158 ruthy gertwagen
small square spritsail forward of the prow. This is the artemon, whose main function
was to help steer the ship—more suited to sailing before the wind than to windward.
Such was the standard rigging of the imperial oneraria from Spain to Asia that, how-
ever, was also able to sail with bowlines abeam. In other words, it was the typical all-
purpose ship and therefore achieved great success. Another, almost vertical type of
foremast was situated much closer to the main mast and was equal to it in height and
diameter. Under the Empire, this type was carried on bigger ships. These features
point to the sailing to windward that was common according to contemporary texts.
Such ships also sailed from Italy to Egypt in winter. By the second century ce bigger
vessels, like the Alexandrian grain ships, also carried a small sail at the stern for maneu-
verability and steering, and to preserve balance, and had a top-flattened triangular sail
on the main mast for open-water sailing (Arnaud, 2011; Wilson, 2011: 44–45).
Another innovation was the use of certain cargoes as ballast. A typical characteristic
for sailing abeam was heeling. For such sailing, even under stronger winds, it was
necessary to lower the ship’s center of gravity with heavy ballast. The location of a
wreck of the late first century ce on the Skerki bank between Tunis and Sicily points
to the conscious choice of upwind routes. Direct, less-segmented routes had sped up
sailing times, despite the risk of strong adverse winds. These were taken by private
shipowners motivated by pure economic calculation. Such direct routes led from
Egypt through North Africa to Rome, instead of the longer one, taken by the grain
ships, along the Levantine coast and through the Aegean. These routes, in fact, indi-
cate a better knowledge of the Mediterranean as a whole based on “mental charts”
since seamanship was conducted by experienced practitioners. In the Greek/
Hellenistic/Roman world, navigation remained an empirical art (Arnaud, 2011;
Ballard et al., 2000). Because of geopolitical constraints, both routes were in use 900
years later by Muslim/Jewish shipping from Egypt, as evidenced by Geniza letters.
The hostile assaults of the Byzantine fleet prevented them from following the longer
route along the Levantine coast through the Aegean to Italy. However, navigational
skills could not save these ships, the technology of which made them apt only as
coasters that could sail very shallow waters, including up river, as their wrecks show
(Gertwagen, 1996).
The third innovation attesting to upwind as well as year-round seafaring was the
bilge pump that collected sea water that seeped through the seams to the bilge, the
lowest point in the hull. Bilge pumps were only found in big ships over 150 tons and
certainly in the grain ships that could ride out storms. By 100 bce the chain pump was
introduced; it was almost identical to the pumps of eighteenth-century ships. From
the seventh century ce, bilge pumps disappeared from the medieval Mediterranean as
ships shrank in size; they reappeared in the mid-fifteenth century. They were, how-
ever, vital equipment for large wooden eighteenth- and nineteenth-century sailing
ships (Wilson, 2011: 222–223).
The evidence of late-antique wrecks shows a transition in hull conception and
construction already around the second century. Wrecks in both basins of the
Mediterranean manifest a shell concept for the hull shape; that is, strakes were installed
before frames, giving the hull its shape and integrity, but with a mixed shell-frame
construction method, evidenced by the small and widely-spaced mortise-and-tenon
joints, loose tenons, and their absence in some planking seams. In other words, mortise-
and-tenon joints lost their role as the main element in a strake-orientated hull and