Marsala A

Country: Italy

Place: South of modern Marsala, Sicily, near Lido Signorino, beneath 40 m of water

Type: Axial timber-built ship, double-ended like Serçe Limanı

Identified: Not identified by name

Dated: Originally dated, together with Marsala B, to the late eleventh to early twelfth centuries based on ceramic assemblages; recent reanalysis concludes the two ships were two separate losses, presumably both built in Sicily, with Marsala A lost around AD 1050.

Number of Masts: At least one; a mast-step partner and stanchion were identified alongside the sister keelsons

History of the shipwreck

After an intense storm off Sicily in 1983, reports of archaeological material exposed along the seabed south of modern Marsala, Italy, prompted swift rescue operations to prevent looting. The source of the uncovered artifacts was a shipwreck, later known as Marsala A, sunk beneath 40 m of water near Lido Signorino. Observations suggested that the wreck was a rich collection of materials with surviving hull remains often covered by shifting sand. Storms in the following years kept exposing the hull, but these same environmental effects would later bury the area in 2 m of sand (Purpura 1985, 129-30).

Interest from the Baglio Anselmi Archaeological Museum led to fieldwork to record and recover the shipwreck for display. The museum was already home to an earlier Punic wreck, and officials were keen on expanding their collection to other periods of Marsala's history, with the hope that the museum would become a regional center for archaeological and conservation expertise. Investigations on Marsala A during the 1985 field season also led to the discovery of a second, smaller wreck site nearby, dubbed Marsala B.

Excavation on Marsala A followed a similar format to that used for the nearby Marsala B, but unfortunately the hull remains were not retrieved; the smaller size of Marsala B and its fewer wooden remains made it easier to recover, and funding dried up before the same could be accomplished on Marsala A (Ferroni and Meucci 1996, 286).

Original dating and provenience of the ceramic assemblage found on Marsala A, together with that of Marsala B, suggested that the ships were from the late eleventh to early twelfth centuries, associated with western Sicily and North Africa. Recent reanalysis of the assemblages emphasizes two ships lost at different times and by coincidence in the same area; both were presumed to be built in Sicily, with Marsala A lost around AD 1050 (Tisseyre 2015, 147; Pisciotta and Garnier 2018, 169).

Description of the site

Marsala A lies south of modern Marsala, Sicily, near Lido Signorino, beneath 40 m of water, in the same area as the nearby Marsala B. The site was periodically exposed and reburied by storm activity, before ultimately being covered by 2 m of shifting sand (Purpura 1985, 129-30).

Cargo

Not described in the source text beyond the ceramic assemblage recovered from the wreck, used to date and provenance the site (Tisseyre 2015, 147; Pisciotta and Garnier 2018, 169).

Hull Remains

Marsala A is an axial timber-built ship with a rectangular, rabbeted keel. Not enough evidence survives to know whether the keel was scarfed to the endposts or simply butted; the in-situ side profile for the wreck suggests the latter, although these might also represent breaks along the keel created by site formation processes. The lower surviving ends of the rounded endposts suggest that Marsala A was double-ended, like Serçe Limanı (Ferroni and Meucci 1996, 297-300).

Frames are composed of alternating L-shaped floor timbers overlapped with short futtocks. Limited evidence suggests vertical flat scarfs with iron nails connect the pieces together; otherwise they were simply diagonally butted. Each frame is nailed to the keel, and their dimensions remained relatively square throughout the hull (Ferroni and Meucci 1996, 297-300).

Elements suggesting a rectangular, rebated keelson positioned along the centerline of the hull are based on surviving wood fragments and the presence of bolts positioned between the floor timbers to connect with the keel. A pair of sister keelsons are adjacent to the keelson, along with a mast-step partner to hold the stanchion for the mast partner beam. Fastener holes at the turn of the bilge suggest the existence of a bilge stringer at this location. Few nail holes were noted on the upper surfaces of the floor timbers, which might indicate transverse ceiling resting on the sister keelsons and bilge stringers (Ferroni and Meucci 1996, 301-2).

On the outside face, at the turn of the bilge, are square bilge keels rebated into the framing. Planking has similar dimensions to Serçe Limanı, which appears to be based on the overall dimensions of the ship. Strakes are attached at each frame station with two or three nails. Little evidence survives regarding the wales that would extend around the upper sides of the hull. Nails are square shanked, while iron bolts are round; there is no mention of treenails used anywhere in the hull construction or repairs (Bonino 2007, 185-6). The overall shape of the hull was originally suggested based on Serçe Limanı's reconstruction; reevaluation of the original data suggests the hull had a flat floor with a slight deadrise and a round bilge (Bonino 2007, 184, fig. 4).

Caulking

Sections of the starboard keel rabbet show that planking seams were caulked with vegetable fiber (Ferroni and Meucci 1996, 298-300, fig. 14).

Size and scantlings

Strakes attached at each frame station with two or three nails.

References

Purpura, G., 1985.

Ferroni, A. M., and Meucci, C., 1996. "I due relitti Arabo-Normanni de Marsala"

Tisseyre, P., 2015.

Pisciotta, F., and Garnier, N., 2018.

Bonino, M., 2007. "Appunti sul Relitto Medioevale"