Everyday Hospitality in Mongolia:

Obligation, Enaction and Projects of Governance.

David Sneath

Introduction: non-reciprocal hospitality

For much of the 20th century ‘exchange’ became a category for universal

anthropological application; used to describe all sorts of material and immaterial

transfers. Following Mauss, the notion of reciprocity was central to the exchange

concept in ethnographic analysis, so that Sahlins’ popular (1972) exchange model

described all sorts of practices, from sharing to theft, in terms of a spectrum of

‘reciprocity’. Davis (1992:29), for example, lists an A-to-Z of 42 types of ‘exchange’

recognised in Britain, from ‘altruism’ to ‘wholesaling’. But such a sweeping

application of the exchange model can be problematic. It implies some sort of

reciprocal relationship between the parties (Gregory and Altman 1989:203-204), and

perhaps a folk theory of human relationality, so that transfers that are unambiguously

one-way represent a sort of contradiction, and the ‘pure gift’ became a problem in

need of explanation (Parry 1986:453-73, Davis 1992:9-27). It has, of course, long

been obvious to anthropologists that a good deal of behaviour could not be treated as

transactional, but, as Hunt (2002:115) notes, the exchange idiom has nevertheless

remained pervasive in discussions of allocation.1

1
Gudeman (2001:80), for example, notes that accepted anthropological wisdom has made reciprocity
the primary building block of community.

1
As Candea and da Col (2012:1) remark, the examination of hospitality offers a

vantage point to explore alternatives to the exchange logic of the Maussian gift.

Unlike the gift, which lends itself rather easily to individual and transactional logics,

hospitality is more clearly bound up with wealth and status within a particular

political order. I argue that, in the light of post-Weberian theories of power, it is the

political order that emerges as the most convenient unit for anthropological

comparison, rather than the essentially 19th-century concepts of a bounded culture or

society.

Mongolian everyday hospitality

Hospitality is an everyday part of life on the Mongolian steppe. Visitors to any rural

ail (encampment) expect to be offered drink, food, and, if need be, a place for the

night.2 Guests can be friends, acquaintances, complete strangers, or foreigners;

neighbours or travellers passing through for one reason or another. In many cases

visitors are from sufficiently far away that all parties understand that there can be no

reciprocal visit; the householders will not make a visit in return. But this is not

important. Old friend or complete stranger, any visitor to a pastoral household can

expect to rest and receive tea, dried curds, and other snacks, and often a meal and

overnight stay if they need them. If consumables are visible, then a guest may simply

help themselves to them, and the way to ask for something is just to ask if there is any

(bainuu?). Since anyone can expect such hospitality without invitation, these practices

2
The ‘unit of hospitality’ is usually, but not necessarily, the encampment as such, but is often a form of
stem family – a parental couple and one or more children with their spouses and children, if any (see
Sneath 1999:139-147). In most cases the encampment is comprised of one such stem family living in
several gers, but may (particularly in summer) include other such families, friends, or dependents.
Generally each stem family is counted as an örkh (household) and could be a unit of hospitality.

2
resemble Adam Chau’s notion (this volume) of xeno-hosting (the hosting of

strangers), although it is also routinely extended to friends and relatives.3

Rather than presuming exchange, we can see these transfers of goods and assistance

as materialisations of the social roles and relations engaged by the visit. As I have

argued elsewhere that (Sneath 2006), rather than as transactions, these routine

material transfers to others can be better seen as enactions of certain aspects of

persons and roles. Hospitality behaviour (zochlomtgoi zan) enacts the role of host or

hostess as an aspect of the householder’s social persona.

Similarly, the guest enacts his or her role as an appreciative recipient by accepting at

least a token of what is offered. This does not resemble the Maussian gift; the

obligations to give and receive emerge from the social settings of host and guest, not

the reattachments of the objects involved, and the transfer itself carries no obligation

to reciprocate. The social setting allows for improvisation and personal expression.

Someone wanting to impress with his or her generosity might choose to act as a

superlative host and provide much more than the usual level of hospitality – offering

luxuries such as cigarettes, vodka, and meat. Conversely, no offence is taken if there

are evident reasons as to why the hosts are unable to provide proper hospitality, but

offer some token like cold tea instead. The main point is that those concerned were

willing to enact their roles, not the value of the transferred materials.

Mongolian everyday hospitality, then, stands in contrast to Pitt-Rivers’ (1968) notion

of ‘turn-taking reciprocity.’ It has more to do with expression than with exchange. In

3
I use the term ‘everyday hospitality’ to indicate the practices directed at all visitors in this way, so as
to set it aside from other kinds of hospitality, such as that directed toward specifically invited guests,
and visitors for special occasions such as the lunar New Year (Tsagaan Sar).

3
her 1992 monograph Inalienable Possessions: The Paradox of Keeping-While-Giving,

Annette Weiner argues that the notion of reciprocity, derived as it was from the

practices of emerging mercantile elites of early-modern Europe, has dominated

anthropological analysis of ‘exchange’ at the expense of considering “other, much

older economic principles in Western history” (Weiner 1992:32). In the case of

Malinowski’s celebrated account of the kula, she argues, Malinowski’s mistake was

to present the analysis in terms of reciprocal exchange as it had been theorised in the

West.4 But, on reflection, the assumption that reciprocity represented a sort of human

universal seems to be the unwarranted product of a particular intellectual tradition.

Rather than the transactionalist logic of reciprocity, Weiner argues, the kula answers

to the competitive logic of aristocratic status and esteem. 5 The most important kula

valuables are symbols of past glory, just as royal regalia – crowns and robes of state –

are in Europe (1992:37). “Fundamentally, kula is not about reciprocating one shell for

another… In essence, these prized shells are like the famous trophies for which Greek

aristocrats vied…” (Weiner 1992:133). Kula is not about the mercantile logics of

transaction; its roots lie in the distinctive privileges of the Trobriand nobility. 6 As a

thought experiment, Candea and da Col (2012) propose an alternative history of

anthropology, in which Mauss wrote a theory of hospitality rather than the gift. My

own project is to imagine what the discipline would look like today if anthropology

had developed the category of aristocracy (the rule of hereditary elites) as a

4
Weiner (1992:43) writes, for example, “Generations of anthropologists, not looking back far enough
into Western economic history, followed Malinowski and…continue to take for granted, as did Smith
and others, that there is an innate, mystical, or natural autonomy in the workings of reciprocity.”
5
Malinowski (1922:88-91) actually did compare kula valuables with crown jewels, in passing, as
Weiner notes.
6
Weiner (1992:138) and Powell agree that “The kula was traditionally forbidden to the commoners”
(Powell 1951:14).

4
comparative theme, rather than the concept of kinship structure. 7 The kula, it turns

out, was a product of the history of the Trobriand nobility. Here I argue that

Mongolian everyday hospitality is a transformed legacy of the country’s pre-

revolutionary aristocratic order, rooted in the historical transformation of obligations

placed upon householders within a system of power relations.

In her 1987 paper ‘The Host and the Guest: one hundred rules of good behaviour in

rural Mongolia’, Caroline Humphrey describes a list of hospitality rules given to her

by a Mongolian teacher friend. They include such things as the customary exchanges

of greetings as a visitor arrives at an encampment and announces their presence by

calling for the inhabitants to control the dogs that frequently guard them; how the host

should look after the guest’s horse and take responsibility for replacing it if attacked

by wolves; what the visitor can take into the dwelling and what items should be left

outside the home; where and how the parties should sit depending on status; how to

offer and place snuff bottles and pipes; what the host should provide by way of drink,

snacks, and food; when to make gifts if there are any; the positions for sleeping

overnight; how to see the guest off when the time comes for them to leave; and so on.

The rules include the proper bodily postures and gestures of the parties depending on

their rank; in the case of adult children visiting their parents, this includes bowing and

other signs of respect that they should make.

Most, but by no means all, of the rules listed resemble practices carried out, or at least

recognised, in contemporary Mongolia to some extent. Some of the rules were clearly

anachronistic, even when Humphrey received them. Rule 96 is that members of the

7
This is the subject of a forthcoming monograph provisionally entitled The Savage Noble: The
anthropology of aristocracy and revisionist readings of power.

5
host’s household should turn prayer-wheels when the guest departs, but these merit-

making Buddhist devices almost entirely disappeared from the Mongolian landscape

during the state socialist era ushered in when Soviet-backed revolutionaries seized

power in 1921. Indeed, the rules have a pre-revolutionary tone to them, referring to

senior lamas (who had not been permitted since the late 1930s), Buddhist holy objects,

and other items of material culture that had become uncommon by the late 20th

century. Some, such as Rule 3 – a host should raise the felt door-flap for the guest, are

never practised, since wooden doors replaced felt flaps in gers many decades ago.

Many of the rules are recognised, but only practised in the most formal of

circumstances, or to make a point about how knowledgeable a host or guest might be

regarding Mongolian custom. Most of the instructions regarding the postures of

respect that juniors should adopt in the presence of seniors have largely dropped out

of everyday practice.

To treat this as a case of ‘the erosion of culture’ would imply a notion of culture as

continuity, a distinctive system of meanings and practices that is somehow outside

historical time. But, as Humphrey (1987:43) notes, “These symbolic acts cannot be

seen as ahistorical, even if many of the categories they are built from seem to have

been present in Mongol culture for centuries… If we look at their social functions it is

clear that they should be seen as obligations… In these rules we see an important

indication of how rural Mongol society is meant to be.”

The rules, then, are an historical artefact, put together from a range of past sources

and describing memories, sayings and probably the writings of particular people at

particular times that were recalled by Humphrey’s teacher friend. Some of the rules

6
must have been invented in the 17th, 18th or 19th century. The rules mention the proper

placement of pipes, for example, and tobacco was not introduced until the 17th

century. The ceremonial presentation of polished snuff bottles, often made of semi-

precious stone, their handling, appreciation, and return, are common greeting rituals

for hosts and visitors, particularly older men. Such bottles were introduced into the

Qing court during the reigns of the Kangxi and Yongzheng emperors, and met with

imperial approval. They spread among aristocrats and commoners in the 18th century

(Olivova 2005:229). These practices, then, have biographies and authors; particular

persons gave rise to popular forms of action and expectation that were taken up by

others in historical time.

Secondly, these rules are about obligation, rather than reciprocity. As Humphrey

(1987:44) pointed out, in Mongolia this sort of hospitality is required of

householders; it is a duty rather than an act of charity. This does not resemble the

grandly abstracted notion of ‘absolute hospitality’ proposed by Derrida in these terms:

“For if I practice hospitality ‘out of duty’ [and not only ‘in conforming with duty’],

this hospitality of paying up is no longer an absolute hospitality, it is no longer

graciously offered beyond debt and economy, offered for the other, a hospitality

invented for the singularity of the new arrival, of the unexpected visitor.” (Derrida

2000:83). Rural Mongolian everyday hospitality is, however, offered out of a sense of

duty, and is offered to unexpected visitors. Arguably, this could mean that either these

acts are not hospitality; or that all visitors in rural Mongolia are, in a sense, expected,

and so do not represent the singularities that Derrida refers to. But this would be to

privilege the imaginary world of Derrida’s thought-experiment over the everyday

meanings of the words ‘hospitality’ and ‘unexpected visitor’. That Derrida has a

7
notion of a more perfect hospitality, judged by his own contingent aesthetic standards,

does not make it something real in the world, just as Kant showed, contra Descartes’

ontological argument for the existence of God, that existence is not a predicate.

Mongolian everyday hospitality does not imply acts of generosity directed toward any

particular person; it is an expression of the status of the householders and their ability

to fulfil a public norm. Just as one or more members of a household will take pride in

a clean and tidy home or well-appointed interior décor, so the snacks and hospitality

for visitors are demonstrative of family propriety. In some respects, such hospitality

resembles Adam Chau’s notion of hosting in China, where “a recognition and

acknowledgement of social worth is communicated between, and co-produced by,

host and guest…” Chau (2014:496). For Chau, hosting is an expression of household

sovereignty, in which the head is transformed into the ruler of the household as a

microcosmic kingdom. In the Mongolian case, as we shall see, the host can, at best, be

attributed a tightly proscribed and nested sovereignty. Nevertheless, there are some

interesting homologies between the position of sovereign and that of host that are

worth exploring.

Hosts and masters

The term used for the host with respect to domestic hospitality is geriin ezen, literally

the ezen of the yurt. The concept ezen (plural ezed) is complex and revealing. Its most

common English translation today is probably ‘owner’, but historically it could be

used to refer to positions we would describe as ‘master’, and ‘lord’. This is a concept

8
that applies to proprietorial authority at different scales. It is used at the level of the

domestic group (the household); at the level of some large resource or enterprise

(land, a herd of animals, or a factory may have an ezen); and in the past it was also

applied at the scale of the whole polity – the common term for the Manchu emperor

was Ezen Khan. Historically, most ezed seem to have been men, but a female head of

household is also referred to as the geriin ezen, and an old name for the mistress of the

house was togoo tulgan ezen (literally ‘ezen of the pot and hearth’).8 The ezen concept

was a fundamental element of the aristocratic order of pre-revolutionary Mongolia.

The lords ruling the khoshuu – the principalities or territorial divisions of the Mongol

lands – were officially termed nutagyn ezen ‘lords of the homeland’ among their other

titles (Natsagdorj 1967:267). Indeed, proprietorial authority was so central to the

notion of social order that to be masterless was to be wild or chaotic. The term

ezengüi baidal (literally ‘a situation without an ezen’) meant ‘anarchy’.

Such authority extended to the spiritual world, in which ezen was used to describe the

spiritual ‘masters’ of the land. The eight classes of spiritual lords of earth and water in

Buddhist cosmology are known as sabdag (from the Tibetan sa bdag), as well as by

the term gazaryn ezed ‘lords of the land’, and were honoured at annual or twice yearly

ceremonies attended by secular and religious authorities in the various administrative

districts throughout the realm. These rituals were conducted at ritual cairns (ovoo) and

reflected a cosmology in which humans do not hold land as they do other mundane

possessions, but must enter into relations with the spiritual powers of the locality to

ensure favourable conditions. Many, but not all, of the gazaryn ezed were associated

8
The phrase ezen bolokh (to become an ezen) means to vouch for something or someone, or to take
responsibility for them. This is the usage in which it most resembles our term ‘patron’ – as someone
who backs a junior. Another example of the similarity of the concepts of ezen and patron is the
Mongolian term ezengüi khüükhed (a child without an ezen), which means an illegitimate child.

9
with sacred mountains, and prayers to them typically sought protection from illness,

plague, drought, storms, cattle-pest, thieves, wolves, and asked for long life, increased

herds, and good fortune (Heissig 1980:105-106; Bawden 1958:38-39; Tatar 1976:26-

33).

In her study of Latin American indigenous politics, de la Cadena (2010:336) makes

use of the term ‘cosmopolitics’, not in the Kantian sense of the politics of global

citizenship, but to mark the intersection between the cosmological and the political,

where non-humans, such as spirits and deities, exist as actors in the political arena.

During the Qing period, the wider order was cosmopolitical in this sense. The

gazaryn ezed local deities were engaged in the project of imperial and aristocratic

rule, within the realms of the Qing emperor, who was himself a bodhisattva (Farquhar

1978). The most important mountain deities held official aristocratic ranks, for

example, and had the corresponding salary allocated to them from the imperial

treasury (Bawden 1968:102). The most senior officials could hold local deities

responsible for their behaviour. One senior Qing resident official in Urga (present-day

Ulaanbaatar) was apparently caught in a heavy storm while carrying out rites at the

nearby Bogd Uul sacred mountain, and responded by punishing the deity, whipping

and depositing fetters on its ovoo; later he fined it by confiscating its horse herd

(Zhambal 1997:16). The spiritual and the political were entirely commensurate

categories that could, and sometimes did, directly interact.9 Although the old order
9
Much of the historical material we have concerning the rituals for local deities concerns senior
secular and ecclesiastical figures. It is more difficult to know how commoners were expected to relate
to spiritual ‘masters of the land’, or how they viewed the ovoo ceremonies. It seems that, in the 17th
century, the rites explicitly reference status hierarchy, with royal, noble, and common peoples
worshipping at different ovoos placed in different locations, although this separation seems to have
faded away in later centuries (Sneath 2007(b)). And it may be that, for commoners and others who
were involved more peripherally in the rites than senior figures, the ways of local deities and how to
treat them was considered specialist knowledge that they were not required to possess. Lattimore
(1941:245) remarks, “Though all Mongols have a deep feeling of reverence for the obo [ovoo] it is a
vague feeling. I have never met a man who could explain in detail the various acts of worship.”

10
was swept away in the revolutionary changes of the 20th century, the rituals for local

deities have been revived since the 1990s, and some have been refashioned as state

ceremonies of national celebration, attended by the president as head of state. The

wider order of these contemporary practices is very different, based on the concept of

a national people, and framed in terms of Mongolian culture and tradition, but it could

also be seen to be cosmopolitical in de la Cadena’s sense, since mountain deities are

addressed by state officials (Sneath 2014).

The concept of ezen, as well as much of the content of the category zochlomtgoi zan

‘hospitality’, is rooted in a pre-revolutionary aristocratic and cosmological order that

resembled European feudalism in many ways.10 The proprietorial authority indicated

by the term could be treated as a form of sovereignty; the jurisdiction of the lord over

his estate, the local deities over the landscape. But if so, this was a nested sovereignty,

not an absolute or unconditional principle, but a limited authority that existed within

the overarching sovereignty of the Qing emperor as ultimate ezen. We do not know if

the term geriin ezen was applied to commoner householders in the pre-revolutionary

period (my guess is that this became a convention in the state socialist period), but if

there was any sense in which most rural householders were sovereign over their tiny

household realms, this must have been a proscribed and limited form. The rights of

ruling nobles over their common subjects were extensive and, by 21st-century

standards, authoritarian.

10
Anthropologists, both Soviet and Western, rejected the use of the term ‘feudalism’ for a number of
‘nomadic pastoral societies’, including Mongolia. Vladimirtsov’s classic 1934 study of Mongolian
society had described it as ‘nomadic feudalism’, but this approach gave way to the advance of a tribal
discourse based on the notion of kinship society that reproduced the old conceptual apartheid by which
the colonised were subject to the primitivist language of tribe and chief. In retrospect, the similarities
between the Mongolian aristocratic order and that of medieval Europe are striking, and are discussed at
length in Sneath 2007(a).

11
But there are a number of ways in which hospitality ceremonial resembles the courtly

forms employed by the aristocracy of the past. The domestic space of the ger (yurt) is,

at least in principle, tightly ordered. The placement of objects and bodies materialises

a social order, both reflecting evaluations of relative status and, to some degree,

producing it – at least within the confines of the home.

The ger is pitched so that the door faces south, or sometimes south-east. This

orientation is so naturalised that the Mongolian words for ‘ahead’, ‘behind’ ‘left’ and

‘right’ are the same as those for south, north, east and west respectively. The space

within the ger is conceptually divided using two principles. Firstly, there is that of

status. High-status people and things are placed in the northernmost part of the ger in

an area known as the khoimor. The more junior an object or individual, the further

south – towards the door – they are placed. The second principle is that the western

hemisphere of the ger is associated with public, external, and to some extent male

items, while the eastern hemisphere is the appropriate place for domestic or inside

things, which in some contexts are associated with women. In diagram (a), positions

are marked 2, 3, 4, and 5 in the west, and B, C and D in the east in order of decreasing

status.

12
In formal situations, there is a recognised system for where people should sit

depending on seniority. The highest status position for both people and objects is [1A]

on the khoimor opposite the door, where there are generally rugs. Guests (of all

genders) will be seated in the west half of the ger. Depending on their seniority, in

formal situations they will be placed nearer the khoimor or the door. Unless the

women know the resident family so well as to be in some way members of it, female

guests will also usually sit in the west half of the ger. The cooking area, however, is a

work zone, and is somewhat separate. Junior and senior women generally use this

portion of the ger, because they are charged with food preparation rather than status

position in an abstract sense, and senior women will often sit on the khoimor. Objects

are also organised spatially: valued and delicate items sit behind the khoimor,

including family photographs, religious items, a radio or TV; kitchen items go in the

south-eastern cooking zone; saddles (if brought into the dwelling at all, which is rare)

13
go in the south-west by the door (although valuable silver-chased bridles are generally

hung behind the khoimor by the rifle, if there is one), and so on. It is a highly practical

set of arrangements: the khoimor with its carpets or rugs is the cleanest and quietest

part of the dwelling, furthest from the doorway, where constant entering and exiting

tramps in dust and dirt. The cooking zone by the door allows for water, waste, and

fuel to be brought in and taken out directly, and guests know to turn to their left upon

entry to keep out of the way of domestic activities.

This spatial orientation and order of the ger appears to be one of the oldest elements

of the contemporary practices. The earliest accounts we have of something akin to this

ordering dates from the 13th century source, the Secret History of the Mongols

(Mongqol-un niucha tobcha’an), which describes a Mongol noblewoman insulted by

being placed near the door (Pelliot 1949:§101).

Once seated, guests are served drink, usually tea, and then often snacks (usually from

a single large bowl) in order of seniority. Snuff bottles may be offered, appreciated,

and returned by people on each side. This will also be done in order of seniority, as

will the offerings of other drinks such as airag (fermented mare’s milk), vodka, or

home-brewed milk alcohol, either in a single bowl (which should be sipped from and

returned) or small shot glasses (typically used for vodka). This pecking order also

appears in the 13th-century Secret History: a prince of the imperial house was insulted

by the order of drinking at a banquet, and the dispute became so serious that it had to

be referred to the emperor for settlement.11

11
Pelliot 1949:114

14
Whether the practices of the Mongolian aristocracy represented a courtly version of

the vernacular practices of their subjects or, as I think more likely, the practices of

ordinary households were imitative of aristocratic ceremonial, we can recognise the

political significance of these forms of social ranking. In such ritual, the household

resembles a miniature court, in which the visitors are treated as envoys; properly

placed and ranked in terms of distance from the sovereign host. In his study of Qing

courtly guest ritual, James Hevia notes: “Ritual practice appears radically historical

when understood as the processes by which embassies from lesser kings were

evaluated, differentiated, and incorporated into an encompassing imperial sovereignty

that included the Cosmos” (Hevia 1995:217).

Returning to Humphrey’s point that, in the obligations placed on host and guest, we

can see an “indication of how rural Mongol society is meant to be”, we can pose some

further questions: who meant Mongol society to be this way, and who placed these

obligations on householders and visitors? We might look perhaps, like Ortner (1978),

to some notion of ‘community’ as the author of these norms and practices. But the

vague and faintly egalitarian notion of ‘community’ is a problematic concept,

critiqued by Young (1986) and Delanty (2003) among others, and best treated as “an

aspiration envisioned as an entity” (Creed 2006:20) rather than a reality. It makes

better sense to assume that the authors of these practices were persons, individuals

and groups, living in historical time. Of course, the identities of such authors are

likely to be forever lost, but we can gain an insight into whose interests were served

by the ubiquity of such hospitality practices, because some of them are recorded in the

law codes by which commoners were governed.

15
Laws of hospitality

The early 18th century Qing-era Mongol law code, the Khalkha Jirum, describes a

wide range of duties of subjects, and the penalties for not meeting them. 12 These

include the legal requirements of householders to provide transport for the travel of

aristocrats and the senior Buddhist reincarnate lama, as well as their

envoy/messengers (elch). Patriarchy was enshrined in law; children were obliged to

honour and obey their parents, and could be punished for insulting them. This

legislation makes the household an explicit site of political regulation by the dynastic

state. The refusal of hospitality for the night was punishable by a fine of one three-

year-old stallion (Riasanovsky 1965 [1937]:114). That such refusals nevertheless took

place, however, is clear from the list of compensation fines that should be levied on a

householder depending upon the frostbite injuries or death of a stranger refused

overnight hospitality.

These laws were not ‘extraneous’ innovations introduced by the Qing. The Khalkha

Jirum closely resembles the pre-Qing Mongol law codes that have survived the

intervening centuries. The 1640 code, the Monggol-Oirad tsaaji, lists the laws agreed

between two sets of steppe rulers, the Mongols on the one hand, and the Oirats on the

other - a Mongolic-speaking aristocracy ruling western Mongolia and part of what is

now Xinjiang. It describes, if anything, even greater powers among the aristocracy

than those found in the Khalkha Jirum. This was at a time when both the Mongol and

Oirat noble houses did not recognise a clear overlord or emperor. Aristocratic power,
12
Commoners were subject to quite tight control; they required permission, for example, to be allowed
to live in an encampment other than the one that they were allocated to (Riasanovsky 1965
[1937]:113).

16
however, is unmistakable: the code makes clear whose rule the code represents – the

distributed sovereignty exercised by the Mongol and Oirat steppe aristocracies.

Section 18 reads: “If nobles holding offices and … [other nobles and officials] … beat

a person for the sake of the lords’ (ejed) administration, law and order, they are not

guilty, even if they beat someone to death.” 13 Here the plural form of ejen (lord) is

used for the “the lords’ administration”; the laws refer to the nobility’s joint

government of subjects.14 Aristocratic power in Mongolia is much older than this,

however, dating from the 3rd century BCE at the least, and the 13th-century dynastic

empire founded by Chinggis Khan shared many of the same features with later

polities. We have only fragmentary and indirect evidence for the Chinggisid law

codes, but the Mamluk source al-Maqrizi (1364-1442) notes that one Mongol law

decreed that any traveller could join those who were eating without permission

(Riasanovsky 1965 [1937]:84).

Although the origin of aristocracy as an institution may be lost in the mists of time,

the emergence of particular aristocracies is certainly not. The Borjigin nobility that

ruled most of Mongolia from the 13th to the 20th centuries were the descendants of the

royal house of the conquest dynasty founded by Chinggis Khan. It was this nobility

that chose to swear fealty to the Qing in the 17th century and intermarried extensively

with the royal Manchu house of Aisin Gioro.

If we can speak of hospitality practices as embodying micro-hierarchies of power

(Zito, this volume), or as part of the habitus of rural Mongolian households, these are
13
The full passage reads: “If nobles holding offices and tabunangs (son-in-law of a noble), junior
nobles and tabunangs, demchi (head of 40 households), shigülengge (head of 20 households) beat a
person for the sake of the lords’ (ejed) administration, law and order, they are not guilty, even if they
beat someone to death.”
14
This distributed aristocracy and what I term ‘the headless state’ is explored in Sneath 2007(a).

17
forms that have been powerfully shaped by historical power relations. To explain their

presence we need not look to the hidden logics of cultures as symbolic systems. The

rulers of Mongolia found it useful to have their subjects offer a certain level of

support to those travelling in their dominions, just as they found it necessary to have

taxes, corvée labour requirements, and orderly relations between their subjects. Of

course, this is an account of only a part of the historical process by which these

particular practices took particular forms; relations of power, or power-knowledge

(pouvoir-savoir) following Foucault, must work with, and through, the affordances of

human psychology. But the heuristic study of power-relations offers the chance of

identifying something resembling structuring structures, since projects of governance

generate explicit normative schemas and techniques for orchestrating compliance, if

not consent.

Since the destruction of aristocracy in Mongolia, hospitality has been reframed as

‘tradition’ and ‘national culture’ (ulamjlal, ündestnii soyol) within the modernist

nation state. The Mongolian People’s Republic received the Leninist version of

modernism and all the trappings of the Soviet version of the European nation-state –

secular education system, health service, urbanisation, industrialisation; together with

the ramifying intellectual apparatus to go with it. This included the modernist

conception of ‘national tradition’, and a state-sponsored intelligentsia set about

editing together elements drawn from pre-revolutionary Mongolian society to form an

official national heritage, including music, art, literature, and material culture. The

100 rules that Humphrey was given fits well within the logic of this project of

national construction – the establishment of definitive Mongolian customs, pictured as

inherent features of a sort of essential national culture.

18
The household (örkh) represents a sort of micro-domain; it is recognised as such in

state discourse, where it is a census and administrative unit. The nested, proscribed

sovereignty of the head of the household now appears within the macro-domain of the

Mongolian nation-state, the ultimate holder of the sacred territory and culture of the

saikhan ekh oron ‘beautiful mother-land’ enshrined in literature, song, poetry, and

art.15 Hospitality is now seen as a national virtue of ‘the Mongol people’, one that

rural residents enact in their everyday treatment of visitors and guests.

Through hospitality ritual, the roles of host, guest, dependents, and attachments are

made manifest through actions and emplacements within domestic space, in

compliance with political authority. As an item of law, hospitality is applied to legal

subjects. When Derrida critiques Kant’s treatment of hospitality in his 1795 essay

Perpetual Peace: A Philosophical Sketch, he notes that Benveniste writes that, in

receiving a foreigner, the host asks their name, so that, Derrida argues, “you are

responsible before the law and before your hosts, you are a subject in law” (Derrida

2000:27). For Derrida this falls short of an absolute hospitality that would offer

“unquestioning welcome” (ibid:28). But the position that Derrida rejects – the notion

of hospitality as a mode of incorporation into the sovereignty of political authorities

and their legal codes – matches the Mongolian case rather well.

Are we perhaps dismissing the notion of traditional culture too readily? Might there

not be a longstanding core to hospitality in Mongolia that was enshrined in law but

15
The influential poem Minii nutag (My Homeland) written by Dashdorjiin Natsagdorj, one of the
founders of Mongolian literature, was published in the journal Mongol Ardyn Ündesnii Soyolyn Zam
(National Cultural Path of the Mongolian People) in 1934, for example, and later work included
Gaitav’s (1978) book Saikhan ekh oron.

19
reflected deeper longstanding cultural forms? Interestingly, it seems that Mongolian

everyday hospitality was in some respects rather different in the Qing period than in

the contemporary era, in ways that directly reflect the legal codes of the time. The

Khalkha Jirum made the offer of overnight hospitality compulsory, but it did not

stipulate that hosts must provide their guests with food. James Gilmour was an

intrepid Scottish missionary who spent many years in Mongolia in the 1870s and

1880s. Gilmour described the hospitality provided by pastoral households when he

was travelling with his local guide.16 They were given space in the ger overnight, but

not food (and sometimes not even tea), which they were expected to bring with them.

The householders, many of whom were very poor, would cook the food and make the

tea for their guests, and would then consume whatever was left over themselves. 17

Gilmour found that his party was:

“better treated, and received with a much warmer welcome, in the tents

of the poor than in the abodes of the rich. A rich man would make us wait

upon his convenience, and expect us to make extra good tea or a meal

which, both as regards quantity and quality, would be in keeping with his

dignity and status, and even then we left feeling that our visit had been

something of an intrusion. In the tents of the poor, on the other hand, we

16
In many ways, the hospitality practices that Gilmour describes resembled those set out in the 100
rules and later practices. He mentions the customary calls of visitors approaching an encampment for
residents to control the guard dogs, the seating of persons depending on status within the ger, the
exchange of snuff bottles, the routine serving of tea, and the hosts offering a plate of dairy products as
snacks, in ways that closely resemble later practice (Gilmour 1883:126-128).
17
Gilmour’s writings also give a sense of the range of people likely to be ‘on the road’ at the time, and
as such the potential recipients of everyday hospitality. Caravans of both camels and ox-carts were
common (1883:125; 1893:77, 78, 88, 91, 93, 94); so were lamas and pilgrims, the poor often on foot
(1883:121; 1893:142, 130). He writes, for example, (1883:121) “a vast amount of foot travelling is
done. A large proportion of the travelling on foot is that of poor men who go on religious pilgrimages.
Foot-travellers, for the most part, trust to the hospitality of the inhabitants of the districts through which
they pass for lodgings…” As is the case today, some visitors were looking for straying livestock
(Gilmour 1893:218).

20
 were warmly welcomed, our tea or food was prepared at once and in all

haste, our animals were looked to as they grazed, the share of food which

we left in the pot was considered a rich reward, and when all was over we

were conducted forth and sent on our way again with many expressions

of friendship and good wishes for the prosperity of our journey.” Gilmour

1883:81.

For Gilmour, the most hospitable hosts were the subjects that ranked lowest in the

social order, those most liable to comply with the requirements of social seniors, and

most in need of leftover food and tea. Although clearly complying with legal

requirement and public expectation, the ‘hospitality’ of the rich appeared grudging

and ungenerous in comparison.18 This echoes, to some degree, the oldest accounts we

have of Mongol hospitality, the Latin texts written in the 13th century by William of

Rubruck and Giovanni da Pian del Carpine, friars who made independent diplomatic

missions to the Mongol imperial court. These both complain of the meanness of their

hosts (Beazley 1903:121, 200), although the keenly observant Rubruck describes the

organisation of interior space in the ger in terms that match that of the present day.

The core continuity that we see here is one of order; the placement and ranking within

social space, which is not unique to Mongolia, but was found historically in many

parts of Inner Asia (Sneath 2007(b)).

The particular requirements of historical hospitality codes appear to have been

products of projects of governance, but there is little doubt that being hospitable was

18
Gilmour’s comments seem to match the remarks of Sechen Jagchid, who grew up in Inner Mongolia
in the 1920s and 1930s. “Good treatment of travelers seem to be a Mongol custom from ancient times,
at least from the period of Chinggis Khan. Still, a stranger who approaches a yurt on the steppe often
gets the impression that a Mongol host is quite cool or reserved compared with people in other parts of
the world.” (Jagchid and Hyer 1979:131).

21
also highly valued and could be seen as a source of pride (see Humphrey, this

volume). Gilmour notes both the obligatory nature of everyday hospitality and the

strong normativity that surrounded it: “Any traveller is at perfect liberty to alight at

any village [encampment] he may wish and demand admittance; and any Mongol who

refuses admittance, or gives a cold welcome even, is at once stigmatised as not a man

but a dog.” (Gilmour 1883:128). He also notes the awareness that this form of

hospitality may not be reciprocated: “Mongols sometimes complain of Chinamen,

who come to Mongolia, enter their tents, and receive their hospitality, but who, when

their Mongol friends go to China, will not let them enter their dwelling-houses”

(ibid:129).

Nevertheless, historical hospitality behaviour was clearly rather different from

contemporary practices. The relative social status of those involved seems to have

been of central concern. As Lattimore (1941:185-6) notes, “a great many travel

writers and travellers go too far in assuming that the nomad air of assurance and

freedom means a level of equality. Far from it. One reason why the average man

behaves with certainty and poise is because he knows exactly his status and your

own.” High-status guests were expected to present the host with a gift, commonly a

khadag – a ceremonial silk scarf. So it is interesting to note that, in the 1920s,

Forbath, the Hungarian traveller, was told that “when you call on a man you must

bring him a chadak [khadag] without fail. That’s the custom in this country” (Forbath

1936:20). But this does not seem to have been a rule for everyone. Jagchid and

Symon (1979:132) make clear that this was an expectation for people of high status

who might be expected to have access to valuables.

22
Overall, we can say that pre-revolutionary everyday hospitality behaviour matched

the stipulations of the historical law codes, and seem to have been, if anything, less

generous than contemporary standards. How is it, then, that hospitality expectations

and practices have come to resemble their current form?

The new practices seem to have emerged alongside the wider transformation ushered

in by the state socialist regime of the Mongolian People’s Republic. The second half

of the 20th century saw increasing levels of material prosperity in the Mongolian

countryside, and the disappearance of the extreme poverty that prevented some

households from offering guests almost anything other than shelter. By the 1960s,

pastoralists were almost all salaried members of collective (negdel) or state farms

(sangiin aj akhui), and, by any standards, much wealthier than all but the richest pre-

revolutionary commoners. The large state-sponsored enterprises that dominated rural

life were explicitly communitarian organisations that promoted collectivist behaviour

among the workforce. They created new roles and relationships, and a new rural

leadership (see Rosenberg 1982) concerned with the management and display of

wealth.19

In particular, from the 1950s until the 1970s, households were subject to a series of

vigorous soyolyn dovtolgoon ‘cultural campaigns’ to remake practices in the home in

line with modernist conceptions of hygiene and cleanliness (Stolpe 2008). The

possession and use of items such as soap, towels, toothbrushes, toothpaste,

19
For an account of the Buryat variant of the collective farm and the role of leadership in managing
wealth, see Humphrey 1983:300-372.

23
washbasins, and white cotton sheets were installed in Mongolian homes and subject to

an inspection regime of spot-checks and penalties.20 The households judged to be

‘best’ by the new criteria received awards, and the ‘worst’ marks of shame and

humiliation. As Stolpe notes (2008:77), socialist competition (sotsialist uraldaan) was

a key principle in these projects of normative engineering, and, surprisingly perhaps,

this competitive aspect came to be popular among most rural people. The most

successful families “greeted the state inspectors by building social ties with them and

creating a festive atmosphere. Family members displayed their newest pastoral

products and dressed in their finest clothes…” (Marzluf 2017:151). Households

became accustomed to the visits of collective officials of all sorts, senior and junior,

as part of the operation of the negdel, as well as other co-workers and friends. The

Mongolian home was reshaped in this era, then, with novel items installed within and

alongside old forms, and householders held to new standards of modernist

respectability, including cleanliness and generous everyday hospitality. The

hospitality etiquette of the old elite may have been the source for the documentation

of ‘tradition’, but the hospitality of everyday practices reflected the values and

conditions of life of the new culture of the rural collectives.

Conclusion: hospitality and the study of political orders

Pitt-Rivers argued that hospitality rituals “might all be considered as ‘rites of

incorporation’, a variety of the rites of passage through which an old status is

20
Stolpe (2008:72) notes: “the 1969-71 cultural campaign…proclaimed the ‘cultured family’ (soyolch
ail, sometimes soyolch suur’) as its motto. Families could earn this title plus a red pennant by scoring
eighteen points for being in possession of items such as a washstand, tooth brushes, towels, a radio, a
home library and newspaper subscriptions. Organized as a competition, this part of the campaign
became very popular.”

24
abandoned and a new one acquired. In this case it is the status of stranger which is lost

and that of community member which is gained” Pitt-Rivers 2012 [1977]:503. This is

a suggestive point. But rather than the incorporation of the universal figure of ‘the

stranger’ into something conceived of as the ‘community’, in the Mongolian case we

might say that hospitality rituals enact the incorporation of the visitor into the micro-

sovereignty of the host within a wider political order. This is a form of ‘hospitality’

produced by forms of power in which subjects of an encompassing sovereignty are

required to enact their respective statuses as respectable householder and respectful

visitor, ranked into seniors and juniors, as part of a wider project of governance aimed

at ensuring orderly public conduct. In the pre-revolutionary period, this wider

cosmopolitical order was Buddhist and aristocratic; in the contemporary era it is

national, and framed in terms of Mongolian culture and tradition. In both eras, order is

manifest in the rituals of domestic hospitality, in which hosts and guests are

physically arranged into positions of relative seniority and subordination, as a mode

of inclusion into the cosmopolitical and sociopolitical orders of the day.

Hospitality, then, can be seen not as a timeless feature of a holistic culture but as a

product and producer of sociopolitical order, and an artefact of historical projects of

governance. This could be thought of, following Ortner (1978), as generative of

integration, but not the integration of a reified ‘culture’ or ‘society’ conceived of in

autonomous or systemic terms.21 Rather it is the product of historical projects of

rulership and governance that ordered social lives in line with the interests of power-

holders. The integration generated by everyday hospitality was shaped by ruling elites

21
For Ortner, the setting for hospitality is very explicitly “the community” (1978:63). However, for
her, this is “a community of relatively self-contained units, each protective of its property and its social
boundaries…” (1978:68), which therefore requires integration. Hospitality, then, is a means of
achieving this.

25
and reflected the wider cosmopolitical orders that supported them. Good subjects

should conduct themselves properly; and should offer the appropriate amenities to

travellers, just as they should honour their parents and respect their seniors.22

The notion that hospitality represents practices constitutive of political subjects

returns us to Kant’s formula for world citizenship critiqued by Derrida, in which

hospitality is a matter “not of philanthropy but of right” (Kant 1903 [1795]:137)

within a project of global governance, in which men are “citizens of the word” (ibid).

In Kant’s vision, universal hospitality is a condition of worldwide ‘cosmopolitan’ law,

and is a legal and constitutional matter rather than an essential human trait.

Hospitality, then, becomes another normative project that elites and power-holders

may seek to initiate, propagate, and regulate. Its exploration offers an insight into the

nature of such ordering projects.

RÉFÉRENCES CITÉES

Bawden, Charles

1958 <Two Mongol texts concerning oboo worship>, Oriens Extremus 5 : 23-41.

1968, The Modern History of Mongolia. London, Weidenfield & Nicolson.

22
This is not to say, of course, that the practices and norms of everyday hospitality have not created
and maintained social networks in countless ways, or that obligations have not been appropriated as
sources of value and satisfaction by those concerned.

26
Chau, Adam

2014 <Household Sovereignty and Religious Subjectification: Comparing the Idiom

of Hosting in Chinese and Christian Religious Cultures>, Studies in Church History

50 : 492-504.

Creed, Gerald

2006 <The Seductions of Community>, in Gerald W. Creed, ed., The Seductions of

Community: Emancipations, Oppressions, Quandaries. Oxford, James Currey : 1-20.

de la Cadena, Marisol

2010 <Indigenous Cosmopolitics in the Andes: Conceptual Reflections Beyond

“Politics”> Cultural Anthropology 25 (2) : 334-370.

Delanty, Gerard

2003 Community. London & New York, Routledge.

Derrida, Jacques

2000 <Step of Hospitality / No Hospitality> in Anne Dufourmantelle and Jacques

Derrida, Of Hospitality: Anne Dufourmantelle invites Jacques Derrida to respond.

Stanford, Stanford University Press : 75-156.

Farquhar, David

1978 <Emperor as Bodhisattva in the Governance of the Ch'ing Empire>, Harvard

journal of Asiatic Studies, 38 (1) : 5-34

27
Forbath, Ladislaus

1936 The New Mongolia (as related by Joseph Geleta), Lawrence Wolfe (trans.)

London: William Heinemann.

Gilmour, James

1883 Among the Mongols. London, The Religious Tract Society.

1893 More About the Mongols. London, The Religious Tract Society.

Heissig, Walther

1980 The Religions of Mongolia. London, Routledge and Kegan Paul.

Hevia, James

1995 Cherishing Men from Afar: Qing Guest Ritual and the Macartney Embassy of

1793. Durham, Duke University Press.

Humphrey, Caroline

1983 Karl Marx Collective: Economy, society and religion in a Siberian collective

farm, Cambridge, Cambridge University Press.

1987, <The Host and the Guest: 100 Rules of Good Behaviour in Rural Mongolia>,

Journal of the Anglo-Mongolian Society 10 (1) : 42-53.

Jagchid, Sechin and Paul Hyer

1979, Mongolia’s Culture and Society. Boulder Colorado, Westview Press.

28
Kant, Immanuel

1903 [1795] Perpetual Peace: A Philosophical Essay. (Trans.) London, Allen &

Unwin.

Lattimore, Owen

1941 Mongol Journeys, London, J. Cape.

Morgan, Catherine

2003 Early Greek States Beyond the Polis. London & New York, Routledge.

Munkh-Erdene, Lhamsuren

2011 <Where Did the Mongol Empire Come From Medieval? Mongol Ideas of

People, State and Empire>, Inner Asia, 13 (2) : 211-237.

Natsagdorj Shagdarjavyn

1967 <The Economic Basis of Feudalism in Mongolia>, Modern Asian Studies, 1 (3) :

265-281

Ortner, Sherry

1978 Sherpas Through Their Rituals. Cambridge, Cambridge University Press.

Olivova, Lucie

2005 <Tobacco Smoking in Qing China>, Asia Major, 18 (1) : 225-260.

Pelliot, Paul

29
1949 Histoire secrète des Mongols: Restitution du texte Mongol et traduction

Française des chapitres I à VI. Paris, Adrien Maisonneuve.

Pitt-Rivers, Julian

2012 [1977] <The law of hospitality> in Pitt-Rivers, Julian The Fate of Shechem or

The Politics of Sex: Essays in the Anthropology of the Mediterranean. Cambridge,

Cambridge University Press : 94-112.

Powell, Harry

1951 <Fourth Fieldwork Report. Kiriwina, Trobriand Islands, January 1st-March 31st,

1951>, unpublished paper, archived http://trobriandsindepth.com/PDFs/POWELL

%204.pdf

Riasanovsky, Valentin

1965 Fundamental Principles of Mongol Law. The Hague, Indiana University

Publications, Mouton & Co.

Sneath, David

1999 <Kinship, networks and residence> in The End of Nomadism? Society, State and

the Environment in Inner Asia, by Sneath, David and Caroline Humphrey, Durham,

Duke University Press : 136-178.

2006 <Transacting and Enacting: Corruption, obligation and the use of monies in

Mongolia>, Ethnos 71 (1) : 89-112.

2007(a) The Headless State: Aristocratic Orders, Kinship Society, and

Misrepresentations of Nomadic Inner Asia. New York, Columbia University Press.

30
2007(b) <Ritual Idioms and Spatial Orders: Comparing the rites for Mongolian and

Tibetan “local deities”> in Bulag, Uradyn and Hildegard Diemberger eds. The

Mongol-Tibet Interface: Opening new research terrains in Inner Asia. Leiden, Brill, :

135-158.

2014 <Nationalising civilisational resources: sacred mountains and cosmopolitical

ritual in Mongolia>’ Asian Ethnicity 15 (4) : 458-472.

Stolpe, Ines

2008 <Display and Performance in Mongolian Cultural Campaigns> in Pirie,

Fernanda and Toni Huber eds. Conflict and Social Order in Tibet and Inner Asia.

Leiden, Brill : 59-84.

Tatar, Magdalena

1976 <Two Mongol Texts Concerning the Cult of the Mountains>, Acta Orientalia 30

(1) : 1-58.

Vladimirtsov, Boris

1948 Le régime social des Mongols : Le féodalisme nomade. Paris, A. Maisonneuve.

Weiner, Anette

1992 Inalienable Possessions: The Paradox of Keeping-While-Giving. Berkeley,

University of California Press.

Zhambal, Boryn

31
1997 Tales of an Old Lama. Damdinsüren, Tsendiin (ed.) Charles Bawden (trans.),

Tring UK, Institute of Buddhist Studies.

32