Printer Friendly

Vera Collum and the excavation of a `Roman' megalithic tomb.

In 1931 Vera Collum excavated the megalithic tomb at Tresse in Brittany, claiming that it had been built during the Roman period and was associated with the cult of the mother goddess. This article traces the course of her excavation and suggests a context for the reuse of Neolithic monuments in that area.

Key-words: megaliths, Brittany, mother goddess, monument re-use

Devotees of the great goddess have yet to discover Vera Collum and her excavations in megalithic tombs. She is not a well-known figure today. Vera Christina Chute Collum is credited with 11 books, all of them published between 1924 and 1940. They include such titles as The Dance of Civa: life's unity and rhythm (Collum 1927), The music of growth (Collum 1933a) and Manifold unity: the ancient world's conception of the divine pattern of harmony and compassion (Collum 1940). Her most ambitious work, published in German, considers The creative mother: goddess of the Celtic-speaking people, their material culture, the mystical word and their beliefs and symbols (Collum 1939). Amongst her publications there are also two excavation reports (Collum 1933b and 1935). The Tresse Iron Age megalithic monument (Collum 1935) can still be found in antiquarian bookshops, but it is rarely paid much attention. That is surprising, as its full title is extraordinary: The Tresse Iron Age megalithic monument (Sir Robert Mond's excavation), its quadruple sculptured breasts and their relation to the mother-goddess cosmic cult.

The second page provides an even longer sub-title (or perhaps it is an abstract) printed in block capitals. In Collum's words this is the report on `a hitherto unexcavated allee couverte of the Gallo-Roman period with realistic sculptures of two double pairs of human breasts and containing a crouched burial in situ, fragments of a one-edged iron short sword, steatite beads and more than 60 pots of Armorican-"dolmenic" and Armorican-Gaulish styles and ceramic techniques' (1935: v). The monument was near St Malo and the report comes `together with an exhaustive study of the cult responsible for the symbolism of the twin pairs of human breasts' (1935: v). The book is dated 1935 and was published by Oxford University Press. Three years later a shortened version appeared in French (Collum 1938).

We can reconstruct some of the background to this project. Sir Robert Mond was a wealthy chemist and industrialist with antiquarian leanings. He had excavated at Thebes and helped to fund Dorothy Garrod's excavation on Mount Carmel. No doubt he paid for the publication of this report. He was also a Francophile with a special interest in megaliths; his work in Brittany is summarised by Giot (1993: 29-31). Vera Collum, whose career is largely undocumented, lived in Wiltshire but she spent part of her time excavating at Camac with Zacharie Le Rouzic. She conducted three excavations which were financed by Robert Mond. Two of these were at chambered tombs: Le Dehus on Guernsey (Collum 1933b) and Tress on the property of his friend, Baron Robert Surcouf. The excavation of Tresse took place in 1931 but it was only part of a larger undertaking, for at the end of her monograph Collum writes (1935: 115):

I desire to place on record my great indebtedness to Sir Robert Mond for having made it possible for me to devote the last five years to research, both in the archaeological field and in that of comparative religion, without the anxiety of earning a living.

What were the findings of that lavishly endowed project? Her text is often obscure, but at one point Collum summarizes her main conclusions. This is a small sample (Collum 1935: 113-14):

The allee couverte at Tresse had not been previously excavated.... It was erected in the first century AD, probably in the reign of Domitian.... It was a native Gaulish tomb that was at the same time a funeral grotto representing an Entrance to the Underworld which symbolised Return to the Womb of the Creatrix.... That goddess was the Great Female Principle (comprehending a Male Principle) both in its unmanifested aspect of Potential Creatrix, and its manifesting aspect as Woman the Lover-Bride and as the all-nourishing Mother, whose cult was widespread in Asia Minor, Syria, Central Asia and NW India, Mesopotamia, Babylonia and Egypt ... Such esoteric teaching was probably first introduced by the poet-seers or `druids' and imparted only to the educated classes.... This Cosmic cult can be demonstrated ... in the archaeological remains of Gaul and Great Britain, ... in the occult poetry and religious epics of Ireland and Wales and in Gaelic hymns surviving orally in the Western Isles of Scotland.

That extract is typical of this passage as a whole. It reads strangely like the contents of similar works today, but in one respect it is quite different, for it comes from an excavation report. How did Collum balance the discipline of field-work with these extravagant claims? Did she find the first Roman megalith in France? And how would we interpret that monument today?

The site itself was certainly an allee couverte or gallery grave. It was set within an oval mound and had probably been closed by a deposit of rubble. The tomb was originally defined by 10 pairs of orthostats, spanned by horizontal capstones. A large slab in the terminal cell was carved in relief and depicts two pairs of breasts; two more are on an orthostat nearby. The breasts are apparently linked by a necklace, a feature that occurs on several tombs of this type in Brittany (L'Helgouac'h et al. 1970). The monument itself is a relatively large one. The covering mound was originally 20 m long and the chambers extended for a length of at least 13 m. Towards the rear one group of orthostats had collapsed and here the roof was missing (FIGURE 1).


There is a complete mismatch between the clarity of the site plans by one of Collum's team and the poor quality of the site photographs, which she took herself. The excavation was carried out by workmen from the Baron's estate, supplemented by professional quarrymen. They were also responsible for rebuilding the monument. Collum's report has a wonderful period flavour and there was an obvious social barrier between the director of the excavation and her work-force, who were not quite as pliable as she believed. This is clear when she describes how the site was reconstructed. The missing orthostats were to be replaced by large concrete blocks, carefully selected for the purpose. A year later the monument had collapsed and Collum discovered that `the "concrete slabs" locally ordered turned out to be boards thinly coated with cement' (1935: 30).

The filling of the monument posed some practical problems. The most recent occupant of the tomb seems to have been a wild animal, and the latest context was defined mainly by its smell: `When we had got down to the yellow tumulus debris beneath Capstone VIII, we were almost overpowered by the stench of wolf' (Collum 1935: 18). Then the rain set in so that work inside the chambers took place beneath a large tarpaulin. It was difficult to see very much under these conditions and the excavated soil was passed to the site foreman who searched it for artefacts. Parts of the original structure were removed by the workmen whilst Collum's attention was distracted. Several stones of the megalithic chamber were leaning and a few had fallen over so that much of the time was spent in levering them back into position. As a result, the floor was very badly disturbed.

Although Collum goes into great detail on the contexts of the individual finds, it is quite clear that, wolf or no wolf, the different levels inside the chamber were hard to distinguish from one another and that all of them had been badly disturbed. Parts of the tomb had apparently been floored by what Collum, with no hint of humour, refers to as `crazy paving'. As she tells us that `tree roots ... have worked their way between the interstices of the pavement, ... heaving them up in places' (1935: 16), there is little reason to suggest that this level remained intact. The entire monument had also been disturbed by burrowing animals and, to judge from the published photographs, Collum's excavation was exceptionally untidy. It is hard to believe that the artefacts that she found at Tresse were properly stratified at all, and yet these provide the entire basis for her interpretation.

The material that she found was in two styles, the Neolithic and the Roman. The Neolithic finds consisted of blades of Grand Pressigny flint, and a series of other artefacts including flint transverse arrowheads and worked chert. Some of the excavated pottery should date from the same period. This material was widely distributed about the site. The sherds were mainly towards the back of the structure but the finds of flint and chert came from every part of the monument.

The Roman artefacts had a more restricted distribution and the beads were about 2 m from the remains of a crouched burial located on the floor towards the back of the chamber. Apart from a quantity of pottery, this material included a coin of Domitian found beneath the eastern part of the cairn. Pieces of iron were recognized in every section of the monument and included one fragment which Collum identified, very optimistically, as a sword and others that she claimed were the tip of a knife and part of a fibula. The Roman pottery came mainly from the northern and middle parts of the tomb and its external cairn. It seems that some of it was in the vicinity of the burial. It was in the same area that the structure seems to have collapsed. This may have happened when the body was introduced.

Collum went into the project well aware that Roman pottery had been found in a number of Breton megaliths, apparently associated with Neolithic artefacts, but she was also influenced by her knowledge of monuments in other parts of the world. As she says (Collum 1935: 10),

Comparative research in the excavation literature of megalithic monuments, and in the finds from them preserved in museums, indicates that the same mixture and the same uncertainty is general, except in North Africa, India, the Caucasus, Korea, Manchuria and Japan, where they have been proven to be of the Iron Age.

This influenced her interpretation of Tresse. She argued that the Roman artefacts, which she associated with the native population of Gaul, were consistently found in the same contexts as those conventionally dated to the Neolithic period. Despite the amount of disturbance experienced by the monument at Tresse, she concluded that both groups of artefacts occurred in primary contexts. Rather than supposing that the apparently younger material was intrusive, she embarked on the bold interpretation that the two collections of artefacts were deposited there at the same time. The allee couverte could only be a Gallo-Roman monument. The crouched burial was a primary deposit together with the metalwork and, in the absence of compelling evidence to the contrary, the `Neolithic' artefacts of Brittany would have to be redated. The carvings of breasts on the tomb at Tresse were central to her argument because other Breton tombs contained pipe-clay figurines of Venus or the Matres. These were found near the entrance together with Roman pottery, whilst coins were incorporated into the covering mounds. Nothing stood in Vera Collum's way as she tracked down the symbols of a cosmic cult of the mother goddess.

This far-reaching interpretation was based on a very literal reading of the stratification at a badly damaged site. We might be able to say that the monument had been disturbed during the Roman period, but had it really been re-used?

If we were limited to the evidence from Tresse, it would be difficult to go any further, but in fact the megalithic tombs found in other parts of Brittany show a rather similar pattern. That not only includes the sites dug by Collum's mentor, Le Rouzic, but also those excavated to a much higher standard during recent years (Andre 1961; Galliou 1989: 31, 151-3; Dark 1993; LeCornec 1994). It seems as if this evidence is of two kinds. First, there are other places where megalithic tombs were chosen as the sites of Roman burials. Tresse is on the edge of their distribution, for most of the examples are found in the Morbihan (Galliou 1989: figure 14), although it is worth remembering that more tombs have been investigated in that area. At the same time, Collum was right in reminding her readers that there were quite a number of megaliths which contained figurines of Venus and the Matres, as well as the customary collections of coins. These artefacts are sometimes found with burials in France but more often they are associated with a shrine (Bemont et al. 1993). The clearest example of this pattern is Le Petit-Mont, Arzon, a site which also included a Roman altar (LeCornec 1994: 71-8).

Why might this have been the case? One possibility is to invert Collum's original argument. Perhaps it was the fact that the tomb at Tresse had been decorated with breasts that encouraged an identification with the cult of the mother goddess, but this did not happen when the tomb was built but during the Roman period when the site was re-used. Could this explain the re-use of other megalithic monuments in this area? The difficulty is that there are not very many tombs with similar carvings and they are not a major feature of the area in which the Roman burials are found (Shee Twohig 1981: 72-3; Galliou 1989: 31). It may be only part of the explanation.

Another possibility is that for the Romans -- and that probably means for the native population of Gaul -- the remains of megalithic tombs were thought of in the same terms as caves and unusual rock formations and were imbued with special powers (Galliou 1984: chapter 11). This is why the figurines occur in rock-shelters as well as megaliths (Bemont et al. 1993: 155-9; Ars 1997). These were numinous places and places of great antiquity, and people were not necessarily able to distinguish between geological features and ruined buildings.

It also seems likely that later deposits in megalithic tombs were associated with fertility as well as death. How might that equation have come about? Perhaps it arose through their striking resemblance to the underground storage structures known as souterrains, which had been a major feature of Iron Age settlement sites until about 100 BC (Giot 1990). Like the storage pits of southern England, these have usually been considered in terms of food production, but this comparison might suggest a more sophisticated interpretation. This is not to suggest that such structures were shrines, but their use as underground food stores recalls the association between death and regeneration which is such a feature of other deposits in late prehistoric Europe (Barrett 1989). Perhaps souterrains, like storage pits, became associated with the fertility of the human population.

Most of the Breton souterrains are rock-cut passages or chambers and do not resemble megalithic tombs, but this is not always the case, for occasionally these features were built in stone (Giot 1990). For later generations they would have been difficult to tell apart from megalithic tombs. In fact, one of the largest deposits of pipe-clay figures was actually found in the secondary levels of one such souterrain near Quimper (Le Men 1868). Here we may have yet another explanation for the link between earlier prehistoric monuments to the dead and Roman shrines dedicated to Venus and the mother goddess. In practice these interpretations can often be combined. Taken together, they suggest why the remains of Neolithic tombs were reinterpreted during the Roman period and how they came to be associated with fertility.

It seems as if Vera Collum was wrong in her main conclusion, but not absolutely wrong. There was no cosmic cult of the mother goddess to be traced from India to France, nor were Breton megaliths built during the Roman period, but after a lengthy interval certain sites were used in the ways that she envisaged. This article has suggested some reasons why they might have been selected, and there are probably others. By misunderstanding the stratigraphy at Tresse she turned French archaeology upside down and her excavation has been ignored as a result. After 60 years it is time to put the pieces back together.

Acknowledgements. This paper was first presented at the 1998 meeting of the Theoretical Archaeology Group, in a session on the re-use of ancient monuments organised by Richard Hingley and Howard Williams. I must thank Mel Costello for the figure drawing and an anonymous referee for some particularly helpful observations.


ANDRE, J. 1961. Les dolmens Morbihanais remployes a l'epoque romaine. Ogam 74-5: 248-54.

ARS, E. 1997. Les figurines gallo-romains en terre cuite du Morbihan. Bulletin et Memoires de la Societe Polymathique du Morbihan 123: 41-54.

BEMONT, C., M. JEANLIN & C. LAHANIER. 1993. Les figurines en terre cuite gallo-romaines. Paris: Editions de la Maison des Sciences de l'Homme.

BARRETT, J. 1989. Food, gender and metal: aspects of social reproduction, in M.L.S. Sorensen & R. Thomas (ed.), The Bronze Age-Iron Age transition in Europe: Aspects of continuity and change in European societies c. 1200 to 1500 B.C.: 127-40. Oxford: British Archaeological Reports. BAR International series S483.

COLLUM, V.C.C. 1927. The Dance of Civa: life's unity and rhythm. London: Kegan Paul & Trench. 1933a. The music of growth. London: Scholar's Press. 1933b. The re-excavation of the Dehus chambered mound at Paradis, Vale, Guernsey, St Peter Port: Societe Guernesiaise. 1935. The Tresse Iron-Age megalithic monument. Oxford: Oxford University Press. 1938. L'allee couverte de Tresse (classe Monument Historique), Ille-et-Vilaine. Paris: Leroux. 1939. Die schopferische Mutter - Gotten der Volker keltischer Sprache, ihr Weltzeig, das mystische `Wort', ihr Kult und ihre Kult-Symbole. Zurich: Rhein Verlag. 1940. Manifold unity: man's perception of the divine pattern of harmony and compassion. London: John Murray.

DARK, K. 1993. Roman period activity at prehistoric ritual monuments in Britain and in the Armorican peninsula. in E. Scott (ed.), Theoretical Roman archaeology: First conference proceedings: 133-46. Aldershot: Avebury.

GALLIOU, P. 1984. L'Armorique romaine. Braspars: Les bibliophiles de Bretagne. 1987. Les tombes romaines d'Armorique. Paris: Editions de la Maison des Sciences de l'Homme.

GIOT, P.-R. 1990. Sonterrains et habitats a l'Age du Per en Armorique, in Les Gaulois d'Armorique: 53-61. Rennes: Archeologie de l'Ouest. Supplement 3. 1993. Chronique de prehistoire et de protohistoire finisteriennes et des archeosciences pour 1992, Bulletin de la societe archeologique du Finistere 122: 11-37.

LECORNEC, J. 1994. Le Petit-Mont, Arzon (Morbihan). Rennes: Documents Archeologiques de l'Ouest,

L'HELGOUAC'H, J., G. BALLANCOURT, C. GALLAIS & J. LECORNEC. 1970, Sepultures et gravures nouvellement decouvertes sur des megalithes de l'Armorique, Bulletin de la Societe Prehistorique Francaise 67: 513-21.

LE MEN, R.F. 1868. Subterranean chambers at La Tourelle, near Quimper, Brittany, Archaeologia Cambrensis (3rd series) 14: 293-311.

SHEE TWOHIG, E. 1981. The megalithic art of western Europe. Oxford: Clarendon Press.

RICHARD BRADLEY, Department of Archaeology, University of Reading, Reading RG5 2AH, England.

Received 4 January 1999, accepted 19 June 1999, revised 18 November 1999.
COPYRIGHT 2000 Antiquity Publications, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Geographic Code:4EUFR
Date:Mar 1, 2000
Previous Article:Burnt mounds in the East Midlands.
Next Article:The International Ancient Egyptian Mummy Tissue Bank at the Manchester Museum.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters