Between Faith and Judgement: Kant’s Dual Conception of Moral Certainty

Abstract

There are two main meanings in Kant’s concept of moral certainty (moralische Gewissheit, certitudo moralis): first, it applies to the kind of certainty embodied in rational faith in the existence of God and a future life; second, it applies to the conscientiousness (Gewissenhaftigkeit) required of an agent in the practice of moral judgement. Despite the growing attention to Kant’s theory of conscience and his concept of conscientiousness, this article is the first to discuss ‘moral certainty’ as the aim of ‘conscientiousness’ and to highlight the relevance of both notions in regard to moral education and the purposes of Kant’s ethical doctrines of method.

But was he really as strongly convinced of such a

revealed doctrine, and also of its meaning, as is required

for daring to destroy a human being on this basis?

(RGV, 6: 186.33–6)

1. Introduction

Quod dubitas, ne feceris! – do not do what you are doubtful about! Kant quotes this adage from Pliny’s Epistulae (I, 18, 5) in the Religion within the Boundaries of Mere Reason in the context of a brief discussion of the role of conscience as a guide in moral decision-making. Kant here presents the principle of refraining from doing something we are to any degree uncertain about as a moral principle that requires no proof and as a ‘postulate of conscience’ (RGV, 6: 185.23 and 186.7; trans. S.D.G. ¹ ). This is contrasted with the principle of ‘probabilism’, that is, the view according to which holding the opinion that an action may well be right is in itself sufficient for justifying the decision to carry it out (cf. RGV, 6: 186.7–9).

On the basis of this juxtaposition, I will investigate the function of Kant’s ethical doctrines of method ² in promoting the passage from opinion, or indeed even from mere persuasion, to firm certainty in the practice of moral judgement. The reasons for focusing on the concept of persuasion (Überredung) will become clearer in the course of the analysis. ³ Suffice it to say for now that it is above all Kant’s recognition of our natural radical tendency towards self-deception that makes it understandable how significant certain fluctuations in judgement must be for a moral theory aiming at an effective correction of moral attitudes. What is in question here is a quite widespread phenomenon in the practice of moral judgement: that of oscillation between an insufficient but still advantageous persuasion of an at least apparent moral value, on the one hand, and one’s fairly clear understanding that the proposed behaviour is in fact morally impermissible, on the other. Within this context, it will be possible to define moral certainty as an agent’s certainty or firm conviction about the rightness of his intentions on the basis of the awareness that the evaluation conducted upon them has consisted in a genuinely conscientious, that is, accurate and truthful, process of moral judgement.

Before embarking on any further analysis in this direction, however, it should first be noted that there is no single meaning in Kant for ‘moral certainty’ (moralische Gewissheit). He rather uses the expression in quite different ways in different contexts. In this regard, Kant essentially conducts his reflections in two distinct theoretical fields, leading to the two main meanings in question, one as part of his theory of conscience, the other in his doctrine of rational faith and the postulates of pure practical reason. Section 2 of this article will briefly discuss the concept in the latter context, noting its novelty with respect to the meaning traditionally attached to the term. We will then return in section 3 to the concept of moral certainty as conscientiousness (Gewissenhaftigkeit) in the practice of moral judgement and the main implications of Kant’s juxtaposition of it, as a postulate of conscience, to the principle of probabilism. As we will see, Kant recognises in probabilism the very paradigm of evil, or at least as the aptest instrument of the human tendency to self-deception as the root of the possibility of acting against better judgement (akrasia). The tendency to quibble in questions of the obedience to be accorded to the law and the guiding reasons of one’s own choices is countered in the agent by the capacity of his conscience to infallibly judge the accuracy and truthfulness of (and thus any potential incautiousness and deceitfulness in) moral decision-making and to spur (or call on) the subject to evaluate his or her choices (more) conscientiously. Section 4 will show that a thorough investigation of this material in the light of Kant’s polemic against the moral-theological doctrine of probabilism would certainly be of considerable importance with regard to the scope of moral education and thus for the understanding of the role Kant assigns to his ethical doctrines of method.

2. Moral certainty as rational faith

The principle of quod dubitas, ne feceris, which in his work on religion Kant calls ‘a postulate of conscience’, is mentioned in several other places in Kant’s published works, as well as in the student notes from his academic lectures and in Kant’s reflections. In particular, a paraphrase of the principle can be found in the Canon of the first Critique. Unlike in the Religion, Kant’s focus there is not on the criteria of judgement in applying moral norms but rather on the degree of certainty of which principles of morality must be capable: ‘[i]n judging from pure reason, to have an opinion is not allowed at all’ (KrV, A822/B850). These judgements, on the contrary, demand ‘full certainty’ (A823/B851). Just as ‘it is absurd to have an opinion in pure mathematics’ – ‘one must know, or else refrain from all judgment’ – so too is it absurd for the ‘principles of morality’ (ibid.). Indeed, as Kant explains by using the Quod dubitas principle, ‘one must not venture an action on the mere opinion that something is allowed, but must know this’ (ibid.).

Unlike in the Religion, the point at issue here is not whether ‘an action which I want to undertake is right’ (RGV, 6: 185.25–186.1) and how certain I am of the judgement I formulate under the specific circumstances of my choice. Rather, the problem referred to in the Canon passage is that of establishing whether on a normative level, independently of any empirical specificity, a certain conduct is in general to be considered permitted, obligatory, or forbidden. It is in this context that Kant’s concept of moral certainty, as most often investigated, takes shape. Its epistemic status is defined in relation to the taxonomy of opinion (Meinung), knowledge (Wissen), and belief (Glaube) and, again, in the perspective of the doctrine of the postulates of pure practical reason. ⁴ However, it is important to emphasise that, although in the passage cited from the Canon Kant clearly raises the issue of the certainty which has to be attained in the knowledge of the principles of morality, he does not actually use the term ‘moral certainty’ in this regard. Thereby, in his understanding of the notion, as well as the epistemic status of the principles of morality, Kant in effect distances himself from the previous tradition of use of the term ‘moral certainty’.

According to a tradition dating back to Descartes, the use of the term ‘moral certainty’ was established to indicate the degree of certainty that can be achieved in the practical sphere, although not specifically in relation to moral issues. In the Principles of Philosophy (1647), Descartes states that that certainty is called ‘moral’ which is ‘sufficient to regulate our morals, or as great as that of things we do not usually doubt, concerning the conduct of life, although we know that it is possible, absolutely speaking, that they are false’ (Descartes 1971 [1647]: § 205, p. 323; trans. S.D.G.). As an example, Descartes mentions the case of someone who, although never having been to Rome, does not therefore doubt that Rome is a city in Italy. In the sense indicated by Descartes, moral certainty ‘denotes a weaker certainty, in comparison with the metaphysical one, but nonetheless useful for everyday life’, which is ‘based on the testimony of other human beings’ (Fonnesu 2011:184). More precisely, as pointed out by Luca Fonnesu, ‘the idea of moral certainty denotes, from Descartes to the Aufklärung, a weaker certainty that can perhaps be useful for ordinary life, but that cannot be connected with ethics in a strict sense’. ⁵

In the wake of the definition elaborated by Descartes, the notion of moral certainty will then be linked, in the Port-Royal Logic (Arnauld and Nicole 1996 and 2014 [1662–1683]; hereafter, Logic and Logique), to the emerging modern notion of probability, understood as a certainty which is only probable (even if to the highest degree) and, once again, not specifically related to moral issues. ⁶ This independence from purely moral questions, together with the close link to the modern notion of probability, can also be found in the definition given by Georg Friedrich Meier in the Excerpt from the Doctrine of Reason (2016 [1752]; hereafter, Excerpt ⁷ ), a work held in high regard at the time and used by Kant as a textbook for his lectures on logic. Meier’s definition of moral certainty fits into the fundamental distinction between a cognition which we hold to be true and to which we therefore give our assent ⁸ and a merely probable cognition (cognitio probabilis, verosimilis), that is, an uncertain cognition that we nevertheless consider as probable, insofar as we cognise ‘more and stronger grounds to accept it than to reject it’ (cf. Excerpt, § 171, p. 40; 16: 416–28). From this follows the definition of moral certainty (moralische Gewißheit, certitudo moralis) as ‘a degree of probability, which in our regular conduct is as good as an extensive certainty’ (Excerpt, § 175, p. 41; 16: 432).

Evidently, Kant cannot but reject the association of such a notion of moral certainty – a certainty that is only probable and generally useful for ordinary practical life – with the level of the principles of morality. And he resolves to abolish all reference to this nexus in his writings: as Fonnesu rightly observes, Kant ‘will almost never – never in the published works – use the expression “moral certainty” for the knowledge of the moral principle, i.e. for the certainty of morality, which is and ought to be an apodictic certainty (as in mathematics)’ (Fonnesu 2011: 188). Although not referring to the foundations of morality, the notion of moral certainty introduced by Kant in the Canon nevertheless has a distinctly moral connotation, as far as it pertains to objects ‘that are morally relevant as (more or less convincing) consequences of the acceptance of the [moral] principle’ (ibid.). In the first Critique, Kant uses the expression ‘moral certainty’ to designate the (subjective) certainty of rational faith (Vernunftglaube) in the existence of God and the immortality of the soul, which is ‘grounded on the presupposition of moral dispositions’ (KrV, A829/B857). For, as Kant explains: ‘no one will be able to boast that he knows that there is a God and a future life’, because such a ‘conviction is not logical but moral certainty’, and, since it depends on subjective grounds (on moral disposition) I must not even say ‘It is morally certain that there is God’, etc., but rather ‘I am morally certain’, etc.’ (A828-9/B856–7). As pointed out in the Vienna Logic (1780 or 1781), ‘rational faith … means logically insufficient holding-to-be-true which is, however, practically sufficient’; ‘it has practical grounds; e.g. that the soul is immortal can drive me to better arrange my life’ (V-Lo/Wiener, 24: 895.1–3, 29f.; trans. S.D.G.).

The association between moral certainty and rational faith may not have been new. ⁹ However, it marks a major step in the history of concepts and an important achievement within Kant’s thought, finding its main textual evidence in the Canon of the first Critique. Influential scholars, starting with Fonnesu, have identified a crucial element of innovation in the connection Kant establishes, in contrast to the traditional use of the term, between the notion of certitudo moralis and questions specifically pertaining to morality – a connection the lack of which in ‘almost all autores’ Kant himself had long complained about. ¹⁰ Commenting on §175 of the Excerpt, in which Meier provides the aforementioned definition of moral certainty, Kant states in his lectures on logic in the early 1770s:

Most, almost all autores are completely unacquainted with moral certainty, and instead they take it in each case to be probability. E.g. It is uncertain whether there are inhabitants on the moon, but nevertheless is still probable. A few accept this as a moral judgement, but this is not the case, for such judicium has no influence at all on behavior. It is a logical probability and a mere speculation.

Nothing is a moral judgement except what has a relation to my actions. (V-Lo/Blomberg, 24: 200.35–201.5).

Due to its overlapping with the notion of rational faith, the concept of moral certainty undeniably acquires in the first Critique a moral significance that it lacks in the tradition dating back to Descartes. Nevertheless, as Fonnesu does not fail to note, Kant’s argument in the Canon is not exactly a moral one: The question ‘What may I hope?’ is here ‘grounded on the practical interest for a cosmic justice, but does not imply consequences for actions’ (Fonnesu 2011: 194f.). Although moral certainty in this relation has ‘nothing to do with theoretical probability’, as it had in the previous tradition, ‘Kant’s thesis about God’s existence and immortality of the soul seems to be a mainly theoretical thesis: the question does concern the existence of conditions of realisation of morality, not its promotion and not, here in the first Critique, the duty to the promotion of highest good as a whole’ (ibid.).

Interestingly, after the first Critique, the expression ‘moral certainty’ no longer occurs precisely where one might perhaps most have expected it, neither in the context of the doctrine of postulates and of the changed horizons for Kant’s reflections on its objects (the existence of God and the immortality of the soul) in the Dialectic of the second Critique (section VIII), nor in the two last sections of the Critique of the Power of Judgement (§§ 90–1) ¹¹ or, finally, in the First Preface to Religion. ¹² All the same, as I have indicated, a further understanding attaching to the expression ‘moral certainty’, which Fonnesu does not mention in his analysis, appears to be of great importance in relation to the sphere of moral agency: moral certainty as conscientiousness in the practice of moral judgement, which will be the subject of section 3.

3. Moral certainty as conscientiousness in the practice of moral judgement

The attention of Kant scholars has so far focused on Kant’s redefinition of the notion of moral certainty in the Canon of the first Critique, where it becomes synonymous with the concept of rational faith. ¹³ As mentioned, however, to this new meaning of the term Kant adds a second one, which has not yet been adequately taken into account in the literature, but which emerges and acquires great prominence in relation to later developments in Kant’s moral psychology, that is, in the entirely different context of the theory of moral conscience and the doctrine of radical evil. Even Kant’s lectures on logic from the 1790s clearly document this new achievement with respect to the point raised some 20 years earlier in the identical context. Here, in the Logic Dohna-Wundlacken (1792), Kant redefines the meaning of moral certainty by connecting it to the specific sphere of conscience and moral judgement. He states: ‘The certainty that belongs to the knowledge of the commandment or prohibition of an action is moral certainty. … Moral certainty refers to conscience. … If one does something at the risk of being wrong, then one is never morally certain’ (V-Lo/Dohna, 24: 734.5–15; trans. S.D.G.). ¹⁴ Let us therefore take a closer look at these issues and the emergence of Kant’s notion of conscientiousness.

A Kantian note on logic presents the two ways of understanding the concept of moral certainty discussed here with regard to their respective conceptual histories, as well as in relation to the different theoretical contexts in which Kant elaborates their meaning and function. Commenting again on Meier’s discussion of the concept of moral certainty in the Excerpt, Kant observes in the margin: ‘What is certain according to laws of the ends of the will is morally certain’ (Refl 2629, 16: 443.2–3; trans. S.D.G.). And: ‘Moral certainty is only subjective, namely: to declare something certain with conscientiousness’ (16: 443.4–5; trans. S.D.G.). Similarly, on this second way of understanding moral certainty Refl 2631 states: ‘One can only say: I am morally certain; for moral certainty expresses only the conscientiousness of the judgement, e.g. there is no conscientiousness in the courts of the Inquisition’ (16: 443.18–20; trans. S.D.G). ¹⁵ The reference to the Inquisition, as we shall see, is particularly significant when compared to Kant’s discussion of the same example in the above-mentioned section of the Religion.

The passage from Kant’s writing on religion in which he examines the role of conscience as a ‘guiding thread in the most perplexing moral decisions’ (RGV, 6: 185.16–17) is the main reference point for an investigation of the concept of moral certainty as conscientiousness in moral judgement – a meaning of the expression to which, again, to my knowledge, no attention has so far been brought among Kant scholars. ¹⁶ Yet it is precisely by using the notion of conscientiousness as a synonym for moral certainty that Kant establishes a novel connection between the traditional notion of certitudo moralis and the specific domain of morality, more exactly: between that notion and the sphere of moral agency, of the execution of the principles of morality. As a consequence, Kant’s understanding of ‘moral certainty’ as conscientiousness cannot be considered analogous to the traditional usage of the term dating back to Descartes: moral certainty for Kant is not a certainty which may be useful in contexts that are mostly morally neutral, but rather the certainty which a (conscientious) decision-maker achieves that there exists for him, in a given situation, the moral obligation to act in a certain way, to pursue certain ends. Moral certainty in this sense is ‘sufficient certainty for the duty’ (cf. Anth, 7: 329.26f.), that is, for the concrete decision to perform actions in accordance with what one (conscientiously) judges to be one’s duty.

‘Conscientiousness’ is a term of central relevance to the Kantian theory of conscience and moral judgement and consequently to the purposes of the ethical doctrines of method and in view of pedagogical applications. For its proper examination, it is necessary to take into account, in addition to the passage from Religion, at least the Concluding Remark of Kant’s Theodicy essay (1791) as well as important passages from his lectures and notes. In all of these places, the meaning of moral certainty I will investigate most closely is defined by the contrast between the postulate of conscience, expressed through Pliny’s adage or through its reformulations, and an entirely alternative approach to moral judgement, which is referable to the moral-theological doctrine of probabilism. ¹⁷ The theory of conscience thus proves to be the systematic place for the development of a meaning of moral certainty that, unlike the meaning of moral certainty in the Canon of the first Critique, is strictly inherent to the sphere of morality, more precisely: to the concrete application of its principles through the agent’s moral choices. At the same time, by referring to the principle of probabilism, Kant discloses the particular problematic background behind his discussion of moral certainty as a postulate of conscience. In several places in his writings, Kant harshly criticises probabilism’s approach to moral judgement, going so far as to consider probabilism to be the paradigmatic expression of man’s natural radical tendency towards evil. A concise formulation of this thought is offered, for example, in a reflection on morality, dating from the second half of the 1770s: ‘The intentional insincerity of human nature: hence probabilismus, peccatum philosophicum and reservatio mentalis’ (Refl 7180, 19: 265.17–18). ¹⁸ But first and foremost, the very reference to the notion of probabilism in RGV, 6: 185 introduces a meaning of ‘probable’ (and of what is to be understood by a ‘probable opinion’) that has nothing to do with the modern notion of probability or, more specifically, with the simple expectation that one’s opinions map onto reality, regarding the content of truth expressed in them.

The notion of probability to which Kant contrasts certainty as a necessary requirement of moral judgement clearly does not refer to actions as morally neutral events and thus to the (calculation of the) probability that they may or may not occur (or have already occurred in the past). ¹⁹ Quite the opposite, the notion of probability Kant uses here is specifically concerned with the judgement on the moral value of actions: whether such an assessment is worthy of approval or cannot be accepted as valid for the justification of the action to be taken. The notion of probability brought into play by Kant’s reference to the principle of probabilism does not therefore reflect the modern (quantitative, frequentist) conception of probability, which looks at phenomena merely from the point of view of the possibility of their occurrence. On the contrary, Kant’s reference recalls a qualitative conception of probability, belonging to medieval thought. According to the (qualitative, endoxical) scholastic conception of probability, ‘probabilis’ (from ‘probo/probare’, which means ‘approve’, ‘give assent’, ‘accept’) is a ‘qualitative predicate accruing to propositions and opinions’ which means that these propositions are ‘fit for adoption and sufficiently, although not optimally, backed by reasons for truth’. ²⁰ The notion of moral certainty in use in early modern and modern Catholic moral theology also refers to this conception of probability to indicate the certainty required for judging the moral quality of actions. In the same context, the formulation of this sort of judgements was considered to be the specific task of conscience. And it was precisely in order to designate the certainty required for moral judgements of conscience that the use of the expression ‘moral certainty’ (certitudo moralis) had become widespread.

Revisiting Aristotelian insights, medieval thought had built on the view that full certainty cannot be achieved in matters concerning human agency (praxis), ²¹ and that therefore opinion was the most appropriate cognitive state for investigations in this area. ²² The French theologian Jean Gerson (1363–1429) was the first author to use the term ‘certitudo moralis’ to designate the certainty arising from the proper adoption of probable opinions (opiniones probabiles). ²³ More precisely, as Rudolf Schuessler points out, Gerson uses the term with a double meaning: ‘on the one hand, it stands for the maximally attainable epistemic certainty in contexts of human agency, on the other hand, it signifies certainty of avoiding sin’. ²⁴ In the medieval imagination, putting oneself in danger of sinning was itself already considered a capital sin. Therefore ‘the correct avoidance of sin-related risks was a key concern of scholastic handbooks for confessors’. ²⁵ In this respect, the principle of quod dubitas, ne feceris was a proven criterion for overcoming moral uncertainty and, therefore, the fear of incurring sin. ²⁶ That this is precisely the problem-historical background of Kant’s treatment of moral certainty should also be evident from the lexical choices recurring in Kant’s illustrations of the Quod dubitas principle, for example in the Religion: ‘It is a moral principle, requiring no proof, that we ought to venture nothing where there is danger that it might be wrong ²⁷ (quod dubitas, ne feceris! Pliny)’ (RGV, 6: 185.23–5). However, according to Kant, one is obviously not protected from the danger of sin or moral infringement by recurring to a probable opinion for the justification of one’s action, that is, by adopting a position that claims to be worthy of approbation because it is commonly held or endorsed by particularly competent persons (doctores graves). The principle of moral certainty expressed through the Quod dubitas is, on the contrary, opposed by Kant to that of probabilism. ‘With respect to the action that I want to undertake’, he declares, ‘I must not only judge and opine, but also be certain that it is not wrong’ (RGV, 6: 186.4–6; trans. S.D.G.). For Kant, this requirement is a ‘postulate of conscience’ (186.7; trans. S.D.G.), ‘to which is opposed probabilism, i.e., the principle that the mere opinion that an action may well be right is itself sufficient for undertaking it’ (186.7–9).

Thus, Kant’s discussion of conscience and moral certainty is clearly set against a particular traditional background, that of the debate on casuistry and probabilism, which profoundly shaped early modern theology and philosophy. From this background, Kant retrieves the general setting of the problem together with its key concepts. Of the latter, however, Kant radically redefines meaning and function, eventually overturning the very perspectives of the whole debate. This dialectic of retrieval and reformulation can be usefully investigated from the consideration of the relationship between moral certainty and infringement in the practice of moral judgement.

In its understanding of what the violation of a moral law or divine command consists in, Kant’s position is diametrically opposed to that of probabilism, particularly if one has in mind the lax drifts of Jesuit casuistry. On the contrary, it is not at all foreign for Kant to think of moral judgement as constitutively exposed to uncertainty and error. While apodictic certainty is required for moral principles, ‘uncertainty and probability’ often prevail in their application to cases of concrete moral experience. ²⁸ Thus in establishing the requirement of moral certainty as a postulate of conscience and supporting the thesis of the infallibility of conscience – in the Doctrine of Virtue (6: 401.3–13) as well as earlier in the Theodicy essay (8: 268.10–18) – Kant does not claim that moral judgement is or should be safe from any possible error, nor does he suggest that conscience should be able to point out and amend one’s erroneous judgements. This for the simple reason that conscience, for which Kant postulates the principle of moral certainty and for the infallibility of which he argues, is not for him a faculty ‘which judges whether an action is in general right or wrong’ (RGV, 6: 186.1–2). For ‘it is understanding, not conscience’ that judges about this (186.2–3). ²⁹

Unlike his predecessors until and including Baumgarten, Kant does not equate conscience with the faculty of moral judgement. ³⁰ In the Religion he rather defines conscience as ‘the moral faculty of judgement, passing judgement upon itself’ (RGV, 6: 186.10–11), hence as an agent’s self-reflexive power, the judgement of which is thus directed towards the subject and refers only indirectly to the object of its moral evaluations: the action to be performed and its moral quality. ³¹ Conscience for Kant ‘does not pass judgement upon actions as cases that stand under the law, for this is what reason does’ (RGV, 6: 186.13–14). Thus, it does not have the function of subsuming actions under laws nor of re-examining the judgement already formulated by reason (or understanding), as another formula popular in the literature, that of second-order power of judgement (Urteilskraft zweiter Ordnung), might erroneously suggest. If, for Kant, conscience is infallible in its judgements, this is due precisely to the fact that conscience does not judge the actions to be performed, but rather the judging subject herself: the conscientiousness of her judgements. With regard to judgements of the first kind, that is, of an objective kind, error is in fact always possible. Things are different with regard to judgements which are reflexively addressed to the subject herself, and which properly belong to the very scope of conscience.

Comparing Kant’s characterisations of conscience in his different treatises, it is possible to state that conscience refers to the judging subject essentially in two ways. On the one hand, it motivates her to preventively employ all available means in order to avoid the risk of errors of assessment as well as the temptation of guilt (mistake of convenience). On the other hand, conscience retrospectively condemns any kind of short-cut in judgement, thus proving itself infallible. The infallibility that Kant argues for conscience is therefore not in contrast to the possibility of error tout court, let alone to mere oversight. Rather, conscience seems for Kant to be an infallible judge of fallacious, misleading judgements which, literally, lead to mistakes or by which one indulges in error, that is, in moral transgression. As Kant explains in the Doctrine of Virtue:

[A]n erring conscience is an absurdity. For while I can indeed be mistaken at times in my objective judgement as to whether something is a duty or not, I cannot be mistaken in my subjective judgement as to whether I have submitted it to my practical reason (here in its role of judge) for such a judgement; for if I could be mistaken in that, I would have made no practical judgement at all, and in that case there would be neither truth nor error. Unconsciousness is not lack of conscience but rather the propensity to pay no heed to its judgement. But if someone is aware that he has acted in accordance with his conscience, then as far as guilt or innocence is concerned nothing more can be required of him. (TL, 6: 401.5–13)

If, on the contrary, the problem cannot be considered volitional, that is, attributable to a (culpable) resistance of the subject to the warnings of conscience, then the agent can still be required to reinforce relevant (theoretical) competences concerning the knowledge of moral principles, that is, ‘to enlighten his understanding in the matter of what is or is not duty’ (TL, 6: 401.14f). However, as Kant’s juxtaposition makes clear, it is primarily the first type of mistake, ascribable to the subject as her fault, that marks the scope of conscience as well as the limits of its infallibility.

The conscience, as the psychological counterpart of the fact of reason, ³² ineluctably impels the judging subject to self-examination, spurring her to formulate an accurate and truthful judgement on her actions (or to rectify preexisting judgements in this sense). In Kant’s theory of conscience, accuracy and truthfulness constitute two complementary aspects of that conscientiousness which for Kant is synonymous with moral certainty and, as such, a fundamental requirement for judgement and an indispensable postulate of moral conscience. The focus on the first aspect seems to prevail in the Religion. Here, Kant defines the specific function of conscience in relation to the self-reflexive judgement as to whether the assessment of the moral quality of an action has been conducted with due accuracy or caution (Behutsamkeit) (cf. RGV, 6: 186.12–20). This fundamental component of conscientiousness, which Kant illustrates on the basis of the (counter) example of the inquisitor (cf. RGV, 6: 186.21–187.10), consists in a form of circumspection in the formulation of judgement, in the care or caution not to entrust one’s evaluations to preconceived opinions or dogmatic assumptions, and in striving rather to lead the judgement under the critical, autonomous evaluation of one’s own reason. ³³

Kant also discusses Behutsamkeit as the material component of conscientiousness in the Concluding Remark of his essay on theodicy. This ‘material conscientiousness’, as Kant calls it, ‘consists in the caution of not venturing anything on the danger that it might be wrong’ (MpVT, 8: 268.7–8). Conversely, Kant defines as ‘formal conscientiousness’ the agent’s consciousness that she really judged the action she intends to perform with the utmost caution or accuracy (cf. 8: 268.8–10). The focus on this formal component of conscientiousness is particularly marked in this passage. Here, Kant also designates ‘formal conscientiousness’ by the name of ‘truthfulness’ (Wahrhaftigkeit) (8: 268.6–7) and clarifies that it consists in the solicitude to acquire a clear consciousness of one’s holding-to-be-true (Führwahrhalten): of one’s actual belief about the moral quality of an action (268.19–22). This difficult passage is easier to understand when compared to the thesis about the infallibility of conscience which Kant in the same place formulates as follows:

But an erring conscience is an absurdity … I can indeed err in the judgement in which I believe to be right, for this belongs to the understanding which alone judges objectively (rightly or wrongly); but in the judgement whether I in fact believe to be right (or merely pretend it) I absolutely cannot be mistaken, for this judgement – or rather this proposition – merely says that I judge the object in such-and-such a way. (MpVT, 8: 268.13–17)

Formal conscientiousness (or, more precisely, that which it tends towards) therefore consists in the consciousness that I have made my own judgement with Behutsamkeit (cf. MpVT, 8: 268.8–10) and that, by virtue of this, I can believe (and claim to believe) that I am right in my judgement. The alternative to the truthful declaration of one’s holding-to-be-true is therefore to feign beliefs one does not in fact possess, that is, (a) to boast a degree of certainty about the rightness of one’s action that one does not actually hold (I pretend that I do indeed believe I am right, but in fact I am not so deeply convinced of my judgement) or (b) to attempt to deceive oneself and others by giving the appearance that one has a different judgement of her own action from what one actually believes to be the appropriate assessment of the case (I assert a holding-to-be-true of which I am not convinced that it expresses my own judgement or the judgement that should be made of an action).

Even more than in the formulation of the infallibility thesis in the introduction to the Doctrine of Virtue (6: 401.3–11), the absurdity of an erring conscience, that is, of being mistaken in the self-reflective judgement of conscience, should be evident from the analysis of the passage just quoted. If I pretend to be right by stating a judgement that differs from my actual holding-to-be-true, be it in its degree of certainty or even in its actual content, I simply lie in the ethical sense of the term. ³⁴ The possibility to which formal conscientiousness or truthfulness is opposed is then evidently not so much that of an unconscious and hardly avoidable mistake but rather primarily that of (self-) deceit and a deliberate dishonesty in the expression of one’s own thoughts or beliefs, of one’s actual holding-to-be-true. Moreover, whereas the truth of a judgement lies in the relationship between assertion and reality, so that the judgement can be correct or false, truthfulness is a self-reflective relationship which concerns the correspondence between what one claims and what one holds to be true or is actually persuaded of. ³⁵

In light of these premises, Kant’s Concluding Remark becomes particularly interesting, further clarifying the notion of formal conscientiousness as ‘the care of becoming conscious’ of one’s own beliefs or non-beliefs ‘and not pretending to hold anything as true we are not conscious of holding true’ (MpVT, 8: 268.19–21): ‘Human beings’, Kant asserts, ‘feign conviction [Überzeugung] – which is at least not of the kind, or in the degree, as they pretend – even in their inner profession, and this dishonesty … gradually forges actual persuasion [Überredung]’ (8: 268.32–5). The hyperbole of this generalisation to all mankind is not surprising if one takes into account that the discussion of conscientiousness in the Concluding Remark is set in the context of a ‘brief reflection on a big subject, namely sincerity … as contrasted with the propensity to falsehood and impurity which is the principal affliction of human nature’ (8: 267.23–6) – a vice, Kant adds later, which seems to be ‘deeply rooted in human nature’ (8: 269.11); an ‘impurity that lies deep in what is hidden, where the human being knows how to distort even inner declarations before his own conscience’ (8: 270.32–4). The standpoint from which Kant here approaches the issue of conscientiousness is clearly that of humanity’s radical propensity to evil, which finds its essential expression in lying to oneself (self-deception) – and in probabilism its most blatant historical realisation. ³⁶

As the work on religion will further explain, ‘the human being (even the best)’ has a natural tendency to reverse ‘the moral order of his incentives in incorporating them into his maxims’ (RGV, 6: 36.23–5). More precisely, he ‘incorporates the moral law into those maxims, together with the law of self-love; since, however, he realises that the two cannot stand on an equal footing, … he makes the incentives of self-love and their inclinations the condition of compliance with the moral law’ (36.25–30). This tendency not to attend to the moral order of the incentives and to see only to the mere external conformity of one’s actions to the law (cf. 37.26–31) finds its concrete realisation in two complementary strategies of self-deception by which the agent attempts to evade the outcome of a correct moral evaluation, of a ‘better judgement’, of which he would still be capable or that his reason already sets before him with sufficient clarity. ³⁷ On the one hand, the agent indulges in sophistry about how to interpret the moral norm to be applied, that is, he stubbornly tries to ‘deceive himself in the interpretation of the moral law to the detriment of the same’ (42.33–4; trans. S.D.G.). On the other hand, one tends to lie about one’s real intentions, that is, to ‘deceive himself about his own good or evil intentions and, if only his actions do not result in evil, which they could well do according to their maxims, not to trouble himself about his intention, but rather to consider himself justified before the law’ (38.8–12).

Concluding my analysis of conscience and moral certainty in the light of their historical roots in the debate on casuistry and probabilism, I would more particularly draw attention to the extraordinarily interesting fact that Kant associates the human propensity to evil with the notion of probabilism, understood as a refined instrument of self-deception, already in the 1770s. In his lectures on moral philosophy at the time, Kant defines ‘moral probabilism’ as ‘a means whereby man deceives himself, and persuades himself that he has acted rightly and according to principles’ (V-Mo/Kaehler (Stark): 201.10–13, trans. S.D.G.; cf. V-Mo/Collins, 27: 359.19–21), that is, in accordance with moral principles, although his action responded first and foremost to his own self-interest, to the principle of self-love. ³⁸ Choosing a probable opinion to justify an action would ultimately be nothing more than a sophistical game with the mere appearance of truth of the authoritative positions expressed in them, a mean attempt to absolve oneself from the nevertheless inescapable condemnation of one’s conscience. A similar judgement can be found two decades later in the Vigilantius lectures. Here, Kant labels casuistry as a ‘procedure of deceiving or quibbling with conscience by sophistry, insofar as we endeavour to lead it astray; e.g., when we invent good intention in actions that involve a transgression of duty’ (V-MS/Vigil, 27: 620.3–6). Moreover, even in the passage from the Religion which serves as the main reference, Kant incidentally defines casuistry as ‘a kind of dialectic of conscience’ (RGV, 6: 186.15–16). Kant’s definition will be quoted from the fourth edition of the Conversations-Lexicon (1817) onwards, confirming that a critical if not openly caricatured conception of casuistry was a commonplace at the time. ³⁹

4. Pursuing moral certainty: on the aims of Kant’s ethical doctrines of method

Kant’s opposition of the postulate of conscience ⁴⁰ to the principle of probabilism ⁴¹ clearly concerns both aspects of conscientiousness we have discussed so far: accuracy or caution (Behutsamkeit) and truthfulness (Wahrhaftigkeit). According to the analysis in the preceding section, Kant’s (dual) notion of conscientiousness most closely expresses a meaning of moral certainty which specifically pertains to the domain of moral judgement. The conscientiousness that this judgement requires provides, with regard to the first aspect, that the subject does not allow any external guides to determine her own choices, especially the most dubious among them (casus conscientiae), nor does she succumb to the temptation of considering a ‘real transgression’ of the moral law as a trifle, as a ‘bagatelle (peccatillum)’ to be left to the arbitrary judgement of a ‘director of conscience’ (cf. TL, 6: 440.10–19; trans. S.D.G.). Not only is this a polemical reference to the lax use of casuistry in Jesuit confessional practice, ⁴² the opposition to seeking external guides for the formulation of one’s own judgements is at the heart of the concern of the infallibility question, that is, the self-reflexive question by which I ask myself whether I have submitted the case to be judged ‘to my practical reason’ (cf. TL, 6: 401.6–10) – rather than let my judgement depend on other, extra-moral considerations or external authorities. It is therefore a central requirement of conscientiousness as well as the main demand of conscience to compare the contents of one’s own judgement with one’s reason, that is, to subject them to reason’s critical examination and to its authority alone. As mentioned, a striking counterexample to this is provided by the figure of the inquisitor and the Kantian discussion of his propensity to judge according to his statutory faith rather than to conduct a critical reflection on his actions and the meaning of the laws that apply to them (cf. RGV, 6: 186.21–187.10; Refl 2631, 16: 433.18–20). As for conscientiousness in the second sense, as the search for and preservation of a veridical relationship with one’s own moral convictions, I have already sufficiently insisted on its opposition to the profoundly human tendency to interpret the law to one’s advantage and to simultaneously falsify the real intentions of one’s actions, which finds its most fitting historical concretisation in probabilism.

This section aims to briefly illustrate how both aspects of conscientiousness are clearly present in the doctrines of method that Kant sketches out in the second Critique and then in the Doctrine of Virtue – namely, as specific aims of these doctrines. ⁴³ As a result of the analysis conducted thus far, it is possible to affirm that the method Kant intends to establish aims at fostering a conscientious practice of moral judgement, which is capable of achieving in each instance the difficult transition from mere, fallacious persuasion or at least from an unsure opinion, ⁴⁴ to a firm certainty as to the moral quality of the action to be undertaken.

In ethics, Kant states that ‘doctrine of method’ means ‘the way in which one can provide the laws of pure practical reason with access to the human mind and influence in its maxims, that is, the way in which one can make objectively practical reason subjectively practical as well’ (KpV, 5: 151.9–12). The method that Kant outlines consists of two exercises or moments. The first is aimed at strengthening the learner’s theoretical competence, that is, at refining her knowledge of moral principles and ability to apply them correctly. The second exercise focuses on the motivational efficacy of the moral norms, on the possibility that they decisively influence subjective choices, the underlying reasons for one’s actions. ⁴⁵

In the context of moral catechism, the dialogue between teacher and pupil on the meaning and relevance of moral norms makes it possible to strengthen those theoretical skills which are necessary for the correct, critical evaluation of the different cases of moral experience and the normative instances operating in them. Only the judgement of a properly educated understanding – as to what is or is not duty (cf. TL, 6: 401.14–15 and 483.32–36) – can eventually result in a truly accurate judgement, capable of shedding full light on the authentic reasons for the conduct to be undertaken. An exercise on the motivational level is also indispensable to ensure that objective practical reason becomes subjectively practical, providing through its laws sufficient motivation in itself for the effective determination of the will. Whereas an almost exclusive attention to the outward conformity of one’s actions to the law characterises the propensity to evil in human nature, as well as the probabilistic approach to reasons for action, this second exercise, conducted on the example of other people’s virtuous actions, aims precisely at ensuring that the agent pays close attention to the moral order of incentives in the determination of his own choices and succeeds in giving prevalence to the moral incentive over the interest of his self-love. Since, finally, the determination of the order of one’s motives is by no means irrelevant to the possibility of an impartial assessment of the circumstances of the action, the meaning of the norms, and the very content of one’s maxims, the search for a truthful relationship with one’s moral convictions and motives is ultimately an indispensable condition for a moral evaluation which aims to be close to the ‘truth’. The search for as exact a correspondence as possible between judgement and reality is precisely the ultimate goal that Kant attributes to casuistry in its positive role within his ethical doctrine of method. In open contrast to its traditional meaning, against which he himself had bitterly polemicised, Kant understands casuistry to be not a sophistical attempt at making justifications of convenience appear plausible, but rather a continuous, never-ending ‘practice in how to seek truth’ (TL, 6: 411.20–21).

5. Conclusion

According to Kant, establishing moral principles requires apodictic certainty (cf. KrV, A823/B851). The same degree of certainty, however, cannot be achieved when it comes to applying moral principles to the manifold cases of moral experience, where instead ‘uncertainty and probability’ often dominate (cf. V-PP/Powalski, 27: 127.35–128.2). Here, on the level of applying norms, agents are called upon to pursue the highest possible degree of awareness regarding their choices until they acquire a firm conviction of their rightness, which takes the name of ‘moral certainty’. More precisely, Kant calls ‘conscientiousness’ a particular attitude in judgement, which is characterised by the utmost accuracy and truthfulness, and which alone can lead the agent to the required steadfast conviction about the moral value of the actions he or she intends to perform.

The present contribution to Kant’s dual conception of ‘moral certainty’ has mainly focused on this new meaning of the term as it comes to the fore, in particular, in the last section of the Religion, in the context of Kant’s discussion of the role of conscience as a ‘guiding thread in the most perplexing moral decisions’ (RGV, 6: 185.16–17). This meaning is novel in a threefold manner: first, in relation to the way the term ‘moral certainty’ was understood before Kant; second, in comparison with the way Kant understood the same expression in an earlier period of his own philosophical reflection; third, in contrast to the meaning which has so far been primarily investigated by Kant scholars.

As developed in section 2, in contrast to a tradition dating back to Descartes, the expression ‘moral certainty’ acquires a clear moral connotation for the first time with the first Critique, when Kant connects it in the Canon to the notion of rational faith in the existence of God and the immortality of the soul. Surprisingly, this very connection of concepts, on which scholars have particularly focused their attention, is lost in the second Critique, just as Kant comes to clarify the status of the highest good as the ‘whole object’ and the ‘determining ground of the pure will’ (KpV, 5: 109.21, 110.1), and thus the essential link of rational faith with the deepest reasons for moral action. Shortly thereafter, a further, entirely different redefinition of ‘moral certainty’ marks another important development in the conceptual history of the notion. This new meaning, investigated in section 3, emerges in the 1790s in the context of Kant’s theory of conscience, namely in his use of the term ‘moral certainty’ to refer to conscientiousness in the practice of moral judgement. This new sense of the term is explained through the principle of quod dubitas, ne feceris, which Kant expressly defines as opposed to the principle of probabilism (cf. RGV, 6: 185.23–186.9), thus also unambiguously indicating the (moral-theological) tradition to which he is thereby polemically referring. Moreover, since Kant understands moral probabilism as the most appropriate instrument for the human tendency to quibble over the moral law and invent seemingly moral justifications for its transgression, it becomes evident that this new understanding of the notion of moral certainty, together with its historical and systematic background, cannot but be of great interest for the purpose of moral education, and so for the aims of Kant’s ethical doctrines of the method. As a first attempt towards further developments, I briefly sketched these matters in section 4.