Bryn Mawr Classical Review 03.01.04


Lionel Casson, The Ancient Mariners. Princeton: Princeton University Press, 1991. Second edition. Pp. 299. $39.95 (hb). $12.95 (pb). ISBN 0-691-06836-4 (hb). ISBN 0-691-01477-9 (pb).


Reviewed by Ford Weiskittel, Trireme Trust USA.

Readers who know the first edition of this book will be very pleased at the appearance of this updated second edition. Revisions are extensive, to take account of all the new evidence which has come to light in the thirty years since the first edition (1959). It is written in a pleasant colloquial style (Herodotus "the Father of the Travelogue" p. 63) and full of delightful description. There are no footnotes; references and sources for further information are collected together at the end of the book and referenced by chapter and page numbers. Many of these references refer to the second edition of Ships and Seamanship in the Ancient World (Princeton, 1986, but already out of print), also by Casson, and the two books together form a comprehensive study of ancient ships, seamanship, and maritime trade. The glossary of nautical terms, however, which was such a valuable feature of the first edition of The Ancient Mariners, has been omitted.

The Ancient Mariners concentrates on the development and expansion of trade, military control of the seas, and the men who made it all possible. It ranges from the third millennium B.C. in Egypt through the end of the Roman Empire. The author is personally knowledgeable about seamanship and displays an impressive command of his subject. He weaves it all together into a flowing narrative that shows how each phase in maritime development follows from the last. It is difficult to do justice to the wide range of material included in the book, from the imaginative reconstructions of voyages of exploration to the nitty-gritty of the Roman grain trade, but I think the descriptions and explanations of how maritime commerce was conducted are its most valuable parts.

I remain unconvinced by Casson's suggestion that the two-level warship manned on both levels was derived from a earlier version that positioned oarsmen on the upper level while cruising and on the bottom level during battle (p. 77). I think it very unlikely that a warship would have provided such alternative accommodation for the rowers without some very good reason. Such a ship with rowers and marines both on the top-level would have been top-heavy and less seaworthy. Casson suggests the top-level rowers would have been dangerously exposed, and the bottom-level rowers in danger of suffocation. If true, these would have been problems also for a true bireme, but of course they were not. Certainly the bottom level rowers were in no danger of suffocation because the fighting platform was not a true deck but only a central gangway, as Casson correctly describes it (p. 76), and did not close off the circulation of air. I think this is an overly elaborate explanation for a vase painting of a ship that looks like all the other two-level biremes depicted on vases of similar date, except that in this case only the top rowing level is manned (pl. 15). Our experiences rowing the reconstructed trireme Olympias show that rough seas caused the bottom level of rowers to be ineffective, with the result that they would pull in their oars and let the other levels do the work. This would have been true for a bireme as well. Movable panels would not be needed in choppy water because the problem was not in shipping water over the side but in the bottom-level rowers not being able to clear their oars from the waves during the recovery part of the stroke. Given the difficulty of clearly showing both levels rowing simultaneously, I think a preferable explanation is that the artist chose to depict this two-level bireme under upper oarpower alone.

I am also sceptical about the idea that quadriremes, quinqueremes, and the larger ships came in a variety of styles, some one-banked, some two-banked, and others three-banked (p. 131). The Trireme Trust is currently engaged in trying to find some answers to how these larger ships were rowed, but the advantages to be gained by keeping hull width to a minimum would make a one-banked quinquereme, for example, far inferior to a two-banked version, and the admiral who went into battle with the one type against the other would find his fleet at a great disadvantage. The depictions of one-banked warships on Roman coins, the principal evidence, are much too sketchy to support the idea of one-banked quadriremes and quinqueremes.

It still remains to be proven, but I think we will also find that no ship larger than a trireme was three-banked. The three-level oar system of a trireme is only possible by overlapping the rowers vertically, so that the heads of the thalamians are level with the thighs of the zygians, and the heads of the zygians are level with the chests of the thranites. This arrangement is necessary to minimize the angle at which the top oars are worked and to keep the center of gravity of the ship as low as possi ble. For such a system to work with multiply-manned sweeps, however, the rowers on each level would have to be set completely outboard of the next lower level. A quinquereme with two thranites, two zygians, and one thalamian would have to be much wider than a trireme, and such evidence as there is from the ship sheds suggests that quinqueremes were not wider than triremes. Therefore, the only possible solution would be a two-level quinquereme with three men on the top level and two on the bottom, with the rowers on the top level sitting completely above the rowers on the bottom.

Any larger three-level ship would also have to place each level completely above the next level below. Otherwise, the large inboard portion of the oar would make effective gearing impossible. Our experiments with oar angles suggest that three levels of rowers tacked vertically in this way would produce an oar system with the top-level oar at too great an angle to the water to be worked effectively. There is also no convincing evidence that any of these larger ships from quadriremes on up had three banks. It is not true that Ptolemy's "forty" had three banks (p. 131). All we know is that the top rowers were called "thranites"; it is not necessary to conclude from this that these were the top rowers in three banks. If Ptolemy's "forty" had two banks in twin catamaran hulls, it would have needed ten men per oar. Eight-man sweeps were the largest practical in 17th century galleys, but Ptolemy's showpiece was not meant to be practical. The oar length recorded would just accommodate ten men, and two banks of ten men per oar per side in twin hulls would explain the "forty".

Some other minor objections. A warship captain may have left his mast and sail ashore to clear the decks (p. 78), but it would have also cut down on wind resistance and reduced weight, two considerations which weighed heavily with Olympias' crew when undertaking sprinting trials. The zygians of a trireme rowed through oarports and not over the gunwale (p. 85). It would be unlikely that the thranite stroke in a trireme was the most wearing (p. 85 and 87), even if that is true with Olympias. A well-designed oar system would equalize the work among the three layers by fine-tuning the gearing. Accordingly, the extra pay for thranites attested on one occasion was not because they had the hardest stroke. Two more likely explanations might be either that the thranites were in the most exposed position and so got danger pay, or to reward them for their function of providing feedback to the rowers below, since the thranites were the only rowers who could see the oars in the water. Lastly, a trireme commander did not seek to ram at an intermediate speed (p. 90) but at one which was only slightly more than that of his target. In this way, he could spring the timbers of his opponent without risk of having his ram embedded too firmly in the other ship's side.

These are only minor points, however, in an excellent and comprehensive book. The summary judgement of Professor Mabel Lang in her review of the first edition (Phoenix, 1960) is still the most appropriate: "Professor Casson has succeeded in writing a book which can be read with pleasure by those who know nothing of the ancient world and used with profit by classical students innocent of naval knowledge."