about the justification of the respective realizations of certain basic values that they (at the given historical moment) deem to be right. Any normative claim is at best a better or worse approximation of principles that are perhaps really universally justified.

The internal pluralism of interpretations of given norms within ethical reflection or legal systems therefore comes as no surprise. The origins of such diverse opinions are manifold. Legal norms, in particular, pose many problems and are often ambiguous and vague. The difficulty of identifying the content of such norms and the underlying normative principles is one reason for the diversity of legal opinions, even about the content of very fundamental norms. In addition, there are political agendas, interest-based legal interpretations and the like at play.¹⁷⁰

This consciousness of the fallibility of human practical thought has a pragmatic side: It implies the necessity of freedom for experimental ways of living,¹⁷¹ of attempts to explore different ways of realizing certain important normative principles.

This is also true for fundamental rights. How to best protect rights is not always obvious, and it makes good sense to see how certain approaches work in practice. There will be limits to such experiments,¹⁷² but universalism certainly does not categorically exclude the possibility of such searches for normatively justified solutions – to the contrary, the awareness that any normative conclusion will be tentative in fact demands scope for such searches.¹⁷³

The plurality of attempts to approximate a set of well-justified norms is therefore one of the practical consequences of this uneven path of practical cognition. Human rights pluralism is a necessary tribute to the difficulties of the problems posed for practical reflection on human affairs and to the fallibility of human thought.

Another reason why normative universalism is reconcilable with human rights pluralism is very straightforward, namely respect for personal autonomy and democratic choices, another source of the legitimacy of experiments in different ways of living. If autonomy means anything, it means the ability to make such choices, without any control of their content as long as they do not violate basic normative principles, most importantly the rights of others. Any legitimate order must leave scope for such acts of self-determination. This includes acts that may (and with good reason) appear less than wise, acts that are the product of partisan political interests or passing political passions. Human rights orders are not just products deduced from the textbook of normative insights. They are embedded in societies’ political processes, with all of the political upheavals and untidy decision-making that make democracies a living, admirable and demanding form of political life. Such political processes have effects on human rights law, sometimes becoming law, sometimes coloring the interpretation of norms, adding to and modifying the meaning of the pluralism of human rights.

Defending the kind of universalism outlined here therefore does not fall prey to dogmatism, Eurocentrism or moral imperialism. The epistemic status of a proposition is one thing, the practical consequences drawn from the status of such propositions quite another. One may think that certain normative principles have universal validity, such as the equality of women, without implying that this provides sufficient reason to invade other countries to spread this gospel by force in communities that think otherwise. Universalism with humility is a very helpful starting point.¹⁷⁴ The parallel to science may once again be illuminating here: Scientists may have compelling reasons to assume the existence of black holes. This does not mean, however, that they are entitled to muster an army to subdue all those astronomers who think otherwise. The reasons for this are the very rights that universalism defends, prominently including the autonomy of free human thought and decision-making. As this is so, the pluralism of normative orientations is the welcome companion of universalist normative theory. Pluralism, including human rights pluralism, thus not only is reconcilable with universalism, but is in fact its necessary consequence.¹⁷⁵

A further point needs mentioning: Values are not just intellectual positions. A certain view on freedom of expression or religion, equality or human dignity is not the same in nature as a position on the question of whether or not Higgs particles exist. In contrast to descriptive propositions, value statements have emotional dimensions in at least two senses. First, certain emotions are the consequences of moral judgments – say, moral outrage at the sight of gross injustices. As we have seen, the moral judgment is the precondition of such emotions; the emotions do not constitute the moral judgment, as emotivists maintain.¹⁷⁶ Second, emotions have a heuristic function, as we underlined above: They are necessary to fully understand what certain acts mean for human beings. One needs to have fathomed the importance of the goods protected by human rights to give them their proper weight: One will not fully understand what freedom of expression means if one has not experienced the preciousness of the possibility of free speech. Freedom of religion will only be appreciated appropriately if one has a conception of the existential importance that religion holds for many people, irrespective of one’s own religious, atheist or agnostic views. The meaning of prohibitions of discrimination can only be understood if one has a sense of the emotional harm inflicted through degradation by the many forms of discrimination – for example, by racism. Human equality will only acquire its full importance if one has a sense of what respect and disrespect for humans as equals feel like.

This consciousness of the existential meaning of human rights, of how they and their violations are spelled out in concrete human lives, is of central importance not only for ethics, but also for law. It is a key to the weight of rights. The appreciation of the importance of certain rights guides decisions about the inclusion or exclusion of rights in a given human rights catalogue. This appreciation is of great importance in the specifying of rights and rendering them concrete, in particular with regard to weighing and balancing rights in the framework of a proportionality analysis. One’s answer to the question of whether a limitation on religious expression – for instance, a ban on headscarves, kippahs, monks’ or nuns’ habits or turbans – is legitimate or not depends very much on how important one considers freedom of religion to be.

Such sensitivities need time to develop, and they will do so unequally in different societies. They may fade away because certain experiences become remote – for example, the experience of what religious strife may mean for a society or what life in societies without freedom feels like. In addition, powerful political passions influence the course of history, including constitutional history, which in the best case are channeled and put to good use to foster the core ideas of human rights, which is not an easy task: “We are not passion’s slave, but we have to apply cunning for reason’s success.”¹⁷⁷ However, the taming of such passions may not succeed, with potentially severe consequences for certain elements or even for the general architecture of law. Accordingly, normative views will shift, adding, reinforcing, altering the variety of normative positions held on certain issues. All of this can be expressed in human rights norms and interpretations. Given today’s internationally interwoven legal systems, such a variety of approaches will include a transnational dimension of norms with a human rights function. Human rights in legal reality are thus inevitably of a protean nature – heterogeneity and variety will continue to be their mode of existence as law.

These observations lead us to a final point: There can be no teleology in any universalist theory worth considering. Universalism does not imply that universally valid norms will necessarily become the law of humankind. There is nothing in history or theory to support this thought. Human beings have moral insights, but they are driven by many motives, and some of the most powerful such motives are the most destructive. The possibility of human societies at least approximately ruled by universally justified norms, although these norms are always only tentatively identified, is a hope, not a certainty, and is entertained to a large degree despite, not because of human history.

The pluralism of human rights orders may therefore violate universally valid norms of equality, liberty and autonomy, and the equal worth of human beings. It did so in the past, and there is no guarantee that it will not do so in the future. But this may not necessarily be the case. Pluralist orders can be something else, namely the fertile attempts of human beings to come a few steps closer to the ideas of justice, solidarity and respect that are the core attractions of the human rights project.

8.14 A New Case for Universalism?

In sum: Universalism as an epistemological position does not demand a necessary uniformity of the moral judgment of all human beings in the real world with its profound complexities and many influences on moral opinions. Nor does it rule out human rights pluralism. What is, then – to return to the important question asked above – the normative importance of a theory of moral cognition? Does it add something to the justification of the validity of those core moral principles that seem to be sufficiently resilient against doubt to form the heart of the normative argument for human rights? In particular, would a theory of a universal and uniform human moral faculty strengthen the case for a universalism of the kind outlined?

The answer is: A theory of the structure of moral cognition has no normative consequences as such, because normative arguments are required to justify any normative point. This also is true for universalist claims with the qualified content outlined. Even if there is indeed something like a universal moral grammar with justice, altruism and respect as its core principles that specify how moral judgements supervene over facts, this would not add anything to the justificatory argument. One would still need to make the case that these principles ought to guide ethical reasoning and legal norms and institutions and need not be corrected, say, by utility calculations or adherence to the principle of “might is right.” One way to make this case involves the arguments developed above, including the absence of any discernible reasons to doubt the validity of these central principles. There is no good case for assuming that it is morally right to intend to harm others, that it is just to treat equals unequally or that one should disrespect other human beings.¹⁷⁸

John Mikhail has explored the “weak normativity” of empirically given structures of moral thought. He argues that properly describing these structures of moral thought already is a meaningful step towards justifying their validity. This argument draws on Goodman’s theory of induction and its use in Rawls’s theory of justice. Goodman argues concerning the problem of induction:

We no longer demand an explanation for guarantees that we do not have, or seek keys to knowledge that we cannot obtain. It dawns upon us that the traditional smug insistence upon a hard-and-fast line between justifying induction and describing ordinary inductive practice distorts the problem. And we owe belated apologies to Hume. For in dealing with the question how normally accepted inductive judgements are made, he was in fact dealing with the question of inductive validity. The validity of a prediction consisted for him in its arising from habit, and thus in its exemplifying some past regularity. His answer was incomplete and perhaps not entirely correct; but it was not beside the point. The problem of induction is not a problem of demonstration but a problem of defining the difference between valid and invalid predictions.¹⁷⁹

Rawls’ use of the reflective equilibrium can be interpreted along these lines. From this perspective, Rawls provides a descriptively adequate account of the sense of justice, as shown by considered judgments. A descriptively adequate account of the sense of justice is an account that captures correctly the main properties of the sense of justice. Considered judgments are judgments that are not skewed by interest, bias and so on, as explained above. The argument advanced on the basis of this interpretation of Rawls is not to follow directly in Goodman’s footsteps and hold that a descriptively adequate account of the sense of justice already serves as the key to the justification of the principles of justice – the practice of justice in itself is not enough to justify the principles of justice.¹⁸⁰ This is because the principles of justice have to pass the test of rationality. The argument is, rather, that a descriptively adequate account of moral judgment provides presumptive, defeasible reasons for the justifiedness of the principles of justice identified as underlying considered judgments of human beings.¹⁸¹

For the problem of the normative relevance of the findings of moral psychology, the most important point regarding Mikhail’s profound idea is that it also seems to point to the view that, ultimately, it is normative arguments that count for the validity of a moral principle: In the end, such normative arguments are the decisive resource for refuting the defeasible reason for the justifiedness of moral principles that descriptively adequately account for considered judgments. In this sense, it confirms the argument presented here that there is no getting away from normative reasoning.

A very interesting problem looms in the background here: If it is true that human beings have certain mental faculties, the properties of which enable their thought but necessarily at the same time also limit the scope of their thinking, then any normative argument ultimately draws on these given mental resources. In this case, the principles that in fact determine moral judgment ultimately are the principles used to evaluate these very principles that in fact determine moral judgment: The justness of the principles of justice that in fact direct human moral judgment is then based on principles of justice that in fact direct human moral judgment – an obviously circular argument.¹⁸²

This can be formulated more generally as a structural problem of human thought: If it is true that human beings have certain mental faculties, the properties of which enable their thought but necessarily at the same time limit the scope of their thinking, any argument for the truth of a proposition ultimately draws on these very same mental resources. The principles that in fact determine human thought ultimately are used to determine the truth of these very principles that in fact determine human thought – a circular argument, as in the case of morality.

How to escape from this circle?

Referring to a hierarchy of principles does not help. The metaprinciples themselves would need to be evaluated with the resources of human thought whose validity is to be justified, begging the question just asked.

The conclusion to be drawn is a familiar one from human epistemology: There is no way of escaping the limits of human thought. But this does not mean that we have to embrace skepticism and abandon any hope of human insight. One important lesson from the history of skepticism is, after all, that there is no valid skeptical argument proving that what seems to be true to human beings is not really true. Any such assertion would lead skepticism into self-contradiction – the familiar self-contradiction of asserting that the proposition that there is no true proposition is itself a true proposition. Given this situation, there is no alternative but to assume that human thought, in theory and morals, does not lead us astray, but reveals something meaningful about the world. This argument is based on trust in human beings’ faculties of thought and understanding, which form the ultimate foundation of any serious effort to wrest some kind of insight from the vast swathes of what has not and perhaps never will be understood. This trust seems to be warranted by its results: Both science and practical life show that human thought does have a productive relation to the facts of the world – airplanes fly; vaccines offer protection; computers help us in our tasks. In the moral sphere, a world of justice, solidarity and respect is a rather attractive vision. There is thus no particular reason to mistrust human thought in principle. The specter of Descartes’ evil demon having become flesh in the cognitive structures of human beings and frustrating our search for insight should now find its final resting place.¹⁸³ The task is, then, to engage in constructive scientific work, while remaining aware of the limits of human understanding, of the fallibility of any theory and of any theoretical stance’s permanent need for critical revision.

For the normative force of psychological facts, these arguments lead to the following conclusion: Even if the properties of human cognitive abilities, including ethical cognitive abilities, ultimately set limits to human understanding, no normative proposition is valid or invalid simply because it is based on empirically operative principles rooted in the nature of human cognition. It is not such psychological facts that are decisive for the validity of normative propositions: Rather, a normative proposition convinces as a piece of normative argumentation against which no valid normative doubts can be mustered. Normative arguments ultimately count only insofar as they formulate compelling normative reasons, not because they are rooted in the psychological makeup of human beings, even if these normative arguments themselves ultimately are the offspring of certain empirical properties of human cognition.

If one accepts this conclusion, the need for normative theory-building in the defense of human rights needs to be reasserted. Psychological theory is no substitute for it, nor can psychological theories ultimately challenge the normative case for (or against) human rights.

This result does not render a theory of moral cognition superfluous: As we have seen, important points of a theory of human moral cognition are, first, to provide a rich explanatory theory of human moral judgment and, second, to perform the critical function of dispelling doubts stemming from theories of the mind and its evolution about whether an ethics of human rights is possible. Moreover, this theory serves a third and rather intriguing function, one that can be identified as the ultimate perspective of this inquiry: A theory of human moral cognition can show that there may, perhaps surprisingly, be at least a partial convergence between what normative theory tells us is right and just and what humans in actual fact empirically regard to be morally right and just. This is not self-evident: Human beings could, for instance, be the selfish utility maximizers that some theories both past and present imagine them to be, creating a mismatch between the world of moral principle and the facts of human moral cognition. In a profound sense, then, human beings in the realm of human ethical understanding are what they morally ought to be: beings committed to justice, respect and concern for others. Given humanity’s track record of self-inflicted suffering, this is perhaps a precious source of hope, although certainly limited in scope.

Epilogue The Tilted Scales of Justice

This inquiry, with all its limits and flaws, has drawn a partly familiar, partly surprising picture about the place of human rights in human life and the human makeup.

Our conceptual and analytical clarifications have indicated that human rights are best understood as a subcategory of a particular normative position in which human beings find themselves: the position of having a subjective right. This position consists of complex normative relations including a rights-holder’s claims to do or not to do X, a necessarily correlated duty of the addressee of the right to do or not to do Y, a privilege to do or not to do X on the part of the rights-holder and a necessarily correlated no-right of the addressee that the rights-holder does or does not do X. Powers and immunities are often part of the normative content of such a right.

Human rights are those subjective rights that protect specific goods of human beings, in particular dignity, life, certain liberties, equality and the material means for pursuing a dignified life. They are part of ethics and legally enshrined in (national) constitutions and supranational and international law. To limit human rights to human rights guaranteed in international law is a fundamental misunderstanding both of the current global legal architecture of human rights and of the explicit intentions of those constructing it.

The history of the development of the human rights idea and ultimately of the institutions embodying it reveals quite clearly that this idea has deep roots in humans’ social form of life. This does not mean that we can expect to discover a universal declaration of human rights written by a prehistoric cave dweller. The lessons taught by the history of human rights are much more complex than this. This history is not about every aspect of ethics, but it is not just the history of the explicit concept of human rights either. There are various normative phenomena that are not human rights but still are important for the history of human rights, because they contributed to the formation of those building blocks that ultimately became the material for the explicit concept of human rights.

The most plausible account of this trajectory is that the raw material of the human rights idea includes specific, concretely situated not arbitrary, but principled moral judgments about the justness and moral rightness of particular intentions, actions and states of affairs – archetypically exemplified by Creusa’s rebellion against the permissibility of her rape in which she invokes the rights she enjoys, even towards a god. During a long historical process, these specific judgments were slowly objectified as to the goods protected, framed in abstract terms, generalized across cases, universalized across people, purged of persistent patterns of exclusion, ultimately made explicit in political and ethical thought and very recently turned into working institutions of positive law. This process was not a smooth ride but – as always in the history of ideas and social change – a tale of progress and regression, during which generations who lived under new and treacherous ideological stars consigned to oblivion the hard-won insights of their forbears, a history of struggle and of the successful suppression of ideas by force. A good example of this, with little consequence for the author but not entirely insignificant consequences for the history of human rights, is the banning of some of Las Casas’ works by the Spanish Inquisition and the effects that this ban had on Europeans’ self-perception of their role in the world, among many other such cases involving the consignment of unpleasant historical truths to the “memory hole” that Orwell so clear-sightedly described.¹

Some examples have illustrated this conclusion – from the political rights of Athenian noninclusive democracy to Enlightenment theories of rights. The discussion did not seek to provide a full account of the complex social, political, economic, religious and cultural contexts of these examples. Moreover, and importantly, no assumptions were implied about a linear, continuous, triumphant historical process, coherent ideas about rights over millennia or simple causal connections between the thoughts about rights of different epochs. The historical review served limited expository purposes: It pursued merely the modest aim of highlighting some important findings for the specific cognitive interests of our inquiry.

Importantly, our study has made the case that intellectual elitism and cultural myopia, which sometimes even smack of racist bias, must be avoided. If we are to steer clear of such prejudice, investigating indigenous cultures, including oral and acephalous civilizations, is of crucial interest. This book has argued that it is entirely implausible to deny human beings living in such societies and cultures the basic, principled moral intuitions that form the raw material of human rights. The voices of the victims of colonialism or slavery, as far as we know anything about them, speak rather strongly against the assumption that the human beings living in these cultures were just moral blank slates, who might have felt the pain of, say, dying in the desert like the Herero in the German genocidal campaign of 1904, but not its injustice.

There is thus substantial historical evidence to show that these moral intuitions are not limited to white, male Europeans (or any other subgroup of the human species), but form part of the common ethical heritage of human beings. The journey from such intuitions to an explicit concept of human rights is a long one. There is, however, no reason to assume that only one group of people can make this journey. No one group has privileged access to the idea of human rights. It should be noted that Europeans, the “West” or the “Global North” did not travel smoothly down this road either. The idea of human rights met with fierce resistance in Europe as elsewhere, and whatever sway it has over human affairs was wrested from the hands of the leading social and political powers.

The more recent history of the explicit idea of human rights only confirms these findings. The human rights project has been promoted by people from very different cultural, political and religious backgrounds, but with a common idea – the idea that governmental and social power has to be limited, that human beings and their well-being count equally, that humans owe each other concern and respect and that liberty is no minor affair for a human life.

Importantly, the history of human rights illustrates a further point. The problem of the inclusion of all human beings in the sphere of protection afforded by human rights is one of the core issues of human rights history. It lies at the root of major historical tragedies, including racism, slavery and the subjugation of women. This problem has two dimensions: first, the formulation of a proper concept of what makes a human being human and thus entitled to rights; and second, the proper application of this concept to all members of the human species. The former does not necessarily entail the latter. Various theories throughout history had a plausible idea of what human beings were but did not apply it to subsets of the human species, such as indigenous Americans because of racist prejudice or women because of misogynist ideas about women’s inferior capacity. This problem is far from solved – today, for instance, problems of bioethics or the rights of elderly people (which have become matters of life and death during the Covid-19 pandemic) show that the question of who is to be regarded as a (full) human being and bearer of rights is still open to intense debate in many aspects. Moreover, the question of the rights of nonhuman animals indicates yet another frontier of the theory of rights. The question of whether human beings have human rights is, however, independent of the question of whether and in which sense nonhuman animals have rights, too. One can assert the former without precluding the latter.

Our reflection on the justification of human rights has suggested that any justificatory account of human rights needs to formulate a theory of goods essential for human beings that are important enough to be protected by human rights, a political theory of the role of human rights in society that contributes to the enjoyment of these goods and a theory of the normative principles of justice, solidarity and respect for the intrinsic worth of human beings, the principles that are the ultimate sources of the normative position of rights.

In light of what we can reasonably assume about human beings, the protection of dignity, life, liberty, equality and the means to lead a dignified life is well justified by an anthropologically informed theory of human goods. Political theory offers no compelling case against human rights as ethical and political principles and legal institutions. Human rights promise not a perennially blissful Elysium, but something that nevertheless is very precious: an order in which certain basic demands stemming from the principles of justice, solidarity and respect are satisfied to such a degree that human beings can pursue meaningfully whatever their aim in life may turn out to be. Given the experience of the mass appeal and destructive power of totalitarian ideologies propagating the worthlessness of human life and existence as such, doubts about the need to protect basic goods of human beings by means of human rights are politically naive and a slap in the face for the victims of these regimes, who perished because of the mass support for the contempt of their rights.

A theory of human rights cannot rest its case here. Recent research on moral cognition and evolution has put the question of whether human rights are justifiably regarded as universally obligatory normative principles squarely on the table. Perhaps, this research implies, human rights are something quite different, nothing but a cognitive illusion necessarily produced by the structure of the human mind but without any claim to normative justification. Moreover, some evolutionary accounts of morality have asked whether a morality of human rights and correspondingly the institutions of the law are in fact irreconcilable with the kinds of proximate cognitive mechanism that natural selection could have produced in humans and that empirically determine moral judgment – that is, whether human rights do not ultimately demand the impossible of human beings as they really are.

Our survey of current neuroscientific theories of moral cognition and the evolution of the human mind has given ample reason to conclude, however, that nothing in this theoretical field undermines the legitimacy of human rights. There is no compelling empirical evidence or theoretical argument proving that human rights are cognitive illusions or anything of the sort. A sufficiently complex theory of evolution confirms the possibility of human cognitive faculties (whatever they may turn out to be) that do not merely produce the illusion of being directed by altruism, justice and respect, but are in fact determined by these principles.

A promising theory to account for the structure and content of human morality is the mentalist theory of moral cognition, assuming a common framework of human moral judgment, thought and sentiment, generated by a shared faculty of moral cognition – a moral competence with well-defined principles that enables human moral judgment with its rich volitional and emotional consequences. By means of a differentiated account of human cognition, this theory thus reconstructs an assumption that has guided the history of ideas since antiquity, namely that humans are endowed with a moral understanding that is one of the defining features of their human identity.

An empirical theory of the human moral mind provides no normative arguments for the justification of human rights as such. This is also true in the weak Goodmanian sense of taking an epistemic practice as an element of justification. Understanding a descriptively adequate account of human moral cognition as a presumptive, defeasible reason for the justification of the principles identified as empirically determining human moral judgment leads to the same conclusion: Ultimately, arguments for the normative validity of normative propositions must be derived from normative theory, not from psychological facts.

The theory of moral cognition nevertheless fulfills three important constructive functions. First, it helps to form nothing less than a rich, empirically grounded explanatory theory of the cognitive basis of human moral judgment and legal thought, which, if correct, would be a major scientific achievement. Second, it is a tool to assess critically theories that aim to delegitimize ethical precepts such as human rights with the means of moral psychology, neuroscience or evolutionary theory. Third, as a rather intriguing perspective, it provides reason to think that there is at least a partial congruence between normative ideas justified by normative theory and the empirical structures of human moral thought: In certain respects, human beings may turn out to be the creatures they ought to strive to be.

Any argument for the validity of human rights is fallible and open to criticism. But this does not mean that the propositions “It is justified to enslave black people” or “It is justified to rape women” enjoy the same epistemic status as “It is not justified to enslave any human being” or “It is a severe crime to rape women.” The history of skepticism has taught us that there is no argument proving beyond possible doubt that what seems true or right to human beings is in fact true or right. However, it also has long been clarified that there is no argument showing that what seems true or right to human beings is in fact not true or right – the latter being the well-known self-contradiction of a skepticism asserting the truth of the proposition that no proposition is true.

If it thus is possible that human beings do in fact understand something true and right about the object of their reflection when they think that they understand something true and right about an object, there is only one epistemological way forward: to seize this precious epistemic chance and to engage in the constructive, fallible pursuit of insight, based on whatever argument can be mustered in favor of a given proposition. That human beings are bound by the necessarily existing limits of their (moral and legal) understanding forms no reason not to travel as far as these boundaries permit. Looking at concrete arguments may provide some epistemological encouragement to pursue this course: For instance, it seems rather hard to maintain that there are not better arguments for the prohibition of slavery and rape (and the implied rights of people) than for the permissibility of such actions.

In this fallible, albeit meaningful sense, universal human rights are justified. However, the universalism that they imply is open for concrete human rights pluralism. One reason for this is epistemic modesty, which does not confuse the possibility of universally valid insight with the assumption that one oneself (as an individual or as a community) has actually gained this universally valid insight. Abstract principles of human rights can, moreover, be realized in more than one form – there is not just one concrete way to a meaningful protection of human rights. Furthermore, human rights spelled out in concrete terms are about important choices of individuals and communities. Given human autonomy, both individuals and the communities they form retain the right to experiment with new ways of living, including varying ways of rendering human rights concrete that must be respected as long as they do not betray the core promises of human rights. Human rights universalism is therefore not wedded to a rigid system of indubitable content revealed by solitary master thinkers, but is a living aspiration, a quest to critically reappropriate the idea of human rights in endeavors that in the best case are collective and democratic, excluding nobody.

In sum, the argument leads to this encouraging conclusion: Principles of egalitarian justice, of human solidarity, care and respect for human dignity, together with a sufficiently rich concept of human existence and a political theory of the means for human flourishing embedded in a plausible theory of mind and its place in natural history provide good reasons to believe that the idea of human rights is as well-justified as anything ever has been in the history of fallible human thinking about morality and law.

Such a theory of human rights, which has answers to the theoretical challenges formulated by the many kinds of skeptics encountered in this inquiry, constitutes an essential element of the intellectual defense of human rights. This would be no small achievement.

The Greek poet Giorgos Seferis wrote in his poem Santorini:

Naked we found ourselves on the pumice-stone

watching the rising islands

watching the red islands sink

into their sleep, into our sleep.

Here naked we found ourselves, holding

the scales that tilted towards

injustice.²

This is how things are, after about 100,000 years of human civilization, ingenuity, foolhardiness, petty sordidness, enchanting and even sublime nobility, after much longing for and the intermittent reality of justice. We stand naked on the rough pumice-stone of our epoch and the scales are still tilted towards injustice. Human rights represent one attempt to balance these scales, never completely, never forever, never on all accounts, but for long enough to show sufficient respect to what humans owe to the better parts of their own humanity.

Human rights therefore are not trivial. They are more than playthings to satisfy one’s intellectual ludic drive. Human rights are not the means of solving all of the world’s problems. But there is a lot that depends on rights, most importantly the well-being of individuals and sometimes even their dignity and lives.

Moreover, respect for human rights is of great significance not only for those who suffer from human rights violations. It also is of some consequence for those people lucky enough to have their human rights sufficiently respected. At least this is so if they belong to the perhaps not insignificant number of people who, despite their own favorable circumstances, are still not able to breathe freely when they have to witness the ongoing tragedy of folly and pain, of repression and contempt for the equal worth of human beings that already have marked too much of human history because they sense the force of some probably quite common elements of human experience: To feel the spark of generous, liberating magnanimity that adds considerable beauty to humans’ inner life and that makes it natural to care for our fellow mortals, whoever they are, wherever and under whatever stars they spend their short and precious time on Earth; to be enlivened by the unpretentious daily work of justice and human goodness; and thus to long for the profound relief of some fresh air bestowed by any modest steps towards a culture and law of human decency.