Have We Found a Quinquereme? And will it Help Us with the Enormous Fleet and Crew Numbers in Polybius?
For several centuries, from the Hellenistic period until the end of the Roman Republic, the standard battleship of the mediterranean states was the quinquereme— or as it is sometimes called after its Greek name, the pentere, a large galley used to engage other ships with a ram or with a sizable contingent of soldiers ready to board. The ship takes its name in both the Latin and Greek languages from their words for “five,” which indicates that, in some way, squads of five men worked the oars. Expressing what has become a common view (though I disagree with it), some writers assert that quinqueremes were, in effect, enlarged triremes. The trireme had been the standard warship of the classical Greek times and was pretty clearly propelled by oarsmen in clusters of three separated, but not greatly, by their level and by their lateral position in the hull. Each man pulled his own oar. Although the precise arrangement was subject to argument for years, the general arrangement now seems beyond dispute.
Galleys from a fresco found in Pompeii.
In this view the quinquereme, which appeared in the 300s BC, was merely an expanded trireme with two men on each of the upper oars and one man on each of the lower. I disagree for reasons set out in detail in an earlier post, and I think it’s fair to say that while it is a minority view, it isn’t without authoritative supporters, for example Lucien Basch, W. W. Tarn, Lionel Casson and Chester Starr. I agree with them and believe that quinqueremes were rowed by teams of five men drawing large oars called sweeps. (If interested in this question, see my earlier blog entry on this topic.) I reiterate my position on this issue because it’s important to what follows. So, I ask the reader to be open to the idea of the quinquereme as a single-level galley rowed by gangs of five men to each sweep, a system mechanically simpler than that of the trireme and requiring fewer trained oarsmen.
But the structure is not what concerns me here; instead it is the distinct possibility that the remains of a pair of quinqueremes are resting unrecognized as such in a museum in Sicily. In 1969 a fisherman uncovered part of the wreck of an ancient warship off the coast of Marsala, the site of the last great naval battle between Rome and Carthage, the Battle of the Aegates Islands (241 BC). Shortly afterward, the wreck of another similar ship was found nearby. Carbon dating is consistent with their loss at the time of the battle.
After the initial wreck was discovered, it was excavated by marine archaeologist Honor Frost. The remains of the second ship allowed her to make a reconstruction of the first, and this reconstruction, generally known as the “Punic Ship,” is on display at the Archaeological Museum Baglio Anselmi in Sicily. I think it is reasonable to suppose that the Punic Ship is composed of the remains of quinqueremes lost in the battle and, if so, the reconstruction may help solve the problem of ship and crew numbers in Polybius’s account of the First Punic War (264 to 241 BC). What is that problem? Unbelievably large numbers of ships and men.
Polybius (ca. 200 to ca. 118 BC), our primary source for information about the First Punic War, was a Greek born more than a generation after its conclusion. He gives us our principal account of the battle Battle of the Aegates Islands and reports that something near 450 quinqueremes were involved, some 80 of which were sunk. This is merely one example of the general problem of huge fleet numbers: the number of ships involved in this one battle is immense if, as Polybius tells us, 300 men had rowed each quinquereme (I.26). Note that Roman and Carthaginian quinqueremes were much the same: the Romans had built theirs on the model of captured Carthaginian vessels, so finding one would throw light on the other, and we should expect crew numbers to be about the same. If we take Polybius’s account at face value, we must accept as many as 135,000 men were rowing the ships in this battle, a huge figure for the time, and one that doesn’t even take into account the numbers serving at the same time in the Roman and Carthaginian armies.
In his Histories, Polybius describes several other fleet actions in which hundreds of quinqueremes are said to have been involved. That the numbers were inflated or mis-reported was recognized over a hundred years ago by W. W. Tarn, who suggested that the usual Carthaginian or Roman fleet would have numbered 100 or 120 such vessels at any given time. I leave it to the reader to follow Tarn’s interesting argument in “Fleets of the First Punic War,” Journal of Hellenic Studies, Vol 27 (1907). I suggest, however, that even a fleet of 100 vessels would have been a strain for the belligerents to launch if the crews had been as large as supposed.
Consider further that Polybius states, “Moreover the Romans lost in this war [the First Punic War] about seven hundred quinqueremes, inclusive of those that perished in the shipwrecks, and the Carthaginians five hundred.” (Histories, I.63) If the crews were as large as claimed, this stunning loss would have amounted to something on the order of 360,000 oarsmen, a fantastic number even for a war that lasted, as this one did, for 24 years. Solutions to the problem of unbelievable fleet and crew sizes are immediately obvious: the fleets were smaller than reported, the ships were smaller than reported, or both. This is where the Punic Ship comes in.
Frost recognized the wreck as that of a warship, the first to have been discovered from ancient times. The hull was about seven times as long as it was wide, and there had been a ram at the bow; it was clearly a war galley. As reconstructed by Frost, the vessel had been about 110 feet long, had a beam of 16 feet, a displacement of 120 tons and was fitted with 34 oars, 17 to a side. She wrote of it:
“… the Punic ship itself was neither a trireme nor a quinquereme (the battleships of the navies) but a Libernian— a smaller kind of fast ship with seventeen oars on each side, each pulled by two rowers (Frost, 67).”
But is this a correct interpretation? Frost herself noted that the oars set were rather far apart, “… perhaps so that two men could control each oar… (Frost 63).” But were they set far apart so that even more than two men could row?
Let’s take a side jaunt. In 1937 Vice Admiral William Ledyard Rodgers published the book Greek and Roman Naval Warfare; a Study of Strategy, Tactics, and Ship Design from Salamis (480 B. C.) to Actium (31 B. C.). In this book Admiral Rodgers proposed a design for a quinquereme. What is interesting about his proposed design is its remarkable similarity to that of the Punic Ship: its length, beam, tonnage and number of oars are almost the same. It may be an accident, but may be not. Let’s look closer.
Admiral Rodgers, a Naval Academy graduate, and hence an engineer, commanded a number of vessels over the course of a long career, so his proposed designs for a Roman quinquereme and his discussions of various technical aspects of oar-powered vessels are quite interesting and detailed, the product of both theory and experience.
Image adapted from Rodgers, Greek and Roman Naval Warfare; a Study of Strategy, Tactics, and Ship Design from Salamis (480 B. C.) to Actium (31 B. C.), United States Naval Institute, Annapolis: 1937, page 307.
Rodgers’s proposed Roman quinquereme has these characteristics: length at the waterline of 100 ft, beam at the waterline of 18 ft, displacement 116 tons, oars 30 (fifteen to a side). It is striking that the differences between the Punic Ship and Rogers’s proposed quinquereme are minor, apart from the number of oarsmen. Where Frost’s proposed reconstruction has 34 two-man oars for a total of 68 oarsmen, Rogers’s design has 30 five-man sweeps for a total of 150 oarsmen. In other words, the hulls are remarkably similar in dimensions, but the number of oarsmen differ significantly. But, if we accept Rogers’s plan, a hull of the size of the Punic Ship should be able to accommodate the additional oarsmen. The sweeps would have been on some sort of outrigger above the waterline, which outrigger is apparently not among the timbers of the half-hull or so that survives of the Punic Ship. One of Rogers’s illustrations of his design show this arrangement.
Ms. Frost refers to the Punic Ship as “a fast ship,” and perhaps she is right, but can she be right if the rowing arrangement is as she suggests? As Admiral Rodgers states, the speed of a galley depends largely on how many rowers it has per foot along the hull. Therefore, increasing the number of rowers per foot increases the power available to drive the ship. Increasing the number of rowers comes with a cost, of course. While a trireme can employ individual rowers pulling single oars, which is the most efficient plan, there is a limit to the number of levels (three) and the length of the ship.
The obvious solution to the problem of levels is to add more oarsmen to each oar instead of increasing the number of levels, but each additional man, because he must work closer to the thole, contributes less power. Rodgers’s figures state that while an individual oarsman working on an individual oar might be expected to produced 0.14 horsepower for 20 minutes, five men on a large sweep could be expected to produce 0.45 horsepower during that time (Rodgers 208).
Let’s compare the horsepower of the Punic Ship with two oarsmen to an oar and with five to an oar. According to Rodgers, a crew using 34 two-man oars should produce 8.26 horsepower for 20 minutes. Rodgers’s proposed quinquereme of about the same dimensions as the Punic Ship with 150 men on 30 oars should produce 13.5 horsepower for 20 minutes. With the 34 oars posited for the Punic Ship, the horsepower rises to 15.3. The difference is striking. Rodgers writes of his design: “With a wetted surface of 1,695 ft. the full speed under oars would be about 6.6 knots and 4 knots for 2 hours (Rodgers, 306).” Could the Punic Ship, which was of similar dimensions, hope to approach this performance with only about half as much power? Could it be considered a “fast ship”? See the table to compare the features of the Punic Ship reconstruction with Rodgers’s proposed design.
I don’t say there are no difficulties with my suggestion. For one, the discoverer of the wreck herself did not construe it in this way, perhaps because of Polybius’s statement that quinqueremes were rowed by three hundred oarsmen. But is this figure accurate? At least, in the case of the fleets of the First Punic War?
Let’s circle back to the problem mentioned at the start of this article: the impossibility of the fleet sizes reported for the First Punic War both in numbers of ships and of men. For example, Polybius reports that the Romans lost 93 ships alone at the Battle of Drepana. If the greater part of these, say four-fifths, were quinqueremes— the standard “battleship” of the period— the losses in man power would have exceeded 22,000 men in a single battle. That number exceeds that of a consular army. And there were other huge losses in battle and from storms over the course of the war.
The easiest solution to this problem is to recognize that the fleet sizes in Polybius are exaggerated or mis-reported, though how much is difficult to say, but also that the quinqueremes themselves were smaller. The logistical situation alone would suggest this, as Admiral Rogers knew decades ago. Now, however, we have some additional evidence: the “Punic Ship.” If, as seems to me probable, it is reconstructed from the remains of a pair of single-level quinqueremes with five men to an oar, then even if Polybius’s numbers are taken at face value, the losses in men to the combatants are nearly halved, and the costs of the vessels much reduced. Combine smaller fleets with smaller ships and the losses of the First Punic War become credible.
I might remark in closing that Polybius notes that in preparation for the Battle of Ecnomus (256 BC), a contest which the Romans determined to win by boarding, they embarked 120 soldiers on each ship. It is interesting to consider that at 34 oars to a ship (to take this number from Ms. Frost), each manned by 5 oarsmen, we reach a figure of 170 men. Add the 120 soldiers, and the rest of the compliment of sailors and so on, and the crew of a “Rodgers-style” quinquereme comes out very close to 300. It may be that the number or oarsmen reported by Polybius is a mistake for compliment of a smaller quinquereme carrying a large number of boarders.
If you have enjoyed this article, consider buying one of my books, perhaps my latest, the historical novel Fortuna at the Rudder available in paperback, hardback and on Kindle on Amazon.
Sources
Frost, H., How Carthage Lost the Sea, Natural History, December 1987, 58-67.
Polybius, Histories, Book I, (Loeb Edition)
Rodgers, W. L., Greek and Roman Naval Warfare; a Study of Strategy, Tactics, and Ship Design from Salamis (480 B. C.) to Actium (31 B. C.), United States Naval Institute, Annapolis: 1937
Saudi Aramco https://archive.aramcoworld.com/issue/198606/the.punic.warship.htm
Starr, Chester G., “The Ancient Warship,” Classical Philology, Vol 35, (1940)
Tarn, W. W.: “Fleets of the First Punic War,” Journal of Hellenic Studies, Vol 27 (1907)
