The Mound of the Hostages, Tara
Duma na nGiall
- The Mound of the Hostages, Tara by Muiris O’Sullivan.
Report on the archaeological excavations carried out at the Mound of the Hostages on the
Hill of Tara, County Meath, Ireland.
This is a report on the archaeological excavations at the Mound of the Hostages
on the Hill of Tara
, County Meath, Ireland. The excavations were directed
initially by Sean P. O Riordain, who spent two summers at the site (1955 and
1956), and were completed in a third season during the summer of 1959 by
Ruaidhri de Valera, O Riordain's successor as Professor of Archaeology in
University College Dublin.
This report incorporates their findings,
supported by post-excavation analysis. The earliest features recorded at the
site date from about the mid-fourth millennium (cal.) BC and include the
remains of individual fires, spreads of charcoal and a ditch running
partially underneath the cairn. The mound itself is a mantle of soil,
approximately 1m deep, covering the cairn which encloses a passage tomb.
tomb consists of three successive compartments separated by low sill stones,
the roof stones surviving over the two inner compartments. Noteworthy
features of the tomb include the occurrence of megalithic art on two of the
orthostats, the presence of three cist-like structures against the outer
faces of the orthostats, and most remarkably the collection of burnt and
unburnt human bone representing hundreds of individuals distributed
throughout the tomb and cists, accompanied by a rich array of artefacts,
some of which are decorated.
Surrounding the cairn and sometimes located
beneath the earthen mantle, the excavators recorded a ring of seventeen bone
deposits that, like the earliest dated burials in the tomb, have produced
radiocarbon determinations focusing in the period 3350-3100 (cal.) BC. A
ring of fire pits coinciding spatially with the ring of burials has been
radiocarbon dated to more than a millennium later.
Within the tomb the
Neolithic deposits were disturbed as successive Early Bronze Age burials
were incorporated, and in due course the burial activity spread out into the
overlying mound so that approximately nineteen separate burials occurred in
the earthen mantle, some of them accompanied by richly decorated pottery and
other artefacts. Traces of linear palisades located close to the mound
appear to date from about 2500 (cal.) BC. Later features include a Late
Bronze Age ring-ditch located beside the mound, a small number of Iron Age
finds and some post-medieval elements.
Plan of megalithic tomb (Mound of the Hostages)
Carrowkeel pots from the pre-cairn cists at the Mound of the Hostages
Duma na nGiall: The Mound of the Hostages, Tara
Book review by Elizabeth Shee Twohig
In 1952, Séan P. Ó Ríordáin, Professor of Celtic Archaeology at UCD, began
excavations at Tara
and in 1955 and 1956 he excavated
at the most prominent mound on the hilltop, known as the Mound of the
Hostages. He found that the mound covered a small passage tomb, facing
southeast, which was covered by a stone cairn, with an overlying clay
mantle. Sadly, Ó Ríordáin fell ill, and died in 1957. His successor at UCD,
Prof. R. de Valera, completed the excavations in 1959.
The author of the volume under review, UCD lecturer Dr Muiris O’Sullivan,
has been working intermittently since the late 1980s on the excavation
report and has now seen it though to publication, 50 years after the
excavation commenced. In the intervening years, other UCD archaeologists
worked on aspects of the material, notably Prof. Michael Herity (Neolithic
artefacts) and Dr Rhoda Kavanagh (Bronze Age pottery and knives).
One of the strong points of this excavation report is the suite of 58
radiocarbon dates that were processed in Groningen in 2001; these are
presented and analysed here by A.L. Brindley, J.N. Lanting and J. van der
Plicht. These dates have contributed hugely towards one’s understanding of
the sequence of events at the site. Both regular and AMS dating were used as
appropriate (AMS of necessity for cremated bone, and also for the unburnt
bone, to avoid destruction of large quantities of it in advance of any
further anatomical research); note however that the caption for the list of
dates in Table 1 confuses the abbreviations for the two forms (GrN and GrA).
A long sequence of events is now identifiable as a result of these
excavations. Firstly, the area was used to some extent before the
construction of the monument, and some Neolithic sherds and radiocarbon
dates indicate activity c. 3800-3700 BC. A Mesolithic chert flake was also
found, though this single item can hardly be used to indicate that the site
was a 'sacred place' (page 246) for hunter-gatherers, and it is more mundanely
interpreted by lithics specialist Graeme Warren in Appendix 8 as indicating
some kind of Mesolithic activity, or the use of a Mesolithic artefact as an heirloom.
The construction and original use of the tomb has now been radiocarbon dated
to 3350-3100 BC. It is estimated that the tomb contained more than 300
burials including 63 from the three primary cists built against the outside
of the chamber. As at Fourknocks
, many of the
unburnt burials were infants or children; however this part of the report is
not altogether satisfactory due to the fact that the anatomist was unable to
complete it due to ill health. A number of cremations placed around the
perimeter of the mound, were shown by radiocarbon dating to be contemporary
with the burials in the tomb. The tomb and cists also yielded an amazing
range of artefacts, including two intact Carrrowkeel pots (protected in the
cists); bone and antler pins; balls, beads and pendants, the latter made
from unusual stones such as serpentine, while the balls and beads were from
more mundane materials. A collection of unusual bone tubes probably formed a
pair of long necklaces.
There are some indications of later Neolithic activity around the mound, in
the form of linear palisades and a ring of fire pits. The author is hesitant
to assign the large oval enclosure detected recently through Discovery
Programme gradiometry programme (Fenwick, J. and Newman, C. 2002. Geomagnetic
survey on the Hill of Tara, Co. Meath, 1998-9. Discovery Programme Reports 6,
1-17) to this period, and underplays the
use of the site in the earlier third millennium BC, claiming (page 244) that
there was no 'aggiornamento' as in the Boyne Valley at this time.
Be that as it may, the contrast with the Boyne Valley sites is certainly
remarkable for the next phase of activity at the mound, when a series of c.
25-30 burials were inserted into the tomb chamber or its covering clay
mantle. These were accompanied by both bowls (3750-3600 BP) and various
forms of vases (3750-3500/3400 BP), or by collared urns (3570-3360 BP), as
well as a small range of other material such as razors, awls and a
Unfortunately it proved impossible to reconcile the terminology used for the
Early Bronze Age pottery between the main report and the discussion in the
appendix dealing with radiocarbon dates. The main report uses the older term
'Food Vessel', while the appendix employs the more recent nomenclature, such
as bowls, vases, vase urns, and encrusted urns. There are inconsistencies
even within the main report: the pot from Burial 33 is described both as a
vase Food Vessel and as a Food Vessel of bowl/vase form on page 185. It is
called a vase Food Vessel in Appendix 2, a Bowl/vase Food Vessel in both the
general list of artefacts (Appendix 4) and in Table 18, while it is
described as a tripartite bowl in the radiocarbon Appendix (my italics). One
of the pots from Burial 38 is classed as a vase Food Vessel in the report
and as a cord decorated urn in Appendix 7. The well-known inhumation burial
of a young male accompanied by a range of exotic beads (faience, jet, amber
and bronze), a razor and probable awl, but no pottery, published by Ó Ríordáin
in 1955 has now been radiocarbon dated to 1750-1500 BC and was thus
the last burial inserted into the mound.
In bringing all this material to publication, the author had an unenviable
task, inheriting the less than perfect archive of an old excavation, and
with some of the paperwork, artefacts and human remains lost, or being
recovered only as the study advanced. An important field drawing was located
after the report had been edited and had to be added as Appendix 10.
Sensibly, O’Sullivan has identified those areas of the archive and record
where information is lacking or unclear.
For the most part the report is easy to follow, no small achievement under
the circumstances. Unfortunately, the gremlins have attacked what would
otherwise have been a very useful concordance list (Appendix 4), and the
figure numbers have slipped forward 2-4 numbers; a random check of c. 12
figure numbers listed found them to be all incorrect. The illustrations are
of a uniformly high standard, especially the drawings and photographs of the
artefacts; old black and white photographs are also excellent, and though
unattributed, are, perhaps, Jim Bambury’s work.
Dr O’Sullivan has grappled valiantly with the unenviable task of pulling all
this material into shape, and has succeeded in presenting us with as clear a
picture as possible under the circumstances. There is little or no
discussion, but at least we now have a basic document from which further
research at Tara can proceed.
Elizabeth Shee Twohig
Drawing of decorated stone, orthostat L2, Mound of the Hostages by Ursula
Mattenberger from the book Duma na nGiall (The Mound of the Hostages), Tara.
Boyne Valley Private Day Tours
Pick up and return to your accommodation or cruise ship. Suggested day tour:
Newgrange World Heritage site, 10th century High Crosses at Monasterboice,
Hill of Tara the seat of the High Kings of Ireland and the Hill of Slane where St. Patrick let a Paschal fire in 433