TIMBER-LACED AND VITRIFIED WALLS IN IRON AGE FORTS: CAUSES OF VITRIFICATION

Abstract

been set out more than once . The practical demonstration in 1937 by Professor V. G. Childe that a drystone wall with many timber beams inside its rubble core would, if burnt under the right conditions, produce vitrified masses, provided the first really satisfactory explanation of the phenomenon*3 >. Forts with such timber-laced walls are known from the continent and there are a few in Scotland of which the best known is probably Aber nethy in Perthshire*4 >. The theory that vitri fication was a deliberate process undertaken to create a fused rampart of enormous strength, always improbable and lacking any experi mental support, finally became untenable after this experiment. However, a conclusive demonstration of the correctness or otherwise of this theory of the origin of the vitrified forts is hard to find because, while beam-holes remain in unburnt timber-laced walls as at Aberncthy, the pro cess of vitrification usually completely destroys the masonry and causes built wall-faces to collapse, together with any beam-holes in them. Finavon is a good example of this. Even the surviving inner wall-faces are so badly warped that, though there are sometimes suggestions of beam-holes, nothing conclusive is visible(5>. Excavations on the vitrified fort of Dun Lagaidh in Wester Ross in August 1967 uncovered a short stretch of wall standing 6 feet high with both vitrified core and beam holes visible (Pl. i). The wall was in fact an outwork, protecting the eastern end of the oblong hillfort<6) : a mass of fused rubble had burst from the core through the inner face which, since it was still standing vertical, must previously have been buried by collapsing rubble from the upper parts of the wall before the maximum heat had been generated. Two rows of clear beam-holes are visible next to this. This is a conclusive demonstration that Childe's theory is correct and that masses of vitrified rubble are the remains of burnt

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.