Ritual and Archaeology in Early Latium

by Christopher Smith

Introduction
In a recent collection of papers on ritual and archaeology (Garwood et al. 1991), the editors
highlight the absence of a theoretical discussion of ritual in the context of historical
archaeology and of the epistemological and methodological qualities that distinguish prehistory
and history with regard to ritual and religion They go on to say ‘it is the problem of symbolism
and meaning which seems, above all, to distinguish prehistoric and historic approaches to ritual
in archaeology. In the latter it appears to be possible to come to complex understandings of the
particular referents in symbolism and the particular beliefs which ritual expressed, offering a
far richer field for the detailed contextual analysis of relationships between beliefs and human
actions than is imaginable for prehistory’ (ibid: ix)
In this paper I would like to offer some thoughts on this issue, and to raise some questions
about the difficulties, often underplayed, of studying ritual in a historic period. The particular
example which I will be using is that of Rome and its hinterland Latium from around 1000 Be
across the appearance of literacy and relatively reliable written accounts into the Republic (this
period is the subject of my forthcoming book on early Rome and Latium). I would like to begin
by discussing some recent theoretical approaches to the study ofritua!.
Theoretical Approaches
Ritual has been interpreted in various ways over the past decades. In previous times, a
functionalist approach was favoured, implying that the ritual was intended to promote the
solidarity of the group (described in Turner & Maryanski: 1979, as an ‘illegitimate teleology’).
It is difficult to see how this causation worked in practice – solidarity may be an effect but it is
not therefore a cause – and such accounts tend to be reductionist. Similar objections are raised
to interpretations that identify the symbolic contents of a ritual and link them in some kind of
quasi-theological discourse. The ambiguity and complexity of symbols are denied by such
intellectualising approaches, and even subtle accounts such as those of Turner (1967; 1969) and
Geertz (1973: chapter 3) tend to imply that symbols and rituals have some kind of existence of
their own (See for instance Turner 1967: 20: ‘the ritual symbol becomes a factor in social
action, a positive force in an activity field. The symbol becomes associated with human
interests, purposes, ends, and means, whether these are explicitly formulated or have to be
inferred from the observed behavior. The structure and properties of a symbol become those of a
dynamic entity, at least within its appropriate context of action ‘)
This approach was criticised in an important article by Asad, which examined Geertz’s
attempt to find a universal definition of religion (Asad 1983). Asad contrasted the view of
symbols as ‘meaning-carrying objects external to social conditions and mental states’, with his
own conception of them as focused on ‘the relations between socially signifying and
psychologically organising principles’. 58 Christopher Smith
Asad could not find in Geertz’s account anything that ensured that the participants in a
ritual take the symbolic forms in the way that leads to faith (Asad is writing in the context of
Medieval Christianity, so for our pagan purposes, we might replace ‘faith’ with ‘belief). ‘The
connection between religious theory and practice,’ writes Asad, ‘is fundamentally a matter of
power – of disciplines creating religion, interpreting true meanings, forbidding certain
utterances and practices and authorising others’ (Asad 1983: 246).
Ritual symbols and ritual experiences are not separate from social realities, and ritual does
not of itself create beliet rather the disposition must be present in the participants. Such
dispositions are created by knowledge and instruction; authoritative religious practices and
utterances are ‘products of historically distinctive disciplines and forces’ (Asad 1983: 251).
Bloch has gone some way towards identifying the historical conditions behind certain
religious practices in his account of the circumcision ritual of the Merina of Madagascar, by
taking over two centuries of development of the ritual, through a number of changing political
situations (Bloch 1986). The practice of the ritual through these two hundred years remains
remarkably static, but the use of the ritual changes dramatically; it began as a ritual performed
by the elders of localised descent youps, was taken over by the king as a royal ritual when a
powerful state arose, and the descent group was forbidden, and then returned to the descent
groups after the French conquest as a symbol of resistance to colonialism.
In an extremely suggestive article, Harrison has used this case, together, with among
others, the evocatio ritual in Rome as an indication of the way in which rituals can be seen as a
form of intellectual property (Harrison 1992). They are, he suggests, rather like luxuries in that
their uses are rhetorical and social; they can be owned, and the ownership of them is a part of
the social inequality of a society. Rituals are forms of knowledge that have no existence outside
their practice, but this practice can be controlled and manipulated through social and
psychological relations between the actors.
Asad and Bloch present ritual in a way that is significantly different from the presentation
given in the functionalist writers, and the symbolist/intellectualist accounts we have briefly
touched on. Their focus is on historical conditions, and on authorising discourses and the way
in which power creates religion.
This leads me directly to Barrett’s accounts of ritual, which have similar concerns (Barrett
1988; 199 L 1994). Barrett wishes to see material remains as the ‘fields of discourse’ in which
ritual, among many other things, takes place. Together with social relations they form the
conditions within which certain forms of knowledge become possible.
The approach is consonant with Giddens’ focus on the way in which the structured and
knowledgeable actions of human agents can be observed reproducing the institutionalised form
of the social system. The social world is constructed through action, but action is constrai ned by
this created world. The process is thus essentially recursive (Giddens 1984)
If the subject for study is taken then to be structure, what is this structure, and how is it
manifested? Rather like ritual, it does not have an existence independent of action; it is in fact
contained within memory and certain repeated actions, and also, as Bourdieu stressed, in
material locales (Bourdieu 1977: 89). Material space is the place where social practices are
performed by agents who refer constantly to their locale; this sphere is acculturated; it is drawn
upon for meaning and invested with meaning by human action.
So what the archaeologist finds, be it standing stones, sanctuaries or votive deposits must
be interpreted as the locale for knowledgeable action. Moreover, agents move from one area of Ritual and Archaeology in Early Latium 59
time and space into another, and take with them their preconceptions and understandings,
which are modified and changed by the new locale, as well as contributing towards the
knowledge of that locale and the action which takes place there. Although this might seem
evident, it is in fact quite the opposite from an approach where ritual is seen as something of a
quite different self-standing identity to the rest of the world and life. Ritual and religious
knowledges are built out of the same conditions. material and social, as the rest of life, even if
they appear to refer to a different kind of reality.
Ritual is often described as a form of communication (see Lewis 1980: 6-38, for a careful
analysis of this analogy). Communication rests on a communality of expectations, which
emerge from shared cultural conditions. Speaker and listener share certain experiences, and
bring others to the moment of communication in order to be able to understand and be
understood. Part of this shared experience is described by Bourdieu as ‘habitus’. Habitus is
defined as a set of it is learnt as a child, neither unconsciously nor by rote, but by
observing a systematic application of principles which are coherent in practice, for instance the
gift exchange. Once attained these habitus give a repertoire for dealing with the world, which
reproduces the world in which the child has been brought up (Lewis 1980: 72). Asad gives a
similar kind of account when he uses Vygotsky’s work. which indicated that the development
of children’s intellect is dependent on the internalisation of social speech, within which the
symbols that are in recursive use in society are central (Asad 1983: 240).
Ritual must be seen as part of this world of knowledgeable action and culturally derived
expectations. One consequence of this is that it forces us to question the idea that rituals are
used to present a falsehood about the world masked as a truth. which sustains the dominant
ideology; that people are duped. Rather we need to identify the knowledges which reproduce
social asymmetries like inequality. These knowledges are precisely part of the authorising
discourse which Asad sought to identify, and they are reflected in and partly constituted by the
material world within which they are acted out.
There is something special about ritual however, and Bloch’s account of the way a certain
set of actions can remain largely (though not entirely) static whilst the context changes
radically is an important indication of this. Although ritual participates in the discourse of the
real world, in that those who are involved bring their knowledge of that world to the ritual, and
the ritual draws on the material world around, it has a sense of timelessness.
This is really what lies behind the truism that Roman religion is conservative. The ritual
may be conservative, but the context changes over time, and thus the meanings which the ritual
has for the participants their perception of what is happening in the ritual alters. In
fact, it is possible that the static nature of Roman ritual may itself have been overplayed.
One of Barrett’s ideas is to draw a distinction between ‘talk’, which involves immediate
contact, and ‘text’, which introduces the idea of distanciation, that is the existence of a higher
or at least an absent authority which has produced the object to be read (Barrett 1991: 1993:
78). Ritual refers to a different reality, and individuals measure their experience of the world
against this other reality. They have to submit to the discipline of a traditional discourse which
involves everyday values but presents them as emanating from another source.
One problem here is that both the idea of timeless rituals and of ‘text’ tend to underplay
the degree to which ritual is open to processes of change and transformation in and of
themselves. This is an unfortunate though I think unintended consequence of Barrett’s use of
text as a metaphor for understanding rituaL one would have to say that the ‘text’ is recreated 60 Christopher Smith
and transformed by each performance; that although it involves the concept of distanciation,
the text is not necessarily unchanging (lowe this point to comments made at the conference).
Ritual or religious symbols cannot do all the work, as Geertz and Turner seem to imply;
they must work in the context of agents who have disciplined themselves to submit to the
experience and who recreate the experience by their participation.
That the disposition to obedience, wherever its distant origins may lie, comes to be a part
of the observable behaviour of people in all periods, prehistoric and historic, attests the
centrality of habitus, and it is the presence of authorizing discourses as the mode of
perpetuating religion that gives ritual its strength. As Harrison and Bloch show, ritual, or the
specific knowledge that is ritual, can be passed from person to person; it is, as Bloch said,
essentially recoverable for all dominant leaders.
Let us move into historical period and consider the spread of literacy. A recent exhaustive
account (Harris 1989: see also Cornell 1991) suggests that literacy beyond the ability to write
one’s own name may not have extended beyond 15% of the population at any time during
antiquity, and was almost certainly even less common in the early Republic. The major
implication of literacy in this period must have been the widening of social distance: it is
another acquisition, like a lUXUry good, whose uses are rhetorical, social and divisive.
At some stage, some prayers may have been written down: we know this for instance from
the archaic prayer of the Arval Brethren, which survives in inscriptions several centuries later
than its first use (see Sheid 1991: Beard 1985). It was written down when it was practically
incomprehensible, but its continued use is part of the discourse, particularly prevalent in the
early Empire, by which the Romans rooted their social practices in the authoritative past. This
also made it the subject of antiquarian investigation, and this has broader implications.
Paganism did not fade away to leave a gap for mystery religions and Christianity to filL it
remall1ed a coherent system of belief long after Constantine’s conversion (for the continuation
of paganism see, for instance, Cameron 1993: 77ff). It did become the object of an
intellectualist discourse, though, and this changed it. When one reads Ovid’s Fasti, we see the
kinds of .interpretation which were being applied; aetiological stories, false etymologies, and
syncretism (see Porte 1985).
Take for instance the Robigalia festival. which was intended to ward off mildew from the
crops (on the RobigaJia, see Ovid Fasti 4. 905ff: Scullard 1981: 108ff. I hope to expand on this
subject elsewhere). In his account of this, Ovid is remarkably inaccurate. He says that Sirius
was rising on 25th April, when in fact it was setting He is confusing about where the festival
takes place. He explains the sacrifice of a dog, an unusual sacrifice in the Roman world, as
simply connected with the description of Sirius as the Dog Star. The origins of this description
may be Greek rather than Roman, and it is interesting that Ovid fails to point out that dogs
have a number of connotations with protection and with the Underworld. Ovid reports a prayer
made by the priest at the festival, and in it equates the rust on the crops with the rust on
weapons: let weapons rust rather than the crops, for this is an age of peace, is the gist. This is
clearly connected with Augustan imagery of the cessation of war. Instances like this could be
\11utiplied from Ovid and others (Res Gestae Divi A uKUSli I J; Virgil (ieorgic.\’; Fraenkel 1957
239[: Chisholm & Freguson 1981: 50-89: Ovid Fasli 1. 67-68), I do not wish to suggest that
all of these interpretations are untrue, but rather that they betray kinds of thinking about
religion which are worryingly reductionist.
One could also cite Cicero for similarly ‘modern’ comments 011 Roman religion; for
instance his functionalist statement that ‘gods are needed for the maintenance of the social Ritual and /-Jrchaeology in Early Latium 61
without them, society would be a faith, justice and community would all go to
or, ‘private worship may not be satisfactorily performed without the assistance of those
in charge of public for the people’s constant need for the advice and authority of the elite
helps to hold the State together’ (Cicero, Nat. Dear. 1.3,230).
Unlike Christianity, there was no correct doctrinal paganism, but there was an attempt to
explain the rituals. It is remarkable that functionalism and intellectualism are in essence the
modes of interpretation which ancient writers applied to their own religion, but the experience
of Roman religion remained manifold, and the concept of heresy inappropriate. So the study of
ritual in the historical period is in some sense made more difficult by the way in which ancient
writers sought univocal explanations for complex rituals (The unchanging nature of Roman
ritual may have been overstated by Roman writers, which adds to the problem).
Fields of Discourse in Latin Ritual
1 wish to turn now to some specific characteristics of Roman religion, and to analyse them
across the boundary of prehistory and history using as a heuristic device the four analytical
components identified by Barrett as parts of the ‘field of discourse’. Barrett has subsequently
made his analysis more subtle, in particular stressing individual experience, for instance the
memories which each person will bring to a particular ritual event. I accept the relevance of
this change, but the earlier formulation suits my purpose better here. (Barrett 1988: 9-12; on
Roman religion see Dumezil North Beard 1990) The four components are the
temporal frequency with which the field of discourse was routinely the spatial extent
or the geographical space occupied by the the cultural resources which are drawn upon
to define and instigate authoritative discourse, and those used to recognise it; and the
transformations in cultural resources over time.
Temporal frequency
The rhythm of ritual in early Latium seems to have been rooted in the agricultural year; the
festivals which we believe to have been early have connotations of protecting the harvest and
purifying the fields (for agriculture in Roman religion see North 1989 and the works of
Dumezil cited there) With the increase in votive deposition, this may have been complemented
by the rhythms of an individual’s life. and moments of stress like birth or illness, and by the
rhythms of political conflict, since more elaborate votives and the building of monumental
temples are part of the political ‘reallocation of attention’ (see Colonna 1984 on temples and
Bauman 1992. 200, for the phrase). This is not to say that the agricultural rhythm became
irrelevant, but that it was overlaid with other interests.
Spatial extent
Tne consequence of Rome’ s expansion in the late regal and early Republican periods was the
growth of Roman territory. It is interesting to see the way in which ceremonies like the
Ambarvalia follow this growth. This festival was a kind of beating of the bounds, and took as a
canonical boundary the circuit five miles outside the city, which was won, it seems, in the 6th
century Be Thus the ritual comes to be as much a part of the discourse about Rome’s military
power as it is an agricultural festival. and since military expeditions in early Rome were led by
the elite, it is part of the acknowledgement of their power (Cornell 1989: 243ft} 62 Christopher .)’mith
Cultural resources
There are some important continuities in the cultural resources which are drawn upon to define
authoritative demands of discourse. Most Roman ritual is always led by a man. The control of
religion is an aspect of patriarchy, but it is also the preserve of an elite group. Patrician control
of religion lasts significantly longer than patrician control of politics; it may be that this control
is actually a defining characteristic of what it is to be a patrician (Mitchell 1990 attempts this
identification; see also Palmer 1970).
In the 6th century, votive deposits are transformed into stone temples across Latium. A
votive deposit, like a burial, previously took place in the open; it could be seen from all sides.
Temples, however, are axial; they direct vision, and have a clearly defined inside and outside.
Their decoration increasingly becomes a reflection of hierarchies within the cosmological
scheme.
What characterised the attitude of the participants at a sacrifice was silence, to avoid
anyone saying the wrong thing. In other words, the people are silent whilst the experts act and
speak. Silence is a potent acknowldegement of the knowledge of others. In all these ways the
distinction between those with knowledge and those without is enacted and reaffirmed.
Transformations
I wish to pick out only two major transformations. The first is material. Votives became
wealthier over time, from clay vessels to silver and gold, from ‘silent’ objects to ‘speaking’
objects which bear an inscription (on votive deposits see Bartoloni 1987 and 1989-90. On
inscribed objects see Cristofani 1990: 16- 25; Cornell 1991) The locales in which ritual
experience takes place also change; the new temples were built with the resources of the elite,
and were thus a visual reminder of the elite power.
The second is an issue of knowledge. Some rituals never left the control of the elite; if
anything they became more entangled in practical politics. The Ambarvalia is one instance; the
evocatio of a deity from a conquered town or country is another – it is as Harrison pointed out
a transfer of ritual knowledge (Harrison. 1992: 229f) As settlements grow in size, the number
of priestly experts declines as a proportion of the population, and the control of knowledge
becomes centred in a smaller and more powerful group.
When baldly stated, these comments may appear to tend themselves towards a functionalist
reading of the rituals, in that the rituals express the dominance of the elite, and are intended so
to do. To avoid this. we need to think about the authorising discourses and knowledges which
reproduce such social asymmetries, and these may perhaps be identified in the concept of
mediation which Beard has put at the heart of the priestly activity (Beard 1990: 29-34) and
which itself reproduces structures such as patriarchy and authority. Moreover, we must not
discount important aspects of religious life which may exist outside of this control (though not
entirley untouched by it), for instance the private rituals to which Cicero referred in the passage
quoted above which itself attests the insecurity of the elite in the face of personal expressions of
religious knowledge. The votive deposits may be central here. in that it is not clear that the
monumentaJisation of them removed the possibility for private acts outside priestly authority.
The supposedly plebeian temple of Ceres Liber Liberaque as a symbol of political opposition
may also be cited here (North 1989: 620), and in the third century, as patrician control
diminishes, there are a number of Greek and eastern cults which could be regarded as having a
subversive element (North 19&9: 616) Ritual and Archaeology in Early Latium 63
Conclusions
In conclusion, I should like to return to the problems identified by Garwood and his colleagues
at the beginning of this paper. There are differences between the study of ritual in prehistoric
and historical the significance of literacy is the most obvious. However it is not
necessarily the case that historical studies have the edge. To assume this would be to argue for
the validity of ancient written sources as interpretations, whereas they are in fact part of the
historical conditions, part of the power, which creates religion.
My second point is that if it is changes in social conditions which transform religion and
ritual, then perhaps the issue which is most relevant is urbanislll. It is the appearance or the
evolution of complex social structures which we awkwardly define as urban that is centraJ to
the development of Roman religion in the late regal and early Republican periods as something
quite different for those participating in it than the religion of the early Iron Age.
Chronology
Latial Period Date (B.C.) Events
I 1000-900 Beginning of Latial civilisation
IIA 900- 830 Major necropoleis begin
lIB 830-770
III 770-730 Orientalising period; arrival of Greeks in Central ItaJy
IVA 730-630 Wealthiest burials; beginning of votive deposits
IVB 630- 580 Cessation of burials
580-500 Beginning of monumental architecture including temples:
period of Etruscan kings and formation of patriciate; expansion
of Roman foundation of Roman (509 BC)
References
Asad, T 1983. Anthropological conceptions of religion Reflections on Geeltz, Man 18: 237-59.
Barrett, lC. 1988. Fields of discourse: Reconstituting a social archaeology, Critique of Anthropology 7:
5-16
Barrett, lC. 1991. Towards an archaeology of ritual. In Garwood, P. et aI., Sacred and Profane;
Proceedings of a COI?{erence on Archaeology, Ritual and ReligiOn. Oxford.
Barrett, le. 1994. Fragmenls /rom Anliquity: An Archaeology o{Social Life in Brilain, 2900- 1200 BC
Oxford.
Bartoloni, G. 1987. Esibizione di richezza aRoma nel VI e V secolo: Doni votivi e corrredi fun era ri ,
Scienze dell ‘Anlichi/ci I: 143- 59
Bartoloni, G. 1989-90. I depositi votivi di Roma arcaica: Alcune considerazioni, Scienze dell ‘Anl/chila
3-4; 747-59.
Bawnan, Z. 1992. Intimations of Post modernity. Routledge, London.
Beard, M. 1985. Writing and Ritual: A study of diversity and expansion in the Arval Acta, Papers of the
British School at Rome 53: 114-62. 64 Christopher Smith
Beard, M . 1990. Priesthood in the Roman Republic. In Beard, M. & NOlth, 1.A. Pagan Priests: Religion
and Power in the Ancient World, 17-48. London.
Bloch, M. 1986. From Blessing to Violence: History and Ideology in the Circumcision Ritual of the
Merina oI Madagascar Cambridge.
Bourdieu, P. 1977. Outline of a TheolY of Practice. Cambridge.
Cameron, A. 1993. The Later Roman Empire. London.
Chisholm, K. & Ferguson, 1. 1981. Rome: The Augustan Age. Oxford.
ColOlma, G. 1984. I templi del Lazio fino al V secolo compreso, Archeologia Laziale 6: 396-411.
Cornell, T1. 1989. Rome and Latium to 390 B.C. Cambridge Ancient History VIl.22: 243-308.
ComelL TJ. 1991 The tyranny of the evidence: A discllssion of the possible uses of literacy in Etruria
and Latium in the archaic age. In Humphrey, 1. H. (ed) Literacy in the Roman World, 7-33.
loumal of Roman Archaeology SupplimentalY Series 3. Ann Arbor.
Cristofani, M. (ed) 1990. La Grande Roma dei Tarquini. Rome.
Dumezil, G. 1970. Archaic Roman Religion. London.
Fraenkel, E. 1957 Horace Oxford.
GaJWood, P. et ai, 199 J. Sacred and Profane: Proceedinf!.s of a (,onference on Archaeology, Ritual and
Re17gion. Oxford.
Geel1z C 1973. The Intelpretation of Cultures. New York.
Giddens, A. 1984. The Conslitution of Society: Oulline oflhe Theory o/Strucluralion. Cambridge.
Harris, W. V. 1989. Ancient Literacy. Cambridge, Mass.
Harrison, S. 1992. Ritual as intellectual property, Alan 27: 225-44.
Lewis, G. 1980. Day of Shininf!. Red: An Essay on Understanding Rilual. Cambridge.
Mitchell, R.E 1990. Patricians and Plebeians. The Ol’lgins of the Roman Stale. Comel!.
N0l1h, J.A., 1989 Religion in Republican Rome. Cambridge Ancient HistOlY VIl.22; 573- 624
Palmer. R.E.A. 1970. The Archaic (‘ommunity ofthe Romans. Cambridge.
POlte, D. 1985. L ‘Etiologie religieuse dans les Pasles d ‘Ovide. Paris.
Scheid, 1. 1990. Romulus et ses (I’eres: Le college des Ft’eres A,vales, modele du eulle public dans la
Rome des empereut:5. Rome.
Scullard, H. H. 1981. Feslivals and ( ‘eremonies oflhe Roman Republic. London.
Turner, V. 1967. The Foresl Aspecls of Ndembu Ritual. Cornell.
TUl11er, V. 1969. The Ritl/al Process: Slmclure and Anti-Slmclure. Cornell.
TUllier, 1.H. & Maryanski, A. 1979. Funclionalism. Califol11ia.