The truth of the matter seems to be that, as in the case of a find of coins, it is the date of the last piece that fixes the time of the deposit. There may be coins in it a hundred or a thousand years older, but this hoard cannot have been buried before the last piece which it contains was coined. So it is with this barrow. The presence of Roman or post-Roman pottery in an avowedly undisturbed sepulchre fixes, beyond doubt, the age before which the skeletons could not have been deposited where they were found by Dr. Thurnam. The presence of flints and coarse pottery only shows, but it does so most convincingly, how utterly groundless the data are on which antiquaries have hitherto fixed the age of these monuments. It proves certainly that flints and shards were deposited in tombs in Roman or post-Roman times; and if there is no mistake in Dr. Thurnam's data, this one excavation is, by itself, sufficient to prove that the Danish theory of the three ages is little better than the "baseless fabric" of—if not "a vision"—at least of an illusion, which, unless Dr. Thurnam's facts can be explained away, has no solid foundation to rest upon.
If any systematic excavations had been undertaken in the Scandinavian long barrows, it would not, perhaps, be necessary to adduce English examples to illustrate their age or peculiarities. Several are adduced by Sjöborg, but none are reported as opened. This one, for instance, is externally like the long barrow at West Kennet, and, if Sjöborg's information is to be depended upon, is one of several which mark the spot where Frode V. (460-494) landed in Sweden, where a battle was fought, and those who fell in it were buried in these mounds, or where the Bauta stones mark their graves. If this is so, the form of the long barrow with its peristalith was certainly not unknown in the fifth century; and there is no improbability of its being employed in England also in that age. In settling these questions, however, the Scandinavians have an immense advantage over us. All their mounds have names and dates; they may be true or they may be false, but they give a starting-point and an interest to the enquiry which are wanting in this country, but which, it is hoped, will one day enable the Northmen to reconstruct their monumental history on a satisfactory basis.
103. Long Barrow at Wiskehärad, in Halland. From a drawing by Sjöborg.
In most cases antiquaries in this country have been content to appeal to the convenient fiction of secondary interments to account for the perplexing contradictions in which their system everywhere involves them. In the instance of the Kennet long barrow there is no excuse for such a suggestion. All the interments were of one age, and that undoubtedly the age of the chamber in which they were found, and the pottery and flints could not have been there before nor introduced afterwards. Indeed, I do not know a single instance of an undoubtedly secondary interment, unless it is in the age of Canon Greenwell's really prehistoric tumuli. When he publishes his researches, we shall be in a condition to ascertain how far they bear on the theory.[339] In the chambered tumuli secondary interments seem never to occur; and nothing is more unlikely than that they should. As Dr. Thurnam himself states: "In three instances at least Mr. Cunnington and Sir R. C. Hoare found in long barrows skeletons which, from their extended position and the character of the iron weapons accompanying them, were evidently Anglo-Saxon."[340] A simple-minded man would consequently fancy that they were Anglo-Saxon graves, for what can be more improbable than that the proud conquering Saxons would be content to bury their dead in the graves of the hated and despised Celts whom they were busy in exterminating.[341]
If the above reasoning is satisfactory and sufficient to prove that the long barrow at West Kennet is of post-Roman times, it applies also to Rodmarton, Uley, Stoney Littleton, and all the Gloucestershire long barrows which, for reasons above given (ante, [page 164]), we ventured to assign to a post-Roman period; and à fortiori it carries with it King Hildetand's tomb at Lethra. It is true we have not the same direct means of judging of its date as we have of our own monuments. The Danes treat with such supreme contempt any monument that does not at once fall in with their system, that they will not even condescend to explore it. So soon as Worsaae found some "flint wedges" in the tomb, he at once decreed that it was prehistoric, and that it was no use searching farther; and we are consequently left to this fact and its external similarities for our identification. Here, again, is a difficulty. The two drawings above given (woodcuts [Nos. 101] and [102]) may show them too much alike or exaggerate differences. The one is an old drawing from nature, the other a modern restoration; still the essential facts are undoubted. Both are chambered long barrows, ornamented by rows of tall stones, either partially or wholly surrounding their base, and both have external dolmens on their summit, and both contain flint implements. If this is so, the difficulty is rather to account for so little change having taken place in 230 years than to feel any surprise at their not being identical. The point upon which we wish to insist here is that they are both post-Roman, and may consequently belong to any age between Arthur and Charlemagne.
The remaining battle-fields of which representations are given in Sjöborg are scarcely so interesting as that at Braavalla, which with the tomb of the king slain there are landmarks in our enquiry. If those circles on Braavalla Heath do mark the battle-field, and that tomb at Lethra is the one in which the blind old king was laid—neither of which facts I see any reason for doubting—all difficulties based on the assumed improbability of the monuments being so modern as I am inclined to make them are removed, and each case must stand or fall according to the evidence that can be adduced for or against its age. To return, however, to the battle-fields given by Sjöborg. Figures 43 and 44 represent two groups of circles and Bauta stones near Hwitaby, in Malmö. These are said to mark two battle-fields, in which Ragnar Lothbrok gained victories over his rebellious subjects in Scania: Sjöborg says in 750 and 762, as he adopts a chronology fifty years earlier than Suhm. But be this as it may, there does not seem any reason for doubting but that these stones do mark fields where battles were fought in the eighth century, and that Ragnar Lothbrok took part in them. These groups are much less extensive than those at Braavalla, but are so similar that they cannot be distant from them in age.
At Stiklastad, in Norway, in the province of Drontheim, a battle was fought, in 1030, between Knut the Great and Olof the Holy; and close to this is a group of forty-four circles of stones, which Sjöborg seems, but somewhat doubtfully, to connect with this battle. But about the next one (fig. 49) there seems no doubt. The Danish prince Magnus Henricksson killed Erik the Holy, and was slain by Carl Sverkersson, in the year 1161, at Uppland, in Denmark; and the place is marked by twenty stone circles and ovals, most of them enclosing mounds and two square enclosures, 30 to 40 feet in diameter. They are not, consequently, in themselves very important, but are interesting, if the adscription is correct, as showing how this heathenish custom lasted even after Christianity must have been fairly established in the country. Another group (fig. 51) is said to mark the spot where, in 1150, a Swedish heroine, Blenda, overcame the Danish king Swen Grate, and the spot is marked by circles and Bauta stones; one, in front of a tumulus, bears a Runic inscription, though it merely says that Dedrik and Tunne raised the stone to Rumar the Good.