Practical judgment as reflective judgment: On moral salience and Kantian particularist universalism

Moral particularists and generalists alike have struggled over how to incorporate the role of moral salience in ethical reasoning. In this paper, I point to neglected resources in Kant to account for the role of moral salience in maxim formation: Kant's theory of reflective judgment. Kant tasks reflective judgment with picking out salient empirical particulars for formation into maxims, associating it with purposiveness, or intentional activity (action on ends ). The unexpected resources in Kantian reflective judgment suggest the possibility of a particularist universalism , where recalcitrant particulars directly inform, and in some cases revise, moral principles. Such an account improves on particularist accounts of moral salience and moral perception: rather than deriving moral sensitivity solely from an agent's upbringing or cultural resources, the reflective dimension is situated alongside the universalist dimension of moral principles typically identified with Kantian ethics, allowing for a critical approach both to moral universals and to the reception of moral particulars.

The role of moral salience poses a problem to Kantian ethics because the categorical imperative, often conceived as a test, can only determine the moral permissibility of maxims that have already been formulated, once the morally relevant features of situation taken to call for one action rather than another have been factored in. As Dancy puts the objection, the categorical imperative tells us only that it is wrong to act on a maxim that cannot be universalized; it does not act "as a test of whether a consideration is morally relevant" (2004, p. 68). To be able to apply a given moral principle, agents have to know which features of their proposed action are of moral import-or, even more basically, they must first know whether a given situation calls for action or not.
Particularism, on this view, thus demonstrates that the introduction of new moral considerations can alter our ultimate moral conclusions. Take Dancy's characterization of the following groupings of reasons for acting: 1. If you are causing someone pain, you are doing something wrong. (p ! q) 2. If p and the pain is a statutory punishment for a recognized offence, you are not doing something wrong.
3. If p & r and the punishee was unjustly convicted, you are doing something wrong. ([p & r & s] ! q) (Dancy, 2004, pp. 8-9) Dancy takes reasoning chains (1-3) to tell in favor of particularism and against universalism, since, on his view, a particular feature (such as p) can count against an action in one context, such as (1), and in favor in another, such as (2). Ultimately, which one it will be comes down to the context in which one acts, and whether one is sufficiently perceptive to notice each of the features salient to one's course of deliberation.
Another way of capturing the objection is in terms of what the Kantian literature has termed the problem of relevant descriptions (O'Neill, 1975(O'Neill, , 2018, a charge initially formulated by Elizabeth Anscombe. Anscombe (1958) contends that since the permissibility of maxims depends on the description of the action one employs, and since any given action admits of innumerable possible descriptions, the universalization procedure for determining permissibility is "useless" (27). If, to continue using Dancy's example, I choose to describe my action as "I am causing someone pain", I will, following the reasoning in (1), conclude my action to be morally wrong; if I choose to describe my action as "I am inflicting statutory punishment for a recognized offense", I will, following (2), take it to be morally justifiable. Or, in an example Anscombe (1957) herself gives: an agent poisons a group of Nazis by putting toxic chemicals into their water supply. Anscombe argues that this agent can justifiably describe his action as "moving his arm", "operating the pump", "replenishing the water supply", or "poisoning the inhabitants". The same action admits of all four possible descriptions, picking out different salient features of the action-context in each case. And, since the results of the categorical imperative procedure differ depending on the description supplied, the problem of relevant descriptions, much like the problem of moral salience, poses a concern for Kantian ethics. The specific action description "I am moving my arm" can be universalized across all agents, but the more general description 3 "I am fatally poisoning the inhabitants of this house" results in a contradiction, and therefore cannot be universalized. 4 Anscombe concludes that Kant's moral philosophy cannot tell us how to formulate our maxims, but only how to adjudicate the morality of already formulated maxims. 5 Since the same action admits of many possible descriptionsin other words, many possible maxims can apply to the same action-Kantian moral theory cannot speak to important morally relevant content that pertains to the very question of moral admissibility the categorical imperative was to adjudicate. 6 The problem of relevant descriptions bears directly on practical judgments that arise in ordinary deliberation, since not only is it the case that agents can offer varying descriptions of the same action, picking out this or that salient feature of their action-context, but it is also the case, in terms of their action phenomenology, that they often do. As Allen Wood notes, "My intentions can be described-not artificially but quite accurately-in greater or lesser detail, at higher or lower levels of generality, by mentioning or omitting this or that feature of the situation" (1999, p. 103).

| RESPONSES TO THE PROBLEM
There have been two principal avenues available to Kantian ethicists seeking to respond to the particularist challenge: one can either embrace moral salience, or one can reject the problem altogether. Opting for the former, however, prompts a further interpretive dilemma: to accommodate moral salience, one must either account for how moral salience can be derived from Kant's moral philosophy, or alternate resources for it, external to Kant's moral philosophy, must be found.
The first horn has been tackled, most famously, by Barbara Herman, and her account of "rules of moral salience" (RMS). Herman argues that the results of the categorical imperative are subject to what agents take to be relevant content to be incorporated into their deliberative procedure, where the issue of which content is relevant is to be settled through RMS. Herman agrees with the particularist, then, that pure moral principle (including the categorical imperative) is not sufficient; agents must also be able to discern "the moral marks of their actions and circumstances" (1993, p. 78; see also Herman, 1996). Yet RMS are "not themselves moral rules"; they are rules for judging context inculcated through socialization and moral training, and as such are contingent rather than necessary, dependent on culture and education (1993, p. 75).
While the strategy of embracing salience helps to account for moral responsiveness, it lacks textual basis in Kant's own expressed views, posing a problem for the Kantian grounding of RMS. Since this grounding cannot be located in the categorical imperative procedure, which applies only to already formulated maxims, Herman traces RMS to the formulation of the moral law as a fact of reason (5:29). The fact of reason draws our attention, on her view, to our consciousness of ourselves as moral agents, and thus to the conditions of our own moral agency, directing us, in turn to the marks determinate of human agency (1993, p. 85). 7 These, she claims, are aspects of moral experience that are not reducible to the categorical imperative procedure, and that therefore can help to orient a Kantian account of moral salience.
However, Kant seems not to have had RMS in mind when he articulated the idea that the moral law appears to us as a fact of reason. The context in which he does so is by situating it as a premise in an argument leading directly to the formula of universal law, FUL (5:30), one of the formulations of the categorical imperative. Indeed, in this passage FUL is given the title "Fundamental Law of Pure Practical Reason" without any explicit transition between Kant's introduction of the "moral law" and FUL as the "fundamental law", raising the possibility that what he means to pick out with the term "moral law" in this context just is the most basic formulation of the categorical imperative. 8 This undercuts Herman's claim that the moral law can be used to ground RMS independently of the categorical imperative procedure.
The second horn-rejecting the problem altogether-has been advanced by Onora O'Neill (2018), who objects to the invalidity of the point of view on action that moral salience theorists presuppose. O'Neill charges particularists with taking up the stance of spectators or observers rather than doers of action. Thus, such accounts rest on a confusion between theoretical and practical cognition, since morally "perceiving" or evaluating situations involves theoretical judgment rather than practical: "A fair amount of ethical writing has tried to construe ethical judgement as reflective. This is plausible only for ethical judgement about existing or past cases, where the particular to be judged can be given. It is not possible for practical judgements about what to do, which do not seek to 'appraise' or 'attend to' or 'evaluate' existing situations" (124n). 9 Thus, practical judgment is "neither determinant nor reflective", since it involves what O'Neill terms "indeterminate universals" that lack already given particulars: practical judgment "must be undertaken before the relevant particular exists: we cannot pick out future act-tokens" (2018, p. 123). The role of moral salience in practical deliberation is therefore limited to a mere "preliminary" to action, an "early warning system" (1989, p. 184; 2018, p. 18): action strictly speaking is about shaping the world rather than observing or judging it.
But how, then, to understand the agent's generation of maxims, or indeed, formation of intentions? What would it mean to apply an "indeterminate universal", such as humanity or autonomy, in abstraction from specific aspects of one's situation? Is it tenable to conceive of the moment of action as altogether divorced from consideration of context? And how should such a view be brought to bear on the formation of maxims themselves-maxims, which, in O'Neill's original formulation, should "be individuated only by referring to a person, and so to a place and a time", such that one's maxim can be "inappropriate to the situation" (1975, pp. 62, 247)? 10 By way of a partial response, O'Neill claims: "The appraisal of situations, and of their details, is of course of great ethical importance-if we do not notice the bullying we cannot consider whether to desist…. But noticing and appraising a situation is not practical judgement: having noticed the bullying we still need to decide whether to desist or to intervene" (2018, p. 89). In other words, for O'Neill, the question of moral salience, while "ethically important", is one of theoretical judgment rather than practical judgment. But notice that a given perception of a particular moral situation already often guides the moral response: judging a given situation to be "bullying" rather than "teasing" already suggests that the right response is to intervene, not to laugh it off. O'Neill's claim that practical judgment has to do solely with shaping the world rather than perceiving it risks leading to the un-Kantian conclusion that the practical question in such cases concerns, first and foremost, the consequences of action over the crafting of a fitting response.
What alternatives then remain if both strategies fall short-both Herman's attempt to accommodate moral salience within Kant's moral philosophy, and O'Neill's rejection of moral salience altogether? One promising and well-known route, albeit one often ignored among Kantian ethicists, is presented by Hannah Arendt, who takes recourse to alternate resources in Kant to account for the particularist aspects of moral judgment. In particular, where Herman overlooks the role of reflective judgment in ethics, and O'Neill censures it, Arendt famously embraces it. She uses the model of aesthetic judgment to counter what she takes to be Kantianism's overly rigoristic reliance on the application of rules-as in the example of Adolf Eichmann, who took himself to be acting on Kantian principles even when organizing Nazi death camps. What is needed, she argues, is not moral principle, but the exercise of judgment, as in the case of judging a beautiful object for which, Kant claims, a definite concept is lacking: "The faculty of judgment, which one may call, with some justification, the most political of man's mental abilities… is the faculty to judge particulars without subsuming them under those general rules which can be taught and learned until they grow into habits that can be replaced by other habits and rules" (1971, p. 446; see also 1963, pp. 136, 295). Arendt can therefore be taken to concur with a key line of argumentation in particularist accounts of moral perception 11 : genuinely moral action requires the judgment of particulars that should not ultimately be subsumed under an externally imposed "general rule" or habit. Yet Arendt's view, too, faces interpretive and broader philosophical challenges. Arendt reads Kant's account of aesthetic judgment in particularist terms, and ascribes to a model of morality derived from this very particularism (Arendt, 1970, pp. 15, 53-55, 61, 67). As I show below in more detail, however, Kant's account of reflective judgment, whether aesthetic or teleological, is not strictly particularist, nor is it non-conceptual: while it is right to say that aesthetic particulars cannot be subsumed under existing universals, they do play a constructive role in forming new universals. 12 This aspect of reflective judgment is even more pronounced in the case of teleological judgment, which does explicitly involve conceptual determination from the outset: it "proceeds in accordance with concepts" and "with particular principles" (5:194).
Attending to these nuances in Kant's account is instructive: judgment here does not take the form of judging each particular in independence from universals, but demonstrates that particulars can shape universals. The options, in other words, are not restricted to either applying rules that are already given or treating particulars solely as particulars. As I show, the relation between moral particular and moral universal can be understood in terms of a new model: as a two-way dependence rather than mere subsumption 13 or application.

| AN ALTERNATE APPROACH
Taking into view all three proposals for construing Kant's ethics in light of the problem of moral salience, it is clear that, while each is instructive in its own right, more work remains to be done to fully address the problem. I therefore suggest building on all three accounts in the following fashion: I concur with Herman that moral salience has a place in practical reasoning, one not captured by the categorical imperative alone. But rather than attempting to ground moral salience in the moral law as fact of reason, I bite O'Neill's bullet: moral salience is the purview of reflective judgment; it is not the domain of the moral law or categorical imperative (what I will later term the "moral dimension" of practical reasoning). 14 However, in taking recourse to reflective judgment to account for this aspect of ethical deliberation, I do not construe it, as Arendt does, in particularist terms. Instead, on my account, reflective judgment is implicated in the reception of salient features and ensuing formation of universals: moral concepts or (what interests me primarily in this paper) maxims. 15 In so doing, I take the role Kant accords to reflective judgment in both Introductions of the Critique of Judgment to also apply to moral reasoning. 16 Reflective judgment is there defined not in terms of its activity in taking in particulars as such, but in forming new generals or universals on the basis of particulars: To reflect… is to compare and to hold together given representations either with others or with one's faculty of cognition, in relation to a concept thereby made possible. (20:211; First Introduction).
The power of judgment in general is the faculty for thinking of the particular as contained under the universal. If the universal (the rule, the principle, the law) is given, then the power of judgment, which subsumes a particular under it… is determining. If, however, only the particular is given, for which the universal is to be found, then the power of judgment is merely reflective [bloß reflektierend]. (5:179; Published Introduction).
Thus, while determining judgment, one major topic of the first Critique, subsumes the experiential manifold under already given rules (the categories, or existent empirical concepts), reflective judgment forms or finds new "universals" in encountering matter that cannot be subsumed under the existing repertoire of rules. Reflective judgment seeks a universal for a particular that is given, while determining judgment seeks particulars to which universals such as the categories, which are already given, can be applied. 17 In considering the role reflective judgment plays in practical reasoning, it is helpful to take on board a suggestion made in the literature on Kantian theoretical cognition, where the interdependence of reflective and determining judgment has been acknowledged for some time (Bell, 1987;Guyer, 1990;Floyd, 1998;Allison, 2001, p. 20;Ginsborg, 2015). As Longuenesse notes, Kant is careful to characterize the aesthetic and teleological judgments that comprise the focus of the third Critique as "merely [bloß, nur] reflective judgments." 18 This usage suggests that both determining and reflective judgments have a reflective aspect, but only reflective judgments are just reflective: "The peculiar feature of aesthetic and teleological judgments is not that they are reflective judgments (for every judgment on empirical objects as such is reflective); it is rather that they are merely reflective judgments, judgments in which reflection can never arrive at conceptual determination" (Longuenesse, 2001, p. 163). All empirical judgments thereby comprise a reflective dimension, insofar as they presuppose the functions of comparison, reflection, and abstraction Kant describes as necessary "subjective conditions under which we can arrive at concepts" in the Amphiboly chapter of the Critique of Pure Reason (A260/B316). 19 But only "reflective" judgments are uniquely reflective in that they lack the possibility of ultimately subsuming their objects under concepts, retaining only these reflective functions. 20 My proposal in this paper is that a reflective dimension could be extended to the case of moral reasoning, as well as theoretical reasoning. 21 "Merely" or "purely" reflective judgments would thereby be differentiated from judgments that combine reflective and determining, or reflective and moral, aspects. On this reading, aesthetic and teleological judgments, as "merely" or "only" reflective, would constitute paradigm cases for studying the general function of reflective judgment, since both aesthetic and teleological judgments lack a corresponding theoretical or moral dimension. 22 Attributing a broader role in this manner to the reflective in cognition helps to explain why Kant characterizes reflective judgment in both Introductions not narrowly in terms of aesthetic or teleological judgment, but more broadly, as I showed above, in terms of the formation of new universals on the basis of undeterminable particulars (20:211; 5:179).

| Reflective judgment and practical purposiveness
To show that a reflective dimension indeed obtains in practical reasoning as well as in theoretical reasoning, I need to demonstrate two claims: first, that reflective judgment is implicated in practical purposiveness as well as natural purposiveness, and second, that practical purposiveness is crucial for Kantian ethics. I will then return to the question of moral salience to examine how the reflective dimension of practical reasoning can help meet the particularist challenge-improving both on Kantian universalism as traditionally conceived, and on particularisms in ethics. I will have to be necessarily cursory in the interpretation of reflective judgment I advance in doing so: it is not my focus in this paper, given my concern with particularism and maxims of action, to elaborate a full defense of my interpretation Can reflective judgment plausibly be conceived as playing a role in practical purposiveness, as well as natural?
Reflective judgment is governed by the principle of purposiveness: "The power of judgment, which with regard to things under possible (still to be discovered) empirical laws is merely reflective, must think of nature with regard to the latter in accordance with a principle of purposiveness for our faculty of cognition" (5:184). Indeed, purposiveness is often taken to be the central topic of the Critique of Judgment as a whole. 24 More specifically, this principle is reflective judgment's a priori (yet merely subjective) principle of natural purposiveness: that nature is conducive to our cognitive powers and is ordered in an intelligible way (5:185).
Yet Kant defines the notion of "purposiveness" (Zweckmäßigkeit) at issue as, most basically, practical purposiveness, or intentionality in action: action as aimed at a purpose (a telos), or end (Zweck). In §10, "On Purposiveness in General", Kant defines a purpose as "the object of a concept insofar as the latter is regarded as the cause of the former", and purposiveness as the "causality of a concept with regard to its object" (5:219). As Zuckert glosses the point, "Purposes are aims of intentional activity: they are objects (products or actions) conceived by agents and brought about in accord with those intentions" (2021, p. 353; see also 2007, pp. 78-79); Ginsborg concurs that Kant's first sense of purposiveness "appears to be restricted to products of intentional causality" (2015, p. 230). 25 Practical purposiveness, then, refers to intentionality: directedness toward an aim, choosing means to one's ends, guided by one's concept of it, or intention involving it. The Critique of Judgment's close connection to the practical through "purposiveness" is unfortunately sometimes missed in translation; Kant frequently employs the term Zweck (purpose) in his moral philosophy, but this is translated as "end", not "purpose"-most famously in the formula of humanity as an end in itself (Zweck an sich selbst) (4:428-437). Yet Kant stresses that the core notion-Zweck-is the same in both domains, advancing nearly identical definitions in both the Critique of Judgment and the Metaphysics of Morals. 26 Natural purposiveness (as "suitability" or "usefulness", in particular to judgment or cognition, or the "purposiveness of nature for our understanding", 5:187) is a second sense of purposiveness derivative from the first, more basic, sense (Zuckert, 2007, p. 80). Kant claims that "merely" reflective judgment judges nature in analogy to practical purposiveness, or "art": the purposiveness of nature is a "concept… distinct from that of practical purposiveness (of human art as well as of morals), although it is certainly conceived of in analogy with that" (5:181). Thus, in attributing purposiveness to nature, we analogically attribute our own agency and practical rationality to it, as if nature were intentional and agential, like we are. Practical purposiveness plays an "analogical" role in judgments of natural purposiveness because in the latter case, "where objects of nature are merely judged as if their possibility were grounded in art", such judgments "do not determine anything about the constitution of the object nor the way in which to produce it" (20:200-201; my italics). In other words, practically purposive judgments concern the production of objects which are suitable to our practical aims, shaping the world directly through our judging activity; judgments of natural purposiveness instead shape thought about the world, conceiving of phenomenal objects in a fashion suitable to our cognitive aims (namely, to be known by us). 27 On my reading, the pure employment of the power of judgment ("mere" reflective judgment) enables us to isolate the form of purposiveness, of the purposive judging activity common to both merely reflective judgment and to practically purposive activity as such, since, in the former case, no object is produced. "Mere" reflective judgments are therefore purposive without the practice, without the activity, arising in the subject's stance as spectator of empirical objects, rather than as actor. Practically purposive judgments, in contrast, encompass both reflective and practical dimensions, 28 as Kant suggests by devoting the opening section of both Introductions to analyzing them (20:195-201; 5:171-3). Thus, in the taxonomy of the three faculties given in both Introductions, Kant cites practical purposiveness, "art", as the power of judgment's "product" or "application" (20:246, 5:198). 29 Indeed, I have argued elsewhere that the "culture of art", or cultivation of practical purposiveness, is the point of unification of the Critique of Judgment as a whole (Vaccarino Bremner, n.d.).
In the Critique of Judgment, Kant ultimately situates the reflective power of judgment as mediating between the theoretical and the practical. Kant tells us that the power of judgment "can occasionally be annexed to either of them," to theoretical or practical philosophy, "in case of need", constituting an "intermediary" or making possible a "transition" between them (5: 168; 5:196). 30 Reflective judgment, in being positioned as intermediary between the two, suggests that reflective judgment is bound up with both theoretical and moral judgment. Kant never fully spells out what he means by this characterization of reflective judgment, indicating many suggestive points of connection throughout the Critique of Judgment. 31 Yet one way to make sense of the mediating role Kant accords to reflective judgment is that it is called for when it comes to those concepts, such as purposiveness, and those aspects of cognition, such as maxim formation, that combine moral and theoretical modes of reasoning, or that provide the possibility of transitioning from one to the other. To be sure, this argument is never consistently developed in the Critique of Judgment, since, throughout, Kant is principally concerned with "mere" reflective judgment rather than the reflective dimension of cognition as such. Nevertheless, we have begun to see that there are numerous indications that important resources for practical reasoning can be traced to reflective judgment, as the domain for purposive judging activity in general.

| Practical purposiveness and Kantian ethics
Even if it is granted that reflective judgment is implicated in thinking purposiveness in general-practical as well as natural-the objection still arises that practical purposiveness is merely a matter of theoretical cognition, rather than moral. Such an objection might be posed from O'Neill's vantage point in claiming that moral salience is merely theoretical and therefore not relevant to practical cognition as such: moral salience arises only in judgments of how things are; it does not figure in practical judgments that shape the world in accordance with a conception of how it should be. This takes us to the further question of whether practical purposiveness is even relevant to Kantian ethics. After all, Kant opens the Critique of Judgment in both Introductions by claiming that "practical precepts" constitute moral philosophy only "when and if their principle [Prinzipien] is not borrowed from the concept of nature, which is always sensibly conditioned" (5:173). Thus, an intention of action not grounded in a moral "principle" is only a part of theoretical philosophy, not moral philosophy, since in this case the will is "determined through natural incentives", concerning only "the nature of things" as they already are (5:172, 20:198). Yet, even in the Critique of Judgment, Kant nevertheless characterizes practical purposiveness in general as a mode of freedom: for instance, as "the freedom of [man's] causality" or "production through freedom" (5:368, 5:303; see also 5:368, 5:431, 6:213, 6:381, 6:384), including in cases which seemingly have nothing to do with morality, such as adorning clothing with 'colorful bird feathers' (5:368). Kant seemingly undermines himself-newly withdrawing the general domain of practical purposiveness from morality when 'naturally determined' (Papish, 2018) while at the same time attributing to it "freedom". 32 By way of an initial response to this apparent contradiction, I take Kant to be drawing a distinction in these sections between empirical and moral motivating grounds of the will, rather than advancing a more general claim against the role of empirical reflection in moral reasoning as such; indeed, I take his central point to be that the domain of practical intentionality, while ultimately subject to moral evaluation, cannot be reduced to the "moral dimension" of the will's motivation. 33 Indeed, Kant's moral theory admits an important role to considerations of the "nature of things" alongside the command of the law. Kant distinguishes maxims from laws, for instance, in that laws are prescriptive, while maxims have an indispensably descriptive aspect: A maxim is the subjective principle of acting, and must be distinguished from the objective principle, namely the practical law. The former contains the practical rule determined by reason conformably with the conditions of the subject (often his ignorance or also his inclinations), and is therefore the principle in accordance with which the subject acts; but the law is the objective principle valid for every rational being, and the principle in accordance with which he ought to act, that is, an imperative. (4:421n).
Maxims are defined as descriptive 34 : they consider the conditions of the subject, calling on her to pick out her desires (her "inclinations"), as well as her "ignorance". Indeed, this is what I take Kant to mean in drawing the distinction between maxims and laws as one between merely subjective as opposed to objective or universal validity: while a maxim is "the subjective principle of willing", the "objective principle (i.e., the one that would also subjectively serve all rational beings as the practical principle if reason had complete control over the desiderative faculty) is the practical law" (4:400). This is not to say that maxims are not also normative-as Longuenesse claims, the distinction can also be conceived as one "between a rule that we assign to ourselves (what we ought to do) under a condition particular to ourselves, and a rule that is binding for all" (2005, p. 252)-but there is no way to draw such a distinction without considering one's sense of what one's particular condition consists in, thus comprising an assessment of how things currently stand in the world.
It is therefore characteristic of maxims that they encompass our sense of our desires, inclinations, degree of ignorance, and other salient factors that bear on our action, as assessed from our particular standpoint. They are formed in accordance with our (at times deficient, or ignorant) perception of the empirical conditions in which we act, the particulars of our action-context: they reflect how the world shows up to us as finite human reasoners. The subjectively descriptive aspect of maxims distinguishes them from laws: a maxim picks out merely how a given "subject acts"; abstracted from the question of its universalizability, it does not (yet) make a claim to how an agent in general "ought to act". Maxims, in incorporating salient particulars of context, therefore reflect differences between subjects: "A principle that makes certain actions duties is a practical law. A rule that the agent himself makes his principle on subjective grounds is called his maxim; hence different agents can have very different maxims with regard to the same law" (6:225). A maxim is distinct from law in that it is made by a given subject, reflecting how the world appears from her point of view.
Kant claims that we act on purposes or ends, and that our purpose is picked out by the maxim of the action: "All maxims have: (1) a form, which consists in universality… (2) a matter, namely an end [Zweck]" (4:436). Maxims, in setting forth a purpose, are therefore expressions of one's practical purposiveness, such that "ethics can also be defined as the system of the ends [System der Zwecke] of pure practical reason" (6:381). Thus, practical reason must conform to subjective conditions, in particular to one's purpose: "In practical cognition… the principles that one makes for oneself are not yet laws to which one is unavoidably subject, because reason, in the practical, has to do with the subject, namely with his faculty of desire, to whose special constitution the rule can variously conform" (5:20, my emphasis).
Maxims are subjectively normative in that they necessitate the adoption of a given purpose, formed in accordance with the agent's desires, thereby reflecting the agent's sense of what is good or fitting for her to do.
We can see how the reflective and moral dimensions come together if we examine the universalization of maxims. If maxims are merely subjective, laws are merely objective: the former abstracted from the question of universalization, the latter from the empirical conditions of the acting subject, such that "the law contains no condition to which it was limited" (4:420). But the merely subjective condition of a proposed maxim can be taken to "also hold objectively", not just subjectively: "You must therefore first consider your actions in terms of their subjective principles; but you can know whether this principle also holds objectively" by "conceiving yourself as also giving universal law through it" (6:225). In other words, the reflective domain of maxim formation and the moral domain of law cooperate in situated practical deliberation-of maxims conceived as universalizable.
We can construe this process of universalization as not only one of top-down application, as it is often presented, but also as one of bottom-up reflection, insofar as the agent compares her representation of her particular condition against an (indeterminate) universal. It thus meets Kant's definition of reflective judgment as "compar [ing] and hold[ing] together given representations either with others or with one's faculty of cognition, in relation to a concept thereby made possible" (20:211; compare A260/B316). In so doing, the agent forms a maxim, reflecting particular features in context she takes to be salient, which can ultimately be vindicated as a universal-a particularized universal. 35 We can therefore see why Kant adds the qualifiers "bloß" and "nur" not only to the reflective dimension set against theoretical determining judgment (as we noted above), but also to moral reason, to define the scope of moral legislation: "Legislation through the concept of freedom takes place through reason, and is merely [bloß] practical.
Only in the practical alone [nur allein im Praktischen] can reason be legislative" (5:175; my emphasis). Yet this still leaves a domain of practical reasoning, that of practical purposiveness and therefore of maxim formation, not solely or directly governed by reason: as the passage continues, "conversely, however, where rules are practical, reason is not on that account immediately legislative" (5:175). In stressing that universalizing one's maxim consists in asking oneself whether the "maxim can also hold as a universal law", Kant indicates that the subjective normativity of maxim formation is only limited by one's representation of universal law-not replaced by it (6:225; 4:402-403, 4:412).

| The example of maxim formation in teleological judgments
How should we make sense of the cooperation of the reflective and moral dimensions? I have claimed that we can think of "mere" reflective judgment as isolating the form of purposive judging in general, which, in the practical case, is bound up with the motivation of the will. Thus, it's helpful for our purposes to consider one of the central examples Kant advances of merely reflective judgment: how can we make sense of the mere form of purposive judging activity without the possibility of any guidance by the moral law? Here, I will focus on teleological judgment, due to several key features that distinguish it from the other paradigm case of merely reflective judgment-aesthetic judgment. The "teleological power of judgment relates its representation of purposiveness to the object not by means of feelings but through concepts… it relates its reflection entirely to reason (not to feeling)" (20:250-251). That is, where aesthetic judgment is famously taken to lack any relation to concepts (or principles) and instead relates its representations to objects through feeling (as Arendt also claims), teleological judgment begins with a concept or principle and relates this reflection directly to reason. Second, and relatedly, "a teleological judgment compares the concept of a product of nature as it is with one of what it ought to be. Here the judging of its possibility is grounded in a concept (of the end) that precedes it" (20:240). Thus, the concept in question is specifically the concept of an end, which is employed to advance explicit "ought" claims-albeit for the activity of judging, rather than for acting; aesthetic judgment, by contrast, is "purposive… without a purpose", or end (5:306). 36 In each of these respects, teleological judgments share central features with practical judgments.
There is another important reason for examining teleological judgment: one issue in assessing the domain of maxim formation is, as Herissone-Kelly (2018)  To initiate a course of teleological reasoning, Kant claims that we must presuppose a determinate principle, 37 or "maxim" (5:376), a characterization consistent with reflective judgment's definition in general ("a mere faculty for reflecting on a given representation, in accordance with a certain principle", 20:211, my emphasis). This is because, while in the case of determining judgments, a concept, a cognitive rule, can guide the activity of judgment, 38 and in practical judgments there is the moral law, reflective judgments lack any conclusive determination (beyond the formal "concept of an end", 20:240) and must therefore first presuppose a principle.
Kant characterizes these principles of reflective judgment as "maxims" (20:205, 20:219, 5:182, 5:184) and "heuristics" (20:205, 5:411) as opposed to laws. As examples, he cites common sayings, such as "the maxims [Maxime] that everything in the world is good for something, that nothing in it is in vain" (5:379). 39 Such principles "do not say what happens… but rather how [things] ought to be judged": in other words, they are normative, claiming "logical objective necessity" (5:182). However, the "ought" in question is merely "subjectively valid", in that judgment gives itself its own normativity in the course of its reflection (20:225; 5:186). In other words, what Kant means by terming the principles of reflective judgment "heuristics" or "maxims" is that they are subjectively normative, as distinct from the objective normativity of moral and theoretical laws. While the moral law is objective, these teleological maxims only claim objectivity, because, unlike the moral law and the categorical imperative that follows from it, they only govern our own cognitive procedure in judging nature; while "transcendental" and "a priori" for our own exercise of judging, we are not justified, Kant thinks, to apply them directly to nature from the outset (5:182, 5:376). 40 Alongside these transcendental maxims, Kant claims that there are also more specific maxims that emerge in the course of empirical investigation, thus arising from experience. For example, the concept of internal purposiveness, as in an organism, follows from what Kant characterizes as an empirical "principle" (namely, "an organized product of nature is that in which everything is an end and reciprocally a means as well"): "This principle is of course to be derived from experience, that is, experience of the kind that is methodically undertaken and is called observation" (5:376). This empirical maxim is guided by the transcendental maxims Kant refers to earlier: its "universality and necessity… cannot rest merely on grounds in experience, but must have as its ground some sort of a priori principle" (5:376). Thus, the specificity of the principle of internal purposiveness is "empirical", while also grounded in a more basic, and more general, transcendental maxim that imbues it with normativity (5:376).
In sum, Kant argues that teleological judgment follows a normative procedure: presupposing a transcendental a priori maxim as a heuristic, proceeding in the course of empirical investigation on the basis of this maxim, and deriving subsidiary empirical maxims on the basis of the particulars encountered in experience. The empirical maxims are normative rather than merely descriptive due to their grounding in the a priori principle that was first presupposed.
Meanwhile, the a priori maxim, initially just a heuristic, is given form by its (subjective) confirmation in the course of experience. The stipulation of the a priori principle leads to the formation of new, empirical maxims that further specify the content of the initial principle. "Nature does nothing in vain", because, for example, natural organisms show up to us as internally purposive, with every part serving a necessary function relative to the whole. Evidence for the initial maxim, then, is provided by reflecting on a particular feature (in this case, a natural organism), which then gives rise to new, empirically derived maxims, and in turn, these maxims give new form and specificity to the initial, guiding principle.
We can thus think of Kant's account of the procedure of teleological judgment as exhibiting a two-way dependence between universals and particulars. Attributing salience to certain particulars rather than others is guided by an a priori principle, which is itself reciprocally informed by the particulars one encounters. The empirical maxims that result from this process are at once (subjective) vindications of the initial principle and determine it with specific content. But because the starting principle is merely a heuristic, nothing excludes the possibility that it might be disconfirmed by investigation. 41 The moment we take such principles to be unimpeachable rather than perpetually open to disconfirmation, Kant holds, we err into the transcendent, speculative realm (20:236). Indeed, the general procedure of teleological judgment, as Kant characterizes it, picks out the process by which the mechanical "rule" for judging nature comes to be revised in favor of a teleological one (at least when it comes to organic particulars).

| Practical judgment as bottom-up reflection: Self-critique and revising universals
The procedure of teleological judgment Kant outlines allows us to make sense of how maxims can be guided by normative principles but nevertheless remain descriptive, encompassing empirical content. At the outset, we considered how the apparent severance of subjective maxims from objective laws threatened to render the role of the categorical imperative arbitrary, resulting in different conclusions depending on the formulation of the initial maxim. We can now see how the relation between maxims and laws can instead be rethought in terms of a greater degree of reciprocity between the two, where moral maxims are guided, but not determined, by laws, which are in turn informed and imbued with a specific "shape" by particular maxims. Indeed, practical judgments can be fruitfully compared to teleological judgments in that they, too, must be informed by experience to render the objective principle more determinate, generating subjectively given maxims to be employed in action. These subordinate maxims of practical reason, however, remain defeasible; upon attempting to integrate new particulars, they may require revision, or may be articulated in terms of newly available concepts. They also-though a full argument for this claim is beyond my scope in this paper-can flesh out the objective principles, categorical imperatives, with greater determinacy and richer content, such that the principle of humanity need not be understood as mere abstraction, but as issuing in this specific maxim (e.g., to "intervene in bullying"). In so doing, practical judgments exhibit a two-way dependence between particulars and universals comparable to teleological judgments, one that can be attributed to the fact that practical judgments also have a reflective dimension.
Indeed, Kant suggests the purview of practical judgment is to generate maxims, likening its role to reflective judgment in science: The Doctrine of Right has to do only with narrow duties, whereas ethics has to do with wide duties.
Hence the Doctrine of Right, which by its nature must determine duties strictly (precisely), has no more need of general directions (a method) as to how to proceed in judging than does pure mathematics; instead, it certifies its method by what it does. But ethics, because of the latitude it allows in its imperfect duties, inevitably leads to questions that call upon judgment to decide how a maxim is to be applied in particular cases, and indeed in such a way that judgment provides another (subordinate) maxim (and one can always ask for yet another principle for applying this maxim to cases that may arise).
So ethics falls into a casuistry, which has no place in the Doctrine of Right. (6:411; my emphasis).
Kant here contrasts the constitutive, determining method of right (which, like "pure mathematics", "certifies its method by what it does"), with the reflective procedure of ethics, which generates new universals ("maxims") on the basis of "particular cases". Indeed, Kant here establishes a distinction between the topdown, derivative procedure associated with right and the bottom-up, indeterminate process of ethics, which remains open-ended: "one can always ask for yet another principle", such that the process of forming new universals is never complete. 42 Kant goes on to reinforce the comparison between reflective and determining procedures in claiming that "casuistry is, accordingly, neither a science nor a part of a science"-much like reflective judgment in teleology is distinguished from determining judgment in mathematics or physics in that it does not rise to the level of a systematic science (6:411; 5:417). This is because "casuistry is not so much a doctrine about how to find something as rather a practice in how to seek truth. To see how incorporating this bottom-up role for reflection might work in practice, we can consider the following, and apparently banal, case: Marisol is at dinner with her friend, Lamar, who has just been laid off. Marisol spends the whole evening talking excitedly about new career opportunities without noticing Lamar's discomfort.
Marisol's subjective endorsement of her maxim of action ("sharing my recent good news with my friend") may be morally permissible as far as can be determined through the categorical imperative procedure: indeed, she might take this specific maxim to be an expression of her commitment to a more general principle or duty ("being a good friend", 6:469-473). 44 But she has missed salient moral information: Marisol might have noticed that Lamar finds himself in a much more unfortunate situation, such that dwelling at length on her own career news rather than his own feelings of precarity may make him feel slighted. She realizes, in other words, that her behavior at dinner is in conflict with her duty to be a good friend to Lamar. Attending to the situation might lead her to revise her maxim as "bragging" rather than "sharing my news with my friend".
How might we think about this process of self-revision, a process which, as the problem of relevant descriptions reminds us, cannot be sought in the categorical imperative procedure alone? Here, we might consider Kant's account of teleological judgment, where initial heuristic maxims are informed by empirical particulars. In judging a natural organism, for instance, the initial maxim of natural purposiveness (e.g., "nature does nothing in vain") is given newfound specificity and determinacy and gives rise to the new, empirically derived principle of internal purposiveness (with corresponding maxim: judge the organism as if it were purposively arranged). In other words, the starting heuristic is revised once applied to empirical circumstances, in coming to grips with a particular recalcitrant to the agent's prior perception of reality (here, of nature as mechanistically ordered).
We can think of Marisol as being faced with a similar process. What is incumbent on her is to attend to a particular unassimilable to her previous,and perhaps rote, realization of guiding principle (the imperfect duty of friendship) in a specific maxim (sharing her good news). She has to notice, in other words, the conflict between the effects of her behavior on the world and her initial understanding of her behavior, and she has to be willing to revise what she is doing accordingly: to create a new universal, a maxim, that can better relate principle to world. In so doing, the initial principle is imbued with greater specificity and determinacy, as new ethical knowledge is formed in the ongoing practice of moral judgment: to be a good friend means that Marisol ought to attend to Lamar's feelings, to his own life circumstances, and alter her own responsiveness to him accordingly.
The picture I have just sketched is importantly different from particularism as such: it is not a merely reflective picture of moral reasoning, limited to perception of salience or the correct "reading" of situations (McDowell, 1996), but combines both moral and reflective dimensions of practical reasoning. Unlike the merely reflective case of teleological judgment (and unlike the strictly particularist picture), on the view I've elaborated here, Marisol also has the resources of the moral dimension of reasoning at her disposal, such that moral motivation can also help guide her perception of moral salience. The moral idea of humanity, for instance, might bring her attention to human feelings as salient features of her moral circumstances. Particulars can imbue universals with specificity, can lead us to revise our initial maxims or initial scope of duty, but the line of influence can also go the other way: moral universals can guide us in judging which particulars are salient.
As this example begins to bring out, accommodating a reflective dimension of practical judgment alongside the moral allows us to get a grip on the prospect of a critical attention-both toward our assessment of morally salient particulars, and toward the universals brought to bear on them. Indeed, we have seen that maxims are characterized as subjectively normative, against the objective or universal normativity of the moral law: we have to make our own maxims, which will reflect aspects of our situation that cannot be generalized to the conditions of all others ("only I myself can make something my end", 6:381). 45 If "the principles that one makes for oneself are not yet laws to which one is unavoidably subject", such that the "act of freedom" of choosing one's own practical ends is not immediately moral freedom (5:20, 6:381, my emphasis), the domain of ethical subjective normativity is not to be identified with the normativity of the moral law per se. The integration of a reflective dimension of practical reasoning allows us to attribute this subjective normativity to reflective judgment, which bears a distinct mode of "autonomy", one that "is merely subjectively valid", in that judgment gives itself its own normativity in the course of its reflection (20:225; compare 5:186). 46 The autonomy of judgment was a domain of normativity that was newly emerging for Kant: by recognizing judgment as itself an activity or practice, 47 one at work in theoretical spontaneity as well as in the making of moral maxims, it serves as "intermediary" between epistemic normativity and ethical normativity. This new domain of normativity, as the reflective dimension alongside the moral, can be indirectly guided by the moral dimension and its laws, much like empirical maxims of teleological judgment are guided by heuristic transcendental principles, while still generating a domain of subjective normativity in its own right-a normativity, or "freedom", which, if not solely moral freedom, can be construed as a mode of creativity, in which reflective judgment proceeds "not as it were merely mechanically, like an instrument, but artistically [künstlich]" (20:214).
Although the example I've just sketched is relatively ordinary, I take one advantage of it to be the generalizability of the commonplace experience it describes to broader processes of "making a concept possible" (20:211). Indeed, we need not think of such a process as limited to individual instances of deliberation, nor, as Kant makes clear, is it limited to maxims of intention. The creation of new normatively freighted concepts or amelioration of preexisting terms, for instance, can do much to make morally relevant features more explicit. The term "gentrification" was only coined once sweeping processes of displacement in London neighborhoods like Chelsea, Notting Hill, and Paddington were already underway, bringing a previously neutral, overlooked practice newly under moral evaluation (Glass, 1964); "sexual harassment" was coined when the members of a women's collective at Cornell University, in the course of discussing their shared experiences, decided to make it a political issue (Brownmiller, 1999). These The process of revising one's own moral perception may profitably be construed as the purview of the critical self-reflection of judgment. Kant claims that conscience is "the moral faculty of judgment, passing judgment upon itself", and refers to the epistemic maxim of "reflect[ing] on one's own judgment" (6:186, 5:295; see discussion in Ware, 2009). The critical assessment of one's maxims, then, is not limited to the application of a test, nor is it a matter of applying an additional rule (such as a rule of moral salience). Instead, it consists in a self-given exercise of judgment, one informed both by prior moral considerations and by the encounter with new particulars that fit at best uneasily with preexisting hermeneutical resources. Indeed, the unfolding process can perhaps be seen as a feedback loop, where the particulars inform our initial moral considerations, which, once given greater determinacy in this domain of moral experience, can thereby better guide us toward attending to features newly salient to us, and so on in turn. Thus, self-critique can be situated in a conception of moral reasoning as, in part, a reflective process, by which, in reciprocal fashion, particulars can direct scrutiny back toward the guiding universals.
If we carry this insight through, and thereby conceive of moral reflection as encompassing both top-down and bottom-up activities of the power of judgment, we can conclude that such a practice does not leave our guiding moral universals unchanged. 48 Indeed, Kant emphasizes our "subjective" understanding of given moral universals, reflecting their realization in particular conditions, alongside their "objective" status as ideals. 49 For instance, Kant claims: "With regard to perfection as a moral end, it is true that in its idea (objectively) there is only one virtue (as moral strength of one's maxims); but in fact (subjectively) there is a multitude of virtues, made up of several different qualities" (6:447). Or, in referring to virtue as such: "It is always in progress because, considered objectively, it is an ideal and unattainable, while yet constant approximation to it is a duty. That it always starts from the beginning has a subjective basis in human nature, which is affected by inclinations" (6:409). The generation of maxims from the bottom up ultimately impresses on the indeterminate, "in progress", "approximative" moral universals themselves, imbuing them with a particular shape or depth-with greater determinacy or specificity, or with a better-delineated moral scope. This is a view that not only responds to the particularist's objections to Kantianism, but gives them resources that they themselves need to fully accommodate moral salience. Indeed, the Kantian account I've elaborated here of the reflective and moral dimensions of practical reasoning gives us the dual resources of particularism and universalism-resources we would be hard-pressed to find in moral particularism alone. If the Kantian picture of moral reasoning they oppose is one of rote or mechanical application of universals to experience, the picture they risk committing themselves to is one on which moral reasoning is simply a matter of brute perception of "mute sense data or raw particulars" (Kukla, 2002, p. 335).

| RESPONDING TO THE PARTICULARIST
As Herman notes, particularists face "the problem of new saliences", or the problem of how to address the possibility of critiquing one's own moral receptivity: for moral sensitivity "to take itself as the object of its own critical regard" (2007, pp. 26-28). Particularists have no account, in other words, of how a process of something like a critical self-reflection of judgment can get underway. Virtue, on such accounts, is instead largely the result of prior inculcation or socialization. 51 As Jonathan Dancy notes, "for us it is probably too late" to acquire the right kind of moral sensitivity: "As Aristotle held, moral education is the key; for those who are past educating, there is no real remedy" (1993, p. 64).
On the particularist picture, then, there is no distance between education and our considered moral outlook. For the particularist, our capacity to act morally is the result of what we have been taught to believe virtue is and requires of us-and how well we have been taught it. But if our grasp of morality is primarily the product of prior processes of training and education of what content is morally relevant, it seems unclear how we can obtain a critical stance on that training, and with it, on our own constitution as agents in the world. If we are unfortunate enough to receive the wrong training (and it seems inevitable that in a non-ideal social world, some of our training will be wrong), what it would take to morally "consider things aright", McDowell claims, is nothing less than a "conversion", although, as he concedes, "admittedly, it is not straightforwardly obvious how we should think of this" (1998, p. 102). Consequently, an elitist undercurrent threatens the particularist picture of moral rectitude, one that has been easier for the particularist to embrace than to dispel. 52 A reason this remains an open problem for particularism can be traced to the weight it gives to moral receptivity to the exclusion of considering how moral receptivity reinforms our initial normative concepts and principles. Since particularist approaches reject (or at least deemphasize) the role of moral principle altogether, the latter concern is, from the outset, positioned outside their purview. The only remaining option left in their arsenal is the idea of a wholesale transformation of an agent's perception, by analogy to the domain of "religious experience" (McDowell, 1998, p. 102).
In this paper, I have begun to suggest the possibility of a more fine-grained process of self-critique: I have proposed a two-way relation between rules and particulars, a position at odds with the particularist opposition between the particularities of experience and the generalities of a conceptual scheme. To universalize a maxim, I have suggested, requires first that one create this maxim, with a critical awareness of the possibility that it might require revision in the face of newly salient particulars. The incorporation of salient features into a maxim of action, I have claimed with the particularist, is one that holds moral relevance. But, I have argued, this dimension is no less central to the Kantian picture of practical reasoning, and indeed, is one with potential lessons for particularists. It is one important, in other words, not just for defending Kantian universalism against particularism, but for its implications for understanding moral deliberation in its own right.

ACKNOWLEDGMENTS
For insightful input on previous drafts of this paper, I am heavily indebted to Angela Breitenbach, Barbara Herman, "progress in understanding a scheme of concepts", where "as we move and as we look our concepts themselves are changing" (1970, pp. 383, 321-325). Importantly, however, Murdoch's view is ultimately universalist rather than particularist, given the role attributed to "understanding" moral universals (Bagnoli, 2011, pp. 222-223). See also Srinivasan (2018Srinivasan ( ), p. 1415 In calling it more general, I follow Anscombe's characterization of each progressive action description as dependent on "wider circumstances" than the one before it (1957, p. 46). 4 That is, it fails the contradiction in conception test; see O'Neill (1975), pp. 143-173. 5 See also Hegel's empty formalism charge (1807, §134). 6 Kantians have also noted many cases of apparent false negatives and false positives of maxim universalization, such as "I will buy clockwork trains, but not sell them" (O'Neill, 1975, p. 163), save "money by shopping in this year's after-Christmas sales for next year's Christmas presents" (Herman, 1993, p. 138, citing Scanlon ms.), or make "a false promise on Tuesday, August 21, to a person named Hildreth Milton Flitcraft" (Wood, 1999, p. 102). While the problem of false negatives and positives seems to more often arise with highly specific maxims, some intentions are in fact formed at greater levels of specificity, raising problems for interpretations that take maxims to only be formulable at the most general levels of description. On the specificity problem, see O'Neill (1989) and Gressis (2010).
Arendtian approaches which similarly embrace particularism, see Ferrara (1998Ferrara ( , 2008, Lara (2008), and Zerilli (2016). While my focus in what follows is on teleological judgment, I hold that Kantian aesthetic judgment is not straightforwardly particularist, either; in Vaccarino Bremner (2021), I claim that aesthetic judgments, too, lead to the creation or revision of concepts, and thus that aesthetic and teleological judgments are ultimately quite close in function.
13 I take reflective judgment to still have a subsumptive function, albeit not subsumption of particulars under given concepts (for an alternate reading, see Euler, 2018, p. 439): as Kant writes, "the judgment of taste… does not subsume under a concept at all", but instead calls on "the subsumption of the imagination itself… under the condition that the understanding in general advance from intuitions to concepts" (5:286-287). In the case of teleological judgment, the power of judgment gives itself a principle under which it subsumes particulars. Thus, reflective judgment is distinguished from determining judgment by its self-legislative, autonomous function in determining its own procedure of relating particulars to universals (20:225, 5:186), which, I claim below, can help to make sense of the subjective normativity of maxim formation.
14 Nevertheless, by way of the moral dimension's influence on the reflective, we can still construe moral salience as "guided", if indeterminately, by the moral law; below, I explain why I take the guiding role of heuristic principles for teleological judgment to be instructive in this regard.
"practical" sometimes invokes the more restricted sense, and sometimes the more general). Part of my claim, however, is that even "morality" in the strict sense will require the cooperation of reflective judgment in assessing salience.
whereas Kant's full taxonomy in his logic includes "singulars" (picking out exactly one object), "particulars" (of indeterminate extension), and "universals"; see discussion in Lu-Adler (2014). To attempt to keep my discussion as consistent as possible with contemporary discussion, I restrict my usage in what follows to "particulars" and "universals" (where "universals" also include "generals"). I am grateful to Katja Vogt for helping to clarify this point. 18 See 20:220-221, 20:408-409, 20:223-224, cited in Longuenesse (2001), pp. 163-164. See also Kant's claim that "the teleological power of judgment is not a special faculty, but only the reflective power of judgment in general, insofar as it proceeds in accordance with concepts, as is always the case in theoretical cognitions"-or in other words, that reflective judgment is always also operative in theoretical cognition (5:194). 19 There, Kant writes that "before all objective judgments we compare the concepts" implicated in the judgment with one another, and thus that "transcendental reflection… contains the basis for the possibility of objective comparison of representations with one another" (A262/B319, A262/B317); see also Makkreel (2015, pp. 61-66).
20 Some commentators take the paradigmatic instance of reflective judgment, and thus "mere" reflective judgment, to be aesthetic judgment, since in this case no concept is found (e.g., Gorodeisky, 2021, p. 377). I follow Longuenesse (2001, p. 163) and Allison (2001, p. 16), who each refer to both aesthetic and teleological judgments as "mere" reflective judgments, since I take issue with the strictly particularist reading of aesthetic judgment (Vaccarino Bremner, 2021). I am grateful to Ulrich Schlösser for pushing me on this point. 21 Reflective judgment has been related to Kantian morality in accounts including Munzel (1998), Recki (2001), Makkreel (2002, Longuenesse (2005, 237n), Klemme (2013), and Nuzzo (2014). The link in extant interpretations passes through themes not always taken to be central to Kant's moral philosophy, including moral self-orientation (Munzel and Makkreel), political judgment (Nuzzo, following Arendt), the analogy to teleology in the Groundwork (Klemme), moral evaluation of already given or performed actions (Longuenesse), and the role of feeling in Kant's moral psychology (Recki). I locate the relation in the problem of maxim formation, situating reflective judgment at the core of Kant's account of practical deliberation. 22 It is in part for this reason, in my view, that Kant divides his study of reflective judgment, the Critique of Judgment, into two parts on teleology and aesthetics: these are prime case studies for the function of the reflective dimension of cognition more generally, theoretical and practical. 23 I am indebted to Yoon Choi for suggesting this formulation. 24 On the role of purposiveness in unifying the two parts of the Third Critique, see Ginsborg's (2006) proposal of "normative purposiveness"; Zuckert (2007) on "purposiveness without a purpose"; and Allison (2001) on "purposiveness of nature". 25 See also Ginsborg's (2011Ginsborg's ( , 2015 account of "a more fundamental normativity required for intentionality as such" derived from reflective judgment (2015, p. 11). Our two accounts are developed in different directions, but we both can be taken to agree that reflective judgment is somehow implicated in intentional activity.
26 Compare: "a purpose [Zweck] is the object of a concept insofar as the latter is regarded as the cause of the former (the real ground of its possibility)" (Critique of Judgment 5:219-220) and "an end [Zweck] is an object of free choice, the representation of which determines it to an action (by which the object is brought about)" (Metaphysics of Morals 6:384). The main difference seems to be the absence of "action" and "choice", present in the latter definition, from the former, which is consistent with my interpretation of the distinction between mere reflection and morally evaluable action. 27 Due to my restricted scope, I am leaving aside much debated questions in the literature, including how to construe the notion of purposiveness at work in the Critique of Judgment, or even whether there is just one notion at issue; on this, see, for example, Guyer (1993), Allison (2001), Ginsborg (2006), and Zuckert (2007). 28 Where, ultimately, the practical dimension encompasses the moral dimension, in my terminology: that is, the "merely practical" precepts Kant discusses in these sections might not be morally motivated, but are nevertheless subject to moral evaluation; see discussion below. 29 In contradistinction to the understanding's application to nature on the one hand, and reason's application to morals or freedom on the other. Kant devotes §43 to the topic of "art in general", as "production through freedom, that is, through a capacity for choice that grounds its actions in reason" (5:303), an analysis extended in §44 and following (on the beautiful arts)-one of the central topics of the Critique of Aesthetic Judgment. 30 In so doing, the activity of the power of judgment renders conceivable how "the concept of freedom should make the end that is imposed by its laws real in the sensible world", securing a "transition from the manner of thinking in 42 Indeed, Kant suggests that the picture on which practical judgment needs to be supplied with additional rules ultimately admits of a regress, pushing us back to the idea of a self-given exercise of judgment (6:411, compare A133). See also Ginsborg (2015, pp. 151-163). 43 When it comes to imperfect duties, there is not even room for casuistical questions to arise since the actions themselves are left up to free choice; for example, for the imperfect duty of self-perfection: "Which of these natural perfections should take precedence… [is a] matter left for him to choose in accordance with his own rational reflection about what sort of life he would like to lead…. While [this duty] does contain a law for the maxim of actions, it determines nothing about the kind and extent of actions themselves but allows a latitude for free choice" (6:445-446).
44 Doing justice to the degree of particularity or generality of maxims or principles in this respect, as well as their connection to duties, would require a comprehensive account of what Allison terms a "continuum of practical principles" (1990, pp. 93-94); see also Timmermann (2000) and Kant's characterization in the Religion of "the ground of all specific maxims… which is itself a maxim" (6:20).