Moral Heteronomy. History, Proposal, Reasons, Arguments: Introduction

1. Moral Autonomy, Realism in Ethics, and Kant

The basic form of the doctrine of moral autonomy is the claim that human beings have the moral law in themselves, and make use of such a law whenever they reason about what is right and wrong.

From an historical viewpoint, the rise of the autonomous theories of ethics to the international philosophical scene comes at the end of a long trajectory, whose point of departure is medieval political thought.

Mainstream medieval theories of power can be read as a systematic development of the paulino-augustinian principle that nulla potestas nisi a Deo (Rom. XIII.1). The principle is meant to establish that no authority has in itself a right to rule, because any kind of established and legitimated power is just a vehicle of God’s plan towards the world. Accordingly, medieval theories of power commonly asserted that both kings and their subjects are subordinate to the law springing from God’s Will. Now, since God incorporates the contents of his Will into the creation, and these are the ends that all creatures should actualize, it follows that such contents turn out to constitute the natural law of good and evil. Consequently, mainstream medieval theories of power qualify the ontological nature of the moral law in terms of both contingency and necessity: while the fundament of ethics relies in what God wants, the actual contents of the moral law that He commands are essentially necessary given the way the actual world is, and equally rule all earthly creatures.1

The early rationalist agenda questioned such a voluntarist stance of the foundation of ethics, at least on some very natural readings of Spinoza’s and Leibniz’s texts. It is customary to sketch this position in terms of the conjunction of realism, innatism and necessitarianism towards the natural law. By the words of G.A.J. Rogers:

[…] the rationalist holds that it is possible to know many major moral truths by recourse to reason. It is often thought central to this view that true moral propositions are logically necessary, and false ones are either implicit or explicit contradictions. On such a view the law of nature could not change. […] An implication of this view was to encourage a belief in some sort of commitment to innate moral knowledge. Since there are many human activities which are taken to be natural (the love of parents for offspring, and viceversa, for example) then these are also to be thought of as morally right conduct or morally worthy feelings. Conversely, unnatural actions (for example the rejection of the child by the mother) would be known to be unnatural and therefore morally wrong.^[2]

Such an approach constitutes the philosophical ground which governs the assumption that ethics is an autonomous field of human experience, i.e, ethics is fully entitled to its self-justificatory strength because it is a constitutive form of human experience which provides access to a realm of basic facts, and is not reducible to any other source.

The formation of the doctrine of moral sense is a further step beyond. Through the longstanding influence of Spinozism on Seventeenth British moral philosophy,2 moral rationalism slightly shifted to a critique of the traditional accounts of authority, power and morality. As a consequence, the doctrine of moral sense acquires an evident reactive nature, that addresses a plurality of views with little in common.3 It denies that:

a) human beings obey the law because of its proceeding from God (e.g., traditionalist theorists of absolutism as R.Filmer or J.B.Bossuet defended versions of this view by the massive use of scriptural reasons);

b) human beings obey the law because of its having a coercive power (e.g., positivist theorists of absolutism as T.Hobbes argued for the constructivist origin of the coercive properties of the law);

c) moral virtues have a value in reasons of their being decreeed by God (e.g., J.Locke held that ethics has a demonstrative nature, although God could have chosen different principles from the actual ones);

d) human beings’ moral virtues are delusive since what counts as a virtue is nothing more than an oblique way to satisfy an egotistical interest (french moralists as Mointaigne and La Rochefoucauld, and anglophone polemists as Mandeville, provided many phenomenal arguments in support of it).

Finally, the moral sense theory became an essential part of the French Enlightenment project, and was advanced by philosophers as a reason for the antitraditionalist attitudes that constitute the deepest core of the movement. Following Shaftesbury and Hutcheson, Voltaire attributed the capability of morality by human beings to their natural sensibility. He wrote:

[…] agriculturists, manufacturers, artisans, do not go through a regular course of morality; they read neither the De Finibus of Cicero, nor the Ethics of Aristotle; but as soon as they reflect, they are, without knowing it, disciples of Cicero. The Indian dyer, the Tartarian shepherd, and the English seaman, are acquainted with justice and injustice. Confucius did not invent a system of morals, as men construct physical systems. He found his in the hearts of all mankind.4

The critical target of Voltaire’s statement is a conflation of all the positions that are included in points (a)-(d). Traditional philosophers, theologians, and law theorists spent their time by trying to give an account of human agency in terms of absolute goods which precede any moral reasoning. Ends are out there — the old-fashioned intellectual claims -: either they are embodied in what a society holds right and wrong are or they are communicated by the Revelation. Conclusion follows: individuals have no justification to reject traditional ethical commands. Contrary to such a view, the moral autonomy theorist argues for both the redundancy and falsity of any voluntarist, positivist and epicurean approach to ethics. Human beings are not indeed in need of external constraints on their moral experiences, expectations, and aims, because they are perfectly able to rule themselves in terms of their natural capabilities.

Kant bursts on the scene precisely at this point. His notorious condamnation of all heteronomous approaches to morality, and his defence of the claim that just a self-legislated law forces individuals to accomplish their duty, provide an explicit and seminal formulation to the doctrine of moral autonomy.5 This formulation gives a technical expression to the plurality of motives rooted in the secular debates on the nature of authority and moral law. As such it depends on these conceptual contexts, and addresses the most important issues that had emerged along the history of moral and political philosophy.

Particularly, Kant’s theorizing aims at defending a realist viewpoint in ethics which is not grounded on the recourse both to a trascendent and immanent discourse on the origin of the law. Regulating moral reasoning on the constitutive characters of ethical experience should drive human beings towards an agreement on the contents of the law on pure rational grounds.

2. From Kant to deontology and mainstream Kantism

After Kant the autonomy jargon has established itself as the dominating view in ethical, social and political debates. Accordingly, any heteronomous attitude is ordinary felt as basically unsuitable. There is indeed something unpalatable in making the obligation to morality depending on heteronomous motives. Such a resistence is due to the fact that the moral heteronomy theorist is charged with committing a categorical mistake: she introduces a kind of non ethical normativity into the realm of morality, which is commonly conceived as the basic domain of normativity. As a consequence, a heteronomous theorist often appears almost embarassed in arguing for a so shameful doctrine as one which denies that ethics is autonomous.

Naturally, not all positions in contemporary metaethics are Kantians. Nonetheless, in light of the pervasive tendency to construe ethics in terms of the notion of autonomy, non Kantian views commonly abstain from highlighting their grounds in terms of the notion of heteronomy; and contemporary Kantism appears then as the mainstream view in the present debates about the nature of ethics. As such, the expression contemporary Kantism is an umbrella term which covers a wide range of positions somewhat recalling their Kantian ancestry.

The common characteristic to all these is their development of Kant’s doctrine into a form of pure deontology. By detaching Kant’s theorizing about the autonomous source of morality from its realist, innatist and necessitarian features, contemporary Kantians sketch their position as a kind of proceduralism that stands beyond any discussion about the realist view that moral judgements apparently endorse.

Briefly, a contemporary Kantian typically holds a conjuction of the following propositions:

1) the constitutive character of moral agency consists in the ability to rule behavior by the application of universal norms that the agent freely chooses;

2) acting morally for an agent implies that she should give true apt reasons in support of her freely chosen application of universal norms;

3) being justified in the performance of a morally relevant action means that an agent is acting morally.

In the conceptual context of “(1) & (2) & (3)”, autonomy is defined as the capability to access moral goods, means, and ends only by the free act of moral reasoning. On the contrary, any theory which assumes that moral contents precede moral reasoning, and are the objective criteria for sound ethical theorizing, is heteronomous. Consequently, realism is banished from the domain of morality as the main reason conducive to heteronomy, and Kantism is to be understood as a strategy for implementing objectivity and cognition into ethics without assuming demanding claims about what is real and what is not.6 Contemporary Kantism is then a rebuttal both of non cognitive approaches to ethics (i.e., varieties of emotivism) and cognitive ones (i.e., stance independent conceptions of moral truths as any kind of moral realism, and alethically relativist conceptions of moral truths as context-dependent and communitarian theories).7

Evidently, such a view is strongly controversial from a purely Kantian viewpoint. While contemporary Kantians sometime argue for interpretation of Kant which makes a constructivist of him, it is difficult to split Kant’s realism and Kant’s assumption that cognitive moral propositions are necessary and a priori from those deontological intuitions that are developed into contemporary mainstream Kantism. For example, a demanding Kantian rationalist as C.Peacocke objects to Korgaard’s proceduralism that it makes sense only in the case that moral properties are mind-dependent, where natural readings of Kant should attribute him the proposition that moral properties are not mind-dependent.8 A serious rationalism requires indeed that the cognitive entitlement to holding a proposition is entirely an a priori matter, proceduralism being an a posteriori strategy for handling moral phenomena.

Nonetheless, I do not think useful to dispute the received terminology, and I do not think that saying that contemporary Kantism is not really Kantian is a very powerful argument against the autonomy view.

This position is advanced in a wide variety of ways. As Kant’s notion of autonomy contrasts with heteronomous conceptions in metaethics, morality and politics at once, that is, it opposes all the claims of the conjunction “(a) & (b) & (c) & (d)”, also mainstream Kantians defend their approach in different field of inquiry.

Some of them deploy their arguments in order to account philosophically for what ethics is, and where the source of moral experience is to be looked for. Some others aim at establishing moral contents according to a procedural scheme. Finally, most social and political philosophers (as well as epistemologists of social sciences) indwell a discourse on democracy, rights, communities and international relations which finds his most evident source in proceduralist reading of Kant.9

3. A map of the alternatives to moral autonomy

The papers in the present collection draw a radically different scenario from that which the autonomy jargon presumes to be unquestionable. As such, they constitute an endeavour to dismantle the alleged mandatory nature of moral autonomy in metaethics.

This special issue is divided into three parts. The first is historical in nature, and presents approaches to ethics that are not committed to the difficulties raised by the assumption of the autonomy view or that conceptualize the notion of autonomy in an alternative way to the Kantian tradition (see M.S.Vaccarezza on Aquinas and M.Micheletti on the Cambridge Platonists). The second is expository, and provides the reader with an overview of the most important heteronomous doctrines that have been proposed during the last century and are nowadays intensely debated (see E.Grimi on Anscombe, N.Marcucci on Foucault, E.Meade on Levinas, G.Cavallo on the Genealogical Approach to Values, M.Damonte on the Divine Command Theory, and D.Bondi on the Ethics of Ecology). The third is argumentative, and offers reasons in support of moral heteronomy (see C.Huff & A.Furchert on the psychology of moral experience, and A.Aguti on the defence of the heteronomous approach to ethics), or against the alleged mandatory nature of autonomy for metaethics (see S.Kahn on Korsgaard’s arguments for autonomy, D.Bertini against the normativist stance of mainstream Kantism, and A.Conty on the contradictions of the proceduralist account of democratic practices).

Now, saying that the articles argue for a homogeneous doctrine is hard. Very different views on ethics and moral contents, possibly not compatible among them, are available to the reader. Nonetheless, there is something basic that all the theories and the arguments presented in the papers share: all of them moves from a rejection of Kant’s and Kantists’ claim that (moral) normativity relies upon the exercise of the autonomy of the will. Consequently, the articles deploy a plurality of strategy for answering to the challenge that mainstream moral assumptions raise against any non proceduralist metaethics.

I list here some of these reasons. A promising starting point consists in developing a non Kantian semantics for the notion of autonomy. For example, Vaccarezza holds that a good way to account for Aquinas’ views is to read them by making use of M.Rhonheimer’s notion of theonomy. According to the swiss philosopher, practical rationality is an autonomous domain of inquiry, because moral claims cannot be deduced from metaphysical propositions, neither can be reduced to non moral facts. Nonetheless, moral knowledge finds its legitimate measure into the transcendent existence of eternal divine knowledge. Such a claim allows for assuming that the notions of autonomy and heteronomy are complementary. That is, being morally autonomous for a human being is to self-discover her core adherence to God’s law. As a consequence, Kant’s oppositional definition of autonomy and heteronomy is rejected, and both autonomy and heteronomy can play a role in the explanation of normativity and moral experience.

Micheletti argues in a similar manner in relation to the interpretation of the Cambridge Platonists. According to him too, Rhonheimer’s semantics for autonomy is possibly a key to understand how authors as Whichcote, Cudworth, Smith, More, and others, during the early modern times, have construed ethics in terms of the anthropological capability to take part to the realm of God’s ends. This being the case, Micheletti shows that the Cambridge Platonists’s view are relevant to the present metaethical debates that make the philosophical marketplace red-hot.

On the same grounds, Grimi reads the seminal criticism to Kant by E.Anscombe. In her terminology, intention is the sinergy of autonomy and heteronomy. None of them comes alone: a moral reasoner never experiences the actualisation of her autonomous rationality without experiencing heteronomous features of reasoning at the same time, and viceversa. Moral phenomenology explains why autonomy and heteronomy are indivisible: they are united in the act of morally experiencing something, i.e., in having an intention to ethically perform something.

Finally, by the assessment of Foucault’s genealogical method applied to autonomy, Marcucci highlights what is hidden by the propositional surface of the notion. Foucault tries to excavate its historical development in order to discover the meaning of the vocabulary of moral autonomy. Particularly, according to Marcucci’s reading of the French philosopher, what is thought when someone says autonomy is not what Kant claims: far from being a milestone for liberation and freedom, autonomy endorses obedience, and is a radical mutilation of spontaneity and authenticity.

Anyway, most of the authors of the present collection works within an overtly Kantian semantics. Consequently, they qualify the notions of autonomy and heteronomy in a standard way. Evidently, such a task is difficult to tackle: given that Kant is the point of arrest of a longstanding tradition that comes at construing autonomy as a conceptual weapon against the conjunction “(a) & (b) & (c) & (d)”, and sometimes it seems that he defines autonomy in terms of not being heteronomy and, conversely, heteronomy in not being autonomy, it turns out to be natural that the articles address different features of the Kantian topic.

A way to unpack such complexity is to rely on the distinction between the prescriptive and descriptive nature of morality. According to (mainstream) Kantism, the normativity of ethics consists in refuting any relation of implicature between ought and can. It is not a primary issue whether the ontological constitution of human beings affords them to be able to perform actions so and so. Normativity is an autonomous realm of experience because ethics does not deal with what human beings actually experience when they perform something morally relevant: rather, ethics has to do with how they should act.

As a consequence, a palatable exit strategy from Kantism may be identified in showing that ought implies can, and that any prescription is useful only in case that it can be actualized. Thus, descriptive approaches in morality matter at a high degree.

Huff & Furchert provide arguments by assuming such a descriptive attitude: they make use of a big amount of psychological researches on morality in order to establish that Kantists’ approach to agency is simply a chimera; i.e., moral actions are complex, and depend on a multiplicity of heteronomous motives.

A move forward within this context is to prove that arguments for autonomy undermine the production of ethical goods. To speak in a radical manner: if the realm of ethics is transcendentally constituted, it is constituted by heteronomous attitudes and not autonomous ones.

Meade advances a genealogical reading of Levinas, whereby the ontological dependency of acting morally on the relationship between the subject and other agents progressively acquires more and more definiteness within the texts. An agent acting freely from external constraints is an agent unable to provide an ethical entitlement to her actions; because the source of any moral experience is how an agent becomes responsible towards the expectations of those real others she encounters.

Cavallo moves from the traditional objections to Kant that are immanent in the idea that the source of ethics are moral feelings to the outcomes of the genealogical inquiry into the nature of values. According to both C.Taylor and H.Joas, moral values are never negotiated a priori by an abstract reasoning concerning how hypotetical agents could answer the self-legislated behavior of a subject. Indeed, values are in no way facts that come to be accessible to pure practical reasons. On the contrary, we learn why they exercise a normative appeal on us, through investagating how they have become reasonable answers to some pressing social issues. Identity is a matter of which values a subject endorse, because we are socially constituted by what counts as a value in our social context. Values show in this way their inescapable cultural nature. Therefore, a heteronomous conclusion follows.

Finally, Aguti defends the Divine Command Theory from the challenges of the autonomous viewpoint by showing how a moral realism which assumes that ethics is an autonomous domain of human experience and traditional theological voluntarism can be made compatible. Following others theorists, Agusti provides reasons in support of the claim that while moral realism without God’s commands is unwarranted, the fact that God commands to obey necessary true moral propositions enforces and justifies the production of ethical goods.

Other authors adopt an alternative method to demolish the autonomy building, i.e. they argue that reasons for an autonomous approach to morality are unable to provide a justification for its assumption.

Conty addresses how Rawls defends his constructivist moral and political theory. Her main claim is that mainstream Kantism is like an avalanche which devastates any real identity. While proponents of proceduralism suppose that their approach is universal and able to defend individual rights and any kind of social minorities, Conty proposes a reading of them that highlights their particularism and their misleading attitude towards differences. Once such a particularism is unveiled, mainstream Kantism becomes completely unsuitable to keep its promises, and slips into an unreasonable view.

Kahn considers in details Korsgaard’s argument from spontaneity and regress argument in order to make evident that the inner logic of both cannot produce a valuable support relation between evidence and conclusion. According to the Kantian philosopher, the arguments would prove that it is a constitutive feature of morally relevant agency that if an agent acts morally, the agent acts autonomously. Kahn offers sound reasons for a counterexample, that is, for the claim that an agent can morally act even if she does not act autonomously. This being the case, Korsgaard’s arguments should be evaluated as overrated.

Finally, Bertini attacks a core element of contemporary Kantian constructivism, i.e. the normativist stance. According to mainstream Kantism, cutting (1) means deleting any possibility to act morally. A case study explains why (1) is not suit to understanding moral agency and cannot make justice to the ordinary use of moral language.

4. Conclusion

Speaking of heteronomy today implies that a judgement of condemnation is to be uttered. My hope is that the present volume shows why such a category should be used as an interesting heuristic tool in metaethics and moral philosophy.

For instance, the essays by Bondi and Damonte do not develop theoretical insights, but provide an enlightening discussion of trendy doctrines as the Ethics of Ecology and the Divine Command Theory.

Obviously, I am not sure that this special issue will succeed in accomplishing its task. But I am profoundly grateful to all the contributors for their enthusiastic adherence to the project, and their outstanding work on their topics.10


  • Bagnoli, C. (2015), “Moral Objectivity: A Kantian Illusion?”, Journal of Value Inquiry, 49 (1-2), pp. 31-45.
  • Bertini, D. (2005), “L’immaterialismo berkeleiano come teoria religiosa dell’esperienza”, in G.Berkeley, Alcifrone, introduzione, traduzione, note e apparati di D.Bertini, Milano: RCS.
  • Bertini, D. (2010), “La critica berkeleiana all’autonomia morale”, in L.Jaffro, G.Brykman, & C.Schwartz, eds., Berkeley’s Alciphron. English Text and Essays in Interpretation, Hildesheim-Zürich-New York: Georg Olms Verlag.
  • Jaffro, L. (2000), “La formation de la doctrine du sens moral: Burnet, Shaftesbury, Hutcheson”, in L.Jaffro, ed., Le sens moral. Une histoire de la philosophie morale de Locke à Kant, Paris: Presses Universitaires de France.
  • Jezzi, N. (2016), “Rawls on Kantian Constructivism”, Journal for the History of Analytical Philosophy, 4(8).
  • Kant, I. (2002), Groundwork for the Metaphysics of Morals, New Haven & New York: Yale University Press, ed. and trans. by A.W.Wood.
  • Peacocke, C. (2004), The Realm of Reason, Oxford: Clarendon Press.
  • Rogers, G.A.J. (1999), “Locke, Law, and the Laws of Nature”, in M.Atherton, ed., The Empiricists. Critical Essays on Locke, Berkeley, and Hume, Lanham-Boulder-New York-Oxford.
  • Terni, M. (1995), La pianta della sovranità. Teologia e politica tra medioevo ed età moderna, Roma-Bari: Gius.Laterza & Figli Spa.
  • Voltaire (1901), Philosophical Dictionary, in The Works of Voltaire: A Contemporary Version, William F. Fleming (ed. and tr.), 21 vols., New York: E.R. Du Mont.
  • Zolo, D. (2001), I signori della pace. Una critica al globalismo giuridico, Roma: Carocci editore S.p.A.

Table of Contents

The prehistory of moral heteronomy and autonomy

How to be heteronomous in morality

Arguing for an heteronomous approach to normativity

  1. Terni (1995), pp. 20-57. ↩︎

  2. Bertini (2005), pp. 14-44; Bertini (2010). ↩︎

  3. Jaffro (2000), p. 13. ↩︎

  4. Voltaire (1901), vol. 6: p. 177. ↩︎

  5. Kant (2002), pp. 50-51. ↩︎

  6. Bagnoli (2014). ↩︎

  7. Jezzi (2016). ↩︎

  8. Peacocke (2004), pp. 198-207. ↩︎

  9. Zolo (2001), pp. 21-23; 59; 118-119; 134. ↩︎

  10. Particularly, I would like to thank Daniela Almansi for her help in editing the introduction and my paper. ↩︎