Jump to ContentJump to Main Navigation
The Age of TitansThe Rise and Fall of the Great Hellenistic Navies$

William Murray

Print publication date: 2012

Print ISBN-13: 9780195388640

Published to Oxford Scholarship Online: May 2012

DOI: 10.1093/acprof:oso/9780195388640.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 25 February 2017

(p.251) Appendix A Testimonia for “FOURS”

(p.251) Appendix A Testimonia for “FOURS”

The Age of Titans
Oxford University Press

Note: The references here point to where the class is mentioned in the text and do not refer to the entire historical episode. All dates are BCE unless noted otherwise.

Historical Development

Invention (end of fifth century).

Pliny NH 7.208: The Carthaginians were the first to build “fours.”

Clem. Al. Strom. In a list of various “firsts”: the Carthaginians were the first to build a “four”; Bosporus built it from what was at hand (autoschedion).

See Diod. 14.41.3 and 14.42.2 (at “Fourth Century: Syracuse”).

Fourth Century.

Syracuse (beginning of fourth century).

Diod. 14.41.3 and 14.42.2: In the preparations for war with Carthage (dated to 399), Dionysius I built “fours”1 and “fives”—“fives” never having been built before. He gathered skilled workers from Greece, Italy, and even from the Carthaginian realm.

Diod. 14.44.7: The prototype “five” is sent in Spring 397 to pick up a woman betrothed to Dionysius from Locri.

Tyre, Phoenicia, and Cyprus (prior to 332).

(p.252) Arr. Anab. 2.21.9: There were “fours” in the Tyrian fleet at the siege of Tyre in 332. Curt. 4.3.14 reveals that the Macedonians also had “fours” in their fleet at Tyre, but because they do not figure before this point in any source, the Macedonians had certainly received these ships from their new Cyprian and/or Phoenician allies (e.g., Sidon, Byblos or Arados).

Macedonia, reign of Alexander III (332).

See above, “Tyre, Phoenicia, and Cyprus.”

Athens (332/31–325/4).

IG II2 1628.81–85 (326/5): The “four” (restored) Hegemonia, built in the archonship of Niketes (332/31) was assigned to grain convoy duty: lines 37–42. The class seems certain from a similar formula used at line 10, from other mentions of “fours” in the surrounding text, and from line lengths.

IG II2 1627.275–8 (330/29): 18 “fours” are listed among the Athenian fleet. [Note: “Fives” are not added to the fleet until 328/27 at the earliest, see App. B.]

IG II2 1629.808–10, 812 (325/24): The number of “fours” is listed at 50, seven of which were at sea.

Diod. 18.10.2 (323, upon the death of Alexander): The orators gave shape to the wishes of the people and wrote a decree that the people should assume responsibility for the common freedom of the Greeks and liberate the garrisoned cities and prepare 40 “threes” and 200 “fours,” and that all men up to the age of 40 should be enrolled for this purpose.2

Wars of the Diadochoi.

Diod. 19.62.8: The fleet of Antigonus in 315 numbered 240: of these 90 were “fours,” 10 “fives,” three “nines,” 10 “tens” and 30 aphracts (open galleys).

Diod. 20.49.2 (in a sea battle off Cyprian Salamis in 306): Ptolemy had 140 ships, the largest were “fives” and the smallest were “fours.”

Diod. 20.50.3 (in the same battle): Demetrius had 30 Athenian “fours” with him.

Diod. 20.93.4 (during the siege of Rhodes in 304): A Cilician “four” was captured that had on board purple clothes and other gear that Phila, the wife of Demetrius had prepared for him.

Third to First Centuries.

Fleet of Ptolemy II (282–46).

Athen. Deip. 5.203d (TLG, 5.36): In this major fleet with “big ships” it was thought proper to have twice as many “fours” to triemioliai as the larger units. See above, p. 188 with n.51.

Diod. 3.43.5: “Fours” were used by the Ptolemies to go after pirates in the Gulf of Aqaba.

(p.253) Carthage and Rome during the Punic Wars.

Livy 21.22.4: The Carthaginians used a few “fours” and “fives” against the Romans in Spain. The commander Hasdrubal was given 50 “fives,” two “fours,” and five “threes,” although of these only 32 “fives” and five “threes” were outfitted and manned.

Livy 28.45.21: In 205, P. Cornelius Scipio (the future Africanus) has to fight to be given the province of Sicily/Africa. Upon his selection, he raises a fleet of 30 ships from contributions from allied communities. The keels of 20 “fives” and 10 “fours” were laid and work pressed so that the ships were launched fully equipped and rigged just 45 days after the timber was felled. The 30 ships carried some 7000 volunteers.

Rhodian navy: a preferred ship class.

For examples, see Polyb. 16.31.3; 27.7.14; 33.13.2; Livy 31.16.7, 37.16; and 37.23–24.

Roman Civil Wars: Julius Caesar (48).

B. Alex. 13.4: The Egyptians pressed into service some old ships, producing 22 “fours” and five “fives” to which they added a considerable number of smaller, open craft. Caesar had 10 “fives” and “fours” and a number of smaller open ships.

Caes. BC 3.111.3: When Caesar held the harbor area of Alexandria, a fight developed over 50 warships that had been sent to Pompey and had now returned from Thessaly, all of them “fours” and “fives”; there were 22 besides, all decked ships (constratae omnes), normally assigned to guard the port at Alexandria.

Roman Civil Wars: Octavian and Agrippa (36).

App. BC 5.11.107: Near Mylae, the Pompeian commander Papias (probably in a “four”) suffers a fatal ram strike from Agrippa, in a “five” or “six.” See below at “Physical Characteristics. Oarsystem.”

Roman Civil Wars: Marcus Antonius (31, 29?).

Dio 50.23.2: “Fours” among classes in Antony’s fleet at Actium.

Dio 51.1.2: Octavian dedicated on Cape Actium a “three,” a “four,” and one of every class up to “ten” from captured enemy vessels.

Physical Characteristics

Note: These testimonia apply only to the period and region from which they derive. Both regional and chronological variations in design should be expected.

Size and Appearance.

Classed among cataphract galleys.

Polyb. 5.62.3: Antiochus III took possession of Tyre and Ptolemais in 219 and received 40 ships, 20 of them cataphracts, “of these none smaller than ‘fours,’ and the rest were ‘threes,’ dikrota and keletes (two-level and single-level open craft).3

(p.254) Livy 37.23.5: In the Syrian fleet of Antiochus III (190), there were 37 ships “of larger build” and, aside from these, 10 “threes,” implying that “fours” were among the ships “of larger build.”

Cic. Ver. 2.5.86–91 (delivered in 70): A “four” is decked (constrata), fast when supplied with a full crew, and looks “like a city” amongst a bunch of pirate craft (myoparones).

“Fours” are the primary ship of the line for the Rhodians in the battles off Side and Myonessus in 190.

Livy 37.23.4–5: At Side, the Rhodian fleet had 32 “fours” and four “threes.” The fleet at Myonessus (22 vessels; Livy 37.30.2) must have had roughly the same composition.

“Fours” were fitted with turrets or deck towers for fighting; the turrets were also used as lookout positions.

Livy 37.24.6: During a sea battle off Side (190), the admiral Eudamus watched the enemy from a turret (e turri praetoriae navis) on his ship.

“Fours” (both Ptolemaic and Roman) are two level (bireme) galleys that can accommodate deck towers at bow and stern. See “Oarsystem” (below).

Lightly ballasted.

“Fours” were lightly ballasted (as were “fives”) and floated when “sunk.”

Diod. 20.52.6: In a sea battle off Cyprian Salamis (306), Ptolemy’s “destroyed” ships (80 in number) were towed, full of sea water, to the camp before the city. The same might be said of Demetrius’s destroyed ships (20 in number) which were towed to shore, repaired, and used again. He had “fours” and “fives” in his fleet (Diod. 20.49.2).


“Fours” are slightly faster than “fives” and lower in the water (Third Century?, First Century).4

Polyb. 1.47.5: At the siege of Lilybaeum (250), the Romans tried to block up the harbor, but were unable to do so. There was a shallow area that they managed to place a mole on, and one night, an enemy “four” of fine build grounded on it. The Romans used this vessel to catch a blockade runner called “the Rhodian” who used a “five.” He was unable to outrun the “four” and had to put about and fight. The Romans had more deck soldiers, captured the “five,” and then used both ships to intercept other blockade runners.

Livy 30.25.1–8: Three Roman envoys were sent by P. Cornelius Scipio to Carthage in 204 to deliver a protest, accomplished nothing, and were sent away. Fearing for their safe return, the Romans, on a “five,” were escorted by two Carthaginian “threes” to the (p.255) Bagradas River, from which the Roman camp could be seen. As soon as the escort broke off and returned to Carthage, three “fours” from the Carthaginian fleet lying off Utica (30.25.5) attacked the Roman “five” from the seaward side. The attacking vessels failed to ram the Roman “five” because of her speedy avoiding tactics, and her higher freeboard prevented her from being boarded, so long as the supply of missiles held out. When these ran out, however, the “fours” drove the “five” ashore. The envoys escaped harm. [Note: Although the “fours” were not fast enough, relative to the “five,” to damage her steering oars or stern by ramming (as they sailed on the same course), they were able to harass her as she continued on her way and this implies superior speed.]

Polyb. 15.2.12: Describing the same incident as the previous text, Polybius calls the hostile Punic ships “threes” and does not say anything about the relative heights of the vessels.

App. Pun. 34: (TLG, Lib. 143–146): Describing the same incident as the previous two texts, Appian calls the escort vessels “threes” but does not give any details about the envoys’ ship or the ones who pursued and overtook it.

Livy 37.29.9: When the fleets lined up for battle off Myonessus (190), Rhodian “fours” are described as quickest in the fleet. They are more agile and their pilots and rowers more skillful than the Romans whose ships were stronger and manned by more courageous soldiers (37.30.2). While Livy does not explicitly describe the Rhodian ships as “fours,” most were surely from this class; see above, “Size and Appearance. ‘Fours’ are the primary ship of the line for the Rhodians in the battles off Side and Myonessus in 190.”


“Fours” (both Ptolemaic and Roman) are two level (bireme) galleys that can accommodate deck towers at bow and stern.

B. Alex. 13.1, 4; 16.6: During the winter of 48/47, while Caesar was in Alexandria, the Egyptians pressed old hulls into service to fight him: 22 “fours,” five “fives,” and a number of small open ships (on which we would not expect a difference between rowers and combat crews). From this force, Caesar captured in battle one “five” and a bireme with their combat crews and rowers. Conclusion: Ptolemaic “fives” were not biremes, and “fours” were biremes.5

App. BC 5.11.107: Near Mylae in 36, the Pompeian commander Papias (probably in a “four”)6 suffers a fatal ram strike from Agrippa, in a “five” or “six.”7 Agrippa (p.256) struck Papias’s ship beneath the epotis, or cathead, broke into the hull and the collision ejected the men in the towers into the sea (there were towers at bow and stern; see App. BC 5.11.106), which began to flood the ship. The thalamian (i.e., lower) oarsmen were cut off, and the other oarsmen (= i.e., the other of two groupings, heteroi not alloi) broke through the canopy deck and swam away.8

A “four” has double-manned oars.9

IG II2 1629.695 (325/4): The curators receive 415 drachmai for a set of oars from a “four,” characterized as “tarrou argou” (unfinished oar set).10 In 411 (during the Peloponnesian War) a rough hewn spar for a trireme oar (kopeus) was apparently worth 5 drachmai.11 Although we must use prices that are separated by almost nine decades for two different commodities (oar spars for “threes” and for “fours”), we can still get a general idea of the relative numbers involved. The money received for the unworked oars of a “four” would purchase roughly 83 units if they cost 5 drachmai a piece. Even if we are off by a variance of 25% to account for the imprecise nature of our evidence, our calculations still indicate a relatively low number of oars for a “four” (roughly 40 to 50 per side) when compared to a “three,” whose tarros numbered 170 (roughly 85 per side), and this implies that the oars were double manned.12

Ship’s Gear.

The following gear is assigned to each “four” at Athens. It is roughly the same as that assigned to a “three.”

IG II2 1627.138–91, 457–72 (330/29):

Wooden gear: set of oars, steering oars, boarding ladders, boat poles, mast, yard arms.

Hanging gear: undergirds, sail, white screen (pararrhymata leuka), hair screen (pararrhymata trichinia), katablema (side curtain), lines [4 sets each of 8 daktyl and 6 (p.257) daktyl lines = 470–72; coils of lines = merymata kaloidion = lines 149–50 in the inscription], leather straps, double halyards, sheets, braces (hyperai), strap (chalinos).

Anchors: 2 iron anchors.

IG II2 1629.1050–85 (325/24): Standard issue for a “three” includes side screens called hypoblemata and sometimes a light sail (lines 581–82 in the inscription).

In Athens, the total cost for a complete set of gear on a “four” is roughly 50% more than a full set for a “three.”13

IG II2 1631.446–48, 517 ff. (323/2): The total cost of ship’s gear (wooden and hanging) on a “three” is roughly 4100 drachmai on average.

IG II2 1629.636–56 (324/3): The total cost of ship’s gear (wooden and hanging) on a “four” is 6105.5 drachmai on the Aktis built by Epigenes; and 6000 drachmai on the Homonoia built by Archineus.

Ramming Characteristics.

“Fours” are used in the front line of battle and routinely participate in prow-to-prow attacks.

Diod. 20.50.3: For a sea battle off Cyprian Salamis in 306, Demetrius places in his left wing’s front line seven Phoenician “sevens” with 30 Athenian “fours”; these are supported by 10 “sixes” and 10 “fives” in a second line behind them. Ptolemy’s fleet was made up of 140 “fives” and “fours” (Diod. 20.49.2).

A Rhodian “four” unexpectedly sank a “seven” with a single blow (uno ictu) in a battle off Side in 190.

Livy (37.24.3): The fate of the “seven” was unexpected and greatly frightened the enemy.

“Fours” are expected to defeat “threes” which can be used to lure “fours” into a trap.

Caes. BC 3.24.1–3: Antonius, at Brundisium in 48, laid a trap with a number of small vessels inside the harbor and then sent out two “threes” to the mouth of the harbor; Libo saw them come out recklessly and sent five “fours” to intercept them. One “four” was captured by Antony’s small boats.

A “four” is bested by a “five” in a prow-to-prow encounter.

B. Alex. 46.1.4: During the war between Caesar and Pompey in 48, a sea battle was fought off Tauris Island (N. Illyricum = Barrington Atlas, Map 20, D-6) in which a “four” is destroyed by a “five” in a prow-to-prow encounter (ut navis Octaviana rostro discusso ligno contineretur = “so that the ship of Octavius is held fast by its ram, the wood having been shattered”). The “four” eventually sinks to the waterline.

“Fours” (and “fives”) can be defeated by smaller vessels through entanglement and subsequent boarding.

Polyb. 2.10.5: During an Illyrian naval attack on Greece in 230, the Illyrians lashed four of their smaller craft together and let them be rammed by their enemies (p.258) the Achaeans. Once the attackers became entangled at the bows with the smaller vessels, marines swarmed onto the Achaean ships and captured them. In this way, four “fours” were captured and a “five” was destroyed along with her crew.

The Rhodian “fours” used bow-mounted fire pots to punish larger vessels that attacked their bows.

Livy 37.30.3–5 (Battle off Side in 190): In order to avoid the fire, the larger ships of Antiochus avoided prow-to-prow encounters with the Rhodians.14 This did not prevent Rhodian prow-to-prow attacks, as revealed by Livy 37.30.9–10: a Rhodian ship (no doubt a “four”) was captured when its anchor fouled a Sidonian prow following a prow strike.

Additional Characteristics of Usage.

Used to convoy grain shipment to Athens (326/5).

IG II2 1628.37–42 (326/5).

Used as support ships for “sixes” and “royal ships.”

Polyb. 16.6.2 (Battle off Chios in 201): Attalus hastened in his royal ship with two “fours” to come to the aid of a stricken “five.” When the “royal ship” was eventually captured, the two “fours” were captured as well (Polyb. 16.7.2).

“Fours” used for reconnaissance.

Pliny Ep. 6.16.9: “Fours” were used by Pliny the Elder, as Prefect of the Fleet at Misenum, to investigate effects of the Vesuvius eruption in 79 CE with the intention of assisting seaside towns along the Bay of Naples.

Less Certain or Doubtful Characteristics.

“Four” similar in design to a “five”?

Polyb. 1.59.7–8: During the First Punic War (243–42), there being no state funds for the navy, wealthy citizens at Rome undertook to underwrite the expenses of a new fleet of “fives” and in this way built a fleet of 200 “fives.” These “fives” were all built on the superior model of Hannibal “the Rhodian’s” ship that had been captured at Lilybaeum in 249 (Polyb. 1.47.5–9). Despite Morrison’s argument (Morrison 1980, 39) that “the Rhodian’s ship” was a “four,” Polybius’s account makes much more sense if it was a “five.” As he tells it, the Romans first captured a “four” that was slightly faster than a “five,” and then used this ship with a select crew of deck soldiers to capture Hannibal’s vessel. That the latter ship was a “five” is made all but certain by its use as the model on which the Roman fleet of “fives” was thereafter based.

Single-level “four” with four men to an oar?

(p.259) Casson 1995, 105: Once multiple man sweeps were introduced, “there is no reason for their not designing a ‘five’ which simply put five men on each oar—even a ‘four’ with four men on each. We have almost certain evidence that ‘fours’ and ‘fives’ were oared in this fashion from 100 B.C. on….” His view is based on an interpretation of images on coins,15 and does not correspond to the surviving testimonia, which show that “fours” were two-level galleys and were lower in the water than “fives.” (p.260)


(1) . The mss. read “threes” instead of “fours” in both passages, but this conflicts with Diod. 2.5.7; as a result, P. Wesseling (Diodorus 1746) long ago substituted “tetrereis” for “te triereis” in both passages. Morrison 1990 argues, unconvincingly to my mind, that we should keep the original text.

(2) . Morrison (in Gardiner and Morrison 1995, 67) accepts this somewhat surprising number. The editor Wesseling (see n. 1) suggests that the text originally read 200 “threes” and 40 “fours.” Justin 13.5.8 estimates the force as 200 ships in all.

(3) . Although we possess evidence for cataphract “threes” and even smaller vessels in most Hellenistic fleets, “fours” were at the lower end of those classed as cataphracts; see Morrison and Coates 1996, 113 and 257.

(4) . See Morrison 1980, 40: “… whichever account we prefer [i.e., that of Livy, Polybius or Appian], the important thing is what Livy in the first century knew about the relative heights of ‘fours’ and ‘fives’.” The same holds true for the hull speeds of “fours” and “fives.”

(5) . This is the only occasion on which this author uses the term biremis, so we can be certain that he is not using it as a synonym for aperta navis (open or small ship) which he uses three times (B. Alex. 11.2, 13.4, 13.6).

(6) . Papias, a Pompeian commander, would have been on one of the larger units in the fleet, which Appian stresses was composed of ships that were lighter and smaller than those of Octavian. Although Appian does not record its class, Papias’s ship held towers, so it was likely no smaller than a “four.”

(7) . Were the two men on flagships opposite one another?

(8) . See Morrison and Coates 1996, 269.

(9) . Morrison (in Gardiner and Morrison 1995, 71) lists a “four’s” oar numbers at 70 (35 per side), on what evidence, I am not sure. In Morrison and Coates 1996, 269, he concludes that a tarros or full set equals roughly 88 oars. Since he works from the price for a finished set of oars for a “four” (IG II2 1629.685–86 = 665 drachmai) his calculations are less exact than those I have adopted.

(10) . Theophrastus (Lap. 27) uses the adjective argos to describe unpolished gemstones, and a later technical writer named Athenaeus Mechanicus (Athen. Mech. 12) uses the term to mean “unworked,” “raw” or “unfinished” when describing hide coverings for siege machinery; see Whitehead 2004, 98.

(11) . Andocides (2.11) claims that when he provided oars for the fleet at Samos (in 411), he could have sold the rough oar spars for five drachmai a piece.

(12) . For the number 170, see Morrison and Williams 1968, 256 with 272n16. The tarros of a “three” did not include the 30 spare oars commonly assigned to each “three” (see IG II2 1614.30), so we can leave them out of our calculations.

(13) . Gabrielsen 1994, 152–53.

(14) . Cf. Livy’s description of a naval encounter at Panormus in Samos (37.11) when fire pots were first used. Polyb. 21.7.1–4 describes them, as does App. Syr. 24. See Walbank 1999, Vol. III, 97–99. This device was not in regular use in later times—why, we do not know.

(15) . Casson 1995, Appendix to Ch. 7: “Coin Evidence for Single-Banked ‘Fours,’ ‘Fives,’ and Larger Units,” pp. 155–56.