[We are pleased to publish the following review provided to us by the Society for German Idealism and Romanticism Reviews series.]

Pollok’s book brings three distinct areas of Kant’s philosophy – the theoretical (epistemic), the practical (moral) and the aesthetic – under the single heading of the ‘theory of normativity’. Insofar as judgements alone are ‘normatively distinct’ in virtue of being imputable to us (p. 13), one question unifies all three domains of normativity: ‘What are the fundamental norms of the validity of our judgments?’ Such norms, Pollok argues, are none other than Kant’s synthetic a priori principles, which ‘represent the structure of the “space of reason”’ by ‘mak[ing] explicit the conditions of our theoretical, practical, and aesthetic judgments’ validity’ (pp. 1, 3).

This thesis is fully borne out in part III of the book (The Legislation of Pure Reason). In chapter 7 (The Normativity of Judgments of Experience), Pollok argues that ‘the central idea of Kant’s theory of epistemic normativity’ – as developed in the Transcendental Aesthetic and Analytic of the first Critique and summarized in §23 of the Prolegomena (4: 305) – is also what addresses ‘the central problem of his transcendental idealism, namely: How are synthetic judgments a priori possible?’ These judgements are possible and normative for the same reason: along with forms of intuition, categories and schemata, they make explicit the formal conditions and lawfulness of our experience (pp. 239, 248; see also pp. 2, 19).

In chapter 8 (The Normativity of Practical Judgments), Pollok locates the ‘core’ of Kant’s theory of practical normativity in the relation between the idea of a universal practical reason and a finite practical reason such as ours and hence in the transformation of the moral law into a categorical imperative. The latter, which is a synthetic judgment a priori, is the principle that determines the objective validity of our maxims so as to render them lawful in the sense of being justifiable, though not necessarily justified (pp. 255–7, 271–2; see also 2–3, 19–20).

In chapter 9 (The Normativity of Judgments of Taste), Pollok reveals the ‘core’ of Kant’s theory of aesthetic normativity by examining his deduction of pure aesthetic judgements, which eventually takes us to a synthesis of two aspects of the idea of the supersensible (as nature’s purposiveness for our cognitive faculties), namely a supersensible substratum of nature and a supersensible substratum of humanity. The question is how pure aesthetic judgements are possible as universally valid while resting on subjective grounds. The answer, Pollok explains, lies in the synthetic a priori principle of purposiveness (pp. 285, 290–304; see also pp. 3, 21–2).

At this point, one may wonder what we gain by identifying the question of the possibility of synthetic a priori judgements as a ‘normative’ question. After all, it does not seem that any substantive Kantian claims made about such judgements could not be made just as clearly without invoking this label. Are Kant’s three Critiques fitted into a theory of normativity partly because normativity happens to be a trending topic in today’s philosophical discourse?

Pollok anticipates this kind of concern early on in the book, when he tries to fend off, pre-emptively, the charge of ‘grave anachronism’ or of using a ‘contemporary buzzword that has its home in debates on meta-ethics and perhaps some other sub-disciplines of analytic philosophy’. Pollok makes two observations in response. First, while Kant admittedly did not use the term ‘normative’, in eighteenth-century Germany the adjective normativ was already used in the juridical sense of ‘binding’. Second, what we now consider to be linguistic markers of normativity, such as ‘law’, ‘lawfulness’, ‘rule’, ‘objective’ and ‘validity’, are ubiquitous in Kant’s major writings. To Pollok, this fact suggests that problems of normativity may be central to Kant’s Critical philosophy as a whole. And his overall aim in the book is to show that Kant indeed has a theory of normativity that is not only ‘systematic’ – as an ‘account of what it means for judgments to be normative’ in each of the three domains of reason – but also ‘original’ (p. 2).

Pollok foregrounds what he takes to be the original features of Kant’s theory in parts I and II plus the first chapter of part III (chapter 6). In part I (chapters 1–3), he shows how Kant’s theory essentially differs from pre-Kantian alternatives. First, as ‘the cardinal insight of the critical turn’, Kant realized that judgements rather than ideas are ‘the basic unit to which normativity applies’ (p. 56). Accordingly, his theory of normativity initiated a fundamental shift ‘from the clarity of ideas to the validity of judgments’ (title of Part I) as ‘the central normative problem’ (p. 47). For that reason, Pollok dedicates the entire chapter 3 to sorting through an enormously complex taxonomy of judgements on Kant’s behalf, in preparation for ‘a proper understanding of the systematicity of our judgments’ normativity’ (p. 115). Second, Kant rejected the ‘perfectionism’ that characterized earlier rationalist theories of normativity, namely ‘the doctrine that the human mind should try to conform as much as possible to an infinite intellect’ (p. 56). Instead, Kant sought ‘the Archimedean point in the space of reason, as it were, revealing the lawful structures of what is and what ought to be the case, or, in his own terminology, some synthetic judgments a priori’ (p. 45). These judgements ‘serve as the laws in light of which the validity of all sorts of judgments … can be assessed’ (p. 56). And marking the Archimedean point is ‘the concept of a self-conscious and autonomous self, capable of understanding and acting in accordance with the grasp of those laws’ (p. 45).

In part II (Kant’s Transcendental Hylomorphism; chapters 4-5), Pollok explains how Kant’s critical turn from ideas to judgements as the primary medium of normativity, along with his anti-perfectionist view of normativity, was deeply connected with another revolutionary move, namely from ‘ontological hylomorphism’ to ‘transcendental hylomorphism’. The latter comes down to a judgement-centred dichotomy between matter and form or between ‘determinable sensibility’ and ‘intellectual determination’, as is fitting for our discursive intellect (pp. 121–31, 143–95). Since forms are what make our representations ‘normatively visible’ or universally intelligible, valid and assessable, they are the ‘currency of normativity’ (pp. 118–19). To that extent, as Pollok later puts it in the Epilogue, ‘without Kant’s transcendental hylomorphism … normativity, as the dependence of the judgment’s validity on the laws of reason [i.e. synthetic principles a priori], generally speaking, could never arise’ (p. 309).

In chapter 6 (Reason Prescribes Laws to Us), Pollok tackles ‘the problem of normativity’ in the form of this question: ‘Where do the norms of thinking and acting originate, how can they be binding on finite reasoners like us, and who performs that ‘pure act’ [by which discursive intellect determines some matter according to some form]?’ (pp. 199–200). The title of the chapter captures what Pollok takes to be the gist of Kant’s ingenious answer: it is a matter of self-legislation of human reason, understood here as ‘an acknowledgement of the laws of reason’. To substantiate this view, Pollok explores the ‘natural right’ tradition coupled with Kant’s distinction between homo phaenomenon and homo noumenon as two attributes under which a human being can view himself, so that he can assume the roles of both a legislator and a subject of law without contradiction (pp. 206–11).

This brief overview can barely do justice to the immense complexity, breadth and richness of Pollok’s book. His analyses touch upon and contain substantial interpretative claims about nearly every major component of Kant’s philosophical system. Many of those claims merit a close engagement. My main interest here is to get a clearer sense of the big picture. For that purpose, I shall focus on the very notion of normativity.

To begin, I ask: what is the single, overarching concept of normativity, if there must be one, that unifies Pollok’s analyses throughout the book? As I read on, I find it difficult to pin down an exact answer to this question. Consider the following cases.

1. (i) At the very beginning of the book, Pollok describes Kant’s theory of normativity as ‘a systematic account of what it means for judgments to be normative’ (p. 2). ‘Judgments are normative’, Pollok later states, ‘in light of synthetic principles a priori’ (p. 14).

2. (ii) These principles are in turn said to be ‘normative for … certain domains of judgments’ (p. 9) and ‘normative for finite reasoners using their cognitive faculties … to make those judgments’ (p. 10). Either way, Pollok is certain that they play ‘the central normative role’ in Kant’s critical philosophy (p. 211, n. 21).

3. (iii) Kant’s critical investigation of a priori judgements is characterized as a ‘normative enterprise’ that, unlike Locke’s physiological or ‘descriptive inventory of our mental powers’, seeks for ‘a principled demonstration of the legitimacy of their usage’ (p. 90). Similarly, Pollok identifies the ‘normative structures’ underpinning the objective validity of a judgement as the ‘non-psychological’ forms of possible cognition (p. 96).

Besides a standard contrast between what is normative and what is merely descriptive in (iii), we can extract three notions of normativity from (i) and (ii), which are at least prima facie different.

1. (a) Particular kinds of judgements, be they theoretical, practical or aesthetic, are normative just in case they are ‘liable to assessment’ in light of certain synthetic a priori principles (p. 10). In other words, the normativity of judgements consists in their ‘possible conformity to’ (p. 141), or the ‘dependence’ of their validity on (p. 309), those principles as ‘the laws that are formative of the correct use of our reason’ (p. 199).

2. (b) Synthetic a priori principles are normative for judgements in the sense of being ‘criterial’ (p. 61) or ‘serv[ing] as the laws in light of which the validity of all sort of judgments – cognitive, practical, and aesthetic – can be assessed’ (p. 56).

3. (c) The same principles are normative for finite reasoners in that they have ‘binding force on the individual subject’ making judgements (p. 13; see also p. 10).

I suppose that, by calling judgements ‘normative’, Pollok really means that they (unlike ideas or concepts) are normatively distinct. So, I shall set (a) aside and focus on (b) and (c). Since being ‘criterial’ and being ‘binding’ are neither identical nor intrinsically connected (at least not in any obvious way), they cannot be simply folded into one concept of normativity. (Using the word ‘normative’ to express both does not help.) How are these two senses of normativity related, then? What, if anything, unites them? To motivate these questions and see why they matter, let me begin by bringing in some textual materials not mentioned in Pollok’s book.

Pollok, defending his attempt to unite Kant’s three Critiques in a theory of normativity even though Kant himself never explicitly used the expression ‘normative’, points out that Kant nevertheless employed the German term Norm to mean ‘a prescribed rule, or law, that one has to observe strictly and must not contravene’ (p. 2; the quote is from a 1740 German lexicon). We may get a significantly more nuanced picture, however, if we look at how the term or its Latin equivalent, norma, is occasionally used in Kant’s writings, notes and lectures. A preliminary study has led me to the following observations.

(1) Basically, a norm is a model or standard for appraisal or passing judgement (Beurtheilung) (Jäsche Logic [JL below], 9: 15). We can find this notion of norm in all three contexts of Kant’s philosophy:

(2) A norm is often portrayed as an ‘ideal’ or ‘prototype’. Here are three examples. First, an ideal of the structure of a face serves as the norm for judging about specific faces (Anth, 7: 297–8). Second, by certain norms, we judge about cognitions regarding their ‘logical [as opposed to aesthetic] perfection’ (JL, 9: 36) or their veracity (Richtigkeit) qua formal perfection (perfectio formalis) (Refl 2189, 16: 264). Third, the ‘idea of a constitution in harmony with the natural right of human beings’ signifies a ‘Platonic ideal (respublica noumenon)’ as ‘the eternal norm for all civil organization in general’ (CF, 7: 90–1; see MM, 6: 313).

(3) There is an important distinction between norm (norma) and law (Lex). This is evidenced by the following passage:

(4) The distinction between a mere ground of obligation and what is obligating may also be captured by the distinction between a norm in the general sense and a ‘binding norm’. Kant reportedly limits the latter, which is an ‘imperative’ properly so-called, to ‘that class only where a moral necessitation or obligation can gain entry, in that only there is it categorical, and thought of in application to a finite being’ (Eth-V, 27: 517; my italicization).

If we may translate these remarks about norms into the language of normativity, there seem to be two distinct notions at play, roughly formulated as follows. First, a rule may be normative in the criterial sense for a type of activity or thing, as the standard for assessing tokens of that type. Second, a rule may be normative in the imperatival sense in relation to a kind of beings, as categorically binding on them. The latter (but not the former) agrees with how, according to Pollok, the adjective normativ was used in the eighteenth-century Germany.

This preliminary analysis seems to pose two exegetical challenges. First, just because such notions as ‘law’, ‘rule’ and ‘validity’ are ubiquitous in Kant’s writings, it does not follow that he has a single concept of normativity to unify them. So the burden is on Pollok to construct an exact one on Kant’s behalf. Second, given that Pollok’s chief concern in the book is ‘the nature of the bindingness of laws’ in all three domains of Kant’s critical philosophy (p. 4), while Kant himself reportedly limited binding norms and hence imperatival normativity to the moral domain, it is unclear whether or how he would extend the same notion of normativity to non-moral domains – or, for that matter, why a Kantian should be motivated to do so.

Pollok is mindful of both problems. He recognizes ‘the notorious vagueness of the term “normativity”’ (p. 3). And he is particularly keen to answer Clinton Tolley’s challenge as to whether Kant’s notion of normativity, as it applies paradigmatically to moral laws, applies to laws in general (e.g. laws of logic). Here is the main question raised by Tolley’s challenge as Pollok sees it: ‘If a law is constitutive of an activity, can this law still be considered normative?’ (p. 5). Pollok’s initial answer is that ‘Kant’s synthetic a priori principles may be constitutive and normative insofar as our judgments are liable to assessment in light of them’ (p. 9). He then explains: those principles are ‘constitutive, since they are the conditions of the possibility of judgments of experience and mathematics, moral maxims, and the pleasantly exciting judging of beautiful objects’; at the same time, they are normative for us as the judging subjects, i.e. ‘binding on our faculties’, since we can indeed fail to obey them (p. 10). In giving this further explication, Pollok has moved from criterial normativity to imperatival normativity, which are directed at judgements and judging subjects, respectively. What is the rationale behind this move? Is it because there is no real difference between the two notions of normativity? Or, rather, is there a certain transition from one to the other? Also, if the judging subject’s ‘self-understanding’ somehow ‘turns [the relevant principles of judgements] from constitutive to normative’ (pp. 9–10), does the same transformative procedure apply to normativity in both senses of normativity?

Since the basic concern behind Tolley’s challenge to the standard normative interpretations is whether imperatival normativity can be meaningfully extended from Kant’s moral theory to the theoretical domain, let me press the challenge a bit. Tolley’s immediate target is the imperatival normativity of pure-logical rules. Using the moral law as the paradigm of what can serve as an imperatival norm, he explains how on Kant’s account this law can stand in different relations with different kinds of subjects: it is non-normative for a holy being, who simply could not fail to fulfil the demands of morality and so for whom ‘ought’ would be out of place; the same law is imperatival-normative for beings like us, who can fail to obey it. Now Tolley asks

By Tolley’s analysis, the relation in question can only resemble that between the moral law and the holy being. Therefore, logical rules cannot be imperatival-normative for the logical subject (Tolley Reference Tolley2006: 377–86). A similar challenge, mutatis mutandis, may be posed concerning Pollok’s theoretical case: how do we, finite reasoners with a discursive intellect, relate to the principles of pure understanding? If the relation is more like the one between the holy being and the moral law, then the principles cannot be imperatival-normative for us.

Pollok disagrees, of course. He states:

Presumably, ‘normative’ in this context is meant in the imperatival sense. But now it is unclear to me how the principles that are supposed to be imperatival should be formulated and prescribed to us, the reasoners. Do they have exactly the same content as the principles in light of which our theoretical judgements are supposed to be assessable? I am not sure. After all, there is no ‘ought’ in those principles, either explicitly or implicitly.Footnote ¹