N E W S F E E D S >>>
April 1977 IndexBulletins Index

Caithness Field Club

Caithness Field Club Bulletin
1977 - April


J. I. Bramman

The beginner in archaeology frequently regards archaeology as being synonymous with excavation and excavation as being synonymous with digging a site. That this is so is perhaps attributable to some of the subject's early popularisers whose concentration on the material finds often made archaeology sound like a hunt for buried treasure. And, indeed, it is still the buried treasure connotation - the finds at Sutton Hoo or on St. Ninian's Isle - which get extensive journalistic coverage and attract the layman to the subject of archaeology.

And yet there is much more to archaeology than excavation and much more to excavation than sinking a pick-axe or spade into a suitable site. An excellent illustration of this has come from work in the 1950s and 60s by Audrey Henshall of the National Museum of Antiquities in Edinburgh on the chambered tombs of Scotland.

The Scottish chambered tombs have long been recognised to have a predominantly western distribution, with some spreading over to the east particularly in the northern Highlands and then extending to the northern isles of Orkney and Shetland. In itself the recognition of this distribution is a significant contribution, particularly since the topographical siting of tombs in upland areas suggests associated settlement in Neolithic times of the shoulders of hills rather than the valley floors or coastal plains. Together with the long-barrows of south eastern Scotland these tombs plus a couple of settlement sites in Orkney represent the main evidence for pre-Bronze Age settlement. As there are two major groups of chambered tombs, the tombs and long-barrows-suggest settlement by perhaps three culturally distinct groups of people.

The groupings of tombs can be made both geographically and typologically. One group, now called the Clyde Group, of south west Scotland, has been extensively studied by J. G. Scott ('The Clyde Cairns of Scotland' in Megalithic Enquiries in the West of Britain', ed. T. G. Powell et al.; Univ. of Liverpool Press, 1969). The chambers were small rectangular or square ones - with small, more or less circular cairns - or were elongated rectangles under rectangular or trapezoidal cairns, though a wide variety of types between these extremes can be identified. The second major group, in the Hebrides and northern isles and on the northern mainland of Scotland, shows a wide variety of chamber shapes, the only common feature being entrance via a narrow passage, and a similarly wide variety of cairn shapes - round, heel-shaped, squarish or extremely elongated - with, again, a common feature: the cairns are generally much larger than would be needed to simply cover the chamber and entrance passage.

It was from this identification of two major groups of tombs that Miss Henshall, at the suggestion of Professor Stuart Piggott, began, in 1951, a detailed survey of all the chambered tombs of Scotland. The result of this was the publication of her two volumes on 'The Chambered Tombs of Scotland' (Edinburgh University Press), the first volume (1963) dealing with over 400 tombs in the north and east and the second (1972) discussing an additional 200 sites in the south and west. In the case of every excavated tomb details are given of the original excavation and of any associated artefacts.

Though the survey follows a geographical plan it is clear that during the decade between publishing the two volumes there had been a major change of thought. This arose mainly because of the increasing use of radio-carbon dating, which considerably extended the time period associated with the building and use of the tombs. The possibility that tombs may have been built and developed over considerable periods was supported by Scott's typological classification of the Clyde tombs into a series extending from the most simple to the most highly developed, and from a series of excavations - arguably the most importance post-war excavations of prehistoric sites in Britain - by the late John Corcoran.

The first excavation was in the north of Scotland on the shore of Loch Calder in Caithness, in 1961. Three chambered tombs, known locally as Tulach an t'Sionnach (Fig. 5) and the Tullochs of Assery, were to be flooded during reservoir development, and a rescue excavation was undertaken (J. X. W. P. Corcoran. Proc. Soc. Antiquaries Scotland, 98, 1-75, 1967 - a commendable detailed report published only six years after the excavation.

Tulach an t'Sionnach appeared to be a long mound with a bulbous cairn at the south western end. The early stages of excavation showed that the long north-easterly "tail" was, in fact, an addition to an earlier structure which took the form of a heel-shaped cairn. Further excavation showed that even this was not the original form and that the tomb was first constructed with only a "minimal" cairn covering the burial chamber. (The heel-shaped cairn actually blocked the entrance passage of the minimal cairn suggesting new access via the roof, or cessation of use as a tomb and development as a ritual site). Excavation of the neighbouring Tullochs of Assery showed that there too the final form of the covering mound represented a later addition to a simple circular mound covering the chambers. In one of these tombs, Tulloch of Assery A, bones were found in six neat piles, five of them in the main chamber, where they were set on low platforms. The disarticulation of the bones in all but one of the piles led Dr. Corcoran to speculate on the possibility of original containment in bags or wooden containers, suggesting perhaps a cult of veneration of bones as a feature of Neolithic religion.

Dr. Corcoran's subsequent excavations, at Mid Gleniron in Wigtownshire - a Clyde group tomb in which the final mound covered two original mounds and chambers and blocked the entrance to one of them (Trans. Dumfries and Galloway Nat. Hist. and Antiqu. Soc., 46, 29-99, 1969) - and then on passage graves at Balvraid in Glenelg (Inverness-shire), at the Ord in Sutherland and finally, again in Caithness, at Camster Long (these last three unpublished at the time of Dr. Corcoran's death), all yielded evidence of multiperiod construction. The most spectacular example, at Camster Long, showed that the 130m long mound covered small minimal mounds over two chambers discovered during Joseph Anderson's 1866 excavation and over a third previously unexcavated chamber. (It is interesting to note that the neighbouring tomb, Camster Round, re-excavated and restored by Department of the Environment teams in the 1960s has not been reported to be a multiperiod structure, though it's covering mound, over 18m in diameter, is much larger than the minimal cairns observed by Corcoran. Unfortunately no excavation report has yet appeared).

The observation that tombs are multi-period structures leads to the conclusion that artefacts found during excavation need not be associated with the early stages of construction, and, indeed, if one accepts the possibility that use of a tomb might continue long after the construction period, then portable finds need not be related to the construction period at all. In Tulloch of Assery A, John Corcoran found evidence of a very recent use of the chamber, possibly by a tramp who built a fire there! More significantly, Miss Henshall, commenting on the frequent recovery of Beaker pottery sherds from chambered tomb contexts, disposes once and for all of the long-held view that this must imply a temporal overlap of tomb construction and Bronze Age settlement - a view reflected in plaques on D. of E. guardianship sites, which until only a few years ago were dating tombs to the middle of the second millenium B.C. Miss Henshall concludes that Beaker Folk were merely finally responsible for closing up the chambered tombs of peoples who had preceded them.

In lieu of a classification of tombs into a development sequence based on portable finds associated with them, Henshall has shown that an analysis can be based on the two major features of the structure: the chambers and the cairns which cover them. There are two basic chamber types, rectangular and polygonal, the former corresponding to the common chamber type of the Clyde group of "gallery graves" (as they are sometimes called), and the latter to the more varied types found in the northern and north-western passage graves. Following Scott's deduction and Corcoran's observation that some tombs show evidence of long-term development, Miss Henshall extends the argument to the assumption that the majority of tombs will show evidence of development and offers several possible modes beginning with a simple round cairn of the minimum possible size to cover the chamber and support the thrusts of the capstone - the "minimal" cairn referred to above.

The first development, it is suggested, would be by simple enlargement to form a bigger round cairn with access permitted down an extended passage. This extension of the passage is often revealed on a ground plan by a discontinuity in the direction of the passage.

A second stage, perhaps associated with the development of tombs as cult sites, is reflected in further increases in the size of the cairn with perhaps blocking of the earlier access to the chamber. Sometimes forecourts are developed with enhancement by projecting "horns" of stone or earth, giving the so-called "heel-shaped" cairns - formerly thought to be confined to Shetland but shown by Corcoran's early excavation in Caithness to be a developmental variety on the mainland too - and thence leading to the "double-horned" cairn, essentially a square cairn with horns projecting at each corner. Finally, perhaps cross-fertilised by contacts with the barrow builders of south and eastern Scotland, there is the development of the long cairn, sometimes in the north well in excess of 60m in length. Long cairns can be formed either by addition of an extended tail (as at Tulach an t'Sionnach) to an earlier single cairn, or by incorporation of several earlier cairns under one mound (as at Camster Long Fig. 5). Note on the plan of Camster Long the development of "forecourts" with horns at the north-east and south-west ends and the discontinuity in the line of the passage to the north-east chamber which marks the continuation of development.

The classification into chamber and cairn types emphasizes parallel lines of development with, basically, rectangular chamber derivatives in the south-west of Scotland and polygonal derivatives in the north and north-west. The latter are generally now called the "Orkney, Cromarty, Hebridean Group".

Tombs of the Orkney, Cromarty, Hebridean (or O.C.H. type) have roughly polygonal chambers, with, in Caithness, for example, division of the vaguely elliptical chamber by two pairs of upright ortho-stats into three linked sections (giving the so-called "tripartite Camster-type chamber", actually typified by Camster Round but also seen in the second chamber from the north-eastern end of Camster Long). This tri-partite chamber shows development by elongation of the chamber and addition of extra pairs of dividers to form the "stalled" chambers of Orkney with, for example, five sections at Unstan on Mainland and fourteen pairs of stalls at Knowe of Rousay on the island of Rousay. Of this O.C.H. regional group, the Shetland tombs are seen as a related offshoot, though chambers there tend to be developed into a trefoil (rather than tri-partite or stalled) form.

There are in addition two intrusive sub-groups of tombs in the O.C.H. region. A group of nine tombs in Orkney differs markedly from the polygonal-to-stalled line of development in having side cells for the interments, these side cells having only relatively small access ports so as not to detract from the splendour of the main chamber; these are the Maes Howe tombs, named after the most sophisticated example. And there is a second, apparently undeveloped sub-group lying on the southern edge of the region to the south of Inverness, called, after the type site, the Clava tombs. The Maes Howe tombs because of the sophistication of both conception and construction are thought to be a late development, perhaps influenced by Irish prototypes. The Clava tombs are, however, much more controversial. Superficially, they are simple passage graves with circular chambers, thus appearing to be an early type, but their association with a similar type of structure, called a "ring cairn",* and with standing stone settings, finally led Miss Henshall to remove them from the archetypal phases and to suggest that they actually represent a late revival of chamber tomb building in the middle of the second millenium B.C., long after other groups had been abandoned. (At the Clava site itself, of the three monuments in a line the other two are passage grave tombs and the middle one is a ring cairn.)

In parallel with these developments in the north, typography of the Clyde group of tombs shows similar stages and again there is an intrusive sub-group, the Bargrennan group of twelve tombs in the extreme south west, which, like the Maes Howe sub-group in Orkney, are thought to reflect Irish influences. Geographical proximity would, however, favour earlier contact here than in the northern isles.

Thus, in their extended development and in their siting the chambered tombs of Scotland are indicative of long continued settlement of marginal land. Most of them lie on what is at present unenclosed land, and Miss Henshall in visiting the sites noted that even where they now lie in enclosures, the land is generally marginal upland taken in as pasture, and not arable. With animal bones frequently appearing during excavation, the distribution suggests not the conventionally accepted settlement by farmers practising hoe cultivation but one by herdsmen with an economy based mainly on cattle.

Though a typology has been devised, the dating of the tombs is still a matter of speculation. There are still only a handful of radio-carbon dates from the Scottish Neolithic. Miss Henshall has, however, offered a tentative dating beginning sometime in the fourth millennium B.C. for the earliest prototypes, with development continuing through the third millennium. The Maes Howe sub-group in the north is, as mentioned above, thought to show Irish influences and the dating of the tomb at New Grange in the Boyne valley to about 2500 B.O. is thought to date the Maes Howe tombs to later in the third millennium. Some of the more highly developed Orcadian stalled chambers are thought to be coeval. Finally some tombs like those at Clava already mentioned and, to give another example, one associated with the stone settings at Callanish in Lewis, are dated to the second millennium B.C.

The work of Audrey Henshall on typology and the supporting excavations by the late John Corcoran have thus revealed the Scottish chambered tombs in a new light. The chamber is the only indisputably original part of the structure; the cairn shows the modifications and extensions subsequently made, and the artefacts and other objects found in the excavation of a tomb may only date from the later stages of use and final abandonment, though they may also shed light on the economy and ritual observances. With very little evidence from contemporaneous settlement sites, apart from late-dated Orkney examples from Skara Brae and Rinyo, knowledge of the Neolithic period in Scotland is based heavily on the chambered tombs. Fortunately, archaeology is much more than excavation and excavation is much more than digging.


* A ring cairn is circular with a central open area too large to have been roofed as a chamber and, more important, with no passage through the cairn to give access from the outside. Bronze Age Beaker ware seems to be associated with the construction phase.


Bulletin Index