Full text of DOI:10.1017/S0010417500002590

Hierarchical Image and Reality: The
Construction of a Tribal Chiefship
PHILIP CARL SALZMAN
McGill University
Image and reality interplay in the chiefship of the Basseri tribe of southern Iran.
We know the Basseri through the influential and rightly esteemed ethnography,
Nomads of South Persia, by Fredrik Barth. Barth’s account of the Basseri chief
is vivid and striking, and, in placing the chiefship within the complex society
of Iran, was an original contribution to the ethnographic literature.And yet there
appear to be certain subtleties of the Basseri chiefship, which, while implied in
the historical and ethnographic information provided by Barth, are not includ-
ed in his explicit description. As a result, Barth overstates, I believe, the degree
of the Basseri chief’s power, and underestimates the extent to which the chief
depended upon the consent of his tribemen.
Image and reality interplay in social life because people present themselves
and describe others pragmatically, in ways that help them to achieve their objec-
tives and goals. The images that people thusly present of themselves and
others—the “clever student,” the “influential professor,” the “honest used-car
salesman,” the “public-spirited politician,” well known to us all—may be partial,
incomplete, or totally false, but usually involve some degree of misrepresenta-
tion. To the extent that the claims people make in constructing their and others’
images do not correspond to their actions and activities, to their practice and prac-
tices, the images diverge from reality. This is the criterion we apply in our every-
day lives, and I believe it is appropriate for understanding other people’s lives.
But the story does not end with the simple distinction between words and ac-
tions, between image and reality. For while the referent of an image may be all
or partly illusory, the presentation of that image to other people is itself social,
intentional, goal-oriented action; it can and does lead others to act, and so the
construction and management of images is itself part of reality. To this extent,
image presentation shapes the worlds we wish to understand, and is thus a le-
gitimate and necessary focus of our attention. Among the Basseri, “the chief-
ship” exists, on the one hand, in images, and on the other hand, in the realities
of image management and other actions. What I explore in this essay is the ex-
tent to which characteristics of the chiefship are best understood as Basseri im-
age management, and to what extent as political action, and the circumstances
that might explain the discrepancy between image and reality.
0010-4175/00/1386–0141 $9.50 © 2000 Society for Comparative Study of Society and History
49

50 philip carl salzman
the position of the chief of the basseri tribe
In Nomads of South Persia, Fredrik Barth describes the chief, khan, of the
Basseri as “the central, autocratic leader of the tribe” (Barth 1961:71), who had
“great power and privilege” (Barth 1961:74). This “power is conceived as em-
anating from him, rather than delegated to him by his subjects” (Barth 1961:71–
72, emphasis in original). Barth (1961:74)—writing in the “ethnographic pre-
sent” about a period prior to 1956—continues:
The outstanding feature of the chief’s position . . . is his power of decision and auto-
cratic command over this subjects. . . . The right to command, to make decisions on be-
half of persons in other tents than one’s own, is a strictly chiefly prerogative. The mo-
nopolization by the chief of the right to command is a fundamental abstract principle of
Basseri social structure.
There was no command structure intervening between the autocratic Basseri
chief and his subordinate subject: “All Basseri are equal in their direct relation
of subordination to the chief, who at any time may give any person an order
which the latter must obey” (Barth 1961:75). The Basseri was thus led, as are
each of the other nomadic tribes of south Persia, by “the omnipotent Khan or
chief” (Barth 1961:90, emphasis added).
This characterization of the Basseri chief as autocratic and all-powerful is, I
believe, misleading, for—to anticipate my argument below—it is based rather
too much upon style and image, and too little upon substance and reality. In crit-
icizing Barth’s characterization of the Basseri chief, I shall examine the sup-
porting evidence and argument as presented in Nomads of South Persia (here-
after: Nomads) and point out what appear to me to be problems in Barth’s
account. I shall also refer to evidence about chiefs in other tribes. I then present
a reinterpretation of the Basseri chiefship, which I consider to be Barthian in
spirit, for it is consistent with the theoretical approach of Nomads.
the role of the basseri chief
Until the mid-1950s, the chief was the keystone in Basseri tribal architecture.
The dozen Basseri tira, tribal sections, were known to have diverse ethnic and
historical origins, variously Arab, Persian, and Turkish (Barth 1961:50–53).
Basseri tribal unity was defined by a common allegiance to the chief. “It is the
fact of political unity under the Basseri chief which in the eyes of the tribesmen
and outsiders alike constitutes them into a single ‘tribe’ in the Persian sense”
(Barth 1961:71).
The Basseri chief and his close relatives formed a small, wealthy elite stra-
tum above the mass of ordinary nomadic pastoralists who made up, along with
some settled sections, the great bulk of the Basseri population.
The chief and his immediate relatives belong in a category entirely apart from the rest
of the Basseri, both in the minds of the tribesmen, and to an outsider. They are Khavanin,
of the Khan’s dynasty; they are shown respect and granted authority. . . . Most of them
. . . take little part in nomadic life. Particularly the chief and his brothers, one of them

hierarchical image and reality 51
the former chief, are sophisticated members of the élite on a Persian national level; they
maintain houses in [the provincial capital] Shiraz and travel extensively within and out-
side of Persia. In wealth they are also in a class entirely apart from other Basseri, each
owning several villages as well as flocks of many thousand head of sheep and goats
(Barth 1961:73–74).
The chief’s position and status were manifested materially in his urban villa
and the size of his nomadic tent, but above all in offering boundless hospi-
tality, generosity which exhibited his greatness of wealth and spirit (Barth
1961:74). To support his needs, the chief was able to call on his tribesmen to
contribute one, two, or even three sheep in a hundred (Barth 1961:74).
The authority of the Basseri chief was exercised regularly in three fields: “al-
lotting pastures and co-ordinating the migrations of the tribe; settling the dis-
putes that are brought to him; and representing the tribe or any of its members
in politically important dealings with sedentary authorities” (Barth 1961:76).
Of these services, it is the third that was most critical: “Perhaps the chief’s most
important function is to represent the tribe in its relations with the Iranian ad-
ministration, and in conflicts with sedentary communities or persons” (Barth
1961:77). By so doing, the chief, enabled by ruling power within his own group
and elite status among other regional and national elites, bridged the gap be-
tween the partly autonomous Basseri on the one hand, and the Persian govern-
ment, village agricultural communities, urban dwellers, bazaar merchants, and
other segments of the “plural” society of Fars Province and Persia at large on
the other (Barth 1961:77). “The chief’s role in mediating relations with seden-
tary society, in protecting the nomadic herders’interests vis-à-vis the often for-
midable and always confusing organizations that structure parts of their envi-
ronment and encroach on their life, is correlated with a strong feeling of respect
and dependence among the tribesmen” (Barth 1961:80).
the basis of the basseri chiefship
The service that the chief provided for his tribesmen did not, according to Barth,
adequately account for the Basseri chiefship, for “the persistence of an institu-
tion is not exhaustively explained by its usefulness” (Barth 1961:80). Rather,
“[t]he position of autocratic authority occupied by the Basseri chief can only be
successfully maintained and defended if it is supported by enough coercive
power to enforce discipline and suppress opposition from below . . . [P]ositions
of authority can only be stable if the incumbent of such a position is able to mo-
bilize enough force to counter any group that can form within the system to
question his authority” (Barth 1961:80–81). Thus, “[a]n analysis of the politi-
cal organization calls for an analysis of the processes whereby the powers of
the chief are exercised and defended” (Barth 1961:75).
What, then, were the sources of the Basseri chief’s coercive powers? There
were not, in fact, any special institutions of coercion available to the chief, no
tribal police or military unit under his control. The main physical coercion

52 philip carl salzman
directed by the chief was carried out by ordinary tribesmen: “Corporal punish-
ment takes place in the presence of the chief and is specified by him . . . The
beating is not performed by any special category of functionary—any by-
stander who is a member of the tribe may be ordered to do it” (Barth 1961:82).
There were some situations, such as the competition between rivals for a va-
cant chiefship, in which assassination was used or at least threatened (Barth
1961:84). The chief was able to levy fines against tribal segments (Barth
1961:82), and could punish groups through the allocation of poor pasture. Of
course, he could also withhold services desired by tribesmen. On the whole,
however, it seems clear that the Basseri chief did not have at his disposal sub-
stantial coercive means to support his position.
In the absence of coercive powers, how did the chief maintain his position?
Barth’s answer is that Basseri commoners were, for organizational reasons, un-
able to challenge the chief: “The political subjects of the chief are . . . organized
in small, mutually hostile, and weakly led groups, each striving to maintain in-
ternal harmony and unanimity without coercive means” (Barth 1961:81). This
social fragmentation made it impossible for the tribesmen to unite for the pur-
pose of opposing the chief. Because ordinary tribesmen “lack strong leaders on
the level of their own communities, they lack mechanisms for delimiting and
containing the powers of leaders on higher levels of tribal organization; and
they thus become subjects of the chief on terms of autocratic dominance/sub-
mission” (Barth 1961:90). Thus, “[t]he poor development of centrally con-
trolled coercive means reflects this impotence of any potential opposition”
(Barth 1961:81).
In sum, according to Barth, ordinary Basseri tribesmen—the commoners
who made up the great bulk of the pastoral nomads—were socially, economi-
cally, and politically fragmented, effectively leaderless at the local level, and
were thus unable to unite in common political purpose to oppose their chief, to
whom they consequently remained subordinate and obedient. The omnipotent
chief, drawing upon his wealth, status, and contacts with outsiders, command-
ed his tribesmen in an authoritative and autocratic fashion. To his tribesmen,
the chief was “he who must be obeyed.” And in the view of the Basseri tribes-
men, this was proper and desirable: “The tribe without its chief was compared
to a flock without its shepherd and a car without its driver” (Barth 1961:75).
Tribesmen wanted a strong chief (Barth 1961:84–85).
how submissive were basseri tribesmen?
In his account of the relationship between the Basseri chief and ordinary tribes-
men, Barth emphasizes the chief’s dominance and the submission of the tribes-
men. Let us now examine some of the description in Nomads, which suggests
that ordinary Basseri tribesmen might have been less than totally submissive in
relation to their chief.
First, one possible response to a chief, or in this case to a pretender for suc-

hierarchical image and reality 53
cession to the chiefship, was passive resistance. The tribesmen could have ig-
nored instructions and orders from a leader whom they did not accept. “Though
the authority of other pretenders may not be directly challenged by the com-
mon tribesmen, people start dragging their feet and fail to act on the orders of
other pretenders than one, until one day the whole edifice of administrative de-
cisions and imperious manner built up by the unsuccessful pretender collapses,
and nobody listens to him any more” (Barth 1961:84). Conceivably this process
could have taken place with an established chief, if he were to have fallen out
of favor with his followers.
Second, we should consider that one of the resources of the chief, as he ne-
gotiated with outsiders, was the military capacity of his tribe. The Basseri, af-
ter all, was made up from “collectivities of free tribesmen, who for one thing
constitute a mobile, ready-made cavalry force” (Barth 1961:90). Many Basseri
owned horses for riding (Barth 1961:6), and hunting was a popular sport (Barth
1961:9). Firearms were esteemed (Barth 1961:74, 88). Thus the tools of mili-
tary force—firearms and riding horses—were in the hands of ordinary tribes-
men, and the tribesmen were skilled in their uses. This suggests that Basseri
tribesmen, notwithstanding their alleged organizational weakness, were capa-
ble, at least as individuals or small groups, of applying armed coercion. And if
arms could be turned against foreign enemies, could they not also have been
turned against internal enemies? If a chief could have been brought down by an
assassin’s bullet, he could have also been brought down by a rebel’s bullet. So
it is difficult to reconcile the image of an omnipotent, autocratic chief and his
subservient, submissive subjects with the reality of a universal distribution of
the means of armed coercion among skilled and hardy tribesmen. As Behman-
Begi (1945, published in Monteil 1966:134), himself a member of the power-
ful Qashqai tribal confederacy, put it, “Le fait . . . que les nomades soient ar-
més et de caractère vif rend la tyranie (estebdâd) impossible.”
Third, and perhaps most important, was the universal capacity of Basseri
tribesmen to remove themselves from the authority of their chief. The Basseri
enjoyed an almost absolute capability for spatial mobility. Their established no-
madism rested on a technology of portable housing and production equipment,
and their capital resources—livestock—were both spatially mobile and, thanks
to the urban marketplace, financially liquid. This mobility gave them a full ca-
pacity of rapid and complete retreat from threats of any kind. Tribesmen could
have, if they wished, literally walked or ridden away from their chief.
Some large herd owners, fearing rivalry with the chief and expropriation of
their herds, sold their livestock, bought agricultural land, and settled in villages
as landowners (Barth 1961:127). Furthermore, notwithstanding Barth’s argu-
ment that Basseri tribesmen were not well situated to organize independently
of the chief, large groups of tribesmen had settled as units in compact villages
in their traditional summer pastures (Barth 1961:117). Even more striking, large
groups of tribesmen acting as units had left the Basseri entirely:

54 philip carl salzman
Repeatedly in the past, groups have defected from the tribe and the area, to move either
northward to the Isfahan area and beyond, or eastward into Kerman. During the rule of
Agha Jan Beg . . . a number of such defections took place. For example, according to
the traditions of the Il-e-Khas, a total of 300 tents left as a group to join in the new tribe
then being formed by Zel-e-Sultan, a son of Nasr-ud-Din Shah, in Isfahan, while about
30 tents of Il-e-Khas went east to Kerman, and a remnant of 8–10 houses only of that
section remained behind as settlers . . . (Barth 1961:116).
The abandonment of the Basseri by these large groups of tribesmen resulted
from tribesmen’s rejection of their tribal chief. “Defections from the tribe by
larger groups of commoners are caused . . . by the ruthlessness of strong chiefs
. . . ” (Barth 1961:84).
A tribal chief in Fars Province did not have a monopoly of leadership over
his tribesmen. The presence of neighboring tribes and their chiefs presented al-
ternative leadership options to common tribesmen. Weak and ineffective chiefs
could lose their tribesmen to competing chiefs from other tribes. “Any imbal-
ance between tribes in the effectiveness of centralized authority stimulates an
extension of the stronger centre’s claims to authority, and a voluntary flow of
commoners from the weaker to the stronger centre” (Barth 1961:85). If tribes-
men believed that their interests were better served by the chief of a neighbor-
ing tribe, they could have switched allegiance. “Camps, oulads and sections
seek out the strong chief and submit to him; from him they obtain better pro-
tection and by him their interests are best safeguarded” (Barth 1961:85). A trib-
al chief was thus in constant if implicit competition with neighboring chiefs for
the loyalty of his tribesmen.
The terms “omnipotent” and “autocratic,” when applied to the chief, suggest
that the Basseri tribesmen, whom the chief “commanded,” were totally depen-
dent upon him and at his mercy: “ . . . subjects of the chief on terms of auto-
cratic dominance/submission”(Barth 1961:90). Yet Basseri commoners, as we
have seen, had many resources of resistance at their disposal: passive refusal;
threat of or actual violent attack; threat of or actual defection from the tribe
through sedentarization or migration; and threat of or actual shift of allegiance
to another chief. The presence of these weapons of resistance in the hands of
ordinary Basseri tribesmen suggests that they were less than totally at the mer-
cy of their chief, and that they did have effective access to alternatives to obey-
ing the chief.
tribal rule and obedience
The King is King by
grace of the people.
Twsana proverb
(Gulbrandsen 1995:433)
If, as Barth’s material suggests, Basseri commoners did not have to obey their
chief because of an inability to resist, why did they obey him, accepting his or-
ders and judgements? And why did they support his elite status and imperial

hierarchical image and reality 55
manner (Barth 1961:74)? The answer is that the Basseri chief was able to rule
because he had the consent of his tribesmen.
Barth argues that a tribal chief maintained his “authority . . . by its effective
and continuous exercise . . . “(Barth 1961:84). “Authority” is an apt term here,
for it implies legitimate power—that is, power based upon consent. But what
exactly counted as effectiveness in the exercise of authority? From the tribes-
men’s point of view, chiefly effectiveness would include as primary component
the serving of their interests. This is one of the reasons given by Barth for tribes-
men switching allegiance to another chief, to “obtain better protection and
[have] their interests . . . best safeguarded” (Barth 1961:85). It was also an ex-
plicit expectation in assessing the chief’s settlement of disputes: the chief must
“make the decision which he feels is ‘best for the tribe’”(Barth 1961:77). Un-
doubtedly tribesmen were watching, then and later, to see if the decision was
indeed “best for the tribe.”
The ineffective or oppressive tribal chief, if he survived assassination, would
have seen his tribe melt away, his “subjects” having disappeared over the hori-
zon to villages, to other chiefs, to other parts of the country. He would have be-
come an ex-chief, for one could not be a chief without followers. Nomadic
tribesmen, mobile warriors, and free herd owners could not be held against their
will. If they felt that, on balance, their interests were not being served, or that
they were being abused, they would have resisted, rebelled, and departed.
No one knew this better than the successful chief. He knew that he was chief
by virtue of his tribesmen. He knew that their consent was the foundation of his
political power. He knew that he always had to be aware of public opinion and
act within its parameters. For example, speaking of the appointment of sub-
chiefs, Bahman-Begi (1945, in Monteil 1966:134) stresses the sensitivity of
chiefs to the preferences of his tribesmen. “[L]’Il-Khân ne peut nommer
chef de tribu ou de clan que celui des candidats . . . que soutient l’opinion
publique. . . . En général, les nomades peuvent choisir leurs chefs.” The suc-
cessful chief knew that to act outside of public opinion was to risk undermin-
ing the consent upon which his rule was based. Finally, he knew that the truly
effective chief was successful in shaping and leading public opinion.
Let us stop for a moment to recall Barth’s (1961:75) statement that “[a]n
analysis of the political organization calls for an analysis of the processes
whereby the powers of the chief are exercised and defended.” This is one of
many statements in Nomads emphasizing processual analysis as an addition, if
not replacement for, structural and normative analysis. Nomads is more than an
ethnographic description: it is a brilliant theoretical demonstration, illustrating
throughout the relevance and appropriateness of processual analysis, its fruit-
ful emphasis on decision-making and transactions, and its power in showing
the way social forms are generated. In his processual analyses, Barth shows
among many other things that the Basseri camp was constituted demographi-
cally by an aggregate of economic decisions made by individuals (ch. VIII),

56 philip carl salzman
that the tribe maintained demographic balance with its environment through
sedentarization (ch. IX), that the herding camp as an operating social group was
based upon negotiation and purposeful alliances (ch. III), that herding groups
were formed by means of utilitarian contract (22–23), that the Basseri tribe was
a congeries of segments diverse in origin, ethnicity, and history brought to-
gether by conscious political allegiance to the chief (49–54), and that the
“Arab” Khamseh Confederacy, of which the Basseri were part, was invented
for instrumental purposes by powerful urban merchants and sealed by the pro-
vision of arms and other resources (86–89; see also Oberling 1974:65). In short,
Barth demonstrates that “structural” features were generated by decision-mak-
ing and transactional processes, and that these were based at least partly upon
rational and strategic considerations.
Barth’s (1961:80–81) account of the Basseri chief’s ability to maintain com-
mand, and his characterization (1961:90) of chief/commoner relations as one
of “dominance/submission” is, as indicated above, based primarily upon a
structural feature, the political fragmentation of the bulk of tribal commoners.
In criticizing Barth’s formulation by identifying processes of resistance and op-
position among commoners, I am following Barth’s theoretical lead, demon-
strated by him in the analyses of other aspects of Basseri life mentioned above,
in his emphasis on the processual generation of structures. If we can now ac-
cept the argument that the Basseri polity was based upon consent and that the
chief’s actions had to be in tune with tribal public opinion, I think we can go a
little farther by looking for the processes by which the Basseri khan tapped into,
tested, and tried to shape public opinion.
Just as the Basseri chief was constantly collecting and exchanging informa-
tion with other members of the regional and national elite in order to carry out
his role as representative of his tribe, and to act as intermediary and advocate
on its behalf, so too the chief had to be constantly collecting information from
tribal commoners about conditions and opinions within the tribe. One process
by which this happened was the standard Middle Eastern exchange of host and
guest, in which the host provides food, shelter, and gracious reception, and the
guest provides respect, appreciation, and information. No one was better placed
to benefit from this exchange than a chief, who was expected as part of his
chiefly role to be a great host. The Basseri chief had a large tent and “[h]is hos-
pitality should be boundless” (Barth 1961:74). The chief’s invitation to come
and receive hospitality was unlikely to be refused by tribal commoners. In the
course of offering hospitality, the chief was able to draw opinions, collect in-
formation, and plant the seeds of rumor in order to test public opinion. The
process of conflict resolution conducted by the chief also provided an occasion
for him to collect information about current tribal circumstance and opinion;
so, too, was the chief’s reception of commoners seeking assistance with prob-
lems internal or external to the tribe. Finally, the chief’s distribution of valu-
ables to commoners not only exhibited his wealth and generosity—“he should

hierarchical image and reality 57
provide spectacular gifts of weapons, and stallions from his large herd of hors-
es, to his more prominent subjects”(Barth 1961:74)—but provided an optimal
occasion for solicitation of information, as well as support and loyalty. In the
provision of hospitality, as well as in the distribution of valuable gifts, the
chief’s patronage offered material benefits and esteem in exchange for infor-
mation and political support. Here the chief was not only collecting and testing
public opinion, but was trying to shape it by catering to the personal desires of
his tribesmen.
Even in his own decision-making process, the chief may have relied upon a
jelling of opinion among interested parties, his advisors, and local leaders.
Barth (1961:74), in his characterization of chiefly decision-making, stresses the
difference between the indirect, democratic opinion-formation processes of the
camps, and the “autocratic command” of the chief.
Decisions governing collectivities among the Basseri are reached either by discussion
or by command. . . . [D]ecisions governing camp groups are made by their members
through a lengthy process of discussion and mutual persuasion. . . . The right to com-
mand, to make decisions on behalf of persons in other tents than one’s own, is strictly
chiefly prerogative. The monopolization by the chief of the right to command is a fun-
damental abstract principle of Basseri social structure.
Even if we accept this account of chiefly command as “a fundamental abstract
principle of Basseri social structure,” we can still ask, in a Barthian spirit,
whether the process of decision-making did not involve the collection of infor-
mation and tapping of public opinion. In this regard, I offer a description of de-
cision-making by the ilkhani, chief, of the QashqaÃi tribal confederation, the
Basseri’s neighbor and sometime opponent, and a much larger and more pow-
erful polity than the Basseri. According to Beck (1986:218),
A primary feature of the ilkhani’s style of leadership, which reflected political process-
es among the QashqaÃi in general, was the large open gathering in the reception tent,
where he presided but did not explicitly dominate the discussion. The process Barth
(1961:43–46) describes for the Basseri [camps], in which men talk most of the evening
about the possibility of migrating the next morning, and in which a consensus slowly
emerges without any single individual’s having made the decision unilaterally, is simi-
lar to a process found among the QashqaÃi. The ilkhani possessed information, especially
about the state and foreign powers, that others present did not, but he tended to intro-
duce it in general conversation to allow others to consider and discuss it. . . . Meetings
and discussions were usually public, and those attending prepared their statements care-
fully because of the scrutiny of others. All present had the right and opportunity to add
opinions and details. . . . Overall, the nature of the gathering was relatively egalitarian,
and the decision-making process was often informal. People presented their problems
in this open forum, presided over by the ilkhani and his advisors, and decisions could
be reached in the course of group discussion without the ilkhani’s having made any kind
of a definitive statement.
So even among the most powerful chiefs open discussion was encouraged,
a wide range of views were sought, the building of a consensus was favored,
and informal and collective decision-making were common. This suggests

58 philip carl salzman
that command, when rarely invoked, was not sociologically opposed to these
processes, but rather dependent upon them.
the image of the omnipotent chief
If, as I propose, the Basseri political system prior to the mid-1950s was funda-
mentally based upon consent, and the chief was severely constrained by public
opinion amongst his tribesmen, why did the chief present an imperial manner,
why did the tribesmen see power emanating from him rather than being dele-
gated by them, and why was the chief accepted as an omnipotent ruler? I do not
believe that Barth got this image wrong. In fact, I think that the Basseri chief
and Basseri commoners took great pains to validate and reinforce this image of
the chief. They themselves wanted to believe it, and above all they wanted oth-
ers—Persian government officials, sedentary villagers, urban merchants, and
other tribal chiefs and commoners—to believe it. Nothing served the Basseri
better in their complex and plural social environment, where they were sur-
rounded by these other groups, than the picture of a solidary and unified tribe
unquestionably following the commands of their strong and unchallenged ruler.
The Basseri projected this image, broadcast it, asserted it, and may even have
believed it. So, too, did the chief himself. This image of the omnipotent chief
and his obedient subjects, considered in relation to the consent on which the
tribe was actually based, can be appreciated as a conspiratorial misrepresenta-
tion jointly engineered, to whatever degree consciously or unconsciously, by
the chief and the commoners. This “presentation of (collective) self” was the
facade exhibited by the Basseri to the wider world, to encourage respect and
fear and discourage opposition and aggression.
While leaders in electoral democracies may undercommunicate their real
power, presenting themselves rather as “ordinary folks,” it is not uncommon in
tribal societies for leaders to overcommunicate their power, especially for the
consumption of external audiences. Lancaster (1981:82–83) discusses image-
making among the sheikhs of the Rwala Bedouin.
In the dealings between sheikhs and outsiders impression-management comes into
play. . . . The impression they intend to convey is one of power and magnificence. My
earliest contacts with the Rwala sheikhs were at feasts given for visiting officials and
the scene was of almost nineteenth-century splendour—slaves and servants abounded
and there was a continual stream of fruit, sweetmeats, kebabs and drinks. The main
course was two, three or more whole sheep on mounds of rice sprinkled with pine-nuts
and, sometimes, vegetables. The slightest whim of the guests was fulfilled, fruit was
peeled for them, glasses were topped up, cigarettes lit, new brands of tobacco produced;
the guests needed to do nothing for themselves. All these services were performed by
slaves and servants at a sign from the sheikh and every action was carried out prompt-
ly. One would have said that they went in fear of their lives if they disobeyed or were
even careless. As soon as the visitors had left, it became apparent that the whole scene
was a charade designed solely to impress. The slaves lolled in the recently vacated seats,
the sheikhs poured their own tea, the servants ate up the titbits or simply left. Later on,
when it became obvious that I wasn’t worth impressing, I heard the same slaves, those

hierarchical image and reality 59
paragons of domestic service, telling the sheikh that it was too hot to do what he had
asked, he’d have to do it himself. How far anyone is taken in by these phoney displays
of omnipotence is hard to say; presumably other Bedu aren’t, but maybe government of-
ficials and other non-Bedu guests are, to some degree. . . . [M]any public statements (and
formal feasts are statements) are only too apparently impression-management.
In relation to free Rwala tribesmen, the Rwala sheikh cannot be said to have
ruled and commanded (Lancaster 1981:87).
[T]he sheikh’s prime task is mediation and negotiation. . . . [T]he sheikh has no author-
ity to treat on the tribe’s behalf, for there is no legal framework nor coercive force avail-
able to him. All the sheikh can do is represent the tribe and bank on knowing what the
tribe wants in any particular circumstance. He then has to persuade the tribe to accept
the results of his negotiation. . . . The sheikh is never a free agent, he can only act as the
mouthpiece of tribal or group consensus.
What a Rwala sheikh needed to pursue his leadership role was information
(Lancaster 1981:93).
In a political system of this sort, where the “leader” has no coercive powers at all,
his best means of justifying and maintaining his position is by being better informed
than anyone else. Thus the sheikh has to ensure that information pours in from all quar-
ters. This isn’t as difficult as it sounds; it is just very expensive, which is not to say that
information is bought but that the sheikh must keep open house (or rather tent) for all
comers.
Rwala sheikhs are highly aware that their influence is based upon public opin-
ion (Lancaster 1981:96).
Emir Sattam ibn Hamad ShaÃalan . . . is quoted as saying [in 1878]: “I can only do what
my people wish”. . . . He spoke no less than the truth . . . [H]ad he tried to coerce he
would have been deserted. The fundamental principle remains the same today, leader-
ship by consensus. Political power does not exist.
Here we are shown that among the Rwala Bedouin great displays of omnipo-
tence for outsiders veiled a political reality based upon consent and consensus.
While we cannot assume that the Rwala and Basseri political systems are the
same, the existence among the Rwala of displays of chiefly omnipotence, and
also the practice by chiefs of closely following public opinion, provides some
credibility for my similar interpretation of Basseri political life.
Outsiders not infrequently believe that tribal chiefs are more powerful than
they really are. The Blunts (1879, cited in Lancaster 1981), to whom Emir
ShaÃalan spoke the words quoted above, “took him to be weak,” unwilling to
use the power at his disposal. Similarly, General R. E. H. Dyer refused to be-
lieve the protestations of Jiand Han, the sardar, chief, of the Yarahmadzai tribe
of Iranian Baluchistan, about the limitations of his control over “his” tribesmen.
Dyer, later known for his part in the “Amritsar massacre” (Colvin 1929), dur-
ing a punitive campaign against Jiand Han and the Yarahmadzai, saw Jiand Han
as “a very notable man, with enormous power and prestige, not only with his
own tribe, the Yarmohommedzais, but with all the nomad tribes of the district,

60 philip carl salzman
and was regarded as a personage by the Governments of both India and Persia”
(Dyer 1921:62). When “Jiand . . . pleaded that he himself had done his best to
restrain his men from interfering with the British lines of communication, warn-
ing them that it was neither safe nor wise” (Dyer 1921:71), Dyer did not believe
him. “[H]e could not seriously have expected that I would swallow this excuse,
as he was known to be held in such awe by his followers that not one of them
would have dared to dispute his authority” (Dyer 1921:71). But while Dyer was
correct in saying that Jiand Han was “a very notable man, with enormous . . .
prestige,” he was mistaken to believe that Jiand had “enormous power” and that
Yarahmadzai tribesmen would not “have dared to dispute his authority.”
the reality of tribal chiefs: the yarahmadzai baluch case
Jiand Han was a notable and had enormous prestige because he was the recog-
nised and successful leader of the Yarahmadzai tribe, an independent tribe of
fierce raiders and warriors. Regular and repeated predatory raiding of Persian
villages and merchant caravans was (until the Persian conquest of Baluchistan
in 1928–35) an integral part of the Yarahmadzai economy, culture, identity, and
personality (Salzman 1994, in press). Raids were carried out by small groups
and large, led by individuals, lineage leaders, or tribal leaders. But raiding was
not centrally controlled; it was a prerogative of each tribesman. Raiding was in
this respect like most other important activities of tribal life, such as livestock
ownership, herding, migration, camp location, well-digging, and agriculture
and arboriculture; control over raiding and these other activities was decen-
tralized, in the hands of and at the discretion of individuals, families, contrac-
tual residential groups, and small corporate lineages. While the sardar repre-
sented the tribe externally and was its war leader, internally his responsibilities
were limited to mediating disputes. Social control and physical coercion rest-
ed in the hands of corporate lineages, and economic resources were common
property (e.g., pasture and natural water sources), or lineage property (e.g.,
wells), or family property (e.g., livestock). The leadership of the sardar rested
not on his control of necessary resources or on his ability to coerce tribesmen,
but on the kabul, permission or consent, of the tribesmen.
I base my understanding of Jiand Han and his “Yarmohammedzais” of the
first third of this century partly on my research among his and their descendants
over an eight year period in the 1960s and 1970s. The Yarahmadzai (called, un-
der the Shah, the “Shah Nawazi”) had at the time of my arrival been effective-
ly encapsulated by Iranian state authorities for thirty-five years. Perhaps the
main change from Jiand Han’s time was the absence of raiding, which had been
suppressed since encapsulation and replaced in post-WWII decades by migrant
labor and small-scale trading and smuggling (Salzman 1994). Han Mahmud,
the Yarahmadzai/Shah Nawazi sardar of the 1960s and 1970s was heavily in-
volved in acting as middleman between the tribe and state authorities, serving
as intermediary, mediator, and broker, and was able to draw on some state

hierarchical image and reality 61
political and economic resources to bolster his position inside the tribe (Salz-
man 1973, 1974). But with most other tribal arrangements remaining in the de-
centralized status quo ante, the sardar’s influence continued to rest on public
opinion and the consent of his followers, and he was not in a very strong posi-
tion to move them where they did not want to go (Salzman 1983).
One example of the sardar’s intervention amongYarahmadzai tribesmen may
serve to illustrate the relations between him and his tribesmen. It all began on
23 August 1972, when a stray camel belonging to a tribesman from one tribal
section ate the dates off of a palm belonging to a tribesman of a different trib-
al section. The latter tied up the camel and refused to return it to its owner with-
out compensation for the lost dates.An exchange of unpleasantries led to an ex-
change of blows, with the owner of the palm receiving the worst of it. This
incident quickly escalated to hostility between members of the two maximal
tribal sections. Four days later, a group of section-mates of the palm owner ran
across a smaller group from the other section, including two respected elders,
and roughed them up. This intensified the hostility and led to a large-scale mo-
bilization for conflict, one instance of which I observed. Upon hearing of this
conflict, Sardar Han Mahmud rushed to the date palm area to meet with elders
of the contending sections. The following is an exchange between the sardar
and elders of one of the sections, which took place on 1 September and was re-
ported to me by several persons present.
Sardar Han Mahmud: The Rahmatzai and Kamil Hanzai are all brothers, should not be
fighting, and should settle the matter peacefully. You, the Rahmatzai did not call upon
me to settle the matter, which was wrong of you. But if you would give the Kamil Han-
zai five thousand tuman and two girls for marriage without bridewealth payments, the
matter would be settled. After all, you man-handled Abdul Karim, an old and important
man of the Sohorabzai, and this puts the balance of blame on you.
Mahmud Selim (speaking for the Rahmatzai): No, we absolutely will not pay. We are
one kom [lineage group] and the Sohorabzai are one kom and the two are equal. The Rah-
matzai is prepared to fight but not to pay.
Sardar: The Rahmatzai should then give three thousand tuman, and I will see that it is
settled.
Mahmud Selim: No, the Rahmatzai will not pay. And furthermore, you, Han Mahmud,
no longer have our kabul [consent] as our leader. We repudiate you as our sardar.
Sardar: I have come to make peace, even though you did not send for me. If you do not
wish to make a peaceful settlement, you cannot expect me to help you when you find
yourselves in difficulty. I have the support of the dolat [the Government of Iran] and
they will use their force as I say. If anything happens to any of the Rahmatzai, I will not
do anything about it.
Mahmud Selim: We Rahmatzai are Pakistani not Irani, and are thus safe from Iranian
Government interference. As far as the sardar and the Soherabzai are concerned, the
Rahmatzai are a separate tribe.
This conflict, which came to be called jang, war, continued for three years, with
attacks and injuries inflicted by one side and then another. Attempts at settle-
ment, in Baluchi tents and in Persian courts, repeatedly ended in brawls. Final-

62 philip carl salzman
ly, in the latter part of 1975, with the courts and the sardar collaborating in
threatening heavy fines on those involved in further violence, a great diwan, de-
liberative meeting, called by the sardar was successful in calling off the feud
and reuniting the sections of the tribe in peaceful amity. (For a full account see
Salzman, in press.)
sardars and khans
The Yarahmadzai sardar, whether of distant or more recent times, however
enormous his prestige and however great a “personage,” was far from having
the power to command his fellow tribesmen. The sardar could not be dominant,
and his fellow tribesmen were not prepared to be submissive. Of course, the
Yarahmadzai was different in important respects from the Basseri. The Yarah-
madzai was conceived as a descent-based lineage group, and enjoyed strong,
corporate lineages; its territory was far from urban centers and large agricul-
tural populations; and its desert pastoral ecology did not lend itself to central
organization. In contrast, the conception of the Basseri centered on the chief-
ship; lineage segments were not united through common descent, and were in-
vested with more limited powers; and the relatively predictable physio-biotic
environment and crowded social environment made some central coordination
useful. These differences would have strengthened the hand of the Basseri khan
in comparison to the Yarahmadzai sardar.
But commonalities in the circumstances of the Yarahmadzai and Basseri
tribesmen circumscribed the power of both chiefs. All tribesmen were warriors
and owned weapons and riding animals, and had the skills and inclination to
use them. All tribesmen were members of economically self-sufficient families
or micro-lineages; livestock—the main form of capital—rested in the hands of
families. All tribesmen were highly mobile; their families, dwellings, furnish-
ings, tools, and capital resources could be packed in a matter of hours onto fam-
ily-owned baggage animals and transported over rough country to near or dis-
tant destinations. All tribesmen, as new recruits, would be assets to the chiefs
of neighbouring tribes, and would be welcomed into those tribes. As a conse-
quence of these circumstances, Yarahmadzai and Basseri tribal chiefs did not
have a monopoly over the means of coercion, or control over the main economic
resources, or a captive body of subjects. In consequence, their leadership and
“rule” depended upon the consent of their followers, who, as we have seen,
would follow only so far as it suited them. To be sure, the Basseri chief did
“command,” as he was called upon to do by the exigencies of impression man-
agement, but in order to maintain his position through the “effective and con-
tinuous exercise” of his authority (Barth 1961:84) required that he only com-
mand what he knew his tribesmen were willing to obey, to lead only where his
tribesmen were willing to follow.
If my interpretation of the Basseri authority structure is correct, thanks must
go to Barth himself, who provided much of the relevant information. As I have

hierarchical image and reality 63
already indicated, I believe that Barth was correct in depicting the public im-
age of the tribal authority structure, an image encouraged and advanced by trib-
al commoners and elite alike. But why did Barth not sense, as I do, a discrep-
ancy between image and reality, between facade and foundation? One reason
might be that Barth carried out his research, as he makes very clear (1961:71–
72, n. 1 and 96), two years after the Basseri chief had been deposed, when le-
gal authority over the tribe had been vested in a colonel of the Persian Army.
Barth (1961:72, n. 1) comments,
The change to Army rule had not been accepted by the Basseri tribesmen, who contin-
ued to act towards the legally deposed chief as if he were formally in office, while liai-
son between the Colonel and the tribe was poor. This de facto situation was even rec-
ognized by the authorities. . . . The activities and processes I shall describe have thus
been observed in the field.
But it is not made explicit in Nomads which aspects of the chiefly authority
structure Barth was able to observe directly, and to what extent he had to de-
pend upon reports from tribesmen about these matters. It is difficult to imag-
ine that the chief was not constrained in his actions by the new arrangements.
Furthermore, given the unwelcome violation of the tribal political structure
by the Persian government not long before Barth’s research, would not the
Basseri tribesmen have been likely to think and speak positively of their now
threatened, if not extinct, independent chiefly authority? However this may
be, even in observing an intact chiefly authority structure, a researcher would
most likely become aware of the discrepancy between rhetoric and reality
only when a chief mistakenly or imprudently overstepped his mandate, per-
haps repeatedly, and tribesmen began to respond by passive or violent oppo-
sition, and by withdrawal or secession. This might happen relatively rarely,
for savvy and prudent chiefs would have striven purposely to avoid it. That
an anthropologist should be present to observe such occasional events would
be unlikely. All the more useful, then, are historical accounts, such as those
provided for us by Barth (1961:84–85, 116–17) of previous secessions and
defections.
could chiefs have had real power?
We need not assume, of course, that all chiefs and chiefdoms are the same. The
“default setting” for an anthropologist should, in fact, be the assumption of
great variation. Wolf (1982:96) is quite explicit on this point.
The term chief has come into common parlance to denote the recognised leader or head
(from the Latin caput) of a socially organized population. In practice, the term was usu-
ally bestowed by Europeans upon any native person of influence who was in a position
to forward or to hinder their interests. As such, references to chiefs cover different kinds
of recruitment and degrees of authority, and are of little analytic utility.
Drawing on “the concept of mode of production,” Wolf (1982:97) distinguish-
es “two rather different kinds” of chiefdoms.

64 philip carl salzman
[T]hose based on the kin-ordered mode, in which the chief and his followers are still em-
bedded in kinship arrangements and bound by them, and those in which the form and
idiom of kinship may be maintained even as a dominant group transforms divisions of
rank into divisions of class. . . . In this second kind of chiefdom, the chiefly lineage is
in fact an incipient class of surplus takers in the tributary mode.
Wolf (1982:99) elaborates on kinship-based chiefdoms:
The kin-ordered mode inhibits the institutionalization of political power, resting essen-
tially upon the management of consensus among clusters of participants. . . . [T]he ex-
tension and retraction of kin ties create open and shifting boundaries of such societies.
A chief can become a pivot of the power of his kinship group; but if he is sometimes
able to incarnate the kin order, he is also its prisoner.
And (1982:99) on tributary chiefdoms:
[Tributary chiefdoms] require mechanisms of domination to ensure that surpluses are
transferred on a predictable basis from one class to the other. Such domination may in-
volve, at one time or another, a wide panoply of sanctions based on fear, hope, and char-
ity; but it cannot be secured without the development of an apparatus of coercion to
maintain the basic division into classes and to defend the resulting structure against ex-
ternal attack. [The tributary chiefdom is thus] marked by the development and installa-
tion of such an apparatus, namely the state.
Using Wolf’s framework, we would without hesitation identify the Yarahma-
dzai as a kin-dominated chiefdom, in which the sardar was distinguished by
rank rather than by class. While the Yarahmadzai sardar could request contri-
butions from tribesmen to advance one or another project, he could not levy or
enforce taxation.
The Basseri prior to the mid-1950s clearly exhibited some features of a trib-
utary chiefdom. The chief and his kinsmen formed an elite class distinct and
above the ordinary tribesmen. And the Basseri khan could levy from one to
three percent of livestock (Barth 1961:74) (although no information is present-
ed on how frequently and for what purposes this could be done, or how the
tribesmen responded to requests). But to what degree could we say that the
members of the Basseri chiefly class were “surplus takers,” in Wolf’s phrase?
In fact, they were able to draw for their incomes on their own large herds, as
well as sources independent of the tribal economy—notably on landlord fees
and shares from the agricultural land and villages that they owned (Barth
1961:74). It seems that it was not necessary for the Basseri tribal elite to be ex-
tracting surplus from tribesmen on a regular basis.And this is fortunate, for they
would probably not have succeeded, as they did not have the necessary “appa-
ratus of coercion” at their disposal. Without the apparatus of coercion, without
a state apparatus, a true tributary chiefdom was not possible. For Wolf, then,
the Basseri, notwithstanding its class division and limited surplus transfer,
could not be considered a true tributary chiefdom.
One of the limitations of Wolf’s formulation—ironically, considering the
general thrust of his approach—is his conception of chiefdoms as isolates, as

hierarchical image and reality 65
if they had to evolve on their own. But Middle Eastern (and Indian and Chinese
and other) social formations have evolved for thousands of years under the di-
rect or indirect influence of states and quasi-states, tribes and kin-groupings,
markets and quasi-markets. The relations of local and regional political group-
ings to centers of power and trade were variously and varyingly close or dis-
tant, in alliance or in opposition. But tribes were rarely in isolation. The Basseri
evolved within the ecologically, culturally, socially, economically, and politi-
cally rich and complex environment of Fars Province, a center of Persian civi-
lization, and the dynamics of Basseri structures and politics were shaped by that
environment, rather than by strictly internal exigencies.
The pastoral nomads of the Basseri tribe were themselves an economically
successful, privileged “middle class” of the Fars region (Barth 1961:17, 108–
10; cf. Beck 1986:158, n. 18). The most successful pastoralists converted their
livestock wealth into land to become elite landowners (Barth 1961:105–6). The
main task of the Basseri chief was to maintain good relations—good for the
tribe, that is—with other tribes, sedentary populations, agents of the Persian
government, and bazaar merchants. The Basseri chief was working for his tribe;
the Basseri tribesmen were not working for the chief, as one might expect in a
tribute chiefdom. The Basseri chief did not have special call on supernatural
powers, nor did he redistribute the productive wealth of the society—two chiefly
functions seen in more isolated chiefdoms (Service 1975:75, 78; cf. Salzman
1975). How did the Basseri chiefship, without a coercive apparatus at its dis-
posal, arrive at elite standing and the symbols, if not substance, of command?
Exactly as Barth has told us, through providing the “interface” between the tribe
and the other, relatively balanced segments in the “plural” society of Fars and
Persia. This is why Irons (1979:362), taking into account the particular charac-
teristics of pastoral nomads, has argued that “among pastoral nomadic societies,
hierarchical political institutions are generated only by external political rela-
tions with state societies, and never develop as a result of the internal dynamics
of such societies.” If it were useful to proliferate types, we might add to Wolf’s
kinship chiefdom and tribute chiefdom an intermediate type, labelled the “plur-
al society chiefdom,” of which the Basseri would be an example.
The general lesson to be drawn from this exercise is that what people say, the
way they represent themselves, is an important part of their lives and reality,
but their lives also reflect other motivations, calculations, pressures, con-
straints, and opportunities. The present emphasis in some social analysis on
“voice” and identity privileges words—the comfortable medium of the acade-
mic—and sometimes underrates or ignores the other ways that people “speak”
with their choices, with their strong arms, with their feet.
references cited
Barth, Fredrik. 1961. Nomads of South Persia. Oslo: Scandinavian University Press.
Beck, Lois. 1986. The QashqaÃi of Iran. New Haven: Yale University Press.

66 philip carl salzman
Behman-Begi, Bahman. 1945. Orf-o ÄÂdat dar ÄAshâÃer-e Fârs. Teheran. Translated and
published as “Moeurs et coutumes des tribus du Fârs,” in Les Tribus du Fârs et la Sé-
dentarisation des Nomades. Vincent Monteil. Paris: Mouton. 1966.
Colvin, Ian. 1929. The Life of General Dyer. Edinburgh: Blackwood.
Dyer, Brigadier-General R. E. H. 1921. The Raiders of the Sarhad: Being the Account
of a Campaign of Arms and Bluff against the Brigands of the Persian-Baluchi Bor-
der during the Great War. London: Witherby.
Gulbrandsen, Ornulf. 1995. “The King is King by the Grace of the People: The Exer-
cise and Control of Power in Subject-Ruler Relations,” Comparative Studies in Soci-
ety and History 37(3):432–44.
Irons, William. 1979. “Political Stratification among Pastoral Nomads,” in Pastoral
Production and Society, L’Equipe écologie et anthropologie des sociétié pastorale,
eds. Cambridge: Cambridge University Press.
Lancaster, William. 1981. The Rwala Bedouin Today. Cambridge: Cambridge Universi-
ty Press.
Oberling, Pierre. 1974. The QashqaÃi Nomads of Fars. The Hague: Mouton.
Salzman, Philip Carl. 1973. “Continuity and Change in Baluchi Tribal Leadership,” In-
ternational Journal of Middle East Studies, IV:428–39.
———. 1974. “Tribal Chiefs as Middlemen: The Politics of Encapsulation in the Mid-
dle East,” Anthropological Quarterly 47:203–10.
———. 1975. “Islam and Authority in Tribal Iran: A Comparative Comment,” Muslim
World Quarterly LXV:186–95.
———. 1983. “Why Tribes Have Chiefs: A Case from Baluchistan,” in The Conflict of
Tribe and State in Iran and Afghanistan, R. Tapper, ed. London: Croom Helm, 262–
83.
———. 1994. “Baluchi Nomads in the Market,” in Pastoralists at the Periphery:
Herders in a Capitalist World, C. Chang and H. A. Koster, eds. Tucson: University of
Arizona Press, 165–74.
———. In press. Black Tents of Baluchistan. Washington, D.C.: Smithsonian Institu-
tion Press.
Service, Elman R. 1975. Origins of the State and Civilization. New York: Norton.
Wolf, Eric. 1982. Europe and the People without History. Berkeley: University of Cal-
ifornia Press.