Proponents of Kantian ethics have paid relatively little attention to the question of whether – and if so, to what extent – Kantian ethics is sufficiently context-sensitive to leave room for morally permissible, different cultural ways of life. An exception to this is Katrin Flikschuh, who proposes an interpretation of Kant’s ethics which she refers to as Kantian contextualism. Contrary to standard ‘universalist’ interpretations, Kantian contextualism maintains that the Categorical Imperative can give rise to ‘contextually different substantive principles’, and hence to different moralities for differently situated persons. Flikschuh takes Kantian contextualism, unlike Kantian universalism as it is standardly conceived, to be sufficiently sensitive to different ways of cultural life. In this paper, I provide several arguments for why Flikschuh’s Kantian contextualism should be rejected. Moreover, I will argue that Kantian universalism leaves ample room for cultural pluralism without having to forfeit the idea that the Categorical Imperative puts categorical, universal constraints on action. Keywords: Kantian universalism; Kantian contextualism; cultural pluralism; Katrin Flikschuh.
I call Kantian universalism the family of Kant-inspired philosophical views according to which universal, normative action principles can be derived from a supreme principle of morality that is grounded in practical reason: the Categorical Imperative (CI). Proponents of Kantian ethics have paid relatively little attention to the question of whether – and if so, to what extent – the theory is sufficiently context-sensitive to leave room for morally permissible, different cultural ways of life. This is somewhat surprising, given the amount of literature that depicts Kant’s ethics as one that is parochial, Eurocentric or in some other capacity incapable of addressing the moral concerns or perspectives of certain (marginalized) groups of people (e.g. Velleman 2015; Allen 2013; Benhabib 1992; Mills 2018).
One of the few Kantian ethicists who has positively engaged with
Kant’s ethics in the context of cultural pluralism is Katrin
Flikschuh. In her 2017 book
In this paper, I take issue with Flikschuh’s suggestion that
“substantive” moral universalism in Kantian ethics would render it
insufficiently sensitive to cultural context (Flikschuh 2017, 45). I
will argue,
The paper is structured as follows. In section 1, I present a reconstruction of Flikschuh’s Kantian contextualism. In section 2, I discuss several problems for Kantian contextualism’s potential as a coherent and substantial moral theory. In section 3, I suggest that Kantian universalism, unlike Flikschuh’s Kantian contextualism, is capable of drawing a clear moral line that may not be crossed by rational beings whilst also preserving an ample amount of contextual flexibility.
According to Flikschuh, many philosophers in the global justice
debate (“global theorists”) tend to “conflate global reasoning with
globalizing particular, domestically favoured moral and political
principles” (Flikschuh 2017, ix-x). By this Flikschuh means that
rather than seriously engaging with substantively different, foreign
points of view on moral issues, global theorists, including those who
subscribe to what Flikschuh calls “current normative Kantianism”, have
a tendency to take certain “commonsense” principles for granted
(Flikschuh 2017, 3; 40). These include a strong commitment to moral
individualism and the freedom and equality of persons (Flikschuh 2017,
2). The idea of such principles being “commonsense”, says Flikschuh,
testifies to an implicit presumption that the “liberal way of thought
and life” is in a sense furthest advanced in the moral span of time,
which gives rise to the concern that global theorists are not taking
other, “illiberal” ways of thought and life as seriously as they
should (especially when people with other forms of thought and life
are the subject of their
theorizing).
The overarching aim of Flikschuh’s project is to provide a Kantian
basis for the idea that foreign, “illiberal” moralities need not be
rationally wanting in terms of moral justification, because of which
global theorists ought to engage with these moralities on equal terms.
On Flikschuh’s view, we can recognize that there is a rational basis
for a morality that is substantially different from our own without
believing that there is reason
Flikschuh is optimistic about our capacity for such recognition.
Her optimism stems from her agreement with Kant’s contention that as
self-reflexive, rational beings, we can become aware of the finitude
of our cognitive capacities (Flikschuh 2017, 126-127). More precisely,
we can become aware of the fact that human beings, in addition to a
universally shared “human standpoint”, need a particular “local
standpoint” to orient themselves in the world (Flikschuh 2017,
104-106). As for the human standpoint, Flikschuh joins Kant in
maintaining that human beings equally represent the world to
themselves through the categories of the understanding (e.g.,
quantity, quality relation, modality, etc. (KrV, B129-B169)). What she
adds to this picture is the idea that “differently situated human
persons will respond to their particular contexts by developing
context-specific concepts and practices” (Flikschuh 2017, 104). This
is what Flikschuh means by the local standpoint. “Apart from the
categories of the understanding”, Flikschuh says, “there is […]
nothing fixed or antecedently determined about [the] process of
empirical concept formation” (Flikschuh 2017, 33). Substantively
differently situated people orient themselves by different conceptual
frameworks, leading to different ways of life. People from cultures
that do not have the practice (let alone the concept) of promising or
contracting, for instance, will not make sense of their actions
through these concepts (cf. Flikschuh 2018,
570-572).
Flikschuh maintains that when we come to understand that, globally,
there is great variety among local perspectives, we can come to
understand that people with substantively different local perspectives
could justifiably adopt action principles that we cannot from our own
local perspective (Flikschuh 2017, 104-105). For instance, a Western
European person could come to understand that a village elder of the
Kenyan Kikuyu clan, as a result of her being part of a certain locally
realized way of life, may have reason to practice female genital
cutting (e.g., because of a deeply held belief that this is how girls
are initiated into womanhood), whereas the Kikuyu elder may come to
understand that the Western European person, being part of a locally
realized way of life with a long tradition of “Enlightenment
thinking”, may be justified in regarding this practice as a violation
of bodily integrity. Hence, not only can there be deep
A natural question that arises at this point is how Flikschuh’s endorsement of faultless moral disagreement can be squared with Kant’s moral philosophy. This is where Flikschuh’s “contextualist” interpretation of Kant’s ethics comes in.
According to Flikschuh’s Kantian contextualism, the CI is meant to
give us the universal form of moral judgment. However, this does
[C]urrent readings of Kant’s categorical imperative […] typically
seek to derive generally valid principles of action from a
categorical imperative test that asks each individual agent to
reflect on the moral purity of
Flikschuh puts this point even more strongly in her 2018 article
“Kant’s Contextualism”: “[t]he point of the universalizability test is
not to generate universally valid principles of action but to judge
the moral purity of one’s maxim” (Flikschuh 2018,
562).
The main difference between Kantian universalism and Kantian
contextualism, it seems, is that according to Kantian universalism,
the
Kantian contextualism’s subjective universalism says that our moral
judgments,
Initially, when you stepped off the plane, you thought that polygamy is where you draw the line. It’s just wrong, you thought. But now you find yourself having second thoughts […] [Y]ou come to see that it works, that it has advantages and disadvantages – for males and females alike – that it, too, is part of the normalcy around here. You begin to think that had you not been brought up on a diet of notions of romantic love, but were instead a member of complex clan and extended family relations with rather different patterns of emotional investment and loyalty, polygamy might strike you as a sensible way of going about what truly matters in family life. (Flikschuh 2017, 119-120).
On Flikschuh’s account, the same mode of reasoning extends to the practice of female genital cutting. Provided that we sufficiently grasp the contextual background and local perspective of pro-female genital cutting members of the Kenyan Kikuyu clan, for example, we should come to understand how it is possible for members of this community to rationally endorse the practice of female genital cutting (Flikschuh 2017, 125).
Flikschuh maintains that it is only when we thoroughly put
ourselves in the shoes of foreigners with substantively different
moralities that we acquire the conceptual means to orient ourselves in
the global justice debate in a way that goes beyond our own moral
framework, which we come to regard as one contingent set of moral
values among others. This is so even if a particular moral framework
(e.g. “liberalism”) remains the only viable reason-giving option
What emerges is a highly innovative take on Kantian ethics that
differs in important respects from “standard” universalist
interpretations. Like Kantian universalism, Kantian contextualism
takes the CI to be the supreme principle of morality. According to
both, the CI demands that you act only on “universalizable” maxims.
However, where Kantian universalism takes this to imply that human
beings are only permitted to act on maxims that
According to Flikschuh, “those who look for a universalism in moral
substance overlook the universalism in the form of willing in Kant’s
ethics” (Flikschuh 2018, 573). In contrast to this, I will argue that
we
Insofar as Kantian contextualism is meant as an
A pure maxim, on Flikschuh’s account, is a maxim of a good will
(Flikschuh 2018, 562). It is true that Kant exclusively considers a
good will to have moral worth (AA 4:393), but it does not follow
that the CI test must therefore be designed to reach a judgment on
the
[N]o certain example can be cited of the disposition to act
from pure duty; that, though much may be done
While Kant is clearly pessimistic about our insight into the
“purity” of our maxims, he is rather
By stating that the best we can hope for on Kant’s account is
that we judge and act from purity of will, Flikschuh obscures Kant’s
contention that when we test whether maxims can at the same time be
willed as universal laws, we should arrive at conclusions that we
must accept on pain of rational self-contradiction. Her conception
of the CI test as a purity test for maxims seems to replace Kant’s
“contradiction criterion” with a “sincerity criterion”. That is,
where for Kant the CI is meant to provide rational insight into the
categorical (im) permissibility of a proposed maxim by showing that
a contradiction in the will does or does not
In sum,
Recall that according to Kantian contextualism, we can come to
Suppose I can imagine myself judging that I have good reason to rob banks. One does not need to be a Kantian to see that it would be a stretch to hold that my capacity to empathetically put myself in the shoes of a bank robber renders bank robbing morally permissible, or potentially morally permissible, from certain perspectives. Yet if we are to follow the recognition principle, it seems that this is exactly the conclusion we should arrive at. After all, on Kantian contextualism’s account, we can also come to understand that a person could have legitimate reasons to perform female genital cutting precisely by virtue of our capacity to imagine ourselves, from that person’s local standpoint, sincerely judging that they have reason to perform female genital cutting.
This is a problem, especially because it is sometimes quite easy
to empathetically put ourselves in the shoes of “bad guys”. The best
movie villains are those who empathetically resonate with us on some
level. Suppose that I can perfectly imagine myself sincerely
believing that banks are little more than the cornerstones of the
capitalistic oppression of the working class. It seems that on the
basis of this sincerely held belief, there are quite a few Robin
Hood-like routes available to me (assuming I care about the working
class) to arrive at the sincere judgment that it is permissible, or
perhaps even a duty, for all rational beings to rob banks and
distribute the wealth among the working class. Sure, I may be
capable of recognizing that there are others for whom it is rational
to care about the bank’s continued existence; it just so turns out
that from
Although Kantian contextualism aspires to justify belief in a
supreme principle of morality that has “reasoned authority” in our
moral deliberations (Flikschuh 2018, 558), its conception of the CI
seems to be largely grounded not in reason but in our capacity for
There is, however, a further problem for Kantian contextualism in
addition to its being overly permissive, namely that the combination
of its insistence on the CI as the universal form of moral judgment
and its commitment to the recognition principle renders Kantian
contextualism
Suppose we are concerned about the moral permissibility (or
“purity”) of a maxim we are considering adopting. As good Kantian
contextualists, our method of evaluation is to ask ourselves whether
“in [our] judgment [our] proposed maxim is serviceable as a
Now, what is it that the Kantian contextualist
In short: Kantian contextualism implores us to evaluate the “purity” of our maxims on the basis of their supposed universalizability (CI) whilst simultaneously imploring us to keep in mind that these maxims need not be “truly” universalizable (recognition principle). This is theoretically incoherent.
Perhaps the issues discussed thus far will turn out to be trivial
if Kantian contextualism can still succeed in what it set out to do:
to provide us with an inclusive Kantian theoretical method for
normatively orienting ourselves in the global justice debate.
Suppose that Flikschuh is right in that “our own liberal principles”
are
There are good reasons to doubt that Kantian contextualism can
give us such a compass. Here is why. As is clear by now, Kantian
contextualism says that practices and action principles that qualify
as violations of the CI
The problem for Kantian contextualism, with regard to its
potential as a compass in the global justice debate, lies in its
insistence that “there is nothing over and above the act of
judgement” that can tell agents whether their actions and action
principles are morally justified (Flikschuh 2018, 573; Flikschuh
2017, 45). Provided the empathetic imagination of the “global
theorist” stretches so far as to make it possible for her to
understand
But does Kantian universalism fare any better? If it is to succeed
where Kantian contextualism fails
First of all, it must be emphasized that it is
First of all, it is possible for an agent to reach a judgment
about how she should
Imagine a scenario in which you and I physically encounter a person who is hurt and in dire need of help. The person is within our reach, and there seems to be no one else around. Both of us have individually adopted a general principle of promoting the happiness of others. However, you are a medical professional, whereas I am not. Given our knowledge about our abilities, it would be rational to judge that the same principle requires different actions of us individually. I may enact this principle to the best of my abilities by calling for an ambulance, whereas you may do so by providing the victim with medical aid
Secondly, in order for an agent to have an epistemic basis for
deciding on a
The same goes for the duty to promote the happiness of others, as
social factors such as culture can greatly affect what may or may
not be done to achieve this. To mention one example, in many Western
cultures it is not considered distasteful to place elderly family
members in nursing homes when they are no longer able to take care
of themselves, whereas this practice would be considered outrageous
in many Asian, African and Middle-Eastern societies, where children
are often expected to take care of their parents until the end of
their lives. Since cultural conceptions and norms can impact which
actions generally serve to make people happy (in the broadest
sense), it can be context/culture-dependent whether an act (such as
having one’s elderly family members placed in a nursing home) could
reasonably be conceived as promoting the happiness of
others.
Thus, it can be seen that on a “standard”, universalistic reading of Kant’s ethics, the same moral principle may favour the performance of different actions in different (cultural) contexts. Hence, Kantian universalism can require different actions and omissions from us when we find ourselves among people with customs, beliefs and traditions that are different from those we are used to. Kantian universalism is context-sensitive by design.
Unlike Kantian contextualism, Kantian universalism is not context-sensitive in the sense that different contexts can give rise to different categorical imperatives. It allows for contextual latitude only with regard to how moral duties can, may and ought to be acted upon.
While Kantian universalism prioritizes maxims, it also forbids
certain
What has been insufficiently appreciated in the Kant literature (and arguably by Kant himself) is that even though contextual considerations never trump the requirements of perfect duty, whether certain actions qualify as a violation of perfect duty is context-dependent. This is exemplified by a consideration that Flikschuh is right to point out, namely the fact that differently situated people may have different conceptual frameworks, because of which there may be certain action types that are available to “them” but not to “us” (Flikschuh 2017, 105).
An example may serve to make this clear. In a discussion on the
normativity of culture-specific action types (or “doables”), David
Velleman mentions a speech act in Russia that is called
v
As mentioned above, Kant deems it a perfect duty to abstain from
lying (AA 4:402-403; 4:422). In one of his famous “casuistical
questions” in the
Since
I mention this particular example to underscore the point that
even though practices may
Does this conclusion imply that in Kantian universalism it is
possible for one of two
A worry that may arise at this point is that the
context-sensitivity of perfect duties in Kantian universalism, much
like Kantian contextualism, comes at the cost of normative
indeterminacy. If perfect duties are supposed to prescribe laws for
actions but actions cannot be separated from their underlying
principles, there may be cause for concern about perfect duties
being incapable of drawing a clear moral line for rational agents
after all. Suppose, for instance, that Kant is right in that is a
perfect duty to abstain from mutilation. What should Kantian
universalism say about female genital cutting rituals, given that
proponents of such practices usually do not
A temptation that should be avoided is to search for an “ideal”
definition of mutilation and to proceed by arguing that proponents
of female genital cutting rituals simply fail to understand that,
according to this ideal definition, female genital cutting falls
under its umbrella. For one thing, it is not clear that morally
loaded concepts such as “mutilation” can be wholly defined in “pure”
terms, uninfluenced by societal attitudes and linguistic customs
(cf. Sticker 2018, 181-185). For another, what is important in the
example borrowed from Korsgaard has nothing to do with whether the
“theft maxim” is based on the best possible definition of stealing
(or property, for that matter). Nor is it important whether the
difference between
Now in some cases, an agent’s maxim may already explicitly
contain an action description that would be contrary to duty on
Kant’s account, such as “I will lie/steal/murder when … to
…”.
As with “lying maxims”, it should be equally possible to adopt “mutilation maxims” without “mutilation” actually being the maxim’s action description. However, to understand what a “mutilation maxim” would be on Kant’s account, we first need to know what Kant means by mutilation.
When discussing perfect duties to the self in the
Kant describes
In a paper from 2009, Michael Cholbi asserts that Kant has a
Kant views our duties as subject to a kind of symmetry, according to which if an agent is morally required to treat herself in a particular fashion, then all other things being equal, she is morally required to treat similarly situated others in the same fashion. (Cholbi 2009, 20).
For my current purposes, I take this claim for granted. What this entails, with regard to the perfect duty against self-mutilation, is that a perfect duty against mutilating myself at the same time spells out a perfect duty against mutilating others. This can only be true if Kant’s idea of what it means to mutilate others fundamentally means the same thing as what it means to mutilate myself (either materially, formally or both).
Now that it is clear what the Kantian perfect duty against
mutilation
If it is true that female genital cutting
The main takeaway is as follows. The idea that actions cannot be
separated from maxims in Kantian universalism does
In this section, I hope to have made a plausible case for the
idea that Kantian universalism is a context- sensitive moral theory
that is also capable of prescribing substantive, universal
action-principles. With Kantian universalism, it takes a
contextually situated rational agent to judge how
Kantian universalism is more potent as a context-sensitive moral
theory than Kantian contextualism. The task with regard to evaluating
the moral permissibility of less familiar cultural practices is not,
as Kantian contextualism would have it, to find out whether we are
capable of imagining ourselves sincerely endorsing these practices
under particular circumstances. Instead, the task for Kantians is to
first get sufficiently clear on the underlying context, conception and
purpose of the practices in question
Flikschuh seems to use the opposition between “we” and “they” in such a way that these terms can pertain to any (cultural) group that is perceived as such from the first-person perspective of an agent. I will follow her in this regard.
Some of the philosophers who belong to Flikschuh’s category of “global theorists” range from John Rawls, Charles Beitz and Aaron James to Amartya Sen, Martha Nussbaum, Simon Caney and Mathias Risse (cf. Flikschuh 2017, 3-5).
Flikschuh is not the only one to raise this point (cf. Velleman 2015, 56-74; Korsgaard 2012, 85; Rawls 1955, 27-28).
When Flikschuh speaks of a ‘pure’ maxim, she seems to refer to a maxim that is adopted on the basis of a sincere judgment that it could serve as a universal law (cf. Flikschuh 2018, 562).
It should be noted that it remains unclear, on this formulation of the CI-test, where this leaves maxims that Kant deems at the same time conceivable as laws but not at the same time willable as laws (cf. AA 4:423).
Flikschuh is aware of the “tension between the ineliminably indexical character of reasons for action and rationality’s equally authoritative claim to universal validity” in Kantian contextualism (Flikschuh 2018, 567) but seriously underestimates how deep this problem goes (see also Flikschuh 2017, 120-123).
Culture is of course not the only relevant consideration when it comes to the ethics of placing the elderly in nursing homes. It is not uncommon for elderly people in Western countries to fervently object to being relocated to a nursing home, and some elderly people in Asia or Africa likely dislike the idea of being taken care of by their children, preferring a nursing home instead. However, these are also contextual considerations that should be taken into account by the agent when deliberating on how the principle of promoting other people’s happiness should be applied in their particular circumstances.
The fact that there is a category of duty in Kant that explicitly forbids certain action types is already a strong indication that he puts much more rigidly defined constraints on action than Kantian contextualism seems to be willing to do. Regardless, the idea that the concept of perfect duty should be preserved in Kantian ethics can be maintained even if we disagree with some of Kant’s claims about what perfect duty specifically requires of us. After all, some of these may be an expression of Kant’s own cultural or personal biases rather than being based on rational principles (such as the supposed categorical prohibition against masturba- tion).
Martin Sticker has raised a similar point with regard to the ethics of suicide in Kant’s ethics. He (tentatively) suggests that “[a]t least to some extent it might depend on […] linguistic customs and societal attitude” whether certain forms of suicide (such as euthanasia when terminally ill) should qualify as impermissible “self-murder” in Kant’s ethics (Sticker 2018, 181-182).
In other circumstances there could be a moral difference between vranyo and the casual use of sarcasm or irony, such as when high-ranking military officials use vranyo in briefings to knowingly uphold an upstanding self-image that is not grounded in reality (thereby possibly enabling the continuation of war crimes, mistreatment or other forms of injustice). Even in such circumstances, however, vranyo would not fall under the umbrella of Kant’s conception of lying.
The idea that outwardly identical actions could qualify as fundamentally different action types due to different underlying maxims does not in turn imply that it is impossible for actions with substantively different underlying maxims to qualify as identical action types. For instance, agent A and agent B can both be said to be performing CPR even if agent A goes through the (correct) CPR motions with the sole intent of saving a person’s life and agent B does the same with the intent of gaining the admiration of bystanders. In this example, agent A performs CPR from duty, whereas agent B performs CPR from self-interest.
It must be noted that Korsgaard mentions this example in the context of establishing that on Kant’s account, all maxims consist of “actions and ends” and that the moral quality of a maxim consists in the “functional arrangement” of both parts, rather than the parts themselves. She does not explicitly argue, as I do, that the moral difference between the first action and the second action can be explained in terms of the first’s (but not the second’s) qualifying as theft.
This concern seems to tie into what has come to be known as G.E.M. Anscombe’s “problem of relevant descriptions” objection to Kant’s ethics, according to which “[Kant’s] rule about universalizable maxims is useless without stipulations as to what shall count as a relevant description of an action with a view to constructing a maxim about it” (Anscombe 1958, 2). While Kantian philosophers like Onora O’Neill, Barbara Herman and Jens Timmermann have aptly addressed this objection by arguing that the correct/relevant description of an action in Kant’s ethics is determined by the actual maxim on which the agent wills the action (O’Neill 1975, 94-110; Herman 1990, 68-69; Timmermann 2005, 251), the current concern cannot be resolved by appealing to this argument. After all, the current challenge for Kantian universalism is to make sense of the idea that it is possible for actions to qualify as action types that perfect duty would forbid precisely despite the fact that agents may genuinely not conceive of their actions as such action types.
See, for instance:
Provided that these concepts refer to the types of action that Kant has in mind.
Assuming, of course, that it is a perfect duty not to lie, steal or murder.
Kant also refers to self-mutilation as ‘partial self-murder’ (6:421-423).
Some maintain that the deprivation of sexual pleasure as a result of female genital cutting is exaggerated, as such criticism supposedly ‘presupposes an excessively mechanical picture of sexuality, omitting the role of the mind as the most important sexual organ’ (Wong 2023, 51). This objection is unconvincing for at least two reasons. The first is the bare fact that different people enjoy sex differently. For some people ‘the mind’ may suffice as a source of sexual pleasure, but others may not be so fortunate. The second reason is that this response does not diminish the fact that female genital cutting comes with the substantial risk that the girl’s ‘mechanical’ capacity for sexual pleasure will be destroyed or diminished – a ‘natural power’ of which the girl may not wish to be deprived (either now or later in life).
See, for instance:
I am grateful to the Netherlands Organization for Scientific Research NWO for financial support, and I thank Pauline Kleingeld, Hanno Sauer, Vinicius Carvalho, Mike Gregory, Fiorella Tomassini and Janis Schaab for invaluable comments on earlier drafts of this paper.