Forged Brick-Stamps from Pevensey


D. P. S. Peacock

Antiquity 1973


[138] In Britain, epigraphic mention of the emperor Honorius is restricted to a number of bricks or tiles from the Shore Fort of Pevensey in Sussex (Eph. Epigr. ix, 1281). The bricks stamped HON AUG ANDRIA are supposedly evidence for the refurbishing of the defences during the emperor's reign (395-423 AD). A few fragments in the poems of Claudian praise the victory of Honorius's general, Stilicho, over the Irish and Picts and they mention that the coast was made secure against Saxon attack (Frere, 1967, 363). This took place between 395 and 399 AD and the bricks from Pevensey have been held as the sole archaeological evidence of the campaign. They are thus important documents and it is not surprising that they have been mentioned in numerous books and papers dealing with the end of the Roman occupation. However, evidence has recently come to light which leaves little room for doubt that the stamps were forged in the early twentieth century, a possibility to be discussed below.



On 11 April 1907, Charles Dawson, later to become known as the discoverer of Piltdown man, exhibited before the Society of Antiquaries of London, some inscribed bricks and tiles from the Roman fort at Pevensey (Dawson, 1907). The specimens were shown in two trays, one containing material collected by Dawson himself, the other with fragments lent by the gentlemen in charge of the then current excavations.

Dawson drew attention to a portion of a black brick which he had discovered beneath the arch of the postern gate in the north side of the wall, in the year 1902. The brick, which had the usual semicircular marking, bore an oblong impression with the relief letters HON AUG ANDRIA, a photograph of which is reproduced in the published account. At this point in the text a footnote refers to a red brick, from the eastern part of the wall, bearing the mutilated outline of the same stamp. Dawson continued with a discourse on the possible interpretations of the word ANDRIA, and in conclusion drew attention to a third brick with part of the same impress ion, found during the Pevensey excavations This piece, bearing the letters . . ON AUG . . . NDR . . ., was again referred to in Salzman's report on the 1907 excavation seasons at Pevensey (Salzman, 1908). The fragmentary inscription was read by comparing with Dawson's more perfect example.

There is however, evidence for the existence of four rather than three stamps. The version of Salzman's report circulated to subscribers is accompanied by a plate, later partly reproduced in the Victoria County History of Sussex, Vol. 3. It shows a brick or tile with a semicircular marking in the middle of which is a clear HON AUG ANDRIA stamp. In one corner of the plate is inscribed, 'Stamped tile from Pevensey Castrum C. Dawson F.S.A 1902', and presumably this is the piece upon which Dawson's discourse began. However although the stamp is clearly from the same die as the one illustrated in his paper, since the lettering and texture of the background are identical, the photographs are different and the stamp illustrated in Salzman's paper has a chip removed from the bottom right hand corner. In my opinion the two photographs are unlikely to be of the same stamp for there are many points of detail which do not tally. The chipped specimen is in the British Museum and examination of the original strengthens this suspicion. It is certainly not the mutilated specimen referred to in Dawson's footnote for it is gray-black, not red, in colour. Thus three stamped bricks are specifically mentioned by Dawson while the existence of a fourth is implied by his illustration.



In view of the importance of the material it is perhaps surprising that only two of possibly four pieces can now be located. The near complete stamp illustrated in Salzman's paper [139] (Fig. 1a ) is in the British Museum (Cat. No. 1908, 6-13, 1), while the fragmentary specimen from the Pevensey excavations is in Lewes Museum (Fig. 1b ). Both are in a similar fine, hard, grey fabric, the British Museum example having blue-black surfaces, while the Lewes specimen is slightly brown in places. Thin sectioning reveals fine quartz grains (o.o2 mm. across) set in a matrix of grey optically isotropic fired clay. Both bear traces of mortar and the fractured surfaces are coated with a pale grey deposit, shown by X-ray diffraction to consist principally of finely divided quartz.


Brick Stamp from Pevensey

Fig. 1. Brick-stamps from Pevensey: a , British Museum; b , Lewes Museum. Scale X 1/2.

Careful examination was made of the bricks now preserved in the walls of Pevensey: a variety of fabrics is represented but none of those studied compared with the stamped bricks in either the hand specimen or under the microscope.

Examination of the stamps leaves no doubt that both were produced from the same die. However, the lettering itself is curious: the rather spidery style is difficult to parallel among the general run of Roman military or civil tile stamps from Britain. For example, stamps of the Classis Britannica, fragments of which were found at Pevensey, always have much thicker lettering (Brodribb, 1969)



The bricks are anomalous in both the fabric and in the style of the stamp. This and their association with Charles Dawson, discoverer of the spurious Piltdown man (Weiner, 1955), Is sufficient to cast serious doubt upon their authenticity. It is true that the anomalous features could be explained by their unique date and the find from Salzman's reputable excavations is reassuring, but under the circumstances, it was felt that a thermoluminescence authenticity test would be worthwhile. This was kindly undertaken by Dr S. J. Fleming of the Research Laboratory for Archaeology and the History of Art, University of Oxford, and the British Museum specimen was also examined in the British Museum Research Laboratory. The principles of the technique are now well established and need not be reiterated here (see Fleming, 1970, 1971; Aitken et al, 1971; Fleming et al, 1971; Fleming and Roberts, I97I).

The British Museum specimen was subjected to detailed examination and, using routine methods, Dr Fleming suggests a maximum age of 360 years. However, with his recently developed pre-irradiation technique, he estimates the firing date of between 1900-1940 AD (see Fleming, 1972, for an account of this development). Measurements carried out in the British Museum Research Laboratory indicate ages of about 70 years and 85 years by the conventional and pre-irradiation techniques respectively. These results are clearly in good agreement, within the limits of error of the method, with the results obtained by Dr Fleming. A routine test was also carried out on the Lewes specimen by Dr Fleming and this also proved to be modern.

The thermoluminescent dates could of course be erroneous if the bricks were heated or refired recently. The Lewes example has been joined with wood glue which would doubtless have involved application of heat to the specimen but the temperature would surely not have reached 350o, required to destroy the thermoluminescent output of a specimen fired [140] in antiquity. Apart from this it is extremely unlikely that two valuable specimens housed in separate museums would each have been subjected to considerable heating, either deliberately or accidentally. Considered alongside the anomalies in fabric and style there are very strong grounds for suggesting that the bricks are twentieth-century forgeries and that the piece from Salzman's excavations was planted to strengthen the case for their acceptance as genuine. The stamps should be disregarded in future discussions of the end of the Roman occupation.

The reassessment of the bricks also has implications regarding the authorship of the Piltdown forgery, a question discussed by Weiner (1955) and more recently in an editorial of this journal (Daniel, 1972). It now seems that Dawson was associated with another fraud at a date prior to the Piltdown affair, which firmly points the finger of suspicion in his direction, although there is of course no proof that he was not the innocent dupe of another party on both occasions. It has been suggested that Teilhard de Chardin was the author of the Piltdown hoax, but since he did not meet Dawson until 1908 he could have had no part in the Pevensey forgery, which weighs against this to some extent.

In my opinion the time is now ripe for a full investigation of Dawson's numerous and often bizarre discoveries. From an archaeological point of view it would be particularly useful to know more of the cast iron figurines reputed to have come from the Roman iron-working site at Beauport Park (Dawson, 1903), while geologically, the mammal Plagiaulax dawsoni represented by a molar tooth from a bone bed 'near Hastings', might repay scrutiny.


Acknowledgements: I am particularly indebted to Mr J. Manwaring Baines, Director of Hastings museum, who first drew my attention to the possibility of the finds being spurious. Mr J. Brailsford and Miss C. Johns of the British Musetim gave every facility in the examination of the piece in their care, as did Mr A. Page of the Barbican House Museum, Lewes. Mr N. Bradford prepared the thin sections and assisted with the X-ray diffraction work. Finally, I am particularly grateful to Dr S. J. Fleming for carrying out the thermoluminescence authenticity tests, and to Dr A. E. Werner for kindly communicating the results of measurements carried out in the British Museum Research Laboratory.


AITKEN, M. J., P. s. MOOREY, and P. J. UCKO. 1971.

The authenticity of vessels and figurines in the Haçilar style, Archaeometry, X111, 8-141.

BRODRIBB, G. 1969. Stamped tiles of the 'Classis Britannica', Sussex Arch. Coll., CVII, 102-27.

DANIEL, G. 197Z. Editorial, Antiquity, XLVI, 161-4.

DAWSON, C. 1903. Sussex iron work and pottery, Sussex Arch. Coll., XLVI, 1-62.

1907. Some inscribed bricks and tiles from the Roman Castra at Pevensey, Proc. Soc. Ant., XXI, 410-3.

FLEMING, S. J. 1970. Thermoluminescent dating: refinement of the quartz inclusion method, Archaeometry, Xll, 133-45. 1971. Thermoluminescent authenticity testing of ancient ceramics; the effects of sample drilling, Archaeometry, X111, 59-70.

1972. Thermoluminescence authenticity testing of ancient ceramics using radiation sensitiivity changes in quartz, Naturwissenschaften , LIX, 145-51.-

FLEMING, S. J., H. JUCKER, and J. RIEDERER. 1971. Etruscan wall paintings on terracotta: a study in authenticity, Archaeometry, X111, 143-67.

FLEMING, S. J. and H. S. ROBERTS. 1970. Thermoluminescent authenticity testing of a Pontic amphora, Archaeometry, Xll, 129-31.

FRERE, S. S. 1967. Britannia: A history of Roman Britain (London).

SALZMAN, L. F., 1908. Excavations at Pevensey, 1906-7, Sussex Arch. Coll., Ll, 99-114.

WEINER, J. s. 1955. The Piltdown forgery. (London).

Back to the

Back to the

Back to