173
Kant, Xunzi and the Artificiality of Manners
Anja BERNINGER*
Abstract
Both Chinese and Western philosophers have argued for the ethical importance of man-
ners. Their approaches are sometimes criticized on the grounds that manners are artificial. 
I compare Xunzi’s and Kant’s responses to this claim, and discuss the relevance of both 
positions for the development of a theory of manners. I show that there is no single 
artificiality claim, but rather four different claims: the claim that polite behavior lacks 
spontaneity, the claim that it is insincere, the claim that it goes against human nature, and 
the claim that it is arbitrary. While Kant is mainly concerned with the insincerity claim, 
Xunzi focusses on the claim that manners are arbitrary rules. Because of their different 
understandings of the function of manners both authors only provide a partial answer to 
the artificiality claim. To arrive at a full account of manners both perspectives must be 
combined. 
Keywords: Xunzi, Kant, manners, politeness, artificiality
Kant, Xunzi in izumetničenost pravil vedênja 
Izvleček
Tako kitajski kot zahodni filozofi so poudarjali etični pomen pravil vedênja, četudi so jih 
pogosto kritizirali zaradi njihove umetne narave. Avtorica primerja Xunzijevo in Kan-
tovo stališče o tem problemu in ovrednoti pomen obeh pozicij za razvoj teorije vedênja. 
Pokaže, da stališče o izumetničenosti vedenjskih pravil ni samo eno, temveč gre za štiri 
različna stališča: pomanjkanje spontanosti, neiskrenost, nasprotje s človeško naravo ter 
arbitrarnost. Medtem ko se Kant v glavnem ukvarja s predpostavko neiskrenosti, se Xunzi 
bolj osredotoča na problematiko arbitrarnosti vedenjskih pravil. Zaradi različnega razu-
mevanja funkcije pravil vedênja vsak od njiju izdela samo delne rešitve problema izu-
metničenosti. Če pa želimo izdelati celovitejšo teorijo vedênja, je treba kombinirati obe 
perspektivi. 
Ključne besede: Xunzi, Kant, vedênje, vljudnost, izumetničenost
* Anja BERNINGER, Akademische Rätin/Lecturer;
 Department of Philosophy, University of Stuttgart, Germany. 
 anja.berninger[at]philo.uni-stuttgart.de.
DOI: 10.4312/as.2017.5.1.173-192
Azijske_studije_2017_1_FINAL.indd   173 30.1.2017   14:01:11

174 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
Introduction 
Both Western and Chinese theorists have argued that manners have ethical im-
portance (some recent examples are Buss 1999; Olberding 2016; Stohr 2012). In 
their view, politeness is not just a matter of convention, and thus we should not 
act politely simply because “that’s what people do”. Nor is well-mannered behav-
ior merely of instrumental value. Rather, according to these thinkers, acting in a 
well-behaved manner is an important part of a good life.
Politeness is not a naturally given quality. Even though manners are not just a 
matter of mere convention, well-mannered behavior generally involves following 
certain man-made rules. Politeness thus requires us to acquire certain affective and 
action dispositions that are not always in line with our innate inclinations. These 
and similar aspects of manners have led critics to claim that manners are artificial.
In this paper I will explore two responses to this claim, one given by Kant and the 
other by Xunzi. In spite of the temporal and cultural distance between these two 
philosophers, their approaches to manners display some striking similarities. Both 
believe that manners and politeness can transform people for the better, and both 
are concerned about the claim that manners are artificial. Nevertheless, a close 
examination of their arguments shows that they discuss artificiality in slightly dif-
ferent senses. As I will show, this difference is a natural result of the overall func-
tion that they assign to well-mannered behavior. While Kant stresses that there 
is a communicative function inherent in manners, Xunzi is primarily concerned 
with what one can call the ordering function of manners. 
These differences are instructive for those interested in systematic questions con-
cerning the status of manners within ethical theory. We prima facie have reason 
to think that the artificiality claim poses a challenge to the belief that manners 
should be accorded ethical relevance. Both Kant and Xunzi respond to part of 
this challenge, but because of their respective takes on the function of manners 
each leaves out one central aspect. I will briefly suggest that this problem can be 
remedied by accepting that manners do not have a single function, but rather a 
row of different functions. 
I will start off by first analyzing the term “artificial”. I will show that we can differ-
entiate four meanings, each of which can be used as the basis of a slightly different 
attack against the claim that manners are ethically relevant. I will then go on to 
explore both Kant’s and Xunzi’s positions on manners in detail and discuss which 
of these artificiality arguments they pick up, how they respond to them, and how 
this fits in with their overall theory. At the end of the paper I will then come back 
to the question of what this means for a systematic perspective on manners.
Azijske_studije_2017_1_FINAL.indd   174 30.1.2017   14:01:11

175 Asian Studies V (XXI), 1 (2017), pp. 173–192
Unpacking the Artificiality Claim
In a recent article on manners in the Xunzi, Amy Olberding highlights both the 
importance of the criticism that manners are in some sense artificial, and the 
diverse forms that this claim can take. She describes “artificiality” as a “catch-all 
term embracing a host of sins and ills associated with etiquette” and then claims: 
For example, practicing etiquette may oblige social actors to insincerity 
and hypocrisy, requiring that they suppress expression of unretouched 
opinions or feeling in favor of adopting a more socially pleasing façade. 
Etiquette can demand adherence to norms that are, sometimes rather 
obviously, culturally contingent or seemingly arbitrary. Etiquette can re-
quire that even the simplest endeavors be pursued circuitously, detouring 
practitioners away from straightforward efficiency by demanding they do 
what they do decorously. (Olberding 2015, 145)
From this we can discern at least four different meanings of “artificial” relevant 
to the discussion of manners, which can also be found in a standard dictionary 
definition of the term (Random House W ebster’s Unabridged Dictionary 2001, 119):
(1) We may use the term “artificial” to indicate that something is effortful or that 
it lacks spontaneity. The opposite of the term here is the way we use the term 
“natural” in saying that something comes naturally to someone (i.e., he does 
not have to use much effort to do it). We might then make the additional as-
sumption that everything that is effortful or non-spontaneous is in some sense 
deficient. If we then add the idea that well-mannered behavior will often prove 
effortful, it becomes clear why we should consider manners as problematic.
1
 
(2) “Artificial” can also be used to mean “false”, “insincere”, or “inauthentic”. The 
assumption here is that our natural behavior has a positive quality, namely 
authenticity or sincerity, and that these qualities are lost through our focus on 
well-mannered behavior. Thus, for instance, well-mannered behavior often 
does seem to involve masking one’s true feelings towards others and their 
actions. 
(3) We can also understand “artificial” as simply meaning “non-natural”. One 
might then further assume naturalness is a positive quality. This would be 
the case, for example, if we assumed that human nature and the behavior that 
directly springs from it may be changed for the worse through the installation 
of certain non-natural patterns and rules of behavior (artificial rules). 
1 Zhuangzi’s criticism of human activity (wei 偽) can be read along these lines. For such a reading 
see e.g., Lee 2005, 15–17. For a further discussion of naturalism in the Warring State Period, see 
Eno 1990, 138–43.
Azijske_studije_2017_1_FINAL.indd   175 30.1.2017   14:01:11

176 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
(4) Lastly, the term “artificial” can be used to denote something arbitrary. Thus, in a 
different context, we can say that a given system of classification is artificial or 
that a particular distinction seems artificial. What we mean to say here is that the 
classification or distinction has no grounding in what is (objectively) given. Polite 
behavior will at least sometimes involve differentiations (such as when we host 
an event and greet the guest of honor before the others). Critics could claim that 
the distinctions made in polite behavior often lack any form of proper grounding. 
If the alleged ethical value of manners is to be defensible, it seems one must have 
something to say about all four of these claims. This can take various forms, such 
as a direct rebuttal of the validity of the claim or some argument that manners 
do not in fact instantiate the property in question. In what follows, I will discuss 
two different strategies of dealing with the claims, one suggested by Kant and 
the other by Xunzi. As I will show, both authors answer the claims by making a 
suggestion as to the function of manners, which in turn also delimits the scope of 
what can count as well-mannered behavior.
Kant’s Concept of Manners
Engaging with Kant when discussing the artificiality of manners may seem like a 
surprising choice. Indeed, in contrast to Xunzi, Kant does not give us a full-blown 
account of manners. Nevertheless, we do find some discussion on the importance 
of manners, in the course of which he portrays them as ethically relevant. The 
most prominent work in this respect is his Anthropology from a Pragmatic Point 
of View, which I will focus on in this paper. There, Kant begins his discussion of 
manners with a paragraph that delves right into the heart of the matter:
On the whole, the more civilized human beings are, the more they are ac-
tors. They adopt the illusion of affection, of respect for others, of modesty, 
of unselfishness, without deceiving anyone at all, because it is understood 
by everyone that nothing is meant sincerely by this. And it is also very 
good that this happens in the world. (Kant 2006, 42)
The passage appears baffling at first. Why should it be a good thing that we be-
have like actors, creating the illusion for others that we respect them and wish 
them well, when really we don’t? And why would it make a difference ethically 
that everyone knows that we are only creating an illusion? 
One suggestion that can be constructed on the basis of Kant’s text is that the il-
lusion in question is a necessary evil resulting from serious flaws in human nature. 
Azijske_studije_2017_1_FINAL.indd   176 30.1.2017   14:01:11

177 Asian Studies V (XXI), 1 (2017), pp. 173–192
Kant discusses the question of whether man’s nature is good or evil towards the 
end of the Anthropology. Here, he distinguishes two aspects of human nature: 
man’s rational (intelligible) nature and his sensible nature. The intelligible nature 
allows us to recognize the moral law and act in accordance with it, and therefore 
Kant takes it to be good. He is clearly far less optimistic with respect to man’s sen-
sible nature. Thus he claims: 
But experience nevertheless also shows that in him there is a tendency 
to actively desire what is unlawful, even though he knows that it is un-
lawful; that is a tendency to evil, which stirs as inevitably and as soon as 
he makes use of his freedom and which can therefore be considered as 
innate. Thus, according to his sensible character the human being must 
also be judged as evil (by nature). (Kant 2006, 229) 
One can thus see man’s sensible nature as a kind of obstacle that must be over-
come. According to Kant this can be done, but only through an effortful process 
of cultivation and moralization (ibid. 1968, 324–5). He does not mention polite-
ness in this part of the Anthropology. Nor does he give us a very detailed account 
of the moralization process here. Nevertheless, it does seem plausible to suggest 
that manners could be part of such a process, as they will often demand keeping 
our sensible nature in check, and arguably form an important basis for humans 
living together in a society, which Kant, in turn, does mention in his discussion of 
moralization (ibid., 324).
2
 Through these remarks Kant indirectly counters arti-
ficiality claims (1) and (3), i.e., the claims that manners are artificial in the sense 
of demanding effort, and in that of running counter to human nature. According 
to Kant, manners do indeed run counter to our sensible nature. However, because 
our sensuous inclinations are problematic anyway, this cannot be taken as the 
basis of an argument against politeness. Kant simply denies the claim that the 
whole of human nature is good and worthy of being preserved. Rather, it is only 
the intelligible part that must be upheld. And, with respect to this part of human 
nature, manners seem to play a bolstering instead of a destructive role. They are 
instrumental in letting rationality prevail against our problematic inclinations. 
A similar strategy can also be applied for giving a Kantian answer with respect to 
the first artificiality claim. Kant simply denies that activity and effort are problematic 
qualities. Rather, according to his account, it is passivity on the side of our intelligible 
nature that is problematic. It is only the active countering of our problematic, sensi-
ble inclinations that makes us “worthy of humanity” (ibid., 325); this is also stressed 
2 It is interesting to see that in the same passage Kant also heavily criticizes Rousseau for his pro-nat-
uralism view of human nature (Kant 1968, 326–7).
Azijske_studije_2017_1_FINAL.indd   177 30.1.2017   14:01:11

178 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
in Klemme 1999). Effort, insofar as it is guided by reason, must thus be seen as a 
positive trait, while passivity towards our sensible nature will incline us to act badly. 
In failing to take an active stance, we are responsible for the evil actions that ensue 
(Klemme 1999). Again, though Kant does not explicitly mention politeness, it does 
seem that we could see it as part of such an active endeavor guided by rationality.
Up to now I have focused exclusively on Kant’s understanding of human nature as 
it is conveyed in his Anthropology. Taking Kant’s other writings into account, and 
especially his Religion within the Boundaries of Mere Reason, gives us a different 
angle on human nature and the role of manners with respect to it (Kant 2003).
3
 In 
this work Kant explicitly rejects the sort of mixed view on human nature sketched 
above (DiCenso 2012; Wood 2014; Palmquist 2016; Kant 2003, 43). Rather than 
residing in our sensible nature, evil here is suggested to come about as a result of 
human choice. Thus, as Wood explains, in this view: 
Inclinations themselves, insofar as they do not originate from free choice, 
can never be evil. They become involved in evil only when they have been 
taken as incentives in a freely chosen maxim which is evil. And then it is 
this maxim that is evil, not the inclinations themselves. (Wood 2014, 34)
Although our natural inclinations are not bad as such, we can still make a mis-
take by assigning a wrong role to them. We can furthermore be tempted to do so 
due to the strength of some of our inclinations and to the propensity to evil that 
Kant ascribes to humans. As has been noted in recent publications, we should not 
understand this as implying that human nature is in some sense bad. Rather a 
propensity, according to Kant, is an inclination that exists only in potential form 
at first but grows if it is nurtured by us through engaging in the related activities. 
Thus, according to Kant, the desire for alcohol in the alcoholic might exist as a 
mere potentiality, until the person has actually drunk alcohol. Only through such 
acts of nurturing do we arrive at strong habitual desires to seek out more expe-
riences of the same kind, and there is thus an element of choice ingrained here 
(Stroud 2010, 642–3; Wood 2014, 50–51; Kant 2003, 34–39).
4
 
3 In claiming that Kant is offering two competing interpretations of human nature in these two texts I 
follow Brandt (Brandt 1999, 483–5). As Brandt points out, it is not quite clear why the two positions 
diverge. As he suggests, it may be the case that Kant does not want to go into too much detail in the 
Anthropology and therefore settles for a position he does not deem fully correct. Otherwise, it may 
also be the case that he is in some sense rethinking his position. I leave the question of how these 
differences come about open, as nothing much depends on this issue for the purposes of this article.
4 I do not have the space to go into the details of Kant’s exposition here, and am therefore skimming 
over some of its complexities. For an overview on the matter, see newer commentaries on this sec-
tion (Palmquist 2016; Wood 2014; DiCenso 2012).
Azijske_studije_2017_1_FINAL.indd   178 30.1.2017   14:01:12

179 Asian Studies V (XXI), 1 (2017), pp. 173–192
Kant also assigns the (good) predisposition to us to follow the moral law. This 
predisposition cannot be lost, even if we continuously act badly. What we should 
aim for is to reinstate it to its “rightful place as the sole element influencing our 
moral choices” (Palmquist 2016, 128). And this, in turn, involves viewing our 
inclinations as irrelevant to these choices (ibid.). In requiring us to act as if we 
respect others (independent of our actual feelings), behaving politely can be an 
important part of such a reinstatement. We can thus say that (in spite of their 
differing discussions of human nature) manners can be assigned an important role 
in self-cultivation on the basis of both the Anthropology as well as Religion within 
the Boundaries of Mere Reason.
Kant does not only state that the illusion of benevolence created by politeness is 
a necessary evil for keeping our sensible nature in check or reinstating our good 
predisposition to its rightful place. Rather, he also claims that this illusion can––in 
the long run––turn into something real. Thus he explains: “(…) these signs of be-
nevolence and respect, though empty at first, gradually lead to real dispositions of 
this sort” (Kant 2006, 44). In behaving politely, we start off by faking our respect 
for others, but over time we can acquire real respect by doing so. 
Nevertheless, the fact that Kant suggests that “faking it” is permissible seems puz-
zling. As quoted, he claims that manners are “signs of benevolence and respect” 
(my emphasis). The talk of “signs” is rather vague here, but an earlier passage in the 
Anthropology suggests that Kant directly links the term with acts of communica-
tion. There he lists “signs of gesticulation” and “memetic signs” with language, let-
ters, and “signs of social standing” as “arbitrary” as opposed to “natural signs” (Kant 
2006, 85–86) and highlights their importance in interpersonal communication. 
Thus, by talking about well-mannered behavior as a sign, Kant is suggesting that 
politeness is a communicative, interpersonal act. But this gives rise to a problem. 
If we are engaging in an interpersonal act of communication and are intentionally 
being untruthful in this act, then one could suggest that we are lying. And it seems 
that even if a lie has positive results in the long run, it is nevertheless a morally 
problematic act.
The problem is all the more pressing for Kant, because in other places he suggests 
that lying is not permitted under any circumstances. As Korsgaard has pointed 
out, a reason for Kant’s rigorous stance can be reconstructed from the so-called 
Formula of Humanity, which demands: “So act that you use humanity, whether in 
your own person or in the person of any others, always at the same time as an end, 
never merely as a means” (Kant 1998, 38). Humanity, in turn, is understood here 
as a specific capacity, namely the capacity to rationally determine our own ends 
(Korsgaard 1986; Kant 1998, 44).
Azijske_studije_2017_1_FINAL.indd   179 30.1.2017   14:01:12

180 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
The problem about lying, according to Kant, is that we are putting others in a con-
dition in which we are depriving them of the possibility to exercise that capacity, 
because those we deceive cannot possibly rationally assent to our action. As Kors-
gaard has highlighted, the issue comes up independently of whether the victim of 
the deceit knows what is happening. Thus, if we do not know we are being deceived, 
we cannot of course rationally assent to that act, as we do not know that this is what 
is happening. However, even, if we do know that we are being deceived, it is also not 
within our capacities to assent to the deceit, because in this situation there is no ac-
tual act of deceit (only a failed attempt to engage in such an act) (Korsgaard 1986).
Because of this rigorous rejection of the permissibility of lying, Kant cannot ac-
cept that manners are both ethically valuable and that they are a way of deceiv-
ing others. He needs to offer some other explication of what “empty” signs of 
“benevolence and respect” amount to, and in doing so he must show that we can 
rationally assent to each other’s polite behavior (thus distinguishing it from the 
case of deceit). 
The solution Kant offers lies in claiming that politeness is no act of deceit, but 
merely a benign illusion. Illusions, according to Kant, do not vanish once we are 
informed about the real circumstances, and in this they differ from deceit. Take 
the case of deceit again: Say Peter is a criminal who has just escaped from jail. 
Assume that he is pretending to be a policeman (by wearing a uniform, acting 
in ways associated with that role, and so on). If I do not know Peter, I will have 
the impression of seeing a real policeman. But as soon as I learn about his real 
identity my impression will change dramatically, and in a rather unpleasant way. 
This is different from the case of optical illusions, like the frequently discussed 
Müller-Lyer-illusion. Even once we learn that the two lines in this illusion are 
of equal length, the impression that they are unequal remains, and there is no 
unpleasantness in this at all. 
Kant claims that politeness is an illusion, not a case of deceit, and as such, in his 
reckoning, the problems resulting from deceit vanish. We are not treated as mere 
means to an end by those that behave politely. We are not deceived, because we 
know that people are being friendly for show. Nevertheless, the behavior of those 
engaging in the illusion still has a positive effect on us, we continue to see this 
sort of behavior as appealing and worthy of emulation. The positive impression 
therefore remains.
5
5 The distinction between deceit and illusion and its importance for understanding Kant’s position 
on manners is highlighted by Frierson on whose discussion I rely here. Frierson also provides a 
detailed discussion of how illusions can motivate, which does much to explain and extend Kant’s 
rather short comments on the issue (Frierson 2005). 
Azijske_studije_2017_1_FINAL.indd   180 30.1.2017   14:01:12

181 Asian Studies V (XXI), 1 (2017), pp. 173–192
There is a further point that Kant makes concerning deceit, which deserves men-
tion. While the positive impression that well-mannered behavior makes on others 
may lead them towards virtuous behavior, it is not clear from what has been said 
so far how it can also play a role in improving ourselves. Kant sees politeness as 
an act of “deceiving the deceiver in ourselves” (Kant 1968, 151), and then goes on 
to claim that, again, this is not deceit proper, but actually the mere creation of an 
illusion. Thus one can say that Kant suggests there are two kinds of illusions at 
play here, an outer and an inner illusion created by manners. While it is clear why 
the outer illusion is relevant, one may wonder why this act of inner illusion should 
be necessary. 
As Frierson has stressed, this idea is linked to Kant’s take on the role that self-de-
ceit plays for us.
6
 What Kant is implying is that while we are able to make rational 
choices independent of our inclinations, it is not always easy to do so. One reason 
for this is that at least some of our inclinations find a powerful ally in self-deceit. 
Kant gives us an example of such an act of self-deceit: he explains that we have 
an inclination towards being at ease, which leads us to staying largely passive. The 
self-deceit here is two-fold. First, we may have the impression that in remaining 
passive we are not doing anything evil, and are thus pursuing something that is 
morally good, while this is obviously not the case. Secondly, we are under the 
impression that we are avoiding unpleasant experiences (i.e., the experience of 
effort). But in doing so we are in fact conjuring up a most unpleasant state, name-
ly boredom (Frierson 2005; Kant 1968, 151–2). Manners trick us into avoiding 
this problematic form of passivity by engaging us in pleasurable pursuits such as 
conversation, which, while seemingly effortless, nevertheless keeps our minds oc-
cupied and turns us into more cultured beings (Kant 1968, 152).
7
Where does this leave us with respect to the artificiality claims? In his descrip-
tion of politeness, Kant counters claims (1), (2), and (3). In his replies he seems 
to be especially preoccupied with (2). This, in turn, seems to be the case because 
he understands manners as acts of tricking oneself and others. Through this, the 
problem of deceit looms large, an issue that Kant has to take seriously, in part 
because of his rigorous views on lying. Furthermore, it is striking that Kant does 
not seem to care much at all about the fourth artificiality claim, namely that man-
ners involve arbitrary differentiations. This too is no accident. As explained earlier, 
Kant understands politeness as a sign of benevolence and respect. According to 
Kant, respect is something we owe to everyone, simply because they are persons. 
6 For a further discussion of self-deceit, its relation to passions and importance for Kant’s position 
on human nature, see Wood (Wood 2014, 44)
7 For a more detailed interpretation that also explores the different discussions on self-deception, 
which Kant offers in his Lectures on Anthropology, again see Frierson’s analysis (Frierson 2005).
Azijske_studije_2017_1_FINAL.indd   181 30.1.2017   14:01:12

182 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
Manners, from this perspective, are thus not connected to any sort of distinction 
between people on the basis of achievement or rank, but rather just expressing 
what is owed to all persons qua persons. Therefore, there is no need to worry about 
them potentially containing arbitrary kinds of differentiation. The fourth artifici-
ality claim thus does not seem threatening to Kant’s approach, because he denies 
that manners have a differentiating function.
8
 
Xunzi’s Understanding of Artificiality
While manners and politeness lie at the periphery of Kant’s philosophy, they are at 
the very heart of Xunzi’s thinking. In saying so, one should be aware of the fact that 
Xunzi uses the term li (禮), and thus does not talk about manners in isolation, but 
rather also about rituals, religious customs, and so on. In spite of this broader no-
tion, I will only discuss manners here. Often, li (禮) is translated as “ritual”. To avoid 
confusion, I will replace this with the original term (li 禮) in the translations cited.
Central for Xunzi’s position on manners is his view of human nature. Thus, at the 
very beginning of the Discourse on Ritual he writes: 
From what did li (禮) arise? I say: Humans are born having desires. 
When they have desires but do not get the objects of their desire, then 
they cannot but seek some means of satisfaction. If there is no measure 
or limit to their seeking, then they cannot help but struggle with each 
other. If they struggle with each other then there will be chaos, and if 
there is chaos then they will be impoverished. The former kings hated 
such chaos, and so they established li (禮) and yi (義) in order to divide 
things among people, to nurture their desires, and to satisfy their seeking. 
(Hutton 2014, 201)
In this passage Xunzi indirectly seems to be tackling the third artificiality claim. 
He does so by applying a strategy that is, on first glance, similar to Kant’s. Like 
Kant, Xunzi highlights certain negative aspects of human nature, as well as the 
disastrous effects uncontained expressions of our naturally given desires may have. 
In the passage cited, Xunzi stresses the social instead of the individual consequenc-
es of leaving our desires uncontained. He might therefore be read as worrying not 
so much that the individual may act in morally problematic ways, but rather that 
social order cannot be upheld if everyone were to focus on their own desires. 
8 In the current literature, Sarah Buss presents a largely Kantian view, which also puts strong empha-
sis on this notion of respect (Buss 1999).
Azijske_studije_2017_1_FINAL.indd   182 30.1.2017   14:01:12

183 Asian Studies V (XXI), 1 (2017), pp. 173–192
This, however, does not mean that Xunzi focusses exclusively on these social con-
sequences. He does not only speak of society being chaotic; he thinks that this can 
also be a feature of individuals.
 9
 As such, in his reflections on human nature he 
claims the following:
Thus, crooked wood must await steaming and straightening on the shap-
ing frame, and only then does it become straight. Blunt metal must await 
honing and grinding, and only then does it become sharp. Now since 
people’s nature is bad, they must await teachers and proper models, and 
only then do they become correct. They must obtain li (禮) and yi (義), 
and only then do they become well-ordered. (ibid., 248)
This passage may seem puzzling. What does it mean that people are “well-or-
dered”? In what sense is a well-ordered person different from a badly ordered one? 
Xunzi certainly does not believe that becoming a well-ordered person involves 
extinguishing one’s desires. As he puts it: 
All those who say that good order must await the elimination of desires 
are people who lack the means to guide desires and cannot handle the 
mere having of desires. All those who say good order must await the 
lessening of desires are people who lack the means to restrain desire and 
cannot handle abundance of desires. (ibid., 243)
10
 
Rather than being exterminated, desires must be brought under control in the 
sense of people not automatically following any desire they have. This is dif-
ficult, because we have the natural tendency to act on them.
11
 Nevertheless, 
our psychological makeup also opens other options. Namely, we are also able 
to reflect on our desires through some higher order capacity, i.e. through our 
intellect (xin 心). It is through this capacity that we decide which desires we 
should act upon.
12
 When we fail to put this capacity to use, we remain inter-
nally disordered.
9 The general focus on order in the Xunzi is also highlighted by Machle. Machle stresses that Xunzi 
mostly uses the term zhi 治, which he translates as: “‘to manage’, ‘to put into order’, ‘to heal’”, in-
stead of zheng 政, which he understands as being connected to “‘government’ or ‘to correct by law, 
force, or sanctions’”. (Machle 1993, 47)
10 In other sections of the Xunzi we do, however, find the idea that we can shape and thus change our 
“inborn dispositions” through practice. (See Hutton 2014, 249).
11 This aspect of Xunzi’s theory is often overlooked, but has recently been stressed by Chenyang Li 
(Li 2011).
12 For an interpretation along these lines see Lee 2005, 48–51.
Azijske_studije_2017_1_FINAL.indd   183 30.1.2017   14:01:12

184 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
The idea of there being such a thing as an “internal disorder” becomes clearer 
when we reflect on what it would mean to act on any desire we have without 
further reflection. If we did so, our overall pattern of action would turn out in-
coherent. To give an example: we may have the desire to remain healthy, but also 
a desire to eat large quantities of fatty foods and smoke cigarettes. Clearly, these 
desires cannot all be fulfilled at the same time, and we must choose among them 
if we are to achieve any of our goals. If we decide that staying healthy is the desire 
we should follow, then this gives our actions a distinct direction and stops them 
from oscillating between healthy and unhealthy ways of acting.
13
This reading does not imply that man innately has both good and bad qualities 
or good and bad desires. Rather, the suggestion is that Xunzi does not take the 
desires in themselves to be the problem. It is not, for instance, that we are beset by 
ethically problematic desires and that we therefore continuously strive for what is 
bad. It is rather that blindly following our desires without any further reflection 
leads to both inner and outer turmoil and chaos. It is this chaotic state that Xunzi 
condemns. This strong disapproval lets him declare human nature to be bad, be-
cause it lacks standards that allow for a proper ordering of desires, a step that is 
crucial if we are to avoid disorder.
14
So far, so good, but why should we think that manners are of relevance for creating 
order at the individual level? One way in which manners are essential, according 
to Xunzi, is that they give us some standard on the basis of which we can decide 
which of our desires we should follow. What we thus need is some understanding 
of which of our desires are permissible (ke 可) and which are not (Hutton 2014, 
243). Manners provide us with a standard against which we can judge the appro-
priateness of our desires, and on the basis of which we can decide which of our 
desires to pursue.
We can now see the difference between Xunzi’s and Kant’s conception of human 
nature and how it figures in their reflections on the artificiality of manners. While 
Kant focusses on the fact that we may be tricked into pursuing what is not allowed 
13 I am picking up on a very similar interpretation of the passage suggested by A. C. Graham. Gra-
ham gives us a very fitting description here: “For Hsün-tzu human nature is bad in that desires 
are anarchic, in conflict both within and between individuals” (Graham 1989, 248). For a further 
interpretation along similar lines see Stroud’s text on Xunzi and Kant (Stroud 2010).
14 My interpretation here owes much to Chenyang Li (Li 2011), who also stresses the fact that de-
sires cannot in themselves be declared as moral or immoral. He distinguishes between the talk of 
badness in a world that is still devoid of any standards, and in a world in which the standards are 
already in place. He sees Xunzi’s talk of human nature as bad as meaning that it is naturally bad 
(i.e. bad in the sense that it inhibits the satisfaction of our desires). Once the standards are put into 
place, we can look back at this point in time and declare human nature as bad from our current 
(moral) standpoint (ibid., 60). 
Azijske_studije_2017_1_FINAL.indd   184 30.1.2017   14:01:12

185 Asian Studies V (XXI), 1 (2017), pp. 173–192
in spite of knowing better, Xunzi stresses the chaos into which our uncontrolled 
desires may lead. Xunzi does not claim that in blindly following our desires we 
are acting against better knowledge. Initially, this cannot be the case, because in 
the state of nature he presumes, there is no standard according to which we could 
know better. 
Xunzi thus suggests that without li there are no standards which allow us to 
fend off the anarchy into which our uncontrolled desires lead. The disorderly 
nature (not the inherent badness) of our desires as well as our inclination to 
blindly follow them is what is bad about them, and this is what makes it nec-
essary to install standards which permit us to weigh our desires, sort them into 
good and bad, important and unimportant, and decide upon which of them to 
act. Xunzi’s reflection on human nature can also be understood as an argument 
against the first artificiality claim, namely the idea that manners disrupt our 
spontaneous ways of acting.
15
 This indicates that Xunzi, like Kant, thinks that 
deliberate effort is needed to make up for the problematic aspects of human na-
ture. This also becomes clear in a further passage, where Xunzi explains: “Thus, 
I say that human nature is the original beginning and the raw material, and de-
liberate effort is what makes it patterned, rendered, and exalted” (Hutton 2014, 
210). Deliberate effort is needed to bring order to our own nature and to the 
social world that surrounds us.
16
Xunzi takes manners to be central for the creation of individual and social order, 
because they allow us to differentiate. In the social world, these differentiations 
take the form of distinctions between different social ranks. He thus explains: 
“In li (禮), noble and lowly have their proper ranking, elder and youth have their 
proper distance, poor and rich, humble and eminent, have their proper weights” 
(ibid., 84–85). At the same time, he also suggests that on the individual level li 
allows us to differentiate between what is permissible and impermissible, and also 
between other aspects such as good fortune and bad fortune which may result 
from our desires (ibid., 245). These differentiations in turn allow us to adjust our 
conduct appropriately. 
Here, a crucial difference between Kant and Xunzi comes into view. While Kant 
stresses the communicative and activating function of manners, Xunzi’s concern 
is with manners in so far as they allow us to bring order to an otherwise chaotic 
world. With this focus on manners as crucial for creating social and individual 
15 For a recent publication stressing this aspect of the Xunzi, see Puett and Gross-Loh (Puett and 
Gross-Loh 2016). 
16 The fact that both Kant and Xunzi stress the importance of human activity with respect to our 
desires and inclinations is also stressed by Stroud (Stroud 2010).
Azijske_studije_2017_1_FINAL.indd   185 30.1.2017   14:01:12

186 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
order, artificiality claim number four (the claim that hardly plays any role in Kant’s 
theory) comes to the forefront of our attention. Because Xunzi understands man-
ners as creating order both on the individual and the social levels, he must rebut 
the criticism that the order thus constructed has no grounding in reality. 
To counter this line of criticism, Xunzi must insist that it is not any sort of order-
ing that will do. What must be installed in society and the individual is rather the 
‘right kind’ of order, and this is supposedly an order that has some grounding in 
objective facts. Thus Xunzi claims: “One’s virtue must have a matching position, 
one’s position must have a matching salary and one’s salary must have matching 
uses” (Hutton 2014, 85). Xunzi stresses that one’s position in the social world is 
not without grounding in one’s own character and achievements, but rather ideal-
ly has a basis in one’s own character and virtues.
As Janghee Lee has elegantly shown, this position can be combined with Xunzi’s 
view on the rectification of names. Lee suggests that names in Xunzi are conven-
tional, in the sense that they are agreed upon by the people using them, but are 
nevertheless not to be understood as totally arbitrary. They cannot be, because 
Xunzi takes care to stress that good names (shanming 善名) should not be at “odds 
with the things” (bufu 不拂) (Knoblock 1994, 22.2 g).
17
 Lee stresses that there is 
an analogy here with measurements (an idea also found in the Xunzi). While the 
metric unit we establish in a society is conventional, no metric system we develop 
should be at odds “with the world” (Lee 2005, 61–62). In a similar vein, we might 
say that manners are conventional, but not arbitrary, because the differences they 
establish must be grounded in reality.
Going back to the original discussion of different kinds of artificiality arguments, 
we can see that Xunzi tries to counter arguments (1), (2), and (4). But what about 
argument (3), the claim that manners are artificial in the sense of being deceitful 
or insincere? 
In my view Xunzi is not concerned with this aspect. There is no detailed, ex-
plicit discussion of whether manners must always present us with a truthful ex-
pression of our inner feelings and intentions. And, indeed, the omission makes 
sense in view of what I have presented thus far. One could suggest that Xunzi 
is primarily concerned with the ordering function of manners and not with the 
17 Note that I am using Knoblock’s translation here. Hutton sees the passage as implying that the 
names should “not conflict” with one another (Hutton 2014, 239). I think that the difference be-
tween the two translations is not as great as it may seem at first. Thus, one would generally assume 
that names that conflict with one another are also at odds with the world. Furthermore, it seems 
that while there could be internally coherent systems of naming that have no basis in reality what-
soever, these would be extremely difficult to uphold. As such, it does seem plausible that Xunzi has 
both internal coherence as well as external congruence in mind. 
Azijske_studije_2017_1_FINAL.indd   186 30.1.2017   14:01:12

187 Asian Studies V (XXI), 1 (2017), pp. 173–192
communicative one, and therefore that inauthenticity and insincerity are not 
issues he must deal with.
Unfortunately, however, the issue is not quite as clear cut. Indeed, there are some 
sections of the Xunzi where the author does seem to associate some form of com-
municative function with li, and perhaps even seems to hint at issues connected to 
their potential insincerity. In what follows I will briefly discuss the most striking 
of these passages, and show why they do not necessarily commit us to assuming 
that Xunzi does assign a communicative function to li.
Some newer interpretations of the Xunzi stress the importance of make-believe in 
the discussion on burial rituals.
18
 For example, Xunzi claims that in funeral rites 
one treats the dead “as if still alive” (Hutton 2014, 211). Doesn’t this “as-if” way 
of acting suggest that there is, after all, some form of pretense ingrained in the 
rites themselves? And how does this fit together with the idea that li must not be 
at odds with the things? In treating someone dead as though he were still living 
aren’t we deviating from this norm? 
Though the observation that there is an as-if mode of acting is adequate, these 
are not the right conclusions to draw. Amy Olberding has suggested that Xun-
zi’s discussion of funeral rites is due to a particular dilemma we encounter when 
facing death. On the one hand, the dead person is, for us, still a person, a person 
that we grieve for. On the other hand, the dead person is also a mere thing, a 
slowly decaying corpse. Both perspectives are bound up with different disposi-
tions: the disposition to take care of the person and the disposition to flee from 
the corpse. Xunzi sees burial rites as necessarily paying heed to both aspects (Ol-
berding 2015). It is thus necessary that the dead person is treated both as a person 
and as a corpse, and that the rituals in question reflect this. We can thus say that 
in treating someone dead as if he were still alive we are not acting at odds with 
the things, but rather very much in line with them. This double perspective is 
reinforced through the fact that, while dressing the person as if he were still alive, 
there are subtle differences that are put in place such as not securing the hat with 
a hairpin and not fastening the belt (Hutton 2014, 211). This can be understood as 
an acknowledgement that we are not dealing with a person in the full sense of the 
term anymore, namely we are not dealing with someone who partakes in social 
interactions. The make-believe we are engaged in is not really a case of deception 
18 Berkson highlights the frequent usage of the term (ru 如) in the Xunzi and claims: “Ritual partic-
ipation can be seen, perhaps, as a form of sophisticated pretending or play, artificial and invented, 
yet necessary for conflicted and fragile beings such as ourselves.” (Berkson 2014, 120). Note that 
Berkson also does not assume that we are deceived, but suggests that we are fully aware of what is 
going on. I am not rejecting Berkson’s interpretation, but rather only a conclusion one might be 
tempted to draw from it.
Azijske_studije_2017_1_FINAL.indd   187 30.1.2017   14:01:12

188 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
(nor a case of illusion, in Kant’s sense of the term), but rather an acknowledge-
ment of our complicated relationship with death.
There are some further aspects of Xunzi’s discussion of li, which also suggest an 
interest in their communicative function. Olberding thus suggests that manners 
are supposed to beautify our behavior. This beautification is necessary, because 
the “open display of our base needs and interest” may well generate aversion in 
others and therewith undermine human cooperation (Olberding 2015, 157). On 
her reading of the Xunzi, manners are needed to make our behavior sufficiently 
attractive to others to allow for cooperation. But this again seems to bring up the 
following problem: Doesn’t beautification amount to merely hiding what is base 
about ourselves? And, again, isn’t this an act of deceit?
19
I don’t think this is the case. As I have stressed before, I take Xunzi’s li to be a 
standard which allows us to choose non-arbitrarily among our desires. Choice 
here will mean putting some desires into action while leaving others aside. And 
this is, in turn, a crucial step for liberating us from a situation in which we must 
blindly follow any desire that happens to crop up. But in this act of choice there 
is no room for deceit. We cannot “just pretend” to follow one desire or another. 
Rather, in choosing a desire we act on it and outwardly express it. Expression and 
action are not additional acts, but part of the choice itself. This also becomes clear 
in the following passage:
Li (禮) cuts off what is too long and extends what is too short. It sub-
tracts from what is excessive and adds to what is insufficient. It achieves 
proper form for love and respect, and it brings to perfection the beauty 
of carrying out yi. (…) Thus, when the change in disposition and ap-
pearance are sufficient to differentiate good fortune and ill fortune and 
to make clear the proper measure of noble and lowly, close relations and 
distant relations, then li (禮) stops. To go beyond this is vile, and even 
should it be a feat of amazing difficulty, the gentleman will still consider 
it base. (…) It is not the proper patterning of li and yi, it is not the true 
disposition of a filial son. It is rather the behavior of one acting for ulte-
rior purposes. (Hutton 2014, 210) 
Changes in disposition are only supposed to make the differentiations in line with 
li and yi; our outward expressions are simply an expression of these differentiations. 
There is no room here for other motives. Such other motives and intentions can 
only come up when we are no longer acting in line with li. As a consequence, it 
19 This is not Olberding’s explicit position, but rather only a conclusion one could (but need not) draw 
from her description of the beautifying function of manners. 
Azijske_studije_2017_1_FINAL.indd   188 30.1.2017   14:01:12

189 Asian Studies V (XXI), 1 (2017), pp. 173–192
seems that deceit and illusion are unrelated to li. One can thus conclude that the 
claim that manners are artificial in the sense of giving rise to insincere behavior is 
irrelevant when viewed from Xunzi’s position.
Implications for a Theory of Manners
The different versions of the artificiality claim discussed above highlight potential-
ly problematic aspects of manners, and so pose a threat to the claim that manners 
are ethically relevant. Both Xunzi and Kant counter this claim. From a systematic 
perspective, we can say that they do so by employing a similar mix of strategies. 
They respond first of all by giving an account of human nature and effortful action 
which indicates that manners can be ethically relevant in spite of being effortful, 
and despite (at least in the case of Xunzi and Kant’s discussion in the Anthropol-
ogy) their running counter to some aspects of human nature. However, they also 
respond to the artificiality claims by giving a specific analysis of the function of 
manners. While Kant focusses on the communicative function of manners, Xunzi 
assigns them the function of installing a non-arbitrary social order. Through this 
step both authors indirectly block one specific sort of artificiality claim (i.e., the 
claim that they are deceptive in the case of Xunzi, and the claim that they install 
an arbitrary social order in the case of Kant). 
When considered from a systematic point of view, this step is not unproblematic. 
From an everyday perspective, it seems that manners can be both expressions of 
an egalitarian type of respect (a type of respect we owe to everyone independent of 
their social standing) and a means of constituting a specific social order, an order 
that will also include certain hierarchical elements (i.e., putting someone above 
or below us in social hierarchy). And it also seems that, properly understood, both 
aspects are ethically relevant. I seem to be at fault, for example, when jumping the 
queue at a local coffee shop, because I fail to pay due respect to those others wait-
ing in line.
20
 But I equally seem at fault if I fail to pay due respect to a legitimate 
guest of honor at an official dinner I am hosting, where “legitimate” indicates the 
guest is truly deserving of the honor. Both Kant and Xunzi thus limit the scope 
of manners and analyze their function in a way that leads them to ignore one of 
these central aspects instead of acknowledging its relevance.
There are various ways in which one can respond to this issue. At first glance, 
it may seem as if the easiest solution lies in denying that both authors are in-
deed right. That is, we could construct some argument for why we must either be 
20 I take this example from Karen Stohr (2012).
Azijske_studije_2017_1_FINAL.indd   189 30.1.2017   14:01:12

190 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
Kantians or Xunzians on the issue in question. In my view, however, this is not a 
very promising avenue. A brief look at our everyday way of speaking suggests both 
authors are saying something fundamentally right and worthy of attention. So, 
prima facie, we should focus on the ways in which the two fundamental insights 
could be combined. There are two ways in which we might do so. We can either 
assume that what we call ‘manners’ in the everyday sense of the term actually con-
sists of (at least) two different sets of conventional rules with different functions: 
one expressing egalitarian respect and one assisting in building a hierarchical social 
order. Or we can assume that well-mannered behaviour always consists in express-
ing respect towards others as persons, and that additionally some cases of well-be-
haved interaction assist in building a well-ordered society. Independently of which 
option we choose, it seems we must find some way of combining both Kant’s and 
Xunzi’s insights to arrive at a full-blown picture of manners and their functions.
Conclusion
To conclude the paper, let me briefly sum up the main line of argument. I sug-
gested that there are different artificiality claims voiced in criticisms of manners: 
manners can thus be characterized as effortful, deceitful, contrary to human na-
ture, and arbitrary, depending on the notion of artificiality one has in mind. 
Both Kant and Xunzi share some basic views on manners, and therefore respond 
to some of these claims in a similar fashion. Specifically, they agree manners may 
function to counter the more problematic aspects of human nature. Therefore, 
both highlight these features and argue against the notion that human nature is 
good in its own right. For both of them manners present an effortful but worth-
while means of self-cultivation. 
Nevertheless, there are also crucial differences between the two: Kant has a view of 
manners that stresses their communicative function and their egalitarian aspects. 
Manners are signs of a type of respect, which is owed to all persons independent 
of social rank, virtue, and so on. These aspects demand that Kant reflects on the 
potential deceitfulness of manners. But they also relieve him of the need to argue 
against the claim that manners install an arbitrary social order. Xunzi, on the other 
hand, focusses on the individual and social order established through manners. 
Therefore, he is hardly concerned with their potential to deceive and much more 
with the threat of their installing a merely arbitrary order. 
I concluded the paper by suggesting that both approaches are insightful in their 
own right. Nevertheless, both accounts are problematic in the sense that they 
Azijske_studije_2017_1_FINAL.indd   190 30.1.2017   14:01:12

191 Asian Studies V (XXI), 1 (2017), pp. 173–192
limit the scope of manners in ways that are inconsistent with our everyday usage 
of the term. I thus suggested that we should aim at combining the two views to 
arrive at a plausible response to the artificiality claims and a full-blown account 
of manners.
References
Berkson, Mark. 2014. “Xunzi’s Reinterpretation of Ritual: A Hermeneutic De-
fense of the Confucian Way.” In Ritual and Religion in the Xunzi, edited by T . 
C. Kline III and Justin Tiwald, 107–32. Albany: SUNY Press.
Brandt, Reinhard. 1999. Kommentar zu Kants Anthropologie in pragmatischer Hin-
sicht. Hamburg: Felix Meiner Verlag.
Buss, Sarah. 1999. “Appearing Respectful: The Moral Significance of Manners.” 
Ethics 109: 795–826.
DiCenso, James. 2012. Kant’s Religion within the Boundaries of Mere Reason: A 
Commentary. Cambridge: Cambridge University Press.
Eno, Robert. 1990. The Confucian Creation of Heaven: Philosophy and the Defense of 
Ritual Mastery. Albany: State University of New York Press.
Frierson, Patrick. 2005. “The Moral Importance of Politeness in Kant’s Anthropol-
ogy.” Kantian Review 9: 105–27.
Graham, A. C. 1989. Disputers of the Tao: Philosophical Argument in Ancient China. 
Chicago, La Salle: Open Court.
Hutton, Eric. 2014. Xunzi: The Complete Text. Princeton University Press.
Kant, Immanuel. 1968. Anthropologie in pragmatischer Hinsicht. Kants Werke. 
Akademie Textausgabe VII. Berlin: Walter de Gruyter.
–––. 1998. Groundwork of the Metaphysics of Morals. Translated and edited by Mary 
Gregor. Cambridge: Cambridge University Press.
–––. 2003. Die Religion innerhalb der Grenzen der bloßen V ernunft. Hamburg: Meiner.
–––. 2006. Anthropology from a Pragmatic Point of View. Translated and edited by 
Robert B. Louden. Cambridge: Cambridge University Press.
Klemme, Heiner. 1999. “Freiheit des Willens und Herrschaft des Bösen: Kants 
Lehre vom radikalen Bösen zwischen Moral, Religion und Recht.” In 
Aufklärung und Interpretation: Studien zu Kants Philosophie und ihrem Umkreis, 
edited by Heiner Klemme, Bernd Ludwig, Michael Pauen, and Werner Stark, 
125–51. Würzburg: Königshausen und Neumann.
Knoblock, John. 1994. Xunzi. A Translation and Study of the Complete Works Vol. 3. 
Stanford, California: Stanford University Press.
Korsgaard, Christine. 1986. “The Right to Lie: Kant Dealing with Evil.” Philoso-
phy and Public Affairs 15 (4): 325–49.
Azijske_studije_2017_1_FINAL.indd   191 30.1.2017   14:01:12

192 Anja Berninger: Kant, Xunzi and the Artificiality of Manners
Lee, Janghee. 2005. Xunzi and Early Chinese Naturalism. Albany: State University 
of New York Press.
Li, Chenyang. 2011. “Xunzi on the Origin of Goodness: A New Interpretation.” 
Journal of Chinese Philosophy Supplement to Volume 38: 46–63.
Machle, Edward J. 1993. Nature and Heaven in the Xunzi: A Study of the Tian Lun. 
Albany: State University of New York Press.
Olberding, Amy. 2015. “From Corpses to Courtesy: Xunzi’s Defense of Etiquette.” 
The Journal of Value Inquiry 49 (1–2): 145–59.
–––. 2016. “Etiquette: A Confucian Contribution to Moral Philosophy.” Ethics 
126 ( January): 422–46.
Palmquist, Stephen R. 2016. Comprehensive Commentary on Kant’s Religion within 
the Boundaries of Bare Reason. Chichester: Wiley Blackwell.
Puett, Michael, and Christine Gross-Loh. 2016. The Path: A New Way to Think 
About Everything: Viking.
Random House Webster’s Unabridged Dictionary. 2001. 2nd ed. New York, Toronto, 
London, Sydney, Auckland: Random House.
Stohr, Karen. 2012. On Manners. New York, Abingdon: Routledge.
Stroud, Scott R. 2010. “Desire and the Project of Moral Cultivation: Kant and 
Xunzi on the Inclinations.” In Cultivating Personhood: Kant and Asian Philos-
ophy, edited by Stephen R. Palmquist, 639–52. Berlin, New York: Walter de 
Gruyter.
Wood, Allen W. 2014. “The Evil in Human Nature.” In Kant’s Religion within 
the Boundaries of Mere Reason: A Critical Guide, edited by Gordon Michalson, 
31–57. Cambridge: Cambridge University Press.
Azijske_studije_2017_1_FINAL.indd   192 30.1.2017   14:01:12