Review article:

The dangers of polemic: Is ritual still an

interesting topic of historical study?

G e o f f r e y K o z i o l

The Dangers of Ritual: Between Early Medieval Texts and Social
Scientific Theory. By Philippe Buc. Princeton: Princeton University
Press. 2002. 310 pp. ISBN 0691016046.
Spielregeln der Politik im Mittelalter: Kommunikation in Frieden
und Fehde. By Gerd Althoff. Darmstadt: Primus Verlag. 1997. 360 pp.
ISBN 3 89678 038 7.
Rituals of Power from Late Antiquity to the Early Middle Ages. Edited
by Frans Theuws and Janet L. Nelson. The Transformation of the Roman
World 8. Leiden: Brill. 2000. 503 pp. ISBN 90 04 10902 1.
Medieval and Early Modern Rituals: Formalized Behavior in Europe,
China and Japan. Edited by Joe¨lle Rollo-Koster. Leiden: Brill. 2002.
310 pp. ISBN 9004117490.

A sure way of knowing that a concept has temporarily outlived its
usefulness is that experts begin to polish it off with encyclopedic
treatments whose smooth-cut surfaces leave no traction for the imagina-
tion. So it is with the study of ritual. At least within anthropology, for
reasons quite relevant to this review, the subject has become more than a
little passe´. And at just this moment, Catherine Bell, herself one of the
subject’s leading theorists, published Ritual: Perspectives and Dimensions.

1

An excellent and absolutely comprehensive survey, the book leaves no
theory of ritual unturned – except, glaringly, those involving the Middle
Ages. Not only are no historians of early medieval rituals mentioned, no
medievalists of any period or specialty are cited beyond a tiny, wholly

1

Catherine Bell, Ritual: Perspectives and Dimensions (Oxford, 1997). See also the same author’s
earlier Ritual Theory, Ritual Practice (Oxford, 1992). The study of early modern European rituals
has reached the same encyclopedic closure: see Edward Muir, Ritual in Early Modern Europe
(Cambridge, 1997). Of other recent ethnological treatments, one of the most interesting is the
late Roy A. Rappaport’s Ritual and Religion in the Making of Humanity (Cambridge, 1999), a
difficult book written with tremendous care and dedication that still ends up sounding
remarkably like an updated and more thoughtful version of William Robertson Smith.

Early Medieval Europe 2002 11 (4) 367–388 # Blackwell Publishing Ltd 2002, 9600 Garsington
Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA

unrepresentative handful. Now medieval historians may not be the most
theoretically sophisticated of social scientists, but many are sophisticated
enough, and besides, we certainly have written often enough about
enough different rituals for some of our studies to have penetrated the
world of ethnologists. Yet nothing of this appears either in Bell’s
bibliography or in her text. She discusses West African circumcision
rituals (both male and female), Hopi Kachina cult initiations, Moroccan
fasting and feasting, Sinhalese healings and exorcisms, ancient Babylonese
new year’s rituals, Brazilian carnivals, Greek firewalking, the Jewish Seder,
American football and softball, graduation ceremonies, opera, Emily Post,
children’s play, childbirth, drumming _ The list goes on and on, yet the
only mentions in the entire book of anything having to do with medieval
rituals are glancing references to the Eucharist and allegorical interpreta-
tions of the liturgy, both flawed by residual unexamined Protestant
critiques.

2

Medieval historians in Britain and the United States are

known for reading ethnology, but it is an unreciprocated borrowing,
for ethnologists are not reading back.

Perhaps there is nothing very good for them to read. That is certainly

the impression an uninformed reader might take from Philippe Buc’s
The Dangers of Ritual, for it is a sweeping criticism of the way historians
have applied ethnological treatments of ritual to early medieval society.
Buc argues that an important group of historians, including, quite
prominently, the present author, have applied ethnological models of
ritual uncritically and overenthusiastically. Specifically, we have taken
over without reflection the (allegedly) functionalist, equilibrationist,
communitarian assumptions inherent in this ethnology: the assumptions
that a society is an organism that always seeks homeostasis and that
rituals ‘function’ to minimize conflict and return the society to equili-
brium. We speak of ‘rituals’ as if they had some sort of real existence in
the world that was causally effective and objectively apprehended, and
that can still be objectively described, forgetting that we know about any
given ‘ritual’ only because it was written about by a contemporary or
near contemporary whose accounts of rituals were inherently polemical
and partisan. Indeed, where the ethnology borrowed by historians
presumes homeostasis and privileges consensus, it is clear that early
medieval society was highly combative and that the clerics who described
rituals were highly argumentative. We therefore cannot speak of rituals
in the abstract as having been either active agents or passive representa-
tions of consensus and community. Community and consensus, and

2

Bell, Ritual: Perspectives and Dimensions, pp. 126, 216–7. For her knowledge of the medieval
Eucharist, Bell refers only to an article by Josef Jungmann on the pre-Reformation Mass. The
other scholars she cites on medieval ritual are Talal Asad and Norbert Elias. (She does cite
Caroline Bynum, but only for her article on Victor Turner.)

368

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

rituals as representations of them, were rhetorical tropes. They were
literary conventions used by clerical polemicists to support their prota-
gonists and their claims.

Believing that historians have been able to apply an artificial model of

ritual only by ignoring the authorial intent and artifice of our sources, Buc
proves the error of our ways by taking the opposite tack. The first part of
The Dangers of Ritual substantiates his arguments through close readings
of crucial passages in a selection of sources that describe what historians
have labelled ‘rituals’. Working backwards in time from the tenth century
to the second, Buc repeatedly shows that the consensus and sacrality he
claims historians attribute to rituals was really a fiction created by writers
engaged in partisan polemic. For example, Liutprand of Cremona,
apologist for Ottonian rule in Italy, promoted his kings’ sacrality and
the excellence of their rule by consistently depicting their rituals as
moments of order, consensus, and epiphany. Conversely, Liutprand
needed to delegitimize the rule of Italian kings, all the more because
they, and not the Ottonians, were descendants of the Carolingians.
Liutprand therefore consistently depicted Berengarist rituals as travesties,
shot through with violence and staged with manipulative duplicity.
Historians therefore cannot write about Ottonian rituals as if the rituals
themselves promoted sacrality. The sacrality and consensus of Ottonian
kingship was not an objective state of affairs but a parti pris polemic that
mobilized ritual to make its point. Presumably, had there been a
Berengarist chronicler, he would have described Ottonian rituals as shot
through with violence and cunning, and Berengarist rituals as a repre-
sentation of pontifical kingship.

3

Similar arguments are adduced in the

chapters on Carolingian annals and Gregory of Tours, but the latter
begins to move towards a point even more crucial to Buc. Though
Gregory depicts numerous rituals, he does not treat all the same way. Only
rituals that occur within or under the aegis of the church and its bishops
demonstrate true unity and community. Rituals led by or initiated by
kings invariably undermine unity and community, or replace transpar-
ency (the perfect correspondence of action and truth) with deceptive
manipulation. The obvious example is the trial of Praetextatus, manipu-
lated by Chilperic, but failed or incomplete transparency remains true of
all ‘royal’ rituals, even those of the best of Gregory’s kings, Guntram; for
Guntram’s ‘rituals’ approach the ideal of community only when he is
performing them under the aegis of and with respect for bishops.

4

Here

too we are dealing not with ‘rituals’ but with a partisan description of
them. Yet what is even more important to Buc is that Gregory was
promoting a particular ideal of the true, perfect ritual as a transparent

3

Buc, Dangers, Chapter 1.

4

Ibid., pp. 100–2, 106–7, 110–8.

The dangers of polemic

369

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

revelation of a true and perfect communitas.

5

This ideal, of course, goes

back to early Christianity, the subject of Chapter 4, dealing primarily with
the Passio Perpetuae. Here, again, far from being objectively fixed in
meaning, the meaning of rituals is precisely what was fought over, through
texts, as the Passio ‘hijacks’ a ritual seen by Romans as the execution of
criminals and transforms it into an imitation of Christ’s suffering.

6

And

here we reach the moment that Buc surely intends to be revelatory, the
reason he has chosen his unusual device of working backwards in time to
the age of the martyrs. Almost uniquely in the ancient world (save for
Jews), Christians desired not to assimilate; they therefore consciously and
completely underlined their difference with non-Christian society. Thus,
the emphasis of early Christian apologists on the idea that Christians were
not a factio, as the Romans thought them, but a mystical corpus, alive with
shared understanding (conscientia). Christians formed a coitio, coetus, a
congregatio, a community of the elect realized through shared rites, such as
prayers, acts of charity and, of course, banquets and eucharistic meals.
What Christians did was not ritual; it was sacrament. And their privileged
access to truth allowed them to see through the shell to the truth of Roman
rituals – to see that these were not sacred embodiments of a community at
all but profane and secular, without a shred of the truth but only a blind
and false belief that they were true.

7

The thrust of Buc’s argument is that this Christian cast of mind

continues to ‘haunt’ sociological and ethnological treatments of society
and ritual, for ethnologists and the historians who borrow from them
continue to see ‘good’ rituals – sacred, orderly, and deeply believed – as
signs of a community that is in fact sacred, orderly, and deeply believing,
rather than seeing these attributes for what they are: a partisan stance. The
second part of Dangers of Rituals therefore shows how these ideas
developed in and after the Reformation, but transmuted to create the
modern, secularist, sociological understanding of ritual and society. First,
the Reformation began to criticize Catholic ‘sacraments’ as mere ‘rituals’,
undermining their claim to sacramental truth by presenting them as mere
social inventions for reasons of political utility. Thus began what Buc calls
‘double religion’: those who understand the truth of rituals know that
their only real function is to promote order and obedience, while the mass
of the populace believes in them uncritically, blindly, and therefore
without true knowledge of their own beliefs and actions. In other
words, those who practice rituals have no real understanding of their
truth, which is apprehensible only by outsiders.

8

The utilitarian mode of

5

Ibid., pp. 118–22.

6

Ibid., pp. 134–40.

7

Ibid., pp. 140–7, 156–7.

8

Ibid., pp. 164–76, 188–94.

370

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

understanding rituals was furthered by European contact with non-
European peoples, which not only made Europeans aware that non-
Christian peoples had social activities that looked much like their own
sacred rites, but also led Europeans to treat those activities as merely
secular, so that unbelieving Christian diplomats and missionaries might
participate in them. The result was an increased tendency to discuss non-
European rituals and religion in terms of the merely secular and utilitarian
function of maintaining good order – and therefore to presume that
Europeans understood the true significance of other peoples’ rituals better
than those people did themselves.

9

The French Revolution accentuated

this secular, functionalist interpretation of social rituals by consciously
promoting a variety of festivals solely for the purpose of maintaining
order.

10

Even more important to creating the modern, sociological model

of ritual and social order was the conservative reaction that followed the
Revolution. For the essence of that reaction was a desire to restore to
society the sense of community that the Revolution’s promotion of
individualism and modernity had broken; and rituals were seen, increas-
ingly, as a sign of the sense of community that had been lost.

11

From

Comte to Durkheim and from Durkheim to Geertz, sociology has simply
absorbed, uncritically, the ‘baggage’ of this circular history.

12

Religion and

the rituals of religion are primarily important and understandable in terms
of their social function. That social function is seen merely as maintaining
the homeostasis of the society, preventing change and promoting order.
Being believers, members of the society cannot see this elemental, secular
truth themselves; only Europeans, privileged outside observers, can see it.
In effect, the sociologist stands to the societies he or she studies as
Protestants stood to Catholics, as Christians stood to Jews and pagans.
Thus, when Geertz, in Negara, interprets Balinese rites as promoting an
order rooted in ritual, and even cites Kantorowicz for an historical
analogy, the cogency of his model is vitiated by the fact that he has
unconsciously borrowed it from Christian society. And when a subse-
quent historian applies a Geertzian model to tenth-century West Frankish
society, he is guilty of the same unconscious ignorance.

13

9

Ibid., pp. 176–88, including an interesting discussion of the ‘Chinese Rites Controversy’.

10

Ibid., pp. 203–6.

11

Ibid., pp. 206–19.

12

Ibid., pp. 219–29.

13

Buc thus professes amazement that I do not cite Geertz’s Negara: The Theatre State in Nineteenth-
Century Bali (Princeton, 1980) in my own Begging Pardon and Favor: Ritual and Political Order
in Early Medieval France (Ithaca, 1992), even though I have so clearly been marked by him (p. 227,
n. 119), but this is a misunderstanding of both Geertz and my own work. The reason I do not cite
Geertz’s Negara in Begging Pardon is because the problems Buc sees in applying Geertz’s reading
of Balinese ceremonial to early medieval European ceremonial are obvious. Indeed, the
incompatibility runs deeper than Buc himself is aware. Since the beginning of his career and
his earliest works (for example, Agricultural Involutions: The Process of Ecological Change in
Indonesia (Berkeley, 1963)), Geertz has always emphasized the close coordination between

The dangers of polemic

371

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

The Dangers of Ritual is a dangerous book. Though a skilled writer, Buc

is acerbic and ad hominem in a way that goes beyond respectful profes-
sional disagreement.

14

In fact, what he has written is itself a polemic, for as

he himself states in his preface,

15

he is pushing a point of criticism to the

extreme, without providing an alternative, without particularly caring for
nuance, in order to raise a point he believes important; and like all
polemicists, his remedy is as extreme as his critique: banish the word
‘ritual’ from all historical analyses of early medieval society. But polemic
persuades by reducing its targets to caricatures, even misstating facts if
necessary, to the point that a reader cannot help but be persuaded, since
no person of reasonable intelligence could possibly dispute what the
polemicist says.

Of many problems with this book, this is the most serious. No

reasonable person believes or practices what Buc attributes to them –
no skilled ethnologist, no historian who has read widely enough in
ethnology to know anything about the subject. Buc is able to make his
criticism only by making generalizations that are flatly untrue. It is
therefore puzzling and disconcerting that from his preface on, Buc sets
up Durkheim as the linchpin of his attack, leads us in detailed textual
analysis backwards from the Ottonians and Carolingians to the Mer-
ovingians and martyrs, then forwards through Melanchthon, Navarrete,
Warburton, Boulanger, Dupuis, Bonald, de Maistre, de Lammenais,
Quinet, Constant and Fustel, and then, in the penultimate chapter,

culture and ecology – that is, the interactions of a given physical terrain and human culture.
Given the specificity of such ecologies, cultures are equally specific. This is a constant
presupposition in Geertz. It is fundamental to Negara, with its emphasis on the determinative
nature of Bali’s physical terrain on the political values expressed in Balinese ceremonial. It
remains fundamental in one of his later works, Local Knowledge: Further Essays in Interpretive
Anthropology (New York, 1983), and in his autobiography, After the Fact: Two Countries, Four
Decades, One Anthropologist (Cambridge, MA, 1995). Since all knowledge is local knowledge –
knowledge by the members of a society, but also knowledge of outsiders about them – it is
impossible to take a model developed to interpret one culture and blindly apply it to any other
society. In other words, if one correctly understands Geertz, one knows that one cannot apply his
model of the Balinese Negara anywhere else. If there is therefore any borrowing from Geertz in
my work, it lies in an effort to ground early medieval supplication in its own distinctive political
ecology (see especially Chapter 8). Any similarity in interpreting ‘rituals’ simply results from the
obvious fact that a certain repertoire of interpretation is common to all ethnographic,
historiographical and literary schools. (My own tend to be much more structuralist than
Geertz’s, and more akin to the formalism of the New Criticism.) In any case, far from my use of
the word ‘iconic’ being borrowed from Negara, I use the term in a way very different from Geertz
and very close to Buc’s own usage (cf. Begging Pardon, pp. 165–73, and Dangers, p. 228).
Similarly, Buc’s characterization of my argument as imitating Negara in presenting a simple
unilineal ‘borrowing’ of forms of supplication downwards through the hierarchy of rulership is
precisely what I argue against (pp. 93–103), as, following Foucault, I argue against the existence of
a ‘monistic ‘‘iconic’’ framework’ (see Dangers, pp. 244–5 and Begging Pardon, pp. 109–12, 147–59,
258–67, 304–6, 316–8).

14

Beginning with the preface, p. viii: ‘the hubris of Pharoah Franc¸ois Mitterand (r. 1981–95)’.

15

Dangers, p. viii.

372

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

having finally arrived at Durkheim – says that there is no reason to discuss
Durkheim’s work in any detail because it is so well known!

16

Of course Durkheim’s work is well known. So is the criticism of it and

all sociological and ethnological models descended from it or akin to it.

17

And those criticisms are known best of all precisely to those historians who
are the objects of Buc’s criticisms, simply because they have read widely,
deeply, and thoughtfully in ethnology for decades. Buc makes sweeping
criticisms of the way historians have used ethnology to analyse early
medieval rituals, but I can think of no historian in either Britain or the
United States who actually practises the naive functionalism Buc attri-
butes to them. Ethnologists themselves have not been able to write
ethnography the way Buc claims since 1955, when Levi-Strauss’s Tristes
Tropiques revealed the imbrication of the ‘primitive’ in the Western.

18

True, Levi-Strauss was not the darling of British social anthropologists, so
perhaps Tristes Tropiques missed them; but they did not miss the volume
Talal Asad edited in 1973 on the colonial underpinnings of earlier
ethnography.

19

Since that time, all important writers in ethnology have

made critiques of their own discipline far more devastating than Buc’s.

20

Does Buc really believe that William Miller, Stephen White, Barbara
Rosenwein, and Patrick Geary are unfamiliar with these critiques? How
can that be, when their studies prominently cite the authors of these
critiques to frame their discussions

21

and when the ethnological models

16

See also p. 223 (broadly criticizing Max Gluckman, Mary Douglas, Victor Turner and Clifford
Geertz), ‘At the risk of being allusive, I eschew an exhaustive and exhausting presentation, and
elect instead to effect a triangular confrontation _’

17

In another undue privileging of Christian elements in his account of sociology, Buc ignores the
influence of Durkheim’s early experience as a Jew in Lorraine, which Durkheim himself
privileged. See Steven Lukes, Emile Durkheim: His Life and Works (New York, 1972), pp. 39–40;
Robert A. Nisbet, Emile Durkheim (Englewood Cliffs, NJ, 1965), p. 28; W.S.F. Pickering, ‘The
Enigma of Durkheim’s Jewishness’, in Debating Durkheim, ed. W.S.F. Pickering and H. Martins
(London, 1994).

18

Claude Le´vi-Strauss, Tristes Tropiques, trans. John and Doreen Weightman (Harmondsworth,
1992), first published in French, 1955.

19

Talal Asad (ed.), Anthropology and the Colonial Encounter (London, 1973).

20

Among countless works, see George E. Markus and Michael M.J. Fischer, Anthropology as
Cultural Critique: An Experimental Moment in the Human Sciences (Chicago, 1986); James
Clifford and George E. Marcus (eds), Writing Culture: The Poetics and Politics of Ethnography
(Berkeley, 1986); Paul Rabinow, Reflections on Fieldwork in Morocco (Berkeley, 1977); Kevin
Dwyer, Moroccan Dialogues: Anthropology in Question (Baltimore, 1982); Renato Rosaldo,
Culture and Truth: The Remaking of Social Analysis (Boston, 1989); Johannes Fabian, Time
and the Other: How Anthropology Makes Its Object (New York, 1983).

21

For example, William Ian Miller, ‘Gift, Sale, Payment, Raid: Case Studies in the Negotiation
and Classification of Exchange in Medieval Iceland’, Speculum 61 (1986), pp. 18–50; Stephen D.
White, ‘Proposing the Ordeal and Avoiding it: Strategy and Power in Western French
Litigation, 1050–1110’, in Thomas N. Bisson (ed.), Cultures of Power: Lordship, Status, and
Process in Twelfth-Century Europe (Philadelphia, 1995), pp. 89–123, and the same author’s much
earlier Custom, Kinship, and Gifts to Saints: The ‘Laudatio Parentum’ in Western France, 1050–1150
(Chapel Hill, 1988), pp. 8–9 and 238–9, n. 34 (with Durkheim specifically mentioned and
criticism assumed); Barbara Rosenwein, To Be the Neighbor of Saint Peter: The Social Meaning of
Cluny’s Property (909–1049) (Ithaca, 1989), pp. 125–43; Patrick Geary, ‘Vivre en conflit dans une

The dangers of polemic

373

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

Buc presumes remain orthodoxy had already been supplanted by the mid-
1970s?

22

As for Buc’s most critical point, that historians who use anthro-

pology are unaware that the idea of communitas is overdetermined within
western civilization, I find this difficult to believe. Quite apart from the
fact that it is impossible to read much of Victor Turner or Mary Douglas
without becoming acutely aware that their interpretations of non-western
society, ritual, and religion are problematic because so very Catholic,

23

it is

very hard to be a literate, historically minded intellectual, at least in the
United States, without knowing that modern discussions of community
have a long genealogy predisposed to reproducing the same intellectual
patterns, precisely as Buc alleges.

24

France sans e´tat: typologie des me´canismes de re`glement des conflits (1050–1200)’, Annales,
E.S.C. 41 (1986), pp. 1107–33; also Koziol, Begging Pardon, pp. 93–4 and n. 43, and especially
Chapter 9.

22

For example, in my own graduate seminars on legal anthroplogy at Stanford University in 1975–
76, the anthropology Buc holds out as the norm was so far from being the norm that its great
names (Evans Pritchard, Radcliffe-Brown, Gluckman) were held out as examples of how not
to do ethnography, precisely because of their unexamined functionalist, equilibrist, and
Eurocentric presuppositions and colonial framing (Gluckman being unusually important
because, like his student Turner, both a late exemplar and early critic of the approach). Instead,
study concentrated on what became known as a ‘process’ approach, identified in particular with
the Berkeley anthropologist Laura Nader. (See Law in Culture and Society, ed. Laura Nader
(Chicago, 1969); The Disputing Process: Law in Ten Societies, ed. Laura Nader and Harry F.
Todd, Jr (New York, 1978).) Central to Nader’s writings were the assumptions that law and legal
procedure were fundamentally imbricated in power relations within society; that they were tools,
access to which was unevenly distributed throughout a society; that far from working to restore
homeostasis, they worked to embed existing power relations ever more deeply within a society;
that societies were constantly in a state of disequilibrium and dysfunction. In other words, the
critique Buc claims historians are ignorant of was, in fact, basic to the teaching of the discipline in
the mid-1970s at his own university.

23

C. Clifford Flanigan, ‘Liminality, Carnival, and Social Structure: The Case of Late Medieval
Biblical Drama’, pp. 42–63, at pp. 57–8, and Ronald L. Grimes, ‘Victor Turner’s Definition,
Theory, and Sense of Ritual’, pp. 141–6, at pp. 144–5, both in Kathleen M. Ashley (ed.), Victor
Turner and the Construction of Cultural Criticism: Between Literature and Anthropology
(Bloomington, IN, 1990). See also Edith Turner’s reminiscences of her and her husband’s
conversions to Catholicism in the late 1950s, while still at Manchester, and the importance of
their Catholicism and his membership in the Communist Party to their ethnographic work:
Matthew Engelke, ‘An Interview with Edith Turner’, Current Anthropology 41 (2000), pp. 843–
52, at pp. 847–9.

24

We might draw that genealogy a bit differently than Buc does. He disagrees with Jonathan Z.
Smith’s suggestion that Zwingli’s and Calvin’s critiques of Catholic sacraments were instru-
mental in establishing a secular view of ritual, on the grounds that their movements lacked the
reach of Lutheranism and Anglicanism. Buc therefore believes that the latter were the two truly
important movements in creating the idea of a false religious community bound by rituals
(Dangers, pp. 164–5, n. 3). This, however, is a very parochial – and conservative – narrative of
intellectual history that shows little awareness of British and American history. Though Buc may
not be wrong, Smith was certainly right, for the tension between communities as immanent
moral entities and legally constituted ones and the problematic position of the individual’s
freedom within politico-moral communities does come into the Anglo-American tradition
through Calvin, via English, Scottish and American Puritanism, as reinvigorated and reinter-
preted in the United States by every subsequent Great Awakening and evangelically based call for
the reform of political morality.

374

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

If all this is so well known, how then can Buc present historiographical

work as so retrograde? In the case of his most prominent target (and the
one I am in the best position to know about) as polemicists do, by rather
breathtaking ellipses in his quotations that transform the original meaning
of a sentence out of all recognition.

25

Or again as polemicists do, by

relentless, ever shifting attacks that divert attention from the inconsis-
tencies inherent in the writer’s own position. And Buc’s own position is
deeply inconsistent. Though he criticizes historians for having invented a
‘swollen grab bag’ of rituals,

26

he himself discusses exactly the same

swollen grab bag of practices: anointings, pacts, oaths, burial, adventus,
occursus, supplications and humiliations, royal acts of charity for the poor,
burials ad sanctos, liturgical commemorations for the royal family, tearful
prayer before the Holy Lance, tearful prayer before the relics of saints,
pilgrimages to shrines, Christmas and Easter feasts and Lenten fasts,
translations of relics, among many others. Not only does he discuss the
exact same rituals, he interprets their symbolic meanings in ways that even
Victor Turner would find absolutely unexceptionable. And though he
criticizes us for inventing new rituals, he invents new rituals himself,
suggesting that Childeric’s poetry ‘may have been an element in a royal
liturgy of thanksgiving’.

27

Of course, Buc does not like calling these

activities ‘rituals’. An arch-philosophical nominalist, he would have us
entirely banish the word ‘ritual’ from discussions of religion and society,
both because such modern abstract categories do not accurately reflect the
specificity of contemporaries’ actions and because the word brings with it
from ethnography a host of allegedly unexamined assumptions. But the
same is true of absolutely every abstraction used by historians. ‘Society’,
‘culture’, ‘institutions’, ‘religion’, ‘church’, ‘structure’, ‘law’, ‘feud’, ‘king-
ship’, ‘court’, ‘family’, ‘class’, ‘status’, ‘nobility’, ‘peasantry’ _ ‘the middle
ages’: no matter the extent to which these words descend from medieval
cognates, all are used currently by historians in terms broader and more
abstract than in medieval usage. The struggle to keep such words fresh and
flexible, to use them profitably without reifying them, is elementary to the
historians’ task. The reason it must be a struggle is because we can’t banish
these words. We have to use them.

So it is with the word ‘ritual’. Buc himself repeatedly uses the very word

he would have others banish. And it is not enough for him to pretend that
he only uses the word as if in quotation marks, as shorthand for ‘a
practice twentieth-century historians have identified as ritual’,

28

when he

25

Compare, for example, Buc’s quotation (p. 211) from Begging Pardon (p. 59): in the original, it is
not the rituals that are important in maintaining power but prestige, loyalty, friendship, and
kinship, rituals being a way these are publicly manifested and tested.

26

Dangers, p. 176.

27

Ibid., p. 113.

28

Ibid., p. 2.

The dangers of polemic

375

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

consistently uses it not as the object of a position he dislikes but precisely
as the historians and ethnologists he criticizes have used it, to define
exactly the same practices which he interprets in exactly the same modes.
Nor is one fooled by his use of words like ‘liturgy’, ‘solemnity’, and
‘ceremony’: given the way Buc actually uses them, these are nothing but
synonyms for ‘ritual’, as when he speaks of the ‘liveliness of liturgical
idioms’ in discussing Merovingian kingship, or explains the importance of
processions in Rome by suggesting that ‘litanies served to measure
power’.

29

Does a synonym really offer a profound analytic advance? Do

the quotation marks really make a difference when Buc emphasizes that
the Carolingians consciously created and elaborated ‘solemnities of all
sorts, solemnities that following current historiographic convention we
shall provisorily call ‘‘rituals’’’?

30

How seriously are we to take his own

demand that we abolish the term ‘ritual’ when he writes that by the ninth
century the ‘great political rituals _ were well established’, that ‘competi-
tion and ritualization fed one another’, that the ninth-century parts of
the Liber Pontificalis ‘abound in descriptions of ceremonies’, that ‘rituals
are so present in Gregory of Tours’s works that one might almost fail to see
them’?

31

Can he really criticize historians’ use of the word ‘ritual’ and then

describe tristitia and laetitia as ‘public bodily postures intended to
manifest respectively political hostility and friendship’, when a public
bodily posture intended to manifest anything is one of the most elemen-
tary definitions of what a ritual is (and a particularly naive definition at
that, since it presumes precisely what all recent ethnologists and historians
have taken as the central problem that requires explanation – that is, how a
physical gesture can intend or manifest meaning at all)?

32

Such incon-

sistencies riddle the book. Thus, his chapter on martyrs begins by noting
the importance of public spectacles in Rome, including executions, then
adds, ‘It was indeed desirable that everything should take place according
to a set scenario – hence, patterns, forms, some would say, ritual.’ This is
disingenuous in the extreme, to use the term ‘ritual’ in its ordinary
meaning, apply it to the kinds of activities it ordinarily denotes in ways
that are utterly conventional, yet claim to be doing something novel and
distinctive. When he then continues the sentence by saying that ‘spec-
tacles, either of the stage or of the circus, provided the universally
acknowledged litmus test of civic unity and order’, providing this gloss
without any trace of irony whatsoever – this within a book that castigates
historians who allegedly interpret rituals as litmus tests of civic unity and

29

Ibid., pp. 109, 87.

30

Ibid., p. 57.

31

Ibid., pp. 58, 62, 93.

32

Ibid., p. 66. On the naivety of this implicit definition, problematic because isolating particular
moments and behavioural forms from the cultural and socio-/politico-/economic context that
informs them, see Koziol, Begging Pardon, 289–307.

376

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

order – he forgoes the right to posture as critic.

33

You cannot criticize

historians, then take over their positions and claim them as your own, least
of all simply by putting the word you criticize in quotation marks.

It is astonishing that a writer so intelligent can have such blind spots.

Presuming that he is not as cynically manipulative of texts as he claims
his clerical alter egos were, one can only conclude that Buc really believes
his critique well founded, and that he has therefore somewhere seriously
misapprehended the work of those he criticizes. I believe that three
recurring contentions in Dangers of Ritual point to three such misap-
prehensions. The first is a dismissal of the purposes of social and cultural
history and a resultant misreading of the ways ethnologically minded
historians have used textual sources to write such history. The second is a
misconstrual of the reasons historians speak of order and community.
The third, and most fundamental, is a misunderstanding of the reasons
historians define and use rituals in ways that Buc finds vague and all
encompassing. Almost any book by almost any well-regarded historian
of early medieval ritual could illustrate the misunderstandings. One
might begin with one of the best known and most influential.

Everyone interested enough in early medieval rituals to be reading this

article has probably read Gerd Althoff ’s Spielregeln or the articles the book
gathers together and will recognize his positions in many of Buc’s
criticisms. Althoff does rely on the classic ‘functionalism’ of the British
school of social anthropology, though in a way that is idiosyncratic rather
than systematic. He insists on the existence of ‘unwritten rules’ of
behaviour, where either the rules or the behaviour allows the social
‘system’ to ‘function’, in the sense of return to an equilibrium – a status
quo ante.

34

Furthermore, one of Buc’s most damning criticisms of current

historiographical treatments of ritual is their lack of concern for authorial
intent, and though Althoff relies heavily on primary sources, he is not
particularly interested in the social and political context within which his
sources were produced or in the authorial intent of the writers who
produced them. Finally, in very few but also very important passages
Althoff speaks of ‘representation’ – specifically, of the representational
capacity of rituals – in ways that are related to the conservative usage of
Carl Schmitt that Buc rightly criticizes.

35

Yet one should be not only critical but also generous. It is worth

remembering that when Althoff wrote the early articles and papers
reprinted in Spielregeln, the terrain of ritual studies in Germany was
quite other than it has been since, that it was, in fact, far closer to the

33

Ibid., p. 128; note also the implicit functionalism of Buc’s own analysis of public prostrations
(pp. 104, 106).

34

Spielregeln, pp. 1–17, 283–304 and passim.

35

Ibid., pp. 13, 231, 259.

The dangers of polemic

377

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

representational inheritance of Tellenbach, Schramm, and Erdmann that
Buc criticizes.

36

If it is no longer, that is partly because of what Althoff has

written. In any case, although Buc’s implicit criticisms of Althoff are in
some ways correct, in more important ways they misrepresent his
assumptions, purposes, and sophistication. To begin with, Althoff rarely
speaks of ‘ritual’ in the abstract or of ‘rituals’ collectively – much less often
than Buc, in fact.

37

Indeed, precisely like Buc, this most particularistic of

historians is far more comfortable speaking of and analysing discrete
events in contextually specific, highly nuanced readings: the election and
pre-anointing procession of Conrad II in 1024, Henry IV’s penance at
Canossa, the deditio of the Milanese in 1158, the second Milanese deditio of
1162.

38

To be sure, Althoff rarely discusses the sources he uses, but this is

hardly because he does not realize that authorial intent shapes and
misshapes their accounts of rituals. Quite the contrary, he says quite
clearly that the image of Conrad II as the true lord of freemen has more to
do with Wipo’s norms than with eleventh-century reality.

39

If, therefore,

he ignores problems of authorial intent, it is for good reasons, the same
reasons that William Miller uses Icelandic sagas as sources for Icelandic
behavioural patterns, or that Stephen White and Barbara Rosenwein use
charters to reconstruct networks of alliance, or that I myself use charters
and histories to isolate distinctive linguistic and gestural patterns of
deference to high lords: to discover the patterns of behaviour and belief
that are common to all writers in a society, regardless of their motivations
in writing any specific text.

40

The fact that Buc is an intellectual historian is

no reason for him to fall back into such a radical nominalism that it
discredits the efforts of others to try to use clerically produced texts to
discern the behaviour and beliefs of laymen and laywomen in courts,
assemblies and halls.

Althoff ’s history simply has different purposes from Buc’s, very

specifically and explicitly so. Nearly every article in Spielregeln begins
with a statement of the historiography against which it was written. For

36

For example, Gerd Tellenbach, ‘Ro¨mischer und christlicher Reichsgedanke in der Liturgie des
fru¨hen Mittelalters’, Sitzungsberichte der Heidelberger Akademie der Wissenschaften, Philologisch-
historische Klasse, 1 (1934–35) (Heidelberg, 1934); P.E. Schramm, Kaiser, Ko¨nige, und Pa¨pste:
Gesammelte Aufsa¨tze zur Geschichte des Mittelalters, 4 vols (Stuttgart, 1968–70); idem, Herrschafts-
zeichen und Staatssymbolik, 3 vols (Stuttgart, 1954–56); Carl Erdmann, Forschungen zur politischen
Ideenwelt des Fru¨hmittelalters (Berlin, 1951).

37

Only a little more frequently he will speak of rituals adjectivally (for example, ‘ritualistic’), to
refer to actions that exhibit recurrent patterns of formal behaviour (for example, pp. 29, 53, 125,
301, 303).

38

Spielregeln, pp. 164–6, 240–3, 103–7.

39

Ibid., p. 27.

40

William Ian Miller, Bloodtaking and Peacemaking: Feud, Law, and Society in Saga Iceland
(Chicago, 1990), Chapter 2; Stephen D. White, ‘Proposing the Ordeal’, and ‘Feuding and Peace-
Making in the Touraine around the Year 1100’, Traditio 42 (1986), pp. 195–265; Rosenwein, To
Be the Neighbor of Saint Peter; Koziol, Begging Pardon, Chapters 1, 6, 7, also pp. 147–59 for the
way in which Dudo of Saint-Quntin’s authorial intent conditions his accounts of supplications.

378

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

the early ones, that historiography is nineteenth-century German Verfas-
sungsgeschichte and its presumption that there was an early medieval
Staatlichkeit which, while not that of a modern state was still like enough
to it to bear the ideological weight of nineteenth- and early twentieth-
century German national aspirations. The focus of these historians was
therefore kings and the courts of kings, laws issued in the name of kings,
the fundamental quasi-constitutional principles that governed relations
between kings and nobles, and their abiding belief was that the Car-
olingian administrative order possessed a ‘rationality’ (Rationalismus) of
purpose and means perhaps not as fully realized as that of the modern state
but recognizably embryonic to it. Yet even assuming that this vision
matched Carolingian reality (which it did not), following the decline
of the Carolingians there are no sources that support anything like this
vision of political order – no written laws, few written judgements, little
trace of formal legal proceedings that look anything like earlier or later
legal proceedings. Given this caesura, Althoff writes, historians had two
alternative paths. One, the older, was to assume that tenth- and eleventh-
century law and administration followed the same principles as the
Carolingians’. Since there was no evidence for this and much against,
the second, more recent approach was to see the tenth and eleventh
centuries as a time of violence and disorder between the collapse of the
Carolingians and the appearance of Frederick Barbarossa’s feudalizing
programme. Althoff ’s goal is therefore quite simple: to try to understand
what did the work of ‘law’ in a society that had no formal written laws and
no formal legal institutions.

41

What makes Althoff ’s work most distinctive and valuable is that rather

than begin his search with norms extrinsic to tenth- and eleventh-century
society, he begins with its most distinctive traits. These are: the lack of
power of the Ottonians at the beginning of their rule; the dispersal of
rights and powers among the nobility; and most crucial of all, the
importance elites attached to rank, honour and precedence. The unusual
coherence of Althoff ’s model of ‘rituals’ in tenth- and eleventh-century
German society stems from the fact that everything flows from his
understanding that the nobility was contending over issues of rank,
precedence and honour.

42

Thus, one of his more notable arguments is

that absolutely every public action was carefully prepared and scripted,
that no public action was ever spontaneous. Simply because rank and
order were so important but maintained in such delicate balance, changes
to rank and order threatened to upset everything. The dangers of hasty
action, spontaneous action, inconsiderate action, the dangers of any
action at all were so momentous that everything was done to make sure

41

Spielregeln, pp. 1–17.

42

Ibid., pp. 2, 12, 23, 162–3, 251–2, 279–80.

The dangers of polemic

379

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

there were no surprises.

43

Thus, the details of the deditiones that accom-

panied surrender were, for Althoff, all negotiated beforehand and care-
fully scripted, nothing at all being left to chance.

44

In terms of the

historiographical tradition within which Althoff writes, perhaps even
more important is the distinction this leads to between public and private
consensus-seeking. German ‘constitutional’ historians had always debated
the extent to which the nobility had a recognized right to give counsel and/
or consent, and the extent to which kings could ignore or act against that
counsel. Althoff ’s spin on this debate is to point out that the statements of
consensus we find in chronicles (for example) refer not to a real process of
decision-making but rather to a formal, public, ‘demonstrative’ enact-
ment of counsel, where decisions that had already been reached privately,
through the canvassing and mobilizing of opinion (Willensbildung), were
formally and publicly acted out.

45

In an interesting reversal of Buc’s

priorities, he even argues that literate modes of communication replicated
this fundamental reflex, insofar as the relationship between letters and
Streitschriften corresponds to that between private and public means of
gaining consensus.

46

Though shifting his critique from Verfassungsgeschichte to Norbert Elias

and Ju¨rgen Habermas, the later articles build on the same insights. Thus,
Althoff responds to Elias’s view of early medieval nobles as unable to
control their emotions and therefore in need of ‘civilizing’ by arguing that
showing highly pitched emotions like anger was, in fact, a demonstrative
action, not a ‘ritual’ but a ‘ritualized’ display. A member of the nobility was
not described as flying into a towering rage because he had been unable to
control his emotions. Rather, in a world in which actions counted, the
accompaniments that signified ‘rage’ were intended to demonstrate and
communicate to potential enemies that they had gone too far, that one was
willing to fight, that the next step in the escalation of a conflict would be
bloodshed. In other words, not even emotions were ‘spontaneous’.

47

As for

Habermas, Althoff argues against his modernist assumption that the early
Middle Ages had no sense of public space. The constituent elements of
public space were different: not congresses, city streets, or newpapers but
rather courts, halls, letters, and stories. Yet the early Middle Ages had a very
highly developed sense of public space, for in a sense, all demonstrative
action was geared towards public communication.

48

There are difficulties with Althoff ’s analysis, particularly with its

theoretical superstructure. Althoff believes that tenth- and eleventh-

43

Ibid., pp. 66–7, 248–50, 256, 264, 273–4.

44

Ibid., pp. 101–3, 125.

45

Ibid., pp. 157–84.

46

Ibid., pp. 183–4.

47

Ibid., pp. 258–81.

48

Ibid., pp. 229–57.

380

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

century German society was governed by rules, known by everyone, clear
to everyone, unambiguous even though unwritten. He says that the
purpose of these rules was to allow the continued functioning of the
system, for if there had not been such clear rules, the entire system would
have fallen apart.

49

At the level of both theory and history, this leads to

problems. In what sense were what he identifies as ‘rules’ really rules?
When he actually comes down to stating what the rules were, they do not
seem very rule-like at all: disputes must be processed in clear stages of
escalation; mediators are essential to the settlement of disputes; composi-
tion and satisfaction are necessary to end disputes.

50

I can imagine calling

these ‘principles’, if one wishes to emphasize the normative, or ‘patterns’, if
one wishes to emphasize non-normative praxis, but to call them ‘rules’
seems to be bending the word considerably, if only because they are far too
broad. Indeed, the breadth of these ‘rules’ is manifested by the fact that
Althoff also emphasizes how flexibly they could be applied. Depending on
circumstances, one could make a deditio more or less harsh, do more or less
to work to get oneself reinstated in grace.

51

But where there is so much

flexibility and variation, then it ceases to be useful to think in terms of rules
at all. ‘Code’ and ‘custom’, though far from unproblematic, are perhaps
safer terms, because unlike ‘rule’, they convey both norm and flexibility,
and do so with an implicit assumption of diffuse social reinforcement in
place of coercive governmental enforcement.

Althoff ’s appeal to ‘function’ is also problematic, in a way that

underscores why ethnologists, sociologists, and historians have repu-
diated the term: where everything in the society works to maintaining or
restoring equilibrium, it becomes difficult to explain change. Interest-
ingly, some of Althoff ’s most astute analyses concern just that – dramatic
changes in or violations of the rules. For example, one of his most
important arguments is that the tenth-century Ottonian polity was
different from either the Carolingian or the Staufen, because the relative
lack of power of the new dynasty and the dispersal of power among the
nobility required more consensual interactions between them. So where
Carolingian deditiones, like Ottonian, ended in displays of mercy,
Carolingian mercy, unlike Ottonian, lay in commuting a death sentence
to mere blinding.

52

And if Staufen legal processes were as demonstrative

as Ottonian, what they demonstrated was the kings’ new attention to
asserting their superior power.

53

However, in a society in which everyone

knew the rules and no one would break them, how does one explain these

49

Ibid., pp. 3, 125, 187, 253–4, 287–90, 297, 300, 303–4.

50

Ibid., p. 53; also p. 294: those of highest rank decide whether to introduce issues into public
discussion, and opinions are presented in public gatherings according to individuals’ rank order.

51

Ibid., pp. 108, 111, 122–3; similarly for all rituals, pp. 16, 36–7, 63–5.

52

Ibid., pp. 37, 53–6.

53

Ibid., pp. 38–9, 52–3, 63 (seeing a growing ‘hardness’ already under the Salians), 71–3.

The dangers of polemic

381

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

changes, except by virtue of the fact that someone at some point was
willing to break them?

In the final analysis, this is my real disagreement with Althoff, and with

all those recent historians who have applied a semiotic ethnography to
emotions in a way that routinizes them, in the same way that old
definitions of ritual routinized behaviour. Such analyses cannot explain
a situation in which honour counts so much to an individual who believes
he has suffered an affront that he is willing to violate all the rules, risk
everything, lay hands on the Lord’s anointed, make a scene in a public
gathering, assassinate a court favourite. For Althoff and many others, these
have become nothing but signs.

54

But where all emotion has become

nothing but a sign that one wishes to renegotiate an ongoing, dyadic
relationship, what has happened to the emotions that people felt, to the
ideals they valued, the values that gave them their identity?

55

If honour and

rank were so important to men and women of the tenth century, are we to
believe that they really didn’t get angry when they suffered affronts to their
honour and rank? There is a point, in the recent attention not to ritual but
to semiotics, in which everything has become a sign of something other
than it is. And Buc himself is guilty of this ‘double-psychology’, for
though one part of his argument asks us to take the religion of these people
seriously, another sees rituals (including rituals of religion) as nothing but
a set of rhetorical tropes that cover a writer’s partisan political loyalties.
Ironically, Buc’s emphasis on the rhetorical nature of our sources has
stripped both actors and writers from real belief in the same way he claims
ethnologists have, and he claims the same stance to superior under-
standing he discounts in ethnologists.

In any case, even though Spielregeln is the book that most corresponds

to the functionalist ethnology Buc criticizes, there is little in it that really
corresponds to Buc’s criticisms. It is true that Althoff does speak of the
‘representation of rulership’. He does so, however, only once. And the
phrasing is noteworthy, because elsewhere Althoff criticizes past historians
for speaking of ‘public communication’ entirely in terms of Herrschafts-
repra¨sentation.

56

Althoff ’s public rituals are consciously constructed,

54

Ibid., pp. 11–12, 63, 65–6, 124–5, 228, 232, 245, 252–3, 262 (‘Demonstrationsfunktion, Signal-
charakter’). See also Geary, ‘Vivre en conflit’; also Gerd Althoff, ‘Ira Regis: Prolegomena to a
History of Royal Anger’, pp. 59–74, Richard E. Barton, ‘‘‘Zealous Anger’’ and the Renegotiation
of Aristocratic Relationships in Eleventh- and Twelfth-Century France’, pp. 153–70, and Barbara
H. Rosenwein, ‘Controlling Paradigms’, pp. 233–47, all in Barbara H. Rosenwein (ed.), Anger’s
Past: The Social Uses of an Emotion in the Middle Ages (Ithaca, 1998).

55

Some anthropologists are becoming aware of this dilemma, though there is still no good model
for how to approach it. See Sandra Bell and Simon Coleman (eds), The Anthropology of
Friendship (Oxford, 1999); Joanna Overing and Alan Passes (eds), The Anthropology of Love and
Anger: The Aesthetics of Conviviality in Native Amazonia (London, 2000).

56

Spielregeln, p. 259: ‘in den Situationen, in denen im Mittelalter Politik gemacht und Herrschaft
repra¨sentiert wurde’; otherwise, p. 13.

382

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

tightly negotiated pieces of choreography, highly scripted theatre (Insze-
nierung) – something quite different from those of Carl Schmitt’s
representational model.

57

How could it be otherwise? Althoff ’s analysis

has taken the ground out from under all representational modes of
analysis. Since everything in his society comes down to contending
individuals, kindred alliances – even the king himself is a contender –
he has left us with nothing really to represent, even embryonically or
ideally. For the same reasons, Althoff does not see tenth-century society as
holistic in a Durkheimian sense; far from it, his entire analysis is
predicated on the idea that primary loyalties and experiential commu-
nities were found in highly shifting and particularistic friendship and
kinship networks. He does not ignore contention. On the contrary, every
member of his society is constantly contending. He does not presume or
construct a reified social reality, in which society in the abstract is more
real than the individuals who comprise it. Instead, he gives us a society
comprised of contending, ever shifting groups, those groups themselves
comprised of proud, contentious individuals.

58

As to the role of ‘ritual’ in

this society, rituals do not provide a sham of consensus in the absence of
the reality. Quite the contrary, Althoff ’s rituals are the means of
contending. This is why it is so significant that he rarely speaks of ‘ritual’
at all, but speaks instead of ‘demonstrative action’ and ‘communication’.

59

Simply because tenth-century nobles were driven by concern for their
rank and honour (in both senses of the word), simply because they were
willing to fight – indeed, had to fight – to enforce respect for their rank
and honour and avenge insults to them, changes in rank and honour or any
action that portended such changes were threatening. Such changes and
micro-changes were therefore marked not by ‘rituals’ in any kind of
artificial sense but by a host of particular, very concrete signs. Where one
took a seat, where one was seated, who spoke to whom first, with what
inflections of familiarity – these were ‘demonstratives’, because their
purpose was to publicize rank, order and favour, and changes in them.

And what is true of Althoff is true of all those historians whom Buc

criticizes. He assumes that we presume community. In fact, we presume
the same contention, rivalry, and manipulation that Buc assumes.
Community is not an assumption we begin with, nor is it a simplistic,
lazy solution we impose on the society by using contemporary ethno-
logical models. Community is the problem. Simply because the society
was so contentious, order, community, and consensus become the
problem, not the explanation.

60

57

Ibid., pp. 12–13, 101, 167, and especially 229–57. Cf. Buc, Dangers, pp. 231–7.

58

See especially Spielregeln, pp. 21–84.

59

Spielregeln, pp. 11–14, 35–7, 108, 113, 123–5, 187, 202–3, and especially 229–57.

60

Ibid., pp. 1–3 and 21–84; Koziol, Begging Pardon, pp. xi, 5–7, 59, 110–12, 128–30, 194–8, 254–7;
White, ‘Feuding and Peace-Making’ and ‘Pactum Legem Vincit et Amor Judicium: The

The dangers of polemic

383

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

Rituals of Power, a volume in the series on The Transformation of the

Roman World sponsored by The European Science Foundation, presents
thirteen articles written by scholars from seven European countries
covering a host of different topics from a variety of methodological
stances.

61

The articles are of very mixed quality. At the very least, the

volume as a whole presents interesting material on a wide array of subjects:
Roman funerary rituals; Germanic origin legends; Visigothic and Lom-
bard royal rituals and Herrschaftszeichen; Frankish investiture with arms;
the circulation of weapons in Anglo-Saxon society; Beowulf and the
reproduction of social identity.

62

Four articles are particularly noteworthy.

Mayke de Jong’s contribution tackles the ‘master narrative’ of public
penance that sees a decline in a dominant early Christian scheme of public
penance in the sixth century. De Jong argues that, to the contrary, there
had never been a strongly institutionalized public penance, that in fact
public penance was something of an invented tradition even for the sixth
century, and that much of the evidence for early Christian public penance
results from a programmatic Carolingian reading of these sixth-century
sources. It is an important article, with great ramifications not only for the
history of the sacraments but also for the history of monasticism and
Christian piety.

63

In an article typical of her craftsmanship, Janet Nelson

provides a very useful sketch of Carolingian royal funerals, arguing against
Alain Dierkens that, in fact, the Carolingians devoted substantial atten-
tion to the public, regalian displays of funerals.

64

Christine La Rocca and

Luigi Provero present a close and very subtle reading of the will of
Eberhard of Friuli and his wife Gisela, daughter of Louis the Pious. Every
early medievalist interested in books, writing, the nobility, the family,
inheritance, and the construction of power should read it – which is to say
that every early medievalist should read it.

65

Finally, Frans Theuws and

Settlement of Disputes by Compromise in Eleventh-Century Western France’, American
Journal of Legal History 22 (1978), pp. 281–308; Miller, Bloodtaking and Peacemaking. The
agenda was announced in an influential article by Frederic Cheyette, though written within the
framework of the older functionalist anthropology of law: ‘Suum Cuique Tribuere’, French
Historical Studies 6 (1970), pp. 287–99.

61

On the project, see Ian Wood, ‘Report: The European Science Foundation’s Programme on the
Transformation of the Roman World and Emergence of Early Medieval Europe’, EME 6 (1997),
pp. 217–27.

62

Javier Arce, ‘Imperial Funerals in the Later Roman Empire: Change and Continuity’, pp. 115–29;
Lotte Hedeager, ‘Migration Period Europe: The Formation of a Political Mentality’, pp. 15–57;
Pablo Diaz and M

a

. R. Valverde, ‘The Theoretical Strength and Practical Weakness of the

Visigothic Monarchy of Toledo’, pp. 59–93; Stephan Gasparri, ‘Kingship Rituals and Ideology
in Lombard Italy’, pp. 95–114; Re´gine Le Jan, ‘Frankish Giving of Arms and Rituals of Power:
Continuity and Change in the Carolingian Period’, pp. 281–309; Heinrich Ha¨rke, ‘The
Circulation of Weapons in Anglo-Saxon Society’, pp. 377–99; Jos Bazelmans, ‘Beyond
Power: Ceremonial Exchanges in Beowulf ’, pp. 311–75.

63

‘Transformations of Penance’, pp. 185–224.

64

‘Carolingian Royal Funerals’, pp. 131–84.

65

‘The Dead and their Gifts: The Will of Eberhard, Count of Friuli, and his Wife Gisela, Daughter
of Louis the Pious (863–864)’, pp. 225–80.

384

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

Monica Alkemade’s examination of weapons depositions in northern
Gaul and Belgica is the most suggestive work I have read not only on
weapons depositions, but also on the transformation of the mechanics and
symbolics of power in the sub-Roman period.

66

Even such a bare sketch indicates the breadth of subject matter included

in the volume. It may also hint at an important issue. Many of the authors
assume a very broad definition of ritual. In fact, the usage is often so
broad that several articles have very little to do with ‘rituals’ as they are
commonly understood, to the point that one can read an entire seventy-
page article and wonder where the rituals are. A superficial conclusion
would be that Buc is correct, that these authors have applied the term
‘ritual’ lazily, that the term has been so emptied of meaning that just
writing a will becomes a ‘ritual’. However, that conclusion would itself be
lazy. In fact, it is precisely the most theoretically advanced articles on ritual
in which rituals tend most to disappear into the background.

67

Under-

standing why their authors’ theory has led to such latitudinarian usage is
important, because it is this misunderstanding that I suspect accounts for
the most fundamental of Buc’s misapprehensions about recent studies of
ritual. The ethnological model most frequently resorted to in Rituals of
Power is the model most commonly used by post-structural ethnologists
and post-processual archaeologists. Its assumptions are now so widespread
that its precise author is no longer relevant. In this volume, one usually
meets them in the ‘hierarchy’ of Louis Dumont or the faits sociaux totaux
of Maurice Godelier (via Marcel Mauss); Pierre Bourdieu’s habitus or
even Michel Foucault’s ‘discourse’ would do as well.

68

Indeed, no

authorizing theorist is required at all; the issue should be apparent to
anyone who has studied rituals long enough, or read enough ethnology or
literary criticism, or simply thinks deeply on the problem. A ritual by itself
cannot define its own meanings or its own values. A single action cannot
do that.

69

This is why the linguistic turn was so quickly received by

students of ritual (or, put another way, why Americans in particular
turned from British social ethnography to a more hermeneutically

66

‘A Kind of Mirror for Men: Sword Depositions in Late Antique Northern Gaul’, pp. 401–76.

67

Thus Bazelmans (though the article is flawed by the author’s tendency to essentialize Beowulf
even as he asks us not to essentialize social categories like the individual), and Theuws and
Alkemade. See also Theuws’s ‘Introduction’, pp. 1–13.

68

Maurice Godelier, L’ide´el et le mate´riel: pense´e, e´conomies, socie´te´s (Paris, 1984), trans. Martin
Thom, The Mental and the Material: Thought, Economy and Society (London, 1986) (Godelier
reworking the phrase from Marcel Mauss); Louis Dumont, Homo hierarchicus: essai sur le syste`me
des castes (Paris, 1967), better consulted in the revised translation with the later English preface as
Homo Hierarchicus: The Caste System and Its Implications, trans. Mark Sainsbury, Louis Dumont
and Basia Gulati (Chicago, 1980); Pierre Bourdieu, Esquisse d’une the´orie de la pratique (Geneva,
1972), trans. Michael Nice, Outline of a Theory of Practice (Cambridge, 1977); Michel Foucault,
L’arche´ologie du savoir (Paris, 1967), trans. A.M. Sheridan Smith, The Archaeology of Knowledge
(New York, 1972).

69

See Koziol, Begging Pardon, Chapter 9, ‘How Does a Ritual Mean?’.

The dangers of polemic

385

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

oriented cultural anthropology): the axiom of semiotics – the study of how
meanings are constituted as networks of signs – is that no single sign has
any inherent meaning. And of all signs, it is particularly impossible to
study cultural elements constitutive of ‘power’ – including rituals –
without returning those elements to the matrix of ideational relationships
and differentiations that produce and reproduce the categories that make
a world whose sites of power seem natural and essential, simply because
ramified so incessantly and thoroughly throughout experience as to be
inescapable, to make alternatives unthinkable. One can cite Bourdieu,
Foucault, Dumont, or Godelier, it hardly matters. The essential point is
that to understand not what a ritual means but how a ritual can mean
anything at all, to the point that it matters enough to contemporaries that
they want and need to contend over its meanings, we have to understand
any given ritual and all its component elements in conjunction with all
other parts of the culture. Not just the artefacts used, but the logic with
which they are put together. Not just the words used, but the syntactic
structures and semantic patterns they reproduce. Not just the syntax and
artefacts of the individuals actively participating in the ritual, but also
those who participate on the margins, as audience, and those whose
conspicuous abjection marks the boundaries of the thought-world of
participants.

70

These are the reasons that ‘ritual’ is no longer a subject of intense study

among anthropologists as a discrete category of analysis. Past under-
standings of rituals as discrete events consistently led to hermeneutic
impasses by turning rituals into a unique category of action and language
incommensurate with other actions and discourses. Older models –
especially simple functionalist and structuralist ones – were also increas-
ingly seen as politically naive, because unaware of strategic uses of ritual to
gain power, unaware of the imbrication of ritualized expressions of power
in the relations and distinctions of ordinary actions and language.
Realizing this, ethnologists stopped asking what any given ritual means
and began asking instead how the meaning of rituals is constituted, which
is to say, how social values are constructed, selected, framed, mobilized,
and changed.

71

In other words, ethnologists shifted their attention from

understanding discrete rituals to understanding the cultural ‘practice’ of
which rituals were a coherent, continuous part. Hence, again, the growing
emphasis on practice and strategy (following Bourdieu), on the mutually
reinforcing interrelationships between elements of a cultural system
constructive and productive of values (following Dumont), on the

70

An elegant illustration is Brigitte Bedos-Rezak, ‘Medieval Identity: A Sign and a Concept’,
American Historical Review 105 (2000), pp. 1489–533.

71

The importance of such constructive matrices for understanding specific rituals was brought
home to medieval historians by Richard Trexler, Public Life in Renaissance Florence (New York,
1980).

386

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002

imbrication of all given elements of meaningful action and articulable
thought in a total discourse that creates and recreates diffuse sites of power
(following Foucault), on the impossibility of understanding any single
social fact in isolation (following Godelier). These are the models used by
recent historians of medieval ritual. And it is these models – not the naive
social functionalism Buc alleges – that have led precisely the most
theoretically sophisticated authors in Rituals of Power to assume such a
broad definition of ‘rituals’ that rituals themselves have almost entirely
disappeared. Such an expansive use of the term is not a sign of theoretical
naivety or a blind borrowing of ethnological models. The definition of
ritual has broadened for the same reasons that ethnologists no longer really
care about rituals per se, and therefore write rarely on rituals as a discrete
subject. These authors are not really interested in rituals at all. They are
interested in power, in the construction of collectivities, and in the
dynamic between individuals and collectivity. They are interested in
meaning, its production and reproduction. It is the opposite of a lazy
borrowing of ethnology. Theuws and Alkemade, for example, begin by
criticizing those who would reduce weapon depositions in bogs to a ritual
that is explainable simply according to ‘some irrational, primitive habit’.
They assume that rituals are imbricated in power relations. They write
against those who would interpret actions and objects according to ‘some
ready-made ‘‘symbolism’’’ that comes from outside the society. Criticiz-
ing those who see rituals as ‘passive’ and ‘representational’, Theuws and
Alkemade see them as active components in the construction of meaning
and social roles, not because the ritual itself has agency, but because the
individuals who act are agents in the constitution of meaning for both
actions and objects used in their actions.

72

In other words, these authors

are making the same points Buc makes but in a way that is far broader and
subtler, and far more open to more contention than Buc’s narrow,
exegetical, clerical perspective allows.

It is true that the theoretical language of many of the articles in Rituals

of Power is overly arcane and abstruse. One wishes one could cast a spell
and create the perfect historian of ritual: someone with Althoff ’s instinct
for the political, Nelson’s craftsmanship and empirical common sense,
Theuws’s theoretical sophistication, and maybe even Paul Dutton’s or
Stephen Glosecki’s sense of style.

73

But no spell is necessary. Medieval and

Early Modern Ritual is a collection of articles of even wider geographical
and chronological range than Rituals of Power. Again the quality of the
articles is uneven; the finest articles, however, are on early medieval topics:
Marguerite Ragnow’s ‘Ritual before the Altar’ and Martha Rampton’s

72

‘Sword Depositions’, p. 409.

73

Paul Dutton, The Politics of Dreaming in the Carolingian Empire (Lincoln, NE, 1994); Stephen
Glosecki, Shamanism and Old English Poetry (New York, 1989).

The dangers of polemic

387

#

Blackwell Publishing Ltd 2002

Early Medieval Europe 2002 11 (4)

‘Burchard of Worms and Female Magical Ritual’. Both are superb.
Rampton does the impossible. She fully recognizes the composite nature
of Burchard’s Corrector, the classical traditions that underlie its texts, and
the chimaeric projections

74

of its depictions of women’s magic; yet she

does not surrender to historiographical agnosticism. Instead, her aware-
ness of the problems and complexity of the text make her conclusions
about the magical practices of women contemporary to Burchard all the
more convincing and courageous. Her theory (like most in this volume,
largely Susan Bell’s practice-orientation) is also cogently applied and very
supple. This is also true of Ragnow’s article on donations made upon
altars, which is not only the best article I have read on the subject, but one
of the best articles I have ever read on any medieval ritual. Ragnow
manages to be traditionally rigorous yet say things that are new. And
though she does not flaunt theory, she is not only fully aware of the
theoretical implications of her argument, she also manages to do some-
thing theoretically quite important, in emphasizing the dense concatena-
tion of meanings that attaches to altars and thereby essentializes their
meaningfulness to contemporaries. Quite apart from these virtues,
Ragnow’s application of the idea of ‘historical memory’ to the problem
of the continuity of Roman law is one of the best models for under-
standing the problem I’ve read.

So there is hope. However, it will not be met by polemicists. It will be

met by historians like Ragnow and Rampton who understand the
importance of theory but also understand the distinctive, historically
shaped and socially reproduced habitus that defines the community of
scholars who work on the early Middle Ages: awareness of source
criticism; attention to detail; mistrust of reification and a sense for the
integrity of the concrete and local; and not least, understanding of and
respect for the writing of predecessors and peers.

University of California, Berkeley

74

For this use of ‘chimaeric’, see Gavin I. Langmuir, Toward a Definition of Antisemitism (Berkeley,
1990).

388

Geoffrey Koziol

Early Medieval Europe 2002 11 (4)

#

Blackwell Publishing Ltd 2002