True Need in Kant

1 The Literature on True Needs

Kant famously holds that wide or imperfect duties, which require that agents adopt obligatory ends, allow for “latitude” [latitudo] (MS, AA 06: 390.06–07). It is up to the agent to discover when and how to further obligatory ends, and potentially even how much an obligatory end is to be furthered.^[2] Whilst it is plausible to leave room for discretion and judgement when it comes to the application of general rules to specific cases, if Kantian ethics lacked the resources to unambiguously adjudicate certain intuitively clear cases, then it would be underdetermined and lack appropriate action-guidingness. For instance, it is plausible that I must save a child who is drowning in the shallow pond if it comes at little or no cost to myself,^[3] that I cannot play a funeral march for the drowning child in order to improve my violin skills (and thus further the obligatory end of self-perfection) instead of saving the child,^[4] and that I do not literally have to give my last shirt to the needy but instead can, to some extent, prioritize my own interests and those of my loved ones.^[5] The latter example also shows that there might be an overdemandingness problem if Kantian ethics fails to provide a plausible rationale for why agents are permitted to take care of their own needs and pursue their personal goals.

Kant obviously cannot simply appeal to consequences of actions (how much is at stake for the child in the pond, etc.) to yield intuitive moral verdicts in these cases. Consequences do not ground duties on Kant’s framework, and he warns that, in general, we should be cautious of consequence-driven reasoning, since calculations pertaining to consequences are always riddled with contingency (GMS, AA 04: 418.01–37, KpV, AA 05: 36.28–37.13, MS, AA 06: 215.24–216.06). However, this does not mean that Kant’s ethical framework is unable to provide a theory of priorities that can guide our application of duties to concrete cases.

In an influential essay on duties of aid, Barbara Herman proposes such a theory. She argues that there are certain ends the pursuit of which we cannot rationally will to forego, because they are “necessary to sustain oneself as a rational being […] Insofar as one has ends at all, one has already willed the continued exercise of one’s agency as a rational being”.^[6] These ends “come from what Kant calls the ‘true needs’ of human agents” (Herman 1984, 586). Herman here makes two central claims. Firstly, certain ends have an elevated status, and it can never be rational for agents to give them up or to endanger their satisfaction for the sake of other, less essential, ends. Secondly, these ends are set by what Kant calls “true needs”, the needs that “must be met if [an agent] is to function (or continue to function) as a rational, end-setting agent” (ibid. 597).^[7]

Due to the elevated status of true needs, helping others meet their true needs is not a simple act of kindness but required by a duty of mutual aid. It is thus not optional for me to save the child from the pond if I am in a position to aid her. The concept of true needs can guide our application of imperfect duties. It helps us to discern between situations that merely call for kindness and situations that confront us with more stringent duties of aid because others’ agency is threatened. Moreover, Herman also argues that true needs constitute a limit to how much the duty of mutual aid can demand of an agent, because it would be self-undermining if agents had to forego the satisfaction of their own true needs and sacrifice their own agency for the sake of others’ true needs.^[8]

Herman assumes that true needs represent standards that are more or less the same for every agent^[9] and that agents cannot rationally decide to ignore these standards. This reading of true needs has its source in John Rawls’s Lectures on the History of Moral Philosophy.^[10] Rawls’s proposal is, however, notably more cautious than many current readings of true needs are. He concedes that the meaning of true needs is “not very exact and cannot be gathered from the few places that Kant uses it” (Rawls 2003, 233). He suggests that true needs are “basic – even universal – needs of human beings conceived as finite rational agents in the order of nature” (ibid. 234). These must be satisfied “[w]hatever an agent’s idea of happiness may be” (ibid. 234). Examples are security, order in society and conditions required to develop rational agency.^[11] Rawls believes that true needs can ground a “maxim of mutual aid” (ibid.). However, he does not tell us how, on a Kantian framework, we can transition from the existence of certain needs to a duty. Moreover, he does not indicate that true needs could limit how much we are to do for others. He does, however, think that these needs are universally shared among human beings, since they represent goods agents must desire insofar as they care about accomplishing any of their ends at all. True needs are thus a standard that is not up to every individual agent to fill in as they please, and these needs concern the conditions of the effective exercise of agency. Both of these ideas are exegetically mistaken, as I will show in section 3.

More recently, a number of Kantians have presented readings of Kant and of Kantian duties to aid, drawing on the notion of true needs along the lines of the Rawls-Herman tradition. Pablo Gilabert, for instance, assumes that there are “basic needs” that have a special status on Kant’s framework, as they concern the conditions of the exercise of autonomous agency.^[12] These needs can ground very stringent positive duties of aid. Likewise, Wim Dubbink emphasizes that dangers to the satisfaction of persons’ true needs create “exceptional” moral circumstances that impact the duties bystanders have.^[13] Alice Pinheiro Walla thinks that true needs represent “fundamental human interests”, i. e., presumably interests common to all humans.^[14] Violetta Igneski assumes, explicitly following Herman (1984), that talk about autonomy is identical to talk about true needs.^[15] Moreover, she argues that autonomy or true needs can limit duty, because duties are founded on autonomy and agents must thus be allowed room to set and pursue personal ends or to exercise their autonomy.^[16]

Finally, Karen Stohr – whilst she is critical of Herman’s emphasis on true needs^[17] – believes that the existence of true needs is the reason why beneficence is an imperfect duty: “It is a wide imperfect duty because I am permitted to make judgments about which sacrifices are required, based on my estimation of my ‘true needs’ in view of my sensibilities” (Stohr 2011, 51). Whilst there is some textual evidence for this reading in P1 (see my section 3), the assumption that true needs are responsible for the latitude of some duties cannot be correct. Imperfect duties admit of latitude because they command that we adopt ends that can be applied to concrete situations in very different ways. If the prerogative to prioritize one’s own true needs were the ground of latitude, as Stohr assumes, then we would not have latitude in cases where we have a number of different options to help others, and none of these options would infringe on the satisfaction of our own true needs. But these situations can still confront us with difficult questions of application, and we do have latitude in these cases.

True needs are frequently understood as representing fundamental human concerns, specifically the need for what is required to maintain and exercise agency or autonomy. In addition, they are considered to be able to ground certain duties of aid, or at least to direct our application of duties to especially important cases and to limit how much an agent can be morally required to do. These are big shoes to fill for a concept that Kant discusses only twice (and each time only briefly). Most likely, the fact that Kant’s discussion of true needs is so brief has led authors to project onto them their very own views about what Kant should have said about agency, duties of aid and the limits thereof.

Notably, there are two approaches to true needs that are more cautious than most of the recent literature on the topic. Firstly, Katja Vogt points out that Kant’s remarks about true needs “remain tentative, and are far less integrated into his ethics than one might wish”,^[18] and that “we can only speculate about how Kant would explain [true needs] were we to push him for clarification” (Vogt 2008, 242). I agree that Kant’s remarks are tentative, but I also think that he says enough about them to cast serious doubt on many recent interpretations. In addition, according to the interpretation I propose, true needs turn out to be well integrated into Kant’s overall ethics and in line with his subjectivist conception of happiness.

Secondly, Martin Sticker and Marcel van Ackeren briefly criticize appeals to true needs (specifically by Vogt 2008, 240 f. and Herman 1984, 597) in order to generate “limits on what agents can be reasonably required to sacrifice, at least for the exercise of imperfect duties” (Sticker/van Ackeren 2018, 424). This discussion is part of their broader aim of showing that duties to self-do not significantly moderate the demandingness of duties to others. In the single paragraph they dedicate to true needs, they point out that true needs should be understood as personal priorities and that Kant still owes us a story for why I can prioritize my own true needs over others’ true needs. I will take up these points in section 3, after presenting my reading of true needs based on a detailed textual analysis. Before I do this, I will take a brief look at the historical context of Kant’s discussion of true needs.

2 The Eighteenth Century on True Needs^[19]

The term “true needs” already appears in eighteenth-century books on law, theology, ethics and economy published before the Metaphysics of Morals.^[20] Sometimes true needs are contrasted with needs of the imagination [Bedürfnisse der Einbildung]. The latter are needs for luxury, the former “real needs” [wirkliche Bedürfnisse],^[21] and “indispensable needs” [unentbehrliche Bedürfnisse] (Pagan 1770, 447), such as food (ibid. 446). Sometimes, however, needs can be imagined and true [“wahre, aber eingebildete Bedürfnisse”].^[22] Importantly, the distinction between true and imaginary needs is not meant to suggest that we should abandon or avoid the latter. True needs are not the only object of rational human striving, since imaginary needs have an important function for commerce and society. They force agents to be useful to each other (Pagan 1770, 444). A good government navigates between a population with too few or too basic needs and an excessive proliferation of imaginary needs that leads to idle luxury and laziness.^[23]

The idea here is presumably that it is relatively easy to satisfy one’s true needs, and it invigorates commerce and social utility if agents develop additional needs. This is a common theme in late eighteenth-century economics literature. The jurist Carl Heinrich von Römer, for instance, claims that a country is fortunate if it can financially benefit from true and imaginary needs, but it is even better to focus on commerce that serves to satisfy “real needs”.^[24] He also distinguishes between “indubitable true needs” [unbezweifelte wahre Bedürfnisse] and dispensable goods considered as true needs (Römer 1971, 110). The latter category shows that authors acknowledge that there can be uncertainty regarding what counts as a true need, and that there are potential variations in what different agents consider their true needs. This, as we will see in the next section, is how Kant thinks about true needs.

Besides literature on national economics, true needs also played a role in discussions of beneficence and charity. The pastor and ethicist Johann Heinrich Schulz reminds his readers that not all needs are “true needs” and that we should learn to distinguish between different types of needs.^[25] True needs are concerned with preserving life and health and with what enables human beings to be useful to others. These needs deserve special attention when we practice beneficence, whereas other needs founded on imagination and passions do not. True needs, according to Schulz, should direct our exercise of duties, but unlike on Herman’s view, for instance, they do not constitute agent-prerogatives to do less for others than one could. Schulz even emphasizes that we should be “impartial” [unpartheyisch] amongst true needs and that only the extent of a true need matters, not whether the need is mine (Schulz 1786, 45).

There is no direct evidence that Kant was familiar with these works, but given the prevalence of the term in academic literature and parallels in his own conception, it is likely that he did come across uses of it. A direct influence on Kant, with which he was certainly familiar, were the works of Jean-Jacques Rousseau.^[26] According to mile, true need [vrai besoin] is natural need (Emile 2: 239), or “physical need” [besoin physique] (Emile 4: 1163), and must be distinguished from the need of imagination [besoin de fantaisie] (Emile 2: 239). The former is the object of amour de soi, which “is content when our true needs are satisfied” (Emile 4: 756). The latter are mere whims or caprices [fantaisie], which result from bad education (Emile 2: 239, 2: 393). They “can only be satisfied with the help of others” (Emile 2: 233) and, in fact, might never be fully satisfied, as they are the object of amour-propre, which “is never satisfied and never can be” (Emile 4: 756). An artificial need is a need “to place oneself above others”, which “inspires in all men a base inclination to harm one another” (OC: III, 175).

Despite Rousseau’s claim that true needs are natural or physical needs, Frederick Neuhouser emphasizes that for Rousseau these needs do not simply represent “a historically fixed or biologically determined set of ‘true’ human needs”.^[27] They are, to some extent, context dependent. For a sufficiently corrupted individual, artificial needs can “degenerate into true needs”, where “being deprived of them [becomes] much more cruel than possessing [the object of the need] was sweet” (OC: III, 168). True needs encompass biological needs but also, for civilized humans, “what is required for them to find satisfying recognition from others (satisfaction of their amour propre)”.^[28] Rousseau stresses the role of the social context for determining what counts as a true need.^[29] Kant, as we will see, develops and even radicalizes this notion when he claims that it is up to the individual to decide what their true needs are. These needs are still biological or physical in the sense that agents cannot decide what their needs are, as this is a matter of their sensuous nature, but agents can designate some of their needs as true needs.

We will also see that Kant departs from Rousseau’s conception in a significant way. Rousseau thinks that, in principle, we can all satisfy our respective true needs, because they do not involve comparison (Emile 4: 756) or the desire to be more esteemed than others. This is not the case for artificial needs, the satisfaction of which involves a rat race with winners and losers. Kant, by contrast, thinks that certain acts of helping others (with their artificial or true needs) can infringe upon the satisfaction of my true needs. Kant does not think that true needs are a distinct type of need the nature of which guarantees that an agent can satisfy them without infringing on others’ satisfaction of their true needs. Satisfying my true needs might require appropriating resources at the expense of other agents’ true needs, or maybe even “gain[ing] worth in the opinion of others” and “acquir[ing] superiority for oneself over others” [sich in der Meinung anderer einen Werth zu verschaffen … sie sich über Andere zu erwerben] (RGV, AA 06: 27.07–12). This, of course, raises the question of why these needs deserve special moral attention.

3 Kant on True Need

The extensive reception true needs have had in the recent literature is astonishing, since the concept of true need comes up in only two passages in Kant’s published writings.^[30] This in itself suggests that true needs are not a central concept for him. It is particularly surprising that Kant hardly mentions them in his Anthropology.^[31] If true needs represented the most fundamental needs of all human beings, then we would expect them to be a prominent element (or at least an element) of his anthropology.

Let us now look at the two passages in which Kant himself discusses true needs, the first of which is from the Introduction to the Doctrine of Virtue:

It is significant that Kant nowhere suggests in P1 that true needs could ground a duty of beneficence, let alone of mutual aid. However, we can certainly gather from P1 why Kantians would think that true needs could limit an agent’s duty to others. After all, Kant here claims that “a maxim of promoting others’ happiness at the sacrifice of one’s own happiness, one’s true needs, would conflict with itself if it were made a universal law”. Kant here seems to suggest that a maxim of self-sacrifice, according to which I promote others’ happiness at the expense of my own true needs, would be morally impermissible according to the Categorical Imperative’s Universal Law Formula (FUL) (GMS, AA 04: 421.07–08). This would be to some extent parallel to one of the Groundwork I duties, which illustrate common moral cognition: “[t]o secure one’s own happiness is one’s duty (at least indirectly); for lack of contentment with one’s condition, in the trouble of many worries and amidst unsatisfied needs, could easily become a great temptation to transgress one’s duty” (GMS, AA 04: 399.03–07).^[33] However, Kant does not mention true needs in the Groundwork when he proposes that there is a minimal standard of wellbeing that an agent is to preserve on moral grounds. Likewise, Kant does not mention true needs when he maintains in the Metaphysics of Morals that seeking prosperity can be a duty “indirectly” insofar as this helps to “ward off poverty”, which is “a great temptation to vice” (MS, AA 06: 388.26–30).^[34]

Kant’s statement in P1, which suggests that a maxim of sacrificing one’s own true needs for others’ happiness is impermissible, has inspired commentators to discuss whether and to what extent duties to self limit duties to others.^[35] However, we should bear in mind that much more would need to be said to construct a proper argument or duty against sacrificing satisfaction of one’s true needs. Firstly, Kant claims that a maxim of sacrificing the satisfaction of one’s true needs “would conflict with itself if it were made a universal law”. This is different from how he usually phrases and applies the FUL, which requires that one be able to will simultaneously and without self-contradiction of the will that a maxim be one’s own and a universal law.^[36] Simultaneity is not mentioned here, and thus the case would at least be somewhat underdescribed compared to many other applications of the FUL.

Secondly, it seems that any attempt to spell out the argument would hinge on the assumption that true needs represent universally shared conditions of agency – an assumption that I will reject below. Moreover, even if we grant this assumption, the argument is still difficult to spell out as the exact formulation of the maxim that would conflict with itself if it were made a universal law is left open. It is unclear whether the FUL is here supposed to rule out sometimes sacrificing my own true needs to make others happy or only never paying attention to my own true needs and acting maximally altruistically, or something in between. Depending on how we formulate the maxim here, true needs restrict our imperfect duties to others substantially or only marginally.

Thirdly, Kant explicitly indicates that a maxim of sacrificing one’s true needs for others’ happiness would “conflict with itself if it were made a universal law”. This is most naturally read as referring to a contradiction in conception, not merely in willing. Sacrificing satisfaction of one’s true needs for the sake of others’ happiness would thus be strictly prohibited qua a perfect duty.^[37] This would be a much stronger duty than what we find in Groundwork I – as well as what Kant alludes to elsewhere in the Metaphysics of Morals (MS, AA 06: 388.26–30) – where Kant presents an indirect duty to preserve a baseline of happiness.^[38] Thus, Kant would be departing significantly from claims made elsewhere. Moreover, we might worry that Kant would be declaring acts of supererogatory sacrifice that we might deem heroic and admirable (albeit not morally required) to be a breach of the most stringent kind of duty, akin to false promising.^[39]

I therefore think it is best not to assume that in P1 Kant intends to make a substantive change to his normative system by introducing agent-prerogatives that limit how much we must do for others, or even how much we are permitted to do for them. If he had intended this, he would surely have spent more than a single sentence on the exact maxim under scrutiny and the specific argument and its ramifications. As a matter of fact, the conclusion we should draw from this sentence is made explicit in the very next sentence, which starts with “Hence” [Also], making it clear that this sentence presents the implications of the previous one: The duty to sacrifice part of one’s welfare to others “is only a wide one; the duty has in it a latitude for doing more or less”. It was not Kant’s intention to establish that it would be morally wrong to sacrifice the satisfaction of true need. His point was merely that beneficence is a wide duty, which is relatively uncontroversial on Kant’s framework anyway. In addition, that “no specific limits can be assigned to what should be done” for others strongly indicates that true needs do not represent a list of universally shared conditions of agency. Such conditions could, after all, impose specific limits to what should be done for others, because they might prescribe specific goods or functions that may never be compromised if an individual is to function as a moral agent.

More generally, P1 is significant for a correct understanding of true needs because Kant here identifies true needs with one’s own happiness [“seiner eigenen Glückseligkeit (seiner wahren Bedürfnisse)”], not with anything pertaining to agency or autonomy. Personal happiness and maintaining agency can come apart. There are many activities, such as alcohol and drug consumption, that make us happy, because they are pleasurable, but that can also be detrimental to our agency.^[40] It is important, though, that Kant does not think that happiness and satisfaction of true needs are identical. After all, the very first thing he maintains in P1 is that morality can require that I “sacrifice a part of my welfare” or happiness. Limits to how much I must do for others, if they exist at all, only come into play once this sacrifice impacts the satisfaction of my true needs. True needs are thus a subclass of all the needs the satisfaction of which yields happiness or welfare.

That true needs are a matter of happiness, not of maintaining agency, also becomes apparent when Kant emphasizes that “[h]ow far [duty] should extend depends, in large part, on what each person’s true needs are in view of his sensibilities, and it must be left to each to determine [bestimmen] this for himself” (my emph.). True needs reflect a subjective or agent-relative notion of what matters to agents and what they see as most conducive to their personal happiness.^[41] This has, if at all, only a contingent connection to maintaining agency. We can also see how Kant develops or radicalizes the Rousseauian idea that true needs are context dependent. For Kant, this context is what each individual determines regarding what matters to her.

Admittedly, “bestimmen”/“determine” is ambiguous both in German and English. It can mean both identifying something that exists independently of the agent and making something the case. That Kant here characterizes true needs as a matter of happiness, alongside his subjectivist conception of happiness (see below), strongly suggests the latter meaning. This is also the meaning assumed by Mary Gregor when she translates “bestimmen” as “decide” – a term that, whilst correct here, does not preserve the ambiguity.^[42]

That true needs are not connected to agency in any straightforward way is unsurprising, given that, in the second Critique, Kant explicitly states that all “need is directed to the matter of [someone’s] faculty of desire, that is, something related to a subjective feeling of pleasure or displeasure underlying it by which is determined what he needs in order to be satisfied with his condition” (KpV, AA 05: 25.17–20).^[43] Need is here defined as something subjective that correlates with pleasure or displeasure and satisfaction, not with any objective goods that agents ought to strive for or preserve regardless of their personal preferences. Kant’s idea that the content of true needs is up to the individual to determine fits organically into his overall subjectivist conception of needs. In the Groundwork, Kant even stresses that inclinations are the “sources of need” [Quellen des Bedürfnisses] (GMS, AA 04: 428.14–15, see also GMS, AA 04: 413 fn.). This is striking as we would normally assume that many of our inclinations are based on our needs. That Kant thinks it is the other way around strongly supports a subjectivist reading of needs across the board. There is no set of basic human needs that we all share and that produce desires or inclinations stable across all human agents; rather, each individual’s needs are dependent on their inclinations.^[44]

In fact, Kant emphasizes that happiness for human beings is not a matter of just following one’s instincts or inclinations but rather “a mere idea of a state”, and a human being “outlines this idea himself” [eine bloße Idee eines Zustandes […] Er entwirft sie sich selbst.] (KU, AA 05: 430.08–12). Rational beings like us cannot simply pursue happiness by satisfying whatever our strongest inclination is at the moment. Rather, I have to construct or outline a conception of what would yield “a maximum of well-being, in my present and every future condition” [ein Maximum des Wohlbefindens, in meinem gegenwärtigen und jedem zukünftigen Zustande] (GMS, AA 04: 418.08–09).^[45] Designating certain needs as true is presumably part of this construction. They are the needs that I prioritize over other needs. Whilst it is typically not up to us to decide which needs affect us, true needs can be understood as those of our needs that we endorse in the sense that we think it particularly important that they be met.

In terminology made prominent by Harry Frankfurt, we can explain this endorsement as a second-order desire that elevates certain first-order desires: we want to have them and want to be moved to action by them rather than by other desires that we deem less important or identify with less.^[46] True needs are left to the individual to determine in the sense that they do not simply represent one’s strongest desire(s), the strength of which is not up to agents to determine, but rather those desires the fulfilment of which is particularly central to one’s conception of happiness. True needs pick out those of an agent’s desires, inclinations or needs, whatever they are, that an agent not only has and feels the force of, but also prioritizes. However, this reflective endorsement does not mean that true needs would stop being needs or that they would be endowed with a special moral status.

My interpretation that true needs do not represent a list of universal human concerns tied to agency is also supported by a subtle point in the German, which is mistranslated in the Gregor (1996, 524) translation from which I quoted above. This mistake might have misled Kantians who rely on translations. Kant’s first mention of true need in P1 and the two times the term is mentioned in P2 (see below) are in the singular [wahres Bedürfniß], whereas Gregor translates all of these as the plural true needs.^[47] The plural suggests that Kant has in mind a defined list of basic needs, such as food, shelter and social order.^[48] The singular, by contrast, indicates that he sees true need as an abstract structure that every agent is left to specify for herself, and this is the conception that fits best with Kant’s text and his overall conception of need.

Finally, we should bear in mind that in P1 Kant argues against a maxim of promoting others’ happiness at the expense of our own true need. He does not address maxims of sacrificing my own true need to help others meet their true need. This will become important in what follows.

Let us turn to the second passage, which occurs in the context of Kant’s discussion of the vice of avarice as a violation of a duty to the self:

Kant emphasizes that neglecting one’s true need is morally criticizable.^[50] Once more, he does not mention agency and in fact claims that we leave true need unsatisfied if we restrict our “enjoyment of the means to good living” [Genusses der Mittel zum Wohlleben]. Kant specifies that the purpose of the means to well-living is “enjoyment” [Genuss], something that leads to a positive mental state, such as pleasure or satisfaction. This again shows that true need is a matter of personal happiness as it results in positive mental states and aims at overall pleasure or satisfaction.

Moreover, there is a translation error in P2. Gregor (1996, 555) translates “das Maß des wahren eigenen Bedürfnisses” simply as “own true needs”. This again translates the German singular [wahren […] Bedürfnisses] as a plural, suggesting that there might be a defined list of true needs that an agent must always satisfy. In addition, Kant here talks about the “measure” [Maß] of true need. He thinks that if we fall below the measure or standard imposed by our true need, then this will damage our well-being to a morally criticizable extent and constitute the vice of avarice. He intends to leave open what true need consists of. What is universal is merely that true need imposes a measure or standard for how much agents are required to attend to their own needs.

This reading of P2 is not only in line with the conception of true need in P1 but also fits organically with Kant’s overall conception of happiness and non-moral motivation. In the first Critique, Kant famously defines happiness as “the satisfaction of all of our inclinations (extensive, with regard to their manifoldness, as well as intensive, with regard to degree, and also protensive, with regard to duration)” [Glückseligkeit ist die Befriedigung aller unserer Neigungen (so wohl extensive, der Manigfaltigkeit derselben, als intensive, dem Grade, und auch protensive, der Dauer nach)] (KrV, A/B: 806/834). In the second Critique, Kant claims that “[h]appiness is the state of a rational being in the world in the whole of whose existence everything goes according to his wish and will” [Glückseligkeit ist der Zustand eines vernünftigen Wesens in der Welt, dem es im Ganzen seiner Existenz alles nach Wunsch und Willen geht] (KpV, AA 05: 124.21–23).^[51] Happiness consists of the complete satisfaction of all of one’s inclinations and the pleasure one takes in such satisfaction. In his definitions and explanations of happiness, Kant does not mention true need or, for that matter, any needs that would have a special function for agency. Kant’s conception of happiness and his moral psychology of the non-moral realm is a subjectivist one, according to which there is no universal or a priori law of happiness. Rather, what happiness consists of is relative to each individual agent, and to their respective desires, inclinations and sources of pleasure (KpV, AA 05: 25.20–24). As a matter of fact, Kant is standardly understood as advocating a hedonistic conception of happiness^[52] and is frequently criticized for this supposedly simplistic conception of happiness.^[53]

It is difficult to see how, on such a subjectivist conception, true need could be anything but needs that an agent personally prioritizes in her pursuit of happiness because they promise more pleasure or because the agent cares more about their satisfaction. Moreover, Kant emphasizes that “[a]ll material practical principles as such are, without exception, of one and the same kind and come under the general principle of self-love or one’s own happiness” [Alle materiale praktische Principien sind, als solche, insgesammt von einer und derselben Art und gehören unter das allgemeine Princip der Selbstliebe oder eigenen Glückseligkeit] (KpV, AA 05: 22.06–08). True need must be a matter of personal happiness, since everything that is not a formal principle is subsumed under the general principle of one’s own happiness. According to P2, true need supposedly plays a role in moral self-assessment. However, if it could ground duties or limits thereof, then this would allow the principle of self-love to impact moral deliberation, and this would be a stark departure from Kant’s formal framework. That the drowning child’s true need is at stake must simply mean that, amongst other things, the pursuit of those ends that are particularly important to the child is at stake. It is plausible to assume that true needs encompass life and health, but it is ultimately up to the child to decide what they consist of.

One might respond that it is unfair to Herman and others to assume that they intend true need to represent non-moral concerns, as I have been implicitly assuming they do. If true need pertains to the conditions of rational agency, we should conceive of them as moral concerns from the get-go. The pleasure we get from satisfying them (if any) is beside the point, as satisfying them is morally required. That would render Kant’s subjectivist conception of happiness a moot point, since true need would not fall under material principles.

However, if true need is supposed to play a role in explaining why we have certain duties in the first place or why we have more stringent obligations in certain situations and can give our own needs special consideration in other cases, then we would need to explain how it is possible that there are moral goods or moral concerns, such as true need, independently of or preceding the Categorical Imperative. After all, at least on a standard Kantian account, moral normativity is the outcome of universalization procedures or considerations pertaining to treating ourselves and others as ends in themselves. Clearly, Herman and others do not think that Kantians can just point to the immediate moral status of certain needs to establish a duty of aid or prerogatives to care for oneself. Yet, if their arguments are supposed to rest on moral concerns that impact how Kantians are to establish duties, guide the application of duty or limit duty, they face a steep explanatory burden to account for the moral status of true need. Most likely, they would point to true need’s role in agency. However, I showed that Kant himself does not make this connection, and thus that the sense (if any) in which true need represents a moral concern is an unresolved issue.

In addition, we should bear in mind that any attempt to carve out agent-prerogatives or limit duty on a Kantian framework faces the difficulty that Kant clearly thinks that morality, in the form of perfect duties, can ask us to disregard our true need and sacrifice everything (see e. g. KpV, AA 05: 30). Preserving one’s agency, if a concern for Kant at all, is not an absolute constraint. Any Kantian attempt to limit some duties must be able to explain why other duties remain unlimited and absolute.^[54]

Moreover, there is a general philosophical problem for any appeal to true need in order to establish limits to how much we must do for others. Even if true need had a moral status, this would leave unexplained why we can prioritize our own true need over those of others.^[55] The existence of true need can at most explain why we should prioritize the satisfaction of these needs over the satisfaction of trivial or mundane needs. It does not establish that and why we cannot be obligated to satisfy others’ true need even at the expense of our own true need if this would mean that, overall, more true need is met. That we could be required to meet true need impartially is not a moot position as such a conception was, for instance, advocated by Schulz (see my sec.2), who stressed that we ought not to be partial towards our own true need.

Recall that in P1 Kant dismisses a maxim of furthering others’ happiness at the expense of my own true need. However, he is silent about a maxim of furthering satisfaction of others’ true need at the expense of my own true need. Such a maxim would be in line with granting true need a special moral status. This shows that the existence of true need, on its own, cannot limit duties. We additionally require a rationale for why we can give preferential treatment to ourselves, to our true need. An explanation for why an agent would be justified in prioritizing herself is, however, not provided by Kant’s claims about true need.^[56]

This, of course, does not mean that Kant cannot ground agent-prerogatives at all.^[57] There might be elements of Kant’s theory that afford such a grounding: Duties to develop one’s talents (Vogt 2008), permissible partiality (Baron 2016), the notion that we do not have to be maximally beneficent in each case (van Ackeren/Sticker 2018), or that it is not morally wrong (not a case of “culpability” [Verschuldung] MS, AA 06: 390.20) to fall short of being as beneficent as one could (Formosa, Sticker 2019). We should bear in mind that the question of whether and, if so, why an agent can give herself (and potentially loved ones) special moral regard is a substantive problem for ethicists who seek to avoid overdemandingness problems. Kant and Kantians cannot simply solve this question by introducing a special kind of need. Instead, substantive arguments for certain forms of partiality, a plausible conception of latitude, duties to develop one’s talents, etc., are required.

4 What to make of True Need

True need is not the powerful tool to ground, guide and limit duty that many in the literature make it out to be. It represents subjective preferences that differ across agents and possibly also across one and the same agent at different times. The literature pays insufficient attention to how Kant characterizes true need and does not address how a conception of universal human needs would fit with Kant’s conception of our sensuous self, e. g. his conception of needs and happiness.

There is a more general point looming in the background, namely the question of the role of agency in Kant’s ethics. Kant’s ethics is often understood as according supreme value to (autonomous) agency, and much emphasis is placed on supposed requirements to preserve and develop one’s agency. It is striking that Kant himself never uses a German equivalent for the term “agency”,^[58] and evidence for his supposed focus on agency is circumstantial and often drawn from his general emphasis on freedom and autonomy and remarks about the duty of self-perfection (see for instance MS, AA 06: 427.12–33). Moreover, Kant’s talk of “[t]he hypothetical imperative” [Der hypothetische Imperativ] (GMS, AA 04: 414.32) and his idea that happiness is an “ideal […] of the imagination” (MS, AA 06: 418.35–36), which requires that agents systematize their inclinations, might suggest a unified structure to the non-moral pursuits of finite agents (see also GMS, AA 04: 399.08–09, KpV, AA 05: 73.09–10, RGV, AA 06: 58.01–07). This is sometimes read as evidence for Kant’s interest in a theory of agency across the board (moral as well as non-moral). Importantly, however, these remarks are concerned with happiness, not agency, and it would be a mistake to run the two together, since certain things can make me happy but undermine my agency (e. g. drug consumption). Anything Kant says about the structure of the pursuit of happiness can hardly be evidence for Kant’s supposed overall concern with a unifying structure of action that we would call “agency”, let alone for the notion that this structure has a moral status and that maintaining it imposes obligations on agents.

I do not mean to deny that “agency” and other anachronistic terms can be useful hermeneutical tools for understanding the distinguishing features and potential of Kant’s ethics or for explaining his concerns in contemporary terms. Yet, in the case of true need, the focus on (preserving) agency has misled the literature. We can only understand P1 and P2 correctly if we carefully distinguish between happiness and morality (as Kant never tired of emphasizing that we must) and bear in mind that needs are a matter of our sensuous and animal side and that there is no straightforward way to bring needs to bear on our moral obligations, even if agents greatly care about satisfying certain needs.

Thus far, my argument has been largely negative, correcting misreadings and arguing that true need is of a more limited function than many Kantians assume. One might respond that introducing a notion of true need along the lines of Rawls and Herman into Kant’s framework still makes sense as a way to fix problems that Kant himself did not address adequately. However, as I hope to have shown, doing so would not just mean adding new material or concepts to Kant’s theory. It would have serious implications given that admitting certain needs that come with a special normative status would require Kant to reconsider his conception of our sensuous selves, and especially of happiness. In addition, even if we allow for certain aspects of the sensuous world to be true in a sense that gives them a special normative status, this does not yet ground a prerogative for me to care especially about my true need.

More importantly, I think Kant’s conception of true need, as I presented it, is philosophically attractive. Notably, Kant does not stipulate that there are basic or universal human needs that must be satisfied “[w]hatever an agent’s idea of happiness may be” (Rawls 2003, 234). Rather, it is up to each individual to determine their true need. Kant’s conception is anti-paternalistic, as it gives individuals maximal discretion to prioritize their needs. This is an advantage as it frees Kant from the challenge of coming up with a list of true human needs stable across cultures, epochs and individual agents.^[59] Kant’s conception takes seriously the idea that it is up to each individual agent to fill in her own conception of happiness.

Granted, some agents might have very unusual needs, maybe even needs that are detrimental to their own overall wellbeing. Yet these needs might nonetheless be very important to them, and they might prioritize them over other needs that are more central to their physical and mental well-being. Think of the man who spends his life counting blades of grass. This might turn out to be one of his true needs.^[60] Ought we to help that man by, for instance, planting more grass for him to count? Moreover, does that person have a prerogative never to help anyone if helping were to interfere with his counting?

There is one very important aspect of Kant’s ethics that we should bear in mind and that can help us to apply our imperfect duties to others and avoid overdemandingness problems. Unless agents have a right to our help (because we promised our help, have contractual or role obligations to help, or are culpably responsible for their predicaments), we have discretion over whether to help them to further the ends they have adopted and the needs they take to be their true ones.^[61] As helpers, we can choose the needs of others that we find important and that we think we ought to help them with. Presumably, we will often choose needs that others consider part of their true need, as these will frequently be concerns that strike (almost) everyone as important (health, safety, prospects for the future). However, the mere fact that someone considers something their true need does not necessitate our help or determine our judgment. We are free to make our own assessment of a situation and to tell the grass counter that he may count all the grass in the world, that we wish him the best of luck with this – after all, our capacity for wishing well is unlimited (MS, AA 06: 452.5–9) – and that he can come to us for shelter and a hot meal, but that we will not be planting a front lawn for him to count.^[62]

The discretion to assess others’ needs for ourselves and to act based on this assessment will have to be balanced with another very valuable Kantian insight, namely the notion that we should not humiliate those we are helping (MS, AA 06: 448.22–449.02, 453.17–33). We therefore have to be responsive to the priorities of others and careful when we disagree with them about what is good for them. This difficulty notwithstanding, I find it plausible that, while every agent can pick their own priorities, this prerogative also extends to potential helpers to determine what matters when they exercise beneficence.

The understanding of our duty to help others that I have proposed is admittedly more straightforwardly an answer to the overdemandingness problem than to the underdetermination problem. We do not absolutely have to help others unless we owe them help, but my reading leaves it open to agents to refrain from saving others in acute need because they do not deem what is at stake in the situation important. What we have to do for others is to some extent underdetermined and up to us to decide, and it may well be that the perspectives of the potential helper and the person who needs help sometimes diverge. However, if we assume that an agent is committed to beneficence and is earnest about its exercise, then this will usually not be a problem, whereas an agent who tries to rationalize and wiggle out of their moral obligations will be a problem for any account of morality.

Finally, I have argued that it is difficult to see how true need could constitute agent-prerogatives to limit duty. This is to be welcomed, as it allows us to tell the grass counter that although we are glad he has found his calling in life, that does not get him off the hook morally. Kant’s conception of true need does not provide a rationale for prioritizing grass counting over beneficence, let alone over perfect duty. That grass counting is the counter’s true need does not make this activity special in any ethically interesting sense, no matter how dear it is to the counter.