by Faun Rice
conquests is called alternately Andriandeiloza or Andriandailoza—the prefix here
just meaning “lord” or “king.” Otherwise the name is the same. No indication is
given of his ancestry, but he is said to have made common cause with the last high
king of Imamo, Andrianampoetsakarivo, who led the western resistance against the
Merina king, but then is said finally, on realizing his kingdom would inevitably fall,
to have buried his wealth somewhere in his ancestral lands and “fled to Sakalava.”
The story now attributed to Leiloza, about Mount Ambohitrambo and its bridges of
silk, does appear in the Tantara (ibid.: 573 n. 1), but the imperious prince is referred
to instead as Andriankotomaditra (“lord naughty young man”), though the author
adds, “who some call Andriandahiloza.” (To make matters even more confusing,
Rakotomaditra is nowadays remembered instead as the name of a faithful slave of
Andriantsihanika, the king who gave up his throne, who is buried next to his tomb
and still assists him in doing battle with witches.) So two generations later, the
legend of Leiloza’s bad behavior and consequent fall was already beginning to take
on its current form—but it had not completely done so.
264
ON KINGS
only such as were required to protect them from the envy of their enemies, but
even those they would gladly abandon should it please the king.
Radama assured Hastie he had abundant proof that they were lying in all
these matters, and were guilty of numerous crimes. He ordered their immediate
execution.
When Radama’s commands were communicated to the heroine, she stood up,
and taking her spear and shield, both of which she handled with skill, she at-
tempted to harangue her followers. She used much gesticulation and said that
those who believed her to be the spirit of Triemanosinamamy were right, and tho’
her person might now again suffer, she would still be victorious and she roared
out “never despair never despair.”
When the infatuated woman was conveyed to the place in which she was
to suffer, she requested that she might be dispatched with a spear, as she had a
great antipathy to being shot. Her wish was complied with, and Rabevola suf-
fered at the same time. (Hastie n.d.: 402)
Such, then, was the real ignominious end of the line of Leiloza. Radama or-
dered the entire settlement razed, and the bulk of its population returned to
Valalafotsy.13
Hastie’s text is particularly intriguing because it appears to represent the
very earliest reference we have to spirit mediumship in the Malagasy highlands.
Written in 1824, it refers mainly to events that must have taken place in the
1760s, 1770s, and 1780s. Hastie lacks the language to describe it—hence his
talk of “prophets” and “reincarnations”—but, clearly, this is what he was talking
about. The slave boy in the story was possessed by, and spoke with the voice of,
the ancient king. When acting as his medium, he would have possessed all the
authority the king would have had in life. (Hastie implies it was a permanent
state, but this is very unlikely to have been the case.)
What’s especially interesting for present purposes is the fact that medium-
ship is clearly operating here as a form of political contestation: the people
rally to the boy who speaks with the voice of their ancient king, though the boy
13. Cf. Edland (2006: 103–4) for a brief summary of the campaign and its significance,
which, however, relies only on European sources. Some Manendy remain there to
this day (cf. Raison 1984, I: 267; Raison, incidentally, claims the Manendy, too, were
originally from Valalafotsy, but this seems a misreading of the evidence).
THE PEOPLE AS NURSEMAIDS OF THE KING
265
himself is of the lowest possible social status;14 the current king eventually reacts
with simple violence, but thus delegitimizes his rule. After his fall, yet another
“prophet” appears—this time, not just a former slave but also a woman—who
quickly becomes effective leader of the rambunctious “republic” of Valalafotsy’s
refugees.
EMBLEMATIC LABOR AND THE KING AS CHILD
The historic Leiloza was a tyrant so jealous of the child possessed by the be-
nevolent spirit of his grandfather that in the end he killed him. Thus did he lose
the loyalty of his subjects and precipitate his fall from power. The legendary
Leiloza of today is a tyrannical child who, after his fall, himself became a be-
nevolent spirit that possesses mediums. Between the two stories, it seems to me,
you have all the themes and elements required to start writing a proper history
of Merina attitudes toward their rulers—that is, if one wishes to understand the
crisis of authority that engulfed Merina kings from at least the beginning of the
nineteenth century, these stories tell you precisely where to look.
Let me give an example. One of the striking elements of the Leiloza story as
it’s told around Mount Ambohitrambo today is that it talks about forced labor.
Under the Merina kingdom, fanompoana, or royal service, was the overarching
principle of governance. Every subject was expected to perform some form of
labor for the sovereign. And indeed the story echoes much of what ordinary
subjects seemed to have considered most obnoxious about fanompoana at that
time: the tending of vast herds of royal cattle; industrial projects; transporting
lords too high-and-mighty to walk on the ground (real Merina royalty didn’t
employ silk bridges, but they were regularly carried about in palanquins). All
this is extremely unusual. While stories about kings told nowadays regularly
14. That is, a slave. Hastie emphasizes the African appearance of both “prophets,” but
in fact he seems to be projecting a Euro-American racial bias onto the Malagasy.
At that time, African descent would not in itself have been taken as a sign of servile
status; in fact, while northern highlanders were, then as now, more phenotypically
Asian than coastal populations, they were then more likely to be enslaved by
coastal populations than the other way around. However, it is possible that spirit
mediumship itself reached the highlands from Africa via Malagasy populations on
the coast.
266
ON KINGS
emphasize their injustice and cruelty, those stories almost never have anything
to say about fanompoana.15
Nineteenth-century sources, on the other hand, often seem to talk of noth-
ing else. Foreign observers would regularly remark that the bulk of an adult
Merina subject’s waking hours was spent either performing fanompoana for the
queen or avoiding doing so, and everything from school attendance to military
service was considered a form of royal service. Not only was fanompoana the
central principle of governance, it was key to the status system and the animat-
ing principle of royal ritual as well—if, indeed, royal work and royal ritual can
even be entirely distinguished.
What I’m going to do in this next section, then, is explore the internal logic
of that system of ritual labor that characterized the old Merina kingdom, since
I think it’s the
key to understanding Merina kingship itself.
* * *
Malagasy ritual provides an unusual challenge to the interpreter because ritual
gestures often seem to be saying two quite contradictory things at exactly the
same time. Malagasy rhetoric (or kabary) is quite similar; it often seems that
everything is double-edged, in the sense that it can be read in at least two differ-
ent ways, sometimes even three. Blessings can be curses in disguise, and curses,
blessings; statements of submission are often covert challenges; assertions of
power often take the form of mock self-effacement. Anyone who has spent
much time in a Malagasy community knows this is one of the things audiences
find most enjoyable about good kabary: observing the agility with which skillful
rhetoricians use professions of support or agreement to subtly slice each other
apart. Yet when analysts turn to ritual, they tend to assume, instead, that ritual
statements must all be taken at literal face value.
Merina royal ritual is a perfect case in point, since it’s often subject to this
kind of heavy-handed reading. Most existing literature on the subject (e.g.,
Lejamble 1972; Delivré 1974; Bloch 1977, 1982, 1986, 1989; Berg 1979, 1988,
1995, 1996, 1998; Raison-Jourde 1983b; S. Ellis 1985, 2002) emphasizes that
the legitimacy of kings was bound up with a concept referred to as hasina. Here’s
a fairly typical example:
15. This is largely because fanompoana has now become a euphemism for “slavery”
(Graeber 1996, 2007a: 43).
THE PEOPLE AS NURSEMAIDS OF THE KING
267
Common to those kingdoms in old Madagascar which succeeded one another,
rising and declining in the extent of their influence, was the notion of hasina.
This designates the invisible essence of power and fertility that can be channeled
to human beings, particularly through ancestors. Maintaining this life-force de-
mands respect for ritual obligations and taboos that in effect bind members of
a family or a community to each other, to nature, and to the land. The foremost
principle of political authorities throughout the island was that they should em-
body hasina and bestow it on their subjects. (S. Ellis 2002: 103; cf. Randrianja
and Ellis 2009: 109)
Statements like this are not so much incorrect as extraordinarily crude. They
annihilate all the subtlety and ambiguity that, in Malagasy eyes, give concepts
like hasina their conceptual power. The passage above would be a little like an-
nouncing that the English political system is organized around a notion of
“force” or “power,” then explaining that English people assume force and power
to be basic constituents of the natural universe, and finally concluding that Eng-
lish political and bureaucratic institutions exist primarily to channel force and
power in benevolent directions. None of these statements are exactly false. But
as with hasina, this, first, suggests a very naïve understanding of what are in fact
quite sophisticated concepts, and, second, assumes everything fits together in a
far neater fashion than it actually does. In fact, if one examines how the term
hasina is used in ritual contexts today, it is clear that it is in no sense some kind
of liquid fertility or power that flows one way or another, let alone something
that powerful figures “bestow” on anybody else. Hasina is first and foremost a
way of talking about powers that no one fully understands. Royal rituals always
play on this: subjects are constantly “giving hasina” to the king, in the form of
unbroken silver dollars,16 which express their desire to create a unified kingdom;
but it is utterly unclear, and indeed people seem to have been of very much
two minds, as to whether those desires created the kingdom, or whether it was
the mysterious allure of royal power that created the desire. Royal ritual often
seems to be declaring that the people recognize the power of the king, and that
the people create the power of the king, at exactly the same time. But this was
simply typical of those domains that we would label “magic” or “politics”—they
16. Most cash transactions in highland markets made use of imported silver coins
cut up and weighed: whole coins were unusual, and in this context were said to
represent the unity of the kingdom such acts of allegiance were meant to create.
268
ON KINGS
were precisely those domains where no one could ever completely understand
what was going on.
On ritual labor
How, then, might we attempt a less heavy-handed approach to Malagasy royal
ritual?
Until now, the most creative and insightful work on Merina royal ritual
has surely been that of Maurice Bloch (1977, 1982, 1986, 1999). I suspect his
work stands out not just because of the quality of the theoretical insight, but
also because Bloch is virtually the only writer on the subject who has comes at
it armed with detailed first-hand knowledge of ritual life in the northern high-
lands today.17 Bloch is most famous for his analysis of the Merina circumcision
ritual (1986)—a ceremony that is typically performed on quite small children:
boys are usually circumcised between the ages of two and four. His analysis of
fahasoavina, or circumcision rituals, is his starting point for analyzing the whole
ritual system, royal ritual included. My own ethnographic work has focused
more on mortuary ritual ( famadihana: Graeber 1995) and curing rituals car-
ried out by spirit mediums ( fanasinana: Graeber 2007a, 2007c). But I think it’s
significant that even though these rituals aren’t typically held for children, meta-
phors about children, and the raising of children, do also appear prominently in
both of them as well.
What I’m going to do in this next section, then, is to read eighteenth- and
nineteenth-century royal ritual, as it were, backward—through the kind of ritu-
als I myself witnessed in the Merina countryside.
First, mediums. Mediums, and curers in general, are often referred to as
mpitaiza olona, the nursemaids, nurturers, or carers for others. The verb mitaiza
17. About the only other body of work that sets out from contemporary ethnographic
observation and moves from there to Merina royal ritual is that by archaeologists
Susan Kus and Victor Raharijaona (e.g., 2000, 2001, 2008). Francophone literature
on the subject is largely by historians, and mostly unabashedly diffusionist, more
interested in representing the kingdom as an unstable amalgam of elements of
Austronesian, East African, Arab, or even South Asian origin than as an emergent
totality in itself. (This is most dramatically true of the work of Paul Ottino: e.g.
1983, 1986, which, while often intriguing in its own right, is of quite limited help
in solving the kind of questions being asked here.) This is not to say there have not
been ethnographically based studies of other Malagasy systems of royal ritual, in
French and in English—I am speaking strictly of the northern highlands here.
THE PEOPLE AS NURSEMAIDS OF THE KING
269
is ordinarily used for either breastfeeding an infant, or, by extension, taking care
of a small, dependent child.18
It can also be used for caring for the sick, as here,
but in this context the implication is somewhat broader than simply “taking
care of ”:19 rather, as in the case of a child, it implies the benevolent, nurturant
authority of someone more able and knowledgeable. In the case of mediumistic
curing, of course, there are in effect two levels of such benevolent authority:
the mediums, and the royal spirits that possess them. Ordinarily, mediums are
simply “pressed down” ( tsindriana) by these spirits, which means that while in
trance they remain at least partly conscious, yet hear disembodied voices, or
even see visions, directing them. But if they fall deeper into trance, the metaphor
reverses: they are no longer “pressed down by something” ( tsindrin-javatra) but
“carried by something” ( entin-javatra), and they become the spirit, their ordi-
nary personality entirely effaced.
A similar reversal lies at the center of the famadihana ritual, where the bod-
ies of ancestors are temporarily removed from tombs and rewrapped in native
silk shrouds. (The word famadihana literally means “reversal.”) When people
talk casually about their communities, they often make it sound as if ancestors
are the only real adults—living people are often referred to as ankizy, “children,”
in comparison. (If you ask who are the local elders, you will almost invariably
be told that the real elders are all dead: “Only we children remain.”) Yet during
famadihana, ancestors themselves are turned into children; they are first placed
across women’s laps—the word used for this is miampofo, “to nurse a child car-
ried on one’s lap”—given candies and honey and small change (though also
rum), coddled as one would a child. Only then, once the ancestral relation has
been thus reversed, can the bodies be taken up by everyone, danced with, re-
wrapped, and in the process, largely pulverized before being locked back in the
ancestral tomb, where they can no longer trouble the living (Graeber 1995). A
18. So one current dictionary defines the word “to breast-feed, to take care of a child
not yet capable of taking care of its own needs” (Rajemisa-Raolison 1985: 909),
with nurture, attend to, etc., as secondary derivative meanings. Richardson suggests
“to nurse, to take care of ” (1885: 662); Abinal and Malzac suggest “allaitar, nourrir,
