Sud Lavezzi 5, AD 50–150

Carlos de Juan Fuertes, 2013

Country: France

Place: Near the Lavezzi islet, Strait of Bonifacio (Corse-du-Sud), on a sandy bottom at a depth of 72 m

Coordinates (approximate): 41.32, 9.26

Type: Large Roman merchant ship, possibly a grain carrier

Identified: No

Dated: AD 50–150, based on a Gauloise 5 amphora, late Italian sigillata and a Dragendorff 24/25 bowl (Delauze et al. 2007: 96).

 

Beam: About 6 m.

Keel Length: Unknown.

Length Overall: About 30 m.

Number of Masts: Unknown; notches for a mast step are present.

 

History of the wreck

The Sud Lavezzi 5 wreck was discovered by Comex during the archaeological survey of 2005 in the south of the island. It has not been looted and preserves a substantial part of its hull. It has not been studied or excavated; only sand has been removed with the Blaster thruster robot.1 The wreck has two features that make it interesting. The first is that it has no remains of a main cargo or ballast; this complete absence suggests that it may have been a ship carrying a main cargo of perishable material, grain being suggested by the authors of the study (Delauze et al. 2007: 96), although other possibilities, from slaves to animals, should not be ruled out. The second particularity of this wreck with an atypical cargo2 lies in the notches supporting the mast step,3 which have no known parallel in the literature. It could be an architectural unit showing an architectural signature not known until now, or else a transitional stage between the archetypal technique for supporting the mast step in ships of the Republican architectural family (notched over the framing) and those of the Imperial architectural family (seated on two sister keelsons).

As an architectural unit, the Sud Lavezzi 5, dated between AD 50 and 150, has half-frames offset from the centreline or asymmetrical, a feature of the Imperial architectural family; but the mast step is not supported on two sister keelsons, as is usual in wrecks defining this family, such as the Saint Gervais 3 of the mid-2nd century, but is fitted to the hull by notches very similar to those of wrecks of the Republican architectural family.4 The presence of an entirely particular technique for housing the mast step, together with a framing that alternates floor timbers and offset half-frames following a precise rhythm in the area of the mast step (Figure 2, 2), leads the author to propose, as a working hypothesis, that this may be a wreck in a transitional or hybrid phase between the Republican and Imperial architectural families.

 

Description of the site

The wreck lies on a sandy bottom at a depth of 72 m in the Strait of Bonifacio (Corsica), near the Lavezzi islet. Some twenty metres of hull were exposed and clearly identified (Figure 1).

 

 

Figure 1. Photograph taken by the work ROV after removal of the sediments by the Blaster. The notches for the mast step and the off-centre half-frames can be seen (photo: courtesy of DRASSM).

 

Cargo

No main cargo was found, suggesting a perishable cargo, perhaps grain. Four amphorae and some further fragments were observed from the ROV: a Gauloise 5, late Italian sigillata in the western part of the site and a Dragendorff 24/25 bowl. Some tegulae and a sounding lead found with these materials suggest that they relate to the stern cabin and the crew rather than to the cargo, in line with the proposal of Delauze et al.

 

Ballast

No ballast was found.

 

Anchors

Six to eight iron anchors, 2 to 3 m long, were identified scattered around the site, half-buried.

 

Guns

Not applicable.

 

Iron concretions

Not reported. Nails with their points bent and clenched over the upper face of the frames were observed.

 

Hull Remains

Little information is available on the ship’s structure; some twenty metres were exposed and clearly identified, suggesting a length of around 30 m with a beam of 6 m. Pistons of the ship’s bilge pump were also found, together with wooden parts of the uptake pipes and parts of the pump housing in the hull. The author’s observations of the published photograph and of the DRASSM archive reports5 suggest that the hull lines are fairly flat, although the greater moulded dimension of the floor timbers over the centreline does not allow this hull to be considered as having completely flat lines (Figure 1).

 

Keel, keelson and mast step

In the Sud Lavezzi 5 a series of notches can clearly be seen in the framing along the ship’s centreline, which can readily be related to the mast step, as the authors observed. The floor timbers do not seem to have been worked in any way to fit the mast step, unlike the offset half-frames, which all have a recess to house it. The author therefore interprets that the underside of the mast step would have had perpendicular notches to fit over the floor timbers (Figure 2, 3), which would block longitudinal movement of this piece. For the half-frames, the mast step may have had shallower notches than the former, or none (Figure 2, 4), as seen in the Dramont A. The notch in the half-frames would serve to restrain lateral movement of the mast step. This technique fulfils the same function as the triangular, rectangular or polygonal notches placed facing each other on the upper faces of the frames, documented on many occasions, for example in the Madrague de Giens, Cavalière, Plane 1 or Cap Béar 3 wrecks, among many others, a technique common in the Republican architectural family.

 

 

Figure 2. Top view of the diagram of the placing of the floor timbers and off-centre, offset or asymmetrical half-frames, arranged alternately in relation to the centreline (1). Axonometric view of the same diagram (2). Axonometric view of the underside of a hypothetical mast step, with the groove fitting over the floor timbers marked in purple and the points of contact and lateral support with the half-frames in green, and a hypothetical underside of the mast step based on the Dramont A (3). Side view of the notches of a hypothetical mast step, with fittings for the floor timbers only, or for both floor timbers and half-frames (4) (drawings: C. de Juan Fuertes).

 

Frames

The framing comprises at least 26 timbers, of decidedly robust appearance and with a tight room-and-space (malla).6 There is a fairly regular alternation of floor timbers and offset or asymmetrical half-frames, the position of the joint of the latter alternating on either side of the keel (Figure 2, 1). There are some nails fastening the framing whose points are bent and clenched over the upper face of the frame, as observed in Colonia de Sant Jordi or Fiumicino 1; it cannot be established whether there is a wooden guide plug (treenail) for this purpose, although by analogy the author thinks this must be the case.

 

Planking

Not reported.

 

Ceiling, thick stuff, and wales

Not reported.

 

Caulking

Not reported.

 

Size and scantlings

Some twenty metres of hull were exposed, suggesting a length of around 30 m and a beam of 6 m. The framing is robust, with a tight room-and-space.

 

Table 1. Scantling of the timber remains of the Sud Lavezzi 5 Shipwreck

TimberSided
[cm]
Molded
[cm]
Keel––
Keelson––
Maststep
Mortise
–
–
–
–
Buttresses––
Floor timbers––
Futtocks––
Room-and-space
At central frames
–
Planking––
Ceiling––
Main mast Ø
Chock
–
–
–
–

No timber dimensions have been recorded; the wreck has not been excavated.

 

Wood

Not reported.

 

Bibliography

Delauze, H.-G., Cayol, J.-C., Massy, J.-L., Long, L., Leroy, F. and Rivet, L. 2007. Épave Sud Lavezzi 5. Bilan scientifique du DRASSM 2005, 94–96. Paris.

 

 

Notes

1 Also used on the deep wrecks of Grand Ribaud F and Tiboulen de Maïre.

2 It could be the only known example of a wreck carrying grain or other perishable material not packed in amphorae.

3 The author’s own observation.

4 Allowing for the differences, its hull lines recall those of the Madrague de Giens.

5 The author thanks F. Cibecchini (DRASSM) for the information provided during the author’s participation in the 2012 campaign of the underwater archaeological map of Corsica.

6 The authors of the study give 30 cm (for the room-and-space?), a figure the author doubts when looking at the published photographs (Delauze et al. 2007: 96).