Giraglia 1, c. AD 20
Country: France
Place: Cap Corse (Haute-Corse), 500 m from the coast, at a depth of 20 m
Coordinates (approximate): 43.02, 9.40
Type: Roman merchant ship carrying wine in bulk (dolia ship)
Identified: No
Dated: c. AD 20, based on the Arretine ware (Sciallano and Marlier 2008: 131).
Beam: Estimated at about 7 m.
Keel Length: Unknown.
Length Overall: Estimated at about 20 m.
Number of Masts: Unknown.
History of the wreck
The discovery of the Giraglia 1 wreck was reported to the authorities in 1992 by sport divers. It is one of the few early Imperial wrecks that carried wine in bulk in dolia and doliola and that preserves, even if only very partially, remains of its hull.1 Its planned excavation began in 1997. The presence of a cargo of Tarraconensian Dressel 2-4 wine amphorae has linked the Giraglia 1 to the export of wine from that region. The presence of Arretine ware allowed Sciallano and Marlier (2008: 131) to date the wreck to around AD 20.
Description of the site
The wreck lies at a depth of 20 m off Cap Corse, 500 m from the coast. The site covered an area of 200 m², with seven dolia grouped in a central area. The heavy fragmentation of the amphorae is interpreted as the result of the post-depositional processes undergone by the wreck, with the movement of the large dolia across the site, which literally flattened the areas with wine amphorae. The timber remains of the ship lay in an area of 7 × 2 m and were very poorly preserved (Figure 1).
Figure 1. Plan of the Giraglia 1 (after Sciallano and Marlier 2008: 136).
Cargo
The ship carried wine in bulk in seven dolia and in doliola, together with a cargo of Tarraconensian Dressel 2-4 wine amphorae. Arretine ware was also found.
Ballast
Not reported.
Anchors
Not reported.
Guns
Not applicable.
Iron concretions
Not reported. Square-section iron nails were used profusely in the frames.
Hull Remains
The remains consist of 26 frames, planking strakes, ceiling strakes and stringers, with no remains of the backbone, and correspond to the central part of the hull (Figure 1). Thanks to the lead pipes for draining on deck the bilge water collected by the pump, the position of bow and stern can be established. The position of the small finds on board, normally related to the location of the cabin, can also help, although in this wreck the two do not seem to coincide.
Keel, keelson and mast step
Not preserved.
Frames
The frames, of fairly square section, 8 to 10 cm sided and 10 to 15.5 cm moulded, have a room-and-space (malla) varying between 13 and 22 cm, which is therefore fairly tight, although their pattern cannot be established (floor timbers and half-frames,2 or flat central floor timbers as in the Ladispoli). They were fastened to the hull by treenails and nails, independent of each other, driven from the outside inwards. The treenails are of truncated-cone shape, 1.8–2.2 cm in diameter at the base and 1.6–2.1 cm at the head. In most cases they are driven perpendicularly, reaching the upper face of the frame. In the frames with greater moulded dimensions (e.g. 14 cm), the treenails are driven indiscriminately through the side faces (Figure 2). The iron nails, blind (a punta perdida) and of square section (2–5 mm), are used profusely in the frames at an average spacing of 10 cm (minimum 3.5 cm, maximum 34 cm), without any relationship to the position of the treenails. No central or lateral limber holes were found, unlike in the Ladispoli.
Figure 2. Frame M117, showing the treenails driven through the south side face and never through the north side face (drawing: Marlier, after Sciallano and Marlier 2008: 142).
Planking
The fragments of strakes recovered are 9 to 13 cm wide and 2.9 to 4 cm thick; according to Marlier (2008: 137) they may correspond to the second or the first garboard strake. The mortises are cut in line along the edges of the strakes, spaced 6 to 9 cm apart. The tenons are about 4.8 cm wide and 0.9 cm thick; the length of none of them could be measured completely (≥10.7 cm). As in other wrecks such as the Madrague de Giens or the Ladispoli, the mortises are slightly wider, which makes it easier to correct errors in the position of “opposing” mortises when two strakes are assembled (Tchernia et al. 1978: 80). The locking pegs are of truncated-cone shape and were driven from the inside outwards, with maximum diameters of 1.1–1.4 cm and minimum diameters of 0.9–1.1 cm. In Marlier’s opinion, the ratio between the spacing of the mortises (6 to 9 cm) and their width (4.8 cm) gives an index (1.5) which, compared with that of the Saint Gervais 3, shows a high density of mortises in the Giraglia 1. There is no evidence of any external protection (e.g. lead sheathing) on the outer face of the hull.
Ceiling, thick stuff, and wales
The stringers were fastened to the frames with metal nails. The surviving fragments show that they were 4 to 6 cm thick and 10 cm wide. The removable ceiling strakes, by contrast, of the same thickness, were 11 to 23 cm wide, with no fastening to the frames.
Caulking
Not reported. No external protection was found.
Size and scantlings
The position of the lead bilge pipes allowed the beam of the Giraglia 1 to be estimated approximately at about 7 m; applying a ratio of 1/3 to the overall length gives a length of about 20 m, comparable to the Grand Ribaud D, 18 m long with a burden of 45–50 t.
Table 1. Scantling of the timber remains of the Giraglia 1 Shipwreck
| Timber | Sided [cm] | Molded [cm] |
| Keel | – | – |
| Keelson | – | – |
| Maststep Mortise | – – | – – |
| Buttresses | – | – |
| Frames | 8–10 | 10–15.5 |
| Futtocks | – | – |
| Room-and-space At central frames | 13–22 | |
| Planking | 9–13 | 2.9–4 |
| Ceiling Stringers Removable strakes | 10 11–23 | 4–6 4–6 |
| Main mast Ø Chock | – – | – – |
Frame dimensions are ranges for all frames. Planking and ceiling: sided = width, moulded = thickness.
Wood
The species recorded for the planking strakes (Quercus sp.) is practically unknown in ancient ship construction, in which resinous species were always chosen for the hull (Rival 1991: 53–56; Guibal and Pomey 1998: 433–434; Guibal and Pomey 1999: 19–20). Marlier examines (Sciallano and Marlier 2008: 144) which other wrecks have planking made of oak,3 these being isolated cases: the Archaic César 1 wreck, with some repairs to strakes; the Fiumicino 1 (4th–5th century AD), a fluvio-maritime cargo vessel whose hull was partly made of oak; and the County Hall wreck4 (end of the 3rd century AD), whose planking is of oak.5 Of the hypotheses proposed, the most viable seems to be to relate the choice of oak to the specific transport of dolia, although there is no parallel in the Ladispoli and Diano Marina wrecks, which have planking of resinous species.
As for the species recorded in the framing, oak (Quercus sp.) represents 58.8%, elm (Ulmus campestris Sm.) 23.5% and beech (Fagus silvatica L.) 17.7%. Except for beech, which bends less well, both oak and elm are common in the framing of ships of the period. For the tenons and treenails, oak is notably used, while ash (Fraxinus excelsior L.) was used for the pegs.
Bibliography
Guibal, F. and Pomey, P. 1998. L’utilisation du matériau bois dans la construction navale antique: analyse anatomique et dendrochronologique. In Méditerranée antique. Pêche, navigation, commerce. Actes des 120e et 121e Congrès nationaux des sociétés historiques et scientifiques, Aix-en-Provence 1995, Nice 1996, 159–175. Paris: CTHS.
Guibal, F. and Pomey, P. 1999. Essences et qualité des bois employés dans la construction navale antique: étude anatomique et dendrochronologique. In A. Corvol (ed.), Forêt et Marine, 15–32. Paris: L’Harmattan.
Marsden, P. 1994. Ships of the Port of London. First to Eleventh Centuries AD. English Heritage Archaeological Report 3. London: English Heritage.
Rival, M. 1991. La charpenterie navale romaine. Matériaux, méthodes, moyens. Travaux du Centre Camille Jullian 6. Paris: CNRS.
Sciallano, M. and Marlier, S. 2008. L’épave à dolia de l’île de la Giraglia (Haute-Corse). Archaeonautica 15: 113–151.
Tchernia, A., Pomey, P., Hesnard, A. et al. 1978. L’épave romaine de la Madrague de Giens (Var) (campagnes 1972–1975). Gallia suppl. 34. Paris: CNRS.
Wicha, S. 2010. Caractérisation d’un groupe d’épaves antiques de Méditerranée présentant un assemblage des membrures par ligatures végétales: approche architecturale et paléobotanique. Saarbrücken (doctoral thesis, Université de Provence, Aix-en-Provence, 2005).
Notes
1 Although very partial, these remains are of great interest when studied in comparison with those of the Ladispoli and Giraglia 2. As for the dolia wreck of Diano Marina, its study of the ship’s construction has not been considered relevant and it has not been included in the corpus of this work.
2 In the Diano Marina wreck the framing in the central area was not verified, so it is known only that floor timbers and half-frames alternate at the ends.
3 To this list of ships the author would add the Baie de l’Amitié wreck, whose keel, garboard strakes and planking are made of oak (Wicha 2010: 147).
4 The County Hall wreck is a Roman ship (in design and construction) discovered in 1910 on the south bank of the river Thames, opposite Westminster (London), dated by dendrochronology and by the associated pottery to the end of the 3rd century AD. The central part of the hull was preserved, about 13 m long and 5.5 m wide. It was built entirely of oak and, although clearly different from the Romano-Celtic construction of central and north-western Europe, its construction is local according to the dendromorphological study. The ship had a keel wider than it was high, to which a protective shoe was probably attached. The framing was fastened to the hull by cylindrical oak treenails. In the part corresponding to the bottom, the framing was denser and more robust than in the upperworks, which survived to a height of 1.55 m, just above what was probably the sheer. A wale was preserved at a height of 1.3 m. A stringer or side keelson, of which the ship would originally have had several, had a series of mortises for stanchions supporting the deck (Marsden 1994).
5 Theophrastus, in Historia Plantarum V, 4, 3, notes that oak is a good wood for building ships for rivers and lagoons because of its resistance to rot, unlike seagoing ships, where salt water affects it more quickly.