CHAPTER 18

WAR AT SEA

PHILIP DE SOUZA

IN most cases the fleets of classical antiquity were not intended to engage in naval warfare in the modern sense of ship-to-ship combat. Rather they were designed primarily for use in what modern military specialists would term “amphibious strike operations,” involving the rapid movement of troops by sea to assault specific coastal targets, or to spearhead invasions of hostile territory. Nowadays such operations are typically conducted using aircraft as well as ships and ground forces, but the basic analogy remains valid. Clarifying the distinction between naval warfare and the type of amphibious strike operations that were routine in the classical world is important because it enables us to understand how maritime operations fit into the broader picture of ancient warfare. Once the distinction is systematically applied to the ancient evidence it becomes possible to appreciate that almost all the so-called “naval” personnel involved in the sailing and rowing of ancient warships can be categorized as soldiers, not sailors. Ancient warship crews were not, primarily, oarsmen, but infantrymen who propelled their own transportation vessels. Many of them would have been deployed as what are normally termed “light troops” (Greek psiloi or kouphoi, Latin velites), used to ravage and plunder enemy territory, assault poorly defended settlements, construct fortifications and siege works, engage in skirmishes, and provide missile fire and harrying actions in battle. Ship-to-ship combat, while occasionally important, was only a secondary function and was neither the primary purpose of ancient war fleets, nor the typical manifestation of ancient naval warfare.¹

WARSHIP DEVELOPMENT

The use of multi-oared ships to transport military forces can be traced back to the Egyptian Old Kingdom period (2575–2175 B.C.), when the pharaohs needed highly mobile forces to assert their control along the length of the Nile and overseas in the commercially important Levantine region, a key source of timber (see de Souza 2007 and Fabre 2005 for surveys of Egyptian naval development; on the New Kingdom see Spalinger 2005). These amphibious units became less important in the New Kingdom period (1539–1069), when Egypt’s primary enemies had chariot divisions at the core of their armies. Nevertheless, maritime operations continued to be a feature of Egyptian warfare, especially against the port cities of the Levant. Other Mediterranean maritime cultures also used fleets on occasion to transport their armies or warrior bands, but there is no unequivocal evidence for purpose-built warships outside Egypt, and combat at sea was extremely rare (see further Wachsmann 1998).

The evidence for the next few centuries is very limited, but representations of ships on painted pottery, relief carvings, and three-dimensional models show that by the end of the eighth century B.C. the Greeks were building ships that were specifically designed for raiding or warfare (figures 18.1 and 18.2). These vessels can be distinguished from other seagoing ships by their low, elongated profile, the prominent forefoot and cutwater at the prow, and by the presence of a fighting platform along part of the ship’s length. They are not uniform in design, and would have carried between thirty and a hundred oarsmen seated on benches.² These characteristics suggest that the ships were intended to transport groups of armed men on voyages of short duration. Images of combat between ships are still rare, and those that do survive do not indicate that the ships themselves were used as weapons, being merely vessels from which fighting men engage in combat.³ Ships carrying warriors feature prominently in the Homeric poems, composed circa 750–700B.C., but they do not have rams and do not fight at sea. Modern scholars have often assumed that Homeric ships did have rams, but that they were omitted from the Homeric epics to avoid anachronism (figure 18.2).⁴ There is no clear depiction of a ram in the ship iconography from the eighth and seventh centuries B.C. and a recent discussion of Homeric seafaring has concluded that it was unknown at this time (Mark 2005: 104–14).

Figure 18.1 This late eighth-century B.C. Attic black-figure drinking cup from Eleusis shows an archer and a helmsman on a rowed ship. The single warriors to either side of the ship appear to carry Dipylon-style shields and three spears. It is unclear whether they are attacking or defending the ship, or even if the scene is meant to represent combat. Photo Credit: D. de Souza, by permission.

Figure 18.2 This fragment of a seventh-century B.C. Attic painted plaque comes from Sunium, where there was a major sanctuary of Poseidon. It shows a rowed ship with a helmsman and five warriors, each of whom wears a helmet and carries a shield and two spears. In front of each warrior is an oar, attached to a tholepin by a loop, suggesting that the warriors are a part of a raiding party who row themselves. Photo Credit: D. de Souza, by permission.

One way to increase the capacity and speed of these warships, without making them too long or broad in the beam, was to incorporate a second row of oarsmen, seated higher up. There are representations of galleys rowed by oarsmen sitting on two levels on relief sculptures from the early seventh-century B.C. palace of the Assyrian king Sennacherib (so Casson 1971: figs. 76, 78; Wallinga 1995: 42–3). In the Greek world unambiguous evidence of two-level galleys dates only from the sixth century, although scholars have argued that painted pottery from as early as the eighth century shows ships being rowed on two levels (Casson 1971: 53–60; Morrison and Williams 1968: 38–40; contra Wallinga 1993: 33–65). When a third level of oarsmen was added is a matter of considerable dispute, but it seems likely that a type of warship rowed on three levels, known as the trireme (Greek triērēs), was developed sometime between 600 and 550, perhaps in Egypt, or in Phoenicia, or possibly at Carthage (Wallinga 1993: 103–18 argues for an Egyptian origin).

The case has been made for the invention of the trireme in Greece in the seventh, or even the late eighth century B.C., based on a series of statements about early Greek naval matters made by the historian Thucydides (Thuc. 1.13.2–4; see Morrison 1996: 178–82; Casson 1971: 80–1). According to Thucydides there was a tradition that the first Greek triremes were built at Corinth, and that the Corinthian shipwright Ameinocles built four ships for the Samians around 700. Thucydides also claims that the earliest recorded naval battle, between the Corinthians and the Corcyreans, took place about 640 (on the dates see Hornblower 1991: 42–5). He does not, however, claim that the ships built by Ameinocles were triremes, nor that the naval battle was fought in the manner of naval battles in the Peloponnesian War, which were between fleets of triremes (see Wallinga 1993: 13–16). Furthermore, the earliest literary testimony for the trireme as a ship type is a mocking fragment by the poet Hipponax of Ephesus:

Lazy good-for-nothing Mimnes, never again

paint a snake along a trireme’s many-benched sides

running back from prow to steersman.

For that’ll be disastrous and a bad omen,

you empty-headed son of a slave, for the steersman,

when the snake comes and bites him on the legs!

Hipponax most probably wrote circa 540–520 (FGrH 239 F42; Plin. HN 36.11; Procl. Chrest. 31). If the trireme was a familiar vessel type by this time, then it would be reasonable to assume that it was invented in the first half of the sixth century B.C.⁵ Herodotus says that Polycrates, tyrant of Samos, in addition to his fleet of one hundred penteconters, possessed forty triremes which took part in the invasion of Egypt in 525 by the Persian king Cambyses (Hdt. 3.39; 44). The rapid growth of Persian imperialism under Cambyses (530–522) and his successor Darius (522–486) was the political context for the creation of large fleets of triremes in the eastern Mediterranean. In the west, at the same time, the intense rivalry between Carthage and the Greeks of Sicily encouraged similar developments (on Persia as the catalyst in warship development see Wallinga 1993; on Carthage, Rawlings 2010).

Within a relatively short space of time triremes became the standard warships across the Mediterranean. Like its predecessors the trireme was a long galley, but it could be rowed by about 150 to 170 oarsmen, pulling oars of equal length and arranged in groups of three, sitting one above the other (figure 18.3). The addition of the third tier of oarsmen, who usually rowed through an outrigger, allowed a substantial increase in the total number of men that could be carried by the warship, as well as in the vessel’s size and weight, without drastically reducing its speed.

Figure 18.3 This plaster cast is of a fragment from a fifth century B.C. sculptural relief, known as the Lenormant relief, found on the Athenian Acropolis. It shows the side of an Athenian trireme with all three levels of oars depicted, although the only oarsmen visible are those rowing at the highest level, the thranitai. The ship’s full-width fighting deck forms a canopy over their heads. Photo Credit : P. de Souza, by permission.

The conventional explanation for the invention of the trireme is that it offered a significant advantage in naval combat, especially ramming. There is, however, very little evidence for ramming tactics prior to the early fifth century B.C., so it seems unlikely that this was the decisive factor (cf. Hdt. 1.166 and his account of the battle of Alalia [ca. 540–535]). The fact that the new warship type could also accommodate a higher, more extensive fighting deck than earlier types meant that it packed a greater punch as an assault vessel, because its complement of rowers was larger and it could carry additional heavy infantry and archers. These men also offered advantages in ship-to-ship combat at sea, during the exchange of missiles and the subsequent boarding of enemy vessels.

By the early fifth century B.C. many Greek poleis were equipping themselves with triremes (de Souza 1998: 282–7). They were ideal for the kind of military expeditions that large, prosperous poleis like Aegina, Athens, Corinth, and Corcyra were engaging in as they sought to expand and consolidate their influence overseas. Another reason for investing in triremes was the creation of the large Persian fleets based in Ionia, Cilicia, and Phoenicia (cf. Wallinga 1993: 130–68; de Souza 1998: 285–7; van Wees 2004: 206–9). Thucydides explicitly states that the triremes constructed by the Athenians in the celebrated Themistoclean program in 483 B.C. did not have decks covering the full expanse of the ships (Thuc. 1.14.3), and Plutarch says that they were built to be fast and easy to turn (Plut. Cim. 12.2). The so-called “Decree of Themistocles” gives an allocation of ten hoplites to each ship, plus four archers, which contrasts with the forty or more soldiers on board the decks of the Chian triremes at the Battle of Lade in 494 and the Persian fleet in 480 (Hdt. 6.15.1; 7.184.2, with Jameson 1962, 1963). The Athenian ships probably had a narrow deck running the length of the ship from stern to prow, but not covering the heads of the rowers on each side, offering relatively little space for archers and hoplites. It is likely that Themistocles knew about the failed attempt of Dionysius of Phocaea to drill the Ionian Greeks in ship-to-ship combat techniques (Hdt. 6.12), and had concluded that keeping the new Athenian vessels light and maneuverable would make it easier for them to perform the sorts of circling and ramming maneuvers that the Ionians had tried to use. On the basis of these points it is reasonable to suppose that the Themistoclean triremes were the first warships that were designed, constructed, and fitted out with extensive ship-to-ship combat in mind, probably against local Greek enemies like the Aeginetans as well as the Persians. They formed the core of the Greek fleet that fought with considerable success against the Persians at Artemisium and Salamis in 480 and were instrumental in the dramatic expansion of Athenian power in the mid-fifth century B.C.

In the early 460s the Athenian general Cimon had the triremes of the Themistoclean fleet adapted so that they had a continuous deck extending from gunwale to gunwale along the entire length of the ship stern to bow, above the heads of the oarsmen. In his biography of Cimon, Plutarch describes this as a means of enabling the ships to carry more hoplites, and thus be better equipped for attacking the enemy (Plut. Cim. 12.2). Assuming that the key factor in trireme design was improved ship-to-ship combat ability, modern scholars have argued that Cimon’s innovation was intended to give the Athenians a numerical advantage when boarding enemy ships, at the expense of less speed and maneuverability at sea.⁶ A better interpretation is that the extra hoplites were for combat on land. In other words, by increasing the deck area of the triremes, Cimon made more space for non-rowing soldiers and thus improved the ratio of hoplites to light troops carried on each ship, effectively “beefing up” his amphibious assault forces. Such a move makes sense in the context of Cimon’s military operations and his association with the moderate politics of the hoplite class in Athens.

The trireme was, therefore, a warship type that could be adapted to suit the operational needs of a particular fleet or squadron. By the middle of the fourth century B.C. triremes in the Athenian shipyards could be subdivided into different categories, according to their quality and function. Some were designated as “fast-sailing,” presumably at the expense of total carrying capacity, while others had been specially adapted for “horse-carrying” (Gabrielsen 1994: 126–31). The Athenians also possessed some quadriremes, slightly larger than triremes but with oars on two levels, each pulled by two men, and quinqueremes, much larger ships, crewed by 300 to 350 men, which probably had oars on three levels, with two men pulling the top two. These new types had been invented around 400 and became the standard warships of the Hellenistic kingdoms. A few very big warships were built for the Ptolemaic and Seleucid monarchs, who could mobilize huge numbers of men, in the late fourth and early third centuries B.C. These “polyremes” were slow and cumbersome, but they could carry hundreds or even thousands of oarsmen and marines—forces large enough to capture cities and fight major campaigns inland (see Morrison 1996 for survey and reconstructions and Murray 2012; the Rhodians favored a variant of the trireme called a trihemiolia, consisting of two levels of oars operating through the outrigger; Morrison 1996: 319–21).

NAVAL STRATEGIES AND TACTICS

In order to understand how ancient warships were used it is important that the modern notion of control of the sea, or even of the so-called “sea lanes,” is not simply transposed to the ancient world (Starr 1989, on which see de Souza 1990). Ancient warships were primarily intended to convey large numbers of men swiftly over long distances in order to attack targets on land. In the Archaic period (ca. 800–500 B.C.) the basic aims of warfare were to obtain booty, do harm to the enemy, and gain wealth and status, as exemplified by the Homeric poems (e.g., Od. 9.39–52; 14.222–34; Il. 11.669–761). The period between the composition of the Homeric poems and the fifth century B.C. saw the gradual emergence of more centralized governments and an increased capacity for interstate warfare in the Greek world (van Wees 2004: 232–6). While the development in the sixth and early fifth centuries of stronger, more politically cohesive states, capable of deploying large armies and fleets, saw a focus on more long-term, strategic objectives, the “raid mentality” remained a key element of ancient warfare (cf. de Souza 1999: 25–36; Gabrielsen 2001a, Gabrielsen 2007: 248–64). Raids could have multiple functions, including forcing the enemy to divert forces to the defense of vulnerable coastal areas, inflicting military, economic, and psychological damage, obtaining vital funds and materials for the raiders and boosting their morale (de Souza 2010). While major set-piece battles and sieges dominate the narrative histories, these accounts also contain hundreds of examples of such raids, showing the extent to which naval forces were deployed by those states that could command the necessary resources and manpower. Raiding was not just a maritime phenomenon, but what characterized maritime raids was the mobility that ships provided, making them far harder to defend against, especially if the targets were unprepared (e.g., Thuc. 6.62), or territory to be protected was extensive (e.g., Thuc. 4.53–57).

Warfare waged by large, state-operated fleets led the Greek historians of the fifth century B.C. to develop the idea of sea power, expressed by Thucydides and Herodotus with the Greek word thalassokrateia (“rule of the sea”), conventionally translated as thalassocracy, and its variations (see further de Souza 1998: 277–88; history of the concept: Momigliano 1944). An important articulation of the concept and practice of thalassocracy is provided by the anonymous Athenian pamphleteer of the late fifth or early fourth century known as the Old Oligarch (see further Marr and Rhodes 2008, 6–12). After characterizing the Athenians’ hoplite force as weak, but nevertheless stronger than those of their allies, the Old Oligarch explains that the key to the Athenians’ domination of those allies is that they are thalassocrats (Xen. Ath. Pol. 2.2). In a passage that brings to mind Thucydides’s accounts of the initial aims and operations of the Delian League and the “Periclean” strategy adopted by the Athenians in the first stage of the Peloponnesian War (Thuc. 1.55–56; 62; 96–9; cf. Hornblower 1996: 1064), the Old Oligarch describes the advantages that a thalassocracy has over powers with no significant naval capacity: “the rulers of the sea have the ability to do what the rulers of the land do not have the ability to do.… For it is possible for them to sail along the coast and put in wherever there is no enemy, or where there are only a few, and then, if the enemy forces attack, to re-embark and sail away.” Additionally,

the rulers of the sea have the ability to sail as far as you like away from their own country, whereas the rulers of the land do not have the ability to undertake a journey of many days away from their country.…It is necessary for the one who goes on foot to travel through territory which is friendly, or else to fight to win his way through, whereas it is possible for the seafarer to disembark wherever he is the stronger, and, wherever he is not the stronger, not to disembark…but to sail along the coast until he comes to friendly territory, or to the territory of those weaker than himself. (Xen. [Ath. Pol.] 2.4–5, trans. Marr and Rhodes 2008: 45)

For much of the fifth and fourth centuries B.C. the Athenians deployed hundreds of warships as they strove to maintain, or, after their defeat in 404, to reestablish a maritime empire. Their efforts, and those of other Greek states that built fleets to oppose and later to emulate them, made trireme fleets a central feature of warfare in the Classical period, as the contemporary accounts of Thucydides and Xenophon attest.

In the early Hellenistic period, when former generals of Alexander vied for control of his vast empire, huge fleets, featuring larger “polyreme” warships were used to achieve and maintain control over the Aegean and eastern Mediterranean (on these fleets see Murray 2008, 2012). For example, the fleet that Antigonus Monophthalmus assembled in 315 B.C., as enumerated by Diodorus, seems to have included thirty lemboi, one hundred triremes, ninety quadriremes, and ten quinqueremes, plus a further thirteen “nines” and “tens,” which between them must have carried in excess of fifty thousand men (Diod. 19.62.7–8). The larger ships had space on their decks for hundreds of heavy infantrymen, with siege artillery and other equipment, enabling Antigonus to dispatch a strong force of fifty ships to the aid of his ally Polyperchon in Greece, while the remainder cruised the islands and coastal cities to ensure that they remained loyal to his cause (Diod. 19.62.9).

If an ancient maritime power’s opponents possessed their own warships, they might challenge their attackers at sea, as the Greeks did to the Persian fleet at Artemisium and Salamis in 480 B.C. Ship-to-ship combats were especially frequent during the Peloponnesian War, in which Athens’s maritime empire was at issue. This gave victories at sea an exceptional significance. In 413 the Syracusans, having thwarted the invading Athenians’ attempts at a circumvallation, made their Great Harbor the scene of a desperate naval struggle, their victory in which forced the Athenians into an ultimately disastrous withdrawal (Thuc. 7.36–87). In the latter stages of the war the Spartans and their allies launched several fleets to challenge the Athenians for control in the Aegean and Hellespontine region, eventually winning a decisive naval battle at Aegospotami in 405 (Xen. Hell. 2.1.13–30).

The likelihood of having to engage other warships at sea is reflected in the fact that rams were a standard feature on all warships in the classical and Hellenistic periods, and naval powers developed sophisticated ship-to-ship combat tactics that made the most of their vessels, their sailing skills, and their fighting men. A captain had two main options available when attacking an enemy ship—striking it with enough force to disable it, or grappling and boarding it to overcome the crew and then capture or disable it. Although many rams must have been broken off in combat, and literary sources record that rams were regularly taken from captured ships and displayed as trophies, few have been recovered (see Tusa, Royal, and Buccellato 2011, Murray 2012: 31–68). A complete ram found, at Athlit, off the coast of Israel is 462 kilograms in weight and a little over two meters in length. It has blunt-ended fins that are designed to rupture an enemy vessel’s hull at the waterline, splitting apart the timbers and allowing water pressure to widen the breach (Casson and Steffy 1991). The risk to an attacking ship of penetrating too far and thus becoming dangerously entangled with a stricken opponent could be reduced by placing a projecting buffer a short distance above the ram. A warship’s captain and helmsman would need considerable skill and judgment to time their attacks and the oarsmen would have to be adept at reversing their stroke so that the ship could withdraw and attack another target while its first victim flooded and sank, although the lack of heavy ballast in ancient warships meant that most did not sink completely, but floated on, partially submerged, either upright or heeled over.

For ramming to be a primary tactic warships needed to combine speed with maneuverability. The best attack vector would have been from the side or rear, at a relatively acute angle, which meant outmaneuvering the target in open water. The most popular tactic was to break through the enemy line and then turn sharply around to attack their vulnerable sterns or beams (the so-called diekplous [sailing through] and periplous [sailing around]: see further Lazenby 1987; Whitehead 1987). If an entire fleet was to employ such tactics, training and practice were essential. When the fleet of 353 triremes mustered by the rebellious Ionian Greeks prepared to confront the superior Persian fleet at Lade in 494 B.C. their commander, Dionysius of Phocaea, insisted on drilling the inexperienced crews repeatedly, although after seven days’ hard work they refused to practice further (Hdt. 6.12). More direct ramming tactics were favored by ships that lacked speed and maneuvering capabilities. It was easier to ram head-on against the opponent’s prow, or square-on against the hull, but both approaches ran the risk that the ramming ship would incur serious damage, or be unable to withdraw. A variation involved the use of a ship’s stempost, ram, and the projecting beams at the end of the outrigger at a ship’s prow (epotides) to smash an enemy’s oars and outrigger. The Corinthians and Syracusans modified their triremes to make them strong enough for these tactics in the middle of the Peloponnesian War (Thuc. 7.34, 36). The larger polyremes developed in the fourth centuryB.C. seem to have been designed with frontal ramming in mind (Murray 2012: 3–68).

If fleets were not prepared to ram, or if maneuvering space was restricted, captains would close with their opponents and rely on the troops on deck to attack with arrows, slings, and javelins, often fired or thrown from a crouching position for better stability, to clear space on an opposing ship’s deck, and then try to board their ship. Grappling hooks could be used to secure the enemy ship at close quarters (called “iron hands,” they were routinely deployed in confined spaces; Casson 1986: 120–2). The invention of torsion catapults in the fourth century B.C. introduced a new element into maritime warfare. In 306 Demetrius Poliorcetes placed stone-throwing catapults on his ships, along with arrow-shooting catapults on their prows, for a naval battle against Ptolemy I Soter off the Cypriot city of Salamis (Diod.20.49.4). It is likely that these weapons had earlier been deployed on land against the city itself, and the very large polyreme warships developed at this time seem to have been designed to assault harbors and city walls (Murray 2008; Murray 2012: 69–207; de Souza 2007: 441–4). Catapults would have been most effective against the fighting personnel, and any artillery they may have had, rather than the ships themselves. In 184 B.C. the exiled Carthaginian general Hannibal, commanding the fleet of King Prusias of Bithynia, used his catapults to hurl pots filled with poisonous snakes onto the decks of the warships of Eumenes II of Pergamum (Frontin. Strat. 4.7.10–11).

Whatever the tactics used, crews consisting of well-trained, experienced men would have an advantage over less practiced opponents, which could often compensate for inferior numbers (as Thuc. 2.80–92 reports of the victory of a small Athenian squadron over a larger Peloponnesian fleet in 429 B.C.). The Athenians were felt to be far superior to other Greeks in open sea combat at the start of the Peloponnesian War, but gradually, thanks to the frequency of naval combat in this war, the Peloponnesians and Syracusans acquired the experience and confidence to match or better them. The Rhodians displayed a high level of tactical organization and proficiency in naval engagements in the Hellenistic period. Livy’s account of the Battle of Side in 190 B.C., in which they defeated the Seleucid king Antiochus III’s fleet, commanded by Hannibal, stresses the superior seamanship and sophistication of the smaller Rhodian fleet. At one point the bulk of the Rhodian ships responded immediately to a signal flag ordering them to regroup around the flagship, thereby coming to the rescue of their outnumbered commander (Livy 37.24.4). Such a professional level of discipline and coordination was, however, exceptional and can be directly attributed to the Rhodians’ maintenance of a small, standing fleet that was the core of their military strength (Rice 1991; Gabrielsen 1997: 85–111; de Souza 2002: 84–92).

Although warships were powered by oars, they also had one or two sails, which could be used when speed was not essential. It was standard practice to remove the masts and sails in anticipation of combat, clearing space for soldiers to fight from the decks and making the ships more stable and easier to maneuver (Xen. Hell 2.1.29; cf. Hale 2009: 241). A fleet that did not expect combat could, therefore, be put at a considerable disadvantage by a surprise attack when under sail, as happened to Antigonus Monophthalmus’s fleet under Theodotus, ambushed off Aphrodisias in Cilicia by the Seleucid commander Polyclitus in 315 B.C. (Diod. 19.64.4–8).

NAVAL INFRASTRUCTURES AND COSTS

Unlike merchant vessels, warship hulls were protected only by light coverings of pitch and paint, and risked becoming waterlogged, encrusted with barnacles, and riddled with the burrows of marine borers if they were not regularly removed from the water to be dried, careened, repaired, and re-coated. In the Mediterranean Sea, wherever a suitably shelving strand was available, ships could be partially beached, stern-first, and made fast with cables or anchors to prevent them drifting away. If they were not in use for a long period they could be drawn up out of the water on wooden slipways, with props to keep them upright, but ready for a swift return to the water (evidenced as early as the Homeric poems: e.g., Od. 9.136–51; 10.87–96; Il. 2.149–54).

As states acquired large fleets of warships, however, more permanent facilities were required, including shipyards for construction and shipsheds to house the finished vessels. Piraeus, the fortified port of Athens, boasted the most extensive naval infrastructure in the Greek world (Garland, 1987; de Souza 2007b). It was created almost out of nothing after the defeat of Xerxes’s invasion force in 480/79 B.C. Thucydides claims that Themistocles, the driving force behind the swift completion, had a vision of Athens projecting her power by means of a large fleet based at Piraeus, from where the Athenians could defy the assaults of their Greek neighbors and the Persian king, whose own strategy in the eastern Mediterranean involved the projection of military power by sea (Thuc. 1.93).

The Piraeus peninsula encompassed three natural harbors—Kantharos (the largest), Zea, and Mounichia. Between 459 and 457 B.C. the Athenians constructed the Long Walls, linking the main city with the Piraeus in one vast set of fortifications (Conwell 2008). Themistocles’s original concept was developed into a maritime imperial hub. The harbors were ringed by hundreds of shipsheds—long, roofed slipways where vessels could be dried out, cleaned, repaired, and refitted.⁷ In addition there were construction yards for new vessels and arsenal buildings to store naval equipment. As Athenian economic dominance increased, through maritime trading links with the Black Sea and the eastern Mediterranean, warehouses, offices for merchants and bankers, brothels, and many other businesses grew up around the commercial harbor at Kantharos, employing thousands of people. The potential vulnerability of the Piraeus to seaborne attack was demonstrated by a Spartan raid in 429 B.C. (Thuc. 2.93–94), prompting the Athenians to narrow the harbor entrances, install archery platforms, and close them off with movable chains. In 387 another Spartan raiding party penetrated right into Kantharos harbor, seizing merchants and their wares from the quayside (Xen. Hell. 5.1.13–24).

Fortified harbors, shipsheds, arsenals for the storage of equipment and the other paraphernalia of naval activity were to be found at many great cities across the Mediterranean besides Athens, including Corinth, Rhodes, Samos, Alexandria, Syracuse, Rome, and Carthage. The extensive Rhodian and Carthaginian military harbors have been partially excavated, as have some smaller harbors and sets of shipsheds (see Blackman 2008: 654–60; Hurst 1994; Gabrielsen 1997: 37–42; Rankov 2008).

In spite of their possession of an impressive infrastructure and their extensive use of fleets, both Athenian and allied, there is no indication that the classical Athenians evolved a separate naval command structure. Their ten annually elected strategoi (generals) were a panel of supreme commanders whose authority was recognized over all military forces, regardless of their composition. They were expected to have the appropriate strategic, tactical, and logistical knowledge and skills for all aspects of warfare, although their particular abilities and aptitudes varied. Most other states followed suit, although the Spartans regularly appointed a commander for their naval forces, called the nauarchos, at least after they had begun to make regular use of fleets in the Peloponnesian War. They clearly felt that this position was distinct from commanding an army, which was the traditional prerogative of their kings.

The man with the highest authority on a classical Athenian trireme was the trierarchos, unless a strategos was also on board (e.g., Lys. 21.6–8; see Gabrielsen 1994 for detailed treatment of the trierarchy). The role of the trierarchos was to ensure that the vessel under his command was appropriately crewed, equipped, and operated during a campaign. These were largely financial obligations, so the trierarchoi were chosen by lot from Athens’s wealthiest citizens, whose military and navigational competence was extremely varied. An inexperienced trierarchos could get a great deal of assistance from his kubernetes (helmsman), a professional sailor whose services could command very high wages (e.g., Lys. 21.10). Similar arrangements seem to have been adopted by other Greek states, including Aegina, Rhodes, and Samos; Herodotus mentions Phoenician trierarchoi (Hdt. 8.90).

The financial aspects of war at sea were often as important as the military ones in determining the nature and outcome of a campaign, or even an entire war. Only very wealthy states, with well-developed fiscal systems, could afford to operate fleets over an extended period (for penetrating analysis of classical Athens and Hellenistic Rhodes see Gabrielsen 2001b). The Athenians recognized that their shipsheds were a superb architectural achievement, and were reputed to have cost the enormous sum of one thousand talents. Modern estimates indicate that the resources required to build them were greater than those needed for the Periclean building program on the Athenian Acropolis (Blackman and Rankov, forthcoming). Ancient warships were built almost entirely of wood. The construction of a major fleet involved, therefore, the selection, felling, cutting, seasoning, and transportation of huge amounts of timber. In order to build the two hundred triremes that the Athenians possessed in 480 B.C., some, if not all, of the necessary timber must have been obtained at least two or three years previously. In 483, at Themistocles’s instigation, the Athenians had voted to make use of an unexpectedly large surplus of revenues from their silver mines to pay for a fleet of one hundred triremes, which played a vital role in the defeat of the Persians in 480. Themistocles’s proposal was probably passed by the Athenians in the summer of 483. Although we lack details of its contents, it seems to have envisaged the spending of up to one talent per ship by individual Athenians (Gabrielsen 1994: 27–35). As their imperial resources expanded and their fleets grew, the Athenians developed a system that harnessed both public revenues, from tribute and taxation, and private wealth. The state provided the warships, gear, and crews, but thetrierarchoi were obliged to use their personal fortunes to pay maintenance and repair costs for the ships under their command. There is good evidence that other maritime powers, like the Rhodians and the Carthaginians, also made considerable use of private funding to keep their fleets operational (Rhodes: Gabrielsen 1997; 2001b; Carthage: Lancel 1995: 178–92; Miles 2010: 174–9, 195–6).

Almost all ancient warships were crewed by men who expected to be paid for their service, which meant that to keep a fleet of any size at sea required plenty of money. The realities of this situation were impressed upon the Athenians by Demosthenes when he addressed them on the subject of the means used by their strategoi to fund the fleets with which the Athenians tried to reestablish their maritime power in the mid-fourth century B.C.:

All your generals who have ever sailed from here…take money from the Chians and Erythraeans, from whomsoever, I say, they possibly can among the peoples living in Asia. Those who have only one or two ships exact less than those who have a more powerful fleet. The providers do not give their large or small contributions for nothing…but on the understanding that they will not be harmed when they leave harbour, nor plundered, or that their ships will be escorted.…They speak of favours being granted and that is what they call their gifts. (Dem. 24–5)

The implications of meeting the costs of naval warfare went even further. The enormous costs that were incurred by the Athenians as the Peloponnesian War dragged on for years caused huge political upheavals and prompted radical fiscal reorganization and innovation (Kallet-Marx 1993; Kallet 2001). For the Spartans the price of victory was acknowledgment of the Persian king’s claim to sovereignty over the Greeks of Asia in return for money to pay the crews of the fleets that they needed to challenge and eventually defeat the Athenians. The huge war fleets of the early Hellenistic period were made possible by the availability to the Successors of Alexander of the royal treasuries taken from the Persians. Financial costs certainly played a major role in the conflicts between Rome and Carthage and Rome and the Hellenistic monarchies in the third and second centuries B.C. (Serrati 2007: 464–77, 488–91). Fleets were, of course, only part of the military establishments of these states, but the sheer numbers of men involved, as well as the basic costs of warships and other facilities meant that any attempt to wage war by sea had major economic implications. Indeed, it can be argued that the need to pay such large numbers of men regularly over long periods of time was one of the driving forces behind the emergence of mass production of coinage and the monetization of the economies of the Mediterranean states. This in turn encouraged the more successful, aggressive states to wage wars to acquire, control, and exploit sources of revenue, creating a vicious cycle of war and imperialism from which the Roman Republic emerged as the eventual “winner” (see Trundle 2010 on the Greeks; on Roman imperialism see Harris 1979; Rich 1993; Rauh 2003).

ROMAN NAVAL DEVELOPMENTS

The old view that the early Romans had no naval forces before the outbreak of the First Punic War in 264 B.C. has been successfully challenged, recent scholarship emphasizing Roman maritime development in the early Republican period (so Thiel 1946: 1–31, 1954: 4–29, following generalizations in Livy 7.26.14 and Polyb. 1.20.8; contra Mitchell 1971; Steinby 2007: 29–86). By the end of the fourth century B.C. citizen colonies at strategic coastal locations provided some protection from seaborne raiders (e.g., Livy 7.25–6; 8.26) and from time to time small fleets, organized by elected officials called duoviri navales (“two men in charge of warships”), conducted raids (e.g., Livy 9.30.3–4, 38.2–4). One such operation in 282 B.C. provoked a war with the city of Tarentum which sought the aid of Pyrrhus, king of Epirus (App. Sam. 7–11; Livy Per. 12–14; Dio frr. 39–40). By the mid-third century B.C. the Romans controlled several south Italian cities that had their own fleets, but at the outbreak of the First Punic War with Carthage in 264B.C. they lacked the resources to mount the kind of large-scale naval expeditions needed. Polybius says that the Carthaginians’ capacity to raid the coasts of Sicily and Italy prompted the Romans, in 261 B.C., to “take to the sea (Polyb. 1.20.5–9). In 260 B.C. the Romans built their first major fleet of twenty triremes and one hundred quinqueremes. To counteract Carthaginian superiority in combat at sea (Polyb. 1.51), the quinqueremes were equipped with a new device known to scholars as the corvus (“raven”), a wooden bridge designed to hold an enemy ship alongside while soldiers boarded it (Polyb. 1.22; Polybius uses the Greek word for raven (korax), but no surviving Latin source describes it; cf. de Souza 2007b; Steinby 2007: 87–104). Both Carthage and Rome suffered huge losses, but thanks to their abundant supplies of timber and greater reserves of manpower, the Romans were better able to replace them. In 241 B.C., following the defeat of a Carthaginian fleet sent to relieve western Sicily, the Carthaginians sued for peace.

Early in the Second Punic War the Carthaginians did not even consider mounting a large-scale maritime invasion of Italy. Aside from the logistical problems of transporting an army of adequate size by sea, as well as the risks from storms and interception by Roman naval forces, the overland route from Spain into northern Italy suited the disposition of Hannibal’s experienced, successful army there and offered the possibility of recruiting further allies among the Celtic peoples of southern Gaul and northern Italy. Nevertheless, naval operations were carried out by both sides, including raids on Italy and Sicily by small Carthaginian fleets, dispatched early on in the conflict, and similar attacks by Roman fleets against Carthaginian territory in North Africa and along the Mediterranean coastline of Spain. Their aim was to plunder, ravage, take prisoners and hostages, or, in some cases, exact payments to forestall such actions (e.g., Polyb. 3.96; Livy 22.20; Livy 25.31).

In 217 B.C., while the dictator Quintus Fabius Maximus and his Master of Horse Marcus Minucius Rufus were confronting Hannibal’s army in northern Apulia, the consul Gnaeus Servilius Geminus led 120 warships on a campaign that was partly a response to Carthaginian naval operations around the islands and Italian coast of the Tyrrhenian Sea (Polyb. 3.96.7–14; Livy 22.11.6–9; 22.31.1–7). He began by taking hostages from Sardinia, where the Carthaginians had made a brief landfall, and Corsica. He proceeded to Lilybaeum in Sicily, where he probably embarked additional troops to strengthen his force. Next, his fleet devastated the island of Meninx, off the coast of Tripolitania, before proceeding north to the Cercina islands, whose inhabitants yielded ten talents to avoid a similar fate. Servilius landed the men from his ships on the African coast and dispersed them in raiding parties, some of which were ambushed by local forces, with the loss of about one thousand men, including the quaestor Tiberius Sempronius Blaesus.⁸ The fleet then sailed back to Lilybaeum, stopping at Cossyra to deposit a garrison, and finally returned to Rome. The fact that the expedition was led by one of the consuls at a time when an invading army was menacing southern Italy demonstrates its importance to the Roman war effort. It succeeded in driving a large, troublesome Carthaginian fleet out of Italian waters, secured key islands with garrisons, inflicted considerable harm on communities under Carthaginian protection and amassed a substantial amount of booty (for discussion of Roman naval actions in the Second Punic War see Rankov 1996).

Similar raiding campaigns were carried out against states subject or allied to Philip V of Macedon during the First Macedonian War (214–205 B.C.). The basic Roman strategy was to use coastal raids, in conjunction with their Aetolian allies, to keep Philip V occupied and unable to launch an assault on Italy in support of his Carthaginian allies (e.g., Livy 26.26; Steinby 2007: 143–55). As the Romans’ imperial ambitions drew them into further wars with the Hellenistic kings and their allies, fleets operating as independent amphibious assault groups became common. Rome’s maritime allies, the kingdom of Pergamum and the city-state of Rhodes, operated in a semi-autonomous fashion, while smaller states crewed warships furnished for them by Rome.

Between 198 and 195 B.C., when Rome was at war with Philip V of Macedon and the Spartan tyrant Nabis, a fleet of Roman and allied vessels commanded by Lucius Quinctius Flamininus captured Eretria, Carystus, Leucas, Gytheum, and numerous other coastal towns and cities, eventually cutting the Spartan tyrant off from the sea (Livy 32.16–17, 19–23; 33.16–17; 34.29–30).⁹ No ship-to-ship confrontations are mentioned in this campaign; instead the emphasis was on assaults against towns and cities (see further Livy 32–34; Polyb. 18; Errington 1989: 264–77). It is clear that Lucius Quinctius made the most of his fleet’s mobility to move from one target settlement to another, disembarking his men and mounting strong attacks. He deployed siege engines, artillery, and constructed siege works, the crews of his warships doing the excavation and construction. Livy remarks that when the Rhodian and Pergamene fleets arrived to join Lucius Quinctius’s force to assault the Laconian port of Gytheum the preparations were completed in just a few days thanks to the abundant manpower provided by the three fleets (Livy 34.29.4–5; the Rhodian fleet, probably trihemioliae [similar to triremes]: Rice 1991; Gabrielsen 1997; Eumenes of Pergamon: ten decked warships [possibly triremes], thirty lemboi, and a variety of other small vessels [Livy 34.26.11]). Early in the second century B.C. the Romans subdued the Ligurians, depriving them of their warships. Although the ancient sources present this as the suppression of Ligurian piracy, it was probably more about controlling the military reach of local rivals (Plut. Aem. 6; Livy 40.18, 26.8, 28.4, 42.1–5). A similar concern can be seen in the treaty that concluded Rome’s war with Antiochus III in 188 B.C. The king’s fleet was limited to ten warships and confined to Syrian waters (Livy 38.38).

Figure 18.4 This section of relief sculpture is from the top of a monument honoring Gaius Cartilius Poplicola, who was a prominent citizen and magistrate of Ostia in the late first century B.C. It shows Roman soldiers on and near a warship (not to scale). The warship’s three-pronged ram is raised above the waterline and there is a secondary, lion-headed projection above, possibly a buffer designed to prevent excessive penetration of an opponent’s hull. The warship has three levels of oars and is probably a quinquereme.
Photo-montage: D. de Souza.

Rome had no call to gather large fleets again until 67 B.C., when Pompey the Great assembled around one hundred warships to subdue the independent coastal communities of Pamphylia and Cilicia (Plut. Pomp. 26–8; App. Mith. 95–6; sources present this campaign as the suppression of piracy, but see de Souza 1999: 161–78). Remnants of this fleet may have formed the core of that used by his son Sextus Pompeius Magnus to raid the coast of Italy from his base in Sicily from 42 to 36 B.C. in the struggle with Octavian, Julius Caesar’s heir. Marcus Agrippa, Octavian’s principal commander, drafted twenty thousand freed slaves as crews for an armada of nearly four hundred warships to spearhead an amphibious assault on Sicily in 36 B.C. Agrippa and Octavian eventually defeated Sextus in a series of sea battles (de Souza 1999: 185–95). Agrippa’s fleet consisted mostly of triremes, quadriremes, and quinqueremes, plus a few larger vessels (figure 18.4). Some carried a novel form of catapult-launched grappling hook, attached to a winch, so that enemy ships could be hauled into close quarters for boarding (App. B. Civ. 5.118). Agrippa also constructed the Portus Julius in Campania, a complex of artificial harbors and lagoons, linked by tunnels, to protect his supply convoys and give his raw recruits somewhere to train (Paget 1968).

Figure 18.5 This scene is a detail from a narrow fresco painted along a corridor in the Villa Farnesina, a luxurious house on the banks of the river Tiber at Rome. The villa may have been built for Marcus Vipsanius Agrippa, who commanded the fleet of the future emperor Augustus at the Battle of Actium in 31 B.C. It shows warships in combat at sea; in the foreground a ship has run aground. Photo Credit: D. de Souza, by permission.

In 31 B.C. Octavian and Agrippa assembled a fleet of similar size to challenge Octavian’s remaining rival, Marcus Antonius, and his ally Cleopatra, whose forces, poised to invade Italy from western Greece, included large polyreme-type warships suitable for assaulting coastal cities and harbors. Agrippa drove Antonius’s ships out of the offshore islands and cut off the supply routes to his army, which Octavian had bottled up around the Gulf of Ambracia. Antonius and Cleopatra tried to break out by sea, but their cumbersome fleet of 230 ships was defeated by Agrippa’s fleet of 400 mostly smaller, more mobile vessels off the promontory of Actium (figure 18.5). Most of the remaining troops surrendered.

With the suicides of Antonius and Cleopatra in 30 B.C., Octavian became de facto ruler of the Mediterranean. Over the next few years he consolidated his position and in 27 B.C. formally assumed the name Augustus. He established Rome’s first permanent, professional military forces, which included substantial naval elements (excellent summary of Roman Imperial naval forces in Rankov 1995; for detailed study see Reddé 1986). The core of these were two fleets, the classis Ravennatium, stationed at Ravenna on the northern Adriatic coast, and the classis Misenatium, based at Misenum on the Bay of Naples. These fleets remained in existence until the early fourth century A.D. Although very little survives of their harbors, shipsheds, and barracks, epigraphical records indicate that in the first and second centuries A.D. each fleet consisted of around fifty warships, mostly triremes, although a handful of quinqueremes and quadriremes are documented (Reddé 1986: 665–9; a smaller fleet stationed at Forum Julii [Fréjus in southern France] ceased to exist by the middle of the first century A.D.; Tac. Ann. 4.5; Reddé 1986: 171–7). Further fleets were based in Egypt, at Alexandria on the mouth of the Nile (classis Alexandrina), and Seleucia on the mouth of the Orontes, which was the port for Antioch (classis Syriaca), with squadrons attested on Rhodes and Sicily, and in Libya in the 170s A.D. (Reddé 1986: 212–13; 227–50). Following Claudius’s invasion of Britain in A.D. 43 a classis Britannica was established, consisting mostly of “liburnians” (Mason 2003). The royal fleet of Pontus was the basis for a classis Pontica after the kingdom was annexed in A.D. 64 (Reddé 1986: 253–69).

Augustus’s wars of conquest in Germany from 12 B.C. to A.D. 16 involved the creation of the classis Augusta Germanica to operate along the Rhine and the North Sea coast. By the end of the first century A.D. two fleets were operating on the Danube, the classis Pannonica and the classis Moesiaca. Using small galleys based at or near legionary fortresses, they patrolled the Empire’s northern river frontiers, guarding against barbarian incursions and ferrying troops and supplies during major campaigns, as depicted on the columns of Trajan and Marcus Aurelius in Rome (see further Konen 2000; Bounegru and Zaharaide 1996; Rankov 1995). These fleets furnished the emperors with amphibious strike units for rapid deployment anywhere in the Mediterranean to deal with barbarian incursions, localized revolts, outbreaks of piracy, and similar, small-scale military operations. When required they supported the main legionary and auxiliary armies on large-scale campaigns, securing harbors, ferrying or escorting troops and supplies (e.g., Rankov 1995; Reddé 1986: 502–72). They also had other functions, including transporting members of the Imperial family and high-ranking officials to and from the provinces. Detachments from the two Italian fleets stationed in the city of Rome assisted in the presentation of mock naval battles in specially prepared maritime arenas and operated the huge canvas awnings that shaded spectators in the Colosseum (Tac. Ann. 12.56; SHA Comm. 15.6).

As argued above, the men who rowed ancient warships were expected to participate in fighting on land as light-armed troops, engaging in raids and skirmishes, building siege works, and supporting the legions of heavy infantry (e.g., Livy 9.38.2–4; Diod.23.18.3–5). The Roman Republic normally crewed its warships with precisely the types of men whose social and political status rendered them unsuitable for service in the legions of heavy infantry. Roman citizens of the lowest property class, theproletarii, were expected to serve in the fleets, rather than the legions, as were freedmen, although men from both groups might be recruited into the legions in exceptional circumstances (Polyb. 6.19.3 identifies the cutoff point as below four hundred drachmas (= four hundred asses); but it was not rigidly maintained and ceased to be observed by the end of the second century B.C.; see de Ligt 2007; Brunt: 402–8). During the Second Punic War (218–201 B.C.) the usual wealth qualification for service in the legions was disregarded on more than one occasion to increase the pool of potential recruits. As a consequence of this, men who might have been available to serve in fleets were recruited by the legions and so slaves and even prisoners of war were pressed into service (e.g., Livy 24.11.7–9; 26.47.1–2). The huge size of the fleets deployed in the second and third centuries B.C. meant that the lowest levels of Roman society could not provide enough men, but an extensive supply of recruits was available from the many Italian communities who had treaties of alliance with Rome. They regularly supplied half or more of the men in Rome’s armies, meeting quotas agreed on with the Roman magistrates at the start of each campaign. In addition to the tens of thousands who fought alongside the legions of Roman citizens, large numbers of these “allies” (Latin socii) were assigned to serve in the fleets. They included men from the Greek maritime cities of southern Italy and Sicily, as well as contingents from non-maritime peoples, such as the Samnites (Zon. 8.11; Oros. 4.7.12).¹⁰

The crews of the ships that made up the Imperial fleets were largely recruited from the provinces of the Empire. In keeping with the tradition that had developed during the second and first centuries B.C., it seems to have been the eastern Mediterranean provinces that provided the bulk of the men for the Misenum fleet, with the Danube provinces furnishing most of the recruits for the Ravenna fleet. Initially they were noncitizens, like the soldiers in the various auxiliary cohorts who supplemented the legions. This can also be interpreted as a continuation of the basic pattern found in the Republican period. However, a recent study of the nomenclature found in the surviving epigraphic and papyrological records has demonstrated that, at least from the early 70s to the mid-third century A.D., when the records run out, the men who served in the Misenum and Ravenna fleets were all Roman citizens (Mann 2002). This may reflect a decision of the emperor Vespasian to reward the two Italian fleets for support in the civil wars of A.D. 68/9 with the honorific title praetoria, conferring on their personnel a privileged status analogous to that enjoyed by the Praetorian Guards in relation to the rest of the army. The various provincial fleets were always manned by noncitizens and freedmen from across the Empire, but an exception was made in the case of the Egyptian fleet, whose locally recruited crews were men holding either Roman or Alexandrian citizenship (see Reddé 1986: 474–533).

Although ancient literary sources and even some documents often refer to Roman Imperial fleet personnel as sailors or rowers, they were in fact soldiers, paid at the same rates as other auxiliary troops and subject to military law (Dig. 37.13). This is confirmed by the way that fleet personnel are routinely designated by the word miles (= “soldier”), or similar terms, on their diplomas and funerary monuments (Spaul 2002 lists the most relevant documents; see figure 18.6). The crew of each ship was organized as acenturia, they were equipped and trained to fight as infantry, and they could be brigaded alongside regular troops, or even, in extreme circumstances, into separate legionary units (Tac. Hist. 1.6, 2.11, 17, 22, 3.55).¹¹ The officer in charge of each vessel had the rank of centurion, although he might be called a trierarchus. Greek terms were also regularly used for subordinate officers. Imperial fleets were commanded by men who held the rank of prefect (praefectus) and were appointed directly by the emperor from the equestrian class. Prefect of one of the Italian fleets was a high achievement for an equestrian. Pliny the Elder, who was prefect of the Misenum fleet when he died in the eruption of Vesuvius in A.D. 79, had previously been a military tribune and commander of a squadron of Thracian cavalry in Germany, as well as an Imperial procurator in Spain and Africa. It is clear from the vivid description that the Younger Pliny gives of his uncle’s actions during the eruption that a fleet commander might act on his own initiative in a time of crisis (Plin. Ep. 6.16).

Figure 18.6 This grave monument, found in Athens, commemorates the Roman soldier Quintus Statius Rufinus, who served 18 years under the centurion Claudius Ingenuus in the Misenum fleet, sometime in the second or third century A.D. The Latin text (CIL III 556a) reads: D(is) M(anibus) / Q(uintus) STATIUS RUFINUS M(iles) CLASSIS PR(aetoriae) / MIS(enensis) > (centuria) CLAUDI INGE(n)UI AN(norum) XXXVIII M(ilitavit) AN(nos) XVIII. Translation: To the spirits of the departed / Quintus Statius Rufinus, soldier of the Praetorian Fleet / at Misenum, from the century of Claudius Ingenuus, 38 years of age, served 18 years. Photo Credit: D. de Souza, by permission.

The maintenance of so many fleets may seem extravagant and unnecessary to modern readers who are used to the constant demands of politicians and journalists for reductions in military expenditure. Yet the Italian and provincial fleets seem neither large, nor expensive, when considered alongside the enormous funds and manpower required to maintain twenty-five to thirty legions, plus numerous auxiliary cohorts, as well as extensive military systems such as Hadrian’s wall, or the Germanic limes, or the frontier forts and communications in North Africa and the Near East. Indeed, it may well have been the combination of versatility and mobility, in return for the relatively modest outlay, that ensured the fleets’ survival. Tacitus gives an example of a police-type action involving a detachment from one of the Italian fleets in the area of Brundisium in A.D. 24. An incipient slave uprising, led by a former Praetorian Guard called Curtisius, was crushed by the crews of three warships, commanded by a quaestor called Curtius Lupus (Tac. Ann. 4.27). Tacitus, whose theme at this point in his Annals is the emperor Tiberius’s moral and political decline and his indifference to the well-being of his empire, presents this episode as a series of fortunate coincidences: the ships chanced to put in at the port; the quaestor just happened to be visiting the area. However, the swiftest way for the local authorities in this area to summon military assistance would have been by sea. The size of the force provided by the three warships is difficult to determine. Tacitus describes them as “biremes,” which probably means they were “liburnians,” each carrying around sixty men (Morrison 1996: 316–17). By the time a strong military force arrived from Rome the situation was under control (inscriptions indicate the regular presence of naval personnel in Brundisium around this time, possibly a detachment of the classis Ravennatium; Reddé 1986: 221).

Tacitus also gives us an insight into the value of a fleet as a strike force. In his account of Julius Agricola’s campaigns in northern Britain in A.D. 79–83 he gives some prominence to the classis Britannica, praising Agricola’s use of it to ravage and plunder ahead of the main army, sailing right around the northern coast and even making landfall on the Orkney Islands, thereby causing the Britons to feel highly exposed and vulnerable to Roman military might (Tac. Agr. 10, 24–25, 29, 38). Tacitus makes the threat posed to barbarian liberty by the extent of Roman naval power a feature of the stirring speech on Roman imperialism that he attributes to the British leader Calgacus before the battle at Mons Graupius:

There is no land beyond us and even the sea is no safe refuge when we are threatened by the Roman fleet.…Today the furthest bounds of Britain lie open.… But now there is no people beyond us, nothing but tides and rocks and, more deadly than these, the Romans.…They plunder, they butcher, they ravish, and call it by the lying name of “empire.” They make a desert and call it “peace.” (Tac. Agr. 30, trans. Birley 1999: 21–2)

Despite their military value, the fleets were never core units in the Roman military structure, and their secondary status was probably to blame for their eventual demise. The Rhine and Danube fleets disappeared in the wake of the barbarian incursions of the third century A.D., as did many provincial fleets like the classis Alexandrinum and the classis Pontica, although there is evidence of the presence of small flotillas in several locations in the fourth century A.D. (Reddé 1986: 572–641; see Aßkamp and Schäfer 2008; Schäfer 2008 for discussion of a fourth-century Rhine patrol vessel). The ships and men of the classis Britannica may have been reassigned to the system of defensive forts known as the Saxon Shore (litus Saxonicum), which lasted until the end of the fourth centuryA.D. (Mason 2003: 149–94). The main Italian fleets saw action during the civil wars of the early fourth century A.D. (Zos. 2.12). In A.D. 324, the fleet which the western emperor Constantine assembled to challenge the 350 triremes of his rival Licinius in the eastern Mediterranean consisted of Liburnians and other small vessels (Zos. 2.22–28; Anon. Vales. 5.21–28). Standing fleets of large warships had ceased to be a feature of the Roman Imperial military establishment after the fourth century A.D. (Reddé 1986: 641–52).

BIBLIOGRAPHY

Aßkamp, R., and C. Schäfer. 2008. Projekt Römerschiff: Nachbau und Erprobung für die Ausstellung ‘IMPERIUM KONFLIKT MYTHOS. 2000 Jahre Varusschlacht. Hamburg.

Basch, L. 1969. “Phoenician oared ships.” MM 55: 139–62, 227–45.

Birley, A. R. (trans.). 1999. Tacitus: Agricola and Germania. Oxford.

Blackman, D. J. 1968. “The shipsheds,” in J. S. Morrison and R. T. Williams, Greek oared ships 900–322 B.C. Cambridge, 181–92.

———. 2008. “Sea transport, Part 2: Harbors,” in J. P. Oleson (ed.), The Oxford handbook of engineering and technology in the classical world. Oxford, 638–70.

———, and B. Rankov. Forthcoming. Shipsheds of the ancient Mediterranean.

Bounegru, O., and M. Zahariade. 1996. Les forces navales du Bas Danube et de la Mer Noire. Oxford.

Casson, L. 1986. Ships and seamanship in the ancient world. Princeton.

———, and R. Steffy. 1991. The Athlit ram. College Station.

Conwell, D. H. 2008. Connecting a city to the sea: The history of the Athenian long walls. Leiden.

de Ligt, L. 2007. “Roman manpower and recruitment during the middle Republic,” in Erdkamp 2007: 114–31.

de Souza, P. 1990. “Review of Starr 1989” in CR, ns 40: 506–7.

———. 1998. “Towards thalassocracy? Archaic Greek naval developments,” in N. Fisher and H. van Wees (eds.), Archaic Greece; new approaches and new evidence. London, 271–93.

———. 1999. Piracy in the Graeco-Roman world. Cambridge.

———. 2002. “Beyond the headland: Locating the enemy in ancient naval warfare,” in J. Andreau and C. Virlouvet (eds.), L’information et la mer dans le monde antique. Rome, 69–92.

———. 2007. “Ancient navies: Egypt,” in Hattendorf 2007: 80–3.

———. 2007a. “Athens (Piraeus): naval port,” in Hattendorf 2007: 189–91.

———. 2007b. “Naval Forces (Hellenistic and Roman Republican),” in CHGRW, 1: 357–67.

———. 2007c. “Naval Battles and Sieges (Hellenistic and Roman Republican),” in CHGRW 1: 434–60.

———. 2010. “Raids on the coast of Attica,” in N. Sekunda (ed.), Ergasteria: Works presented to John Ellis Jones on his 80th Birthday. Gdansk, 82–93.

———. Forthcoming. Ancient naval warfare.

Erdkamp, P. (ed.). 2007. A companion to the Roman army. Oxford.

Errington, R. M. 1989. “Rome against Philip and Antiochus,” in CAH² 8: 244–89.

Fabre, D. 2005. Seafaring in ancient Egypt. London.

Fagan, G. G., and M. Trundle (eds.). 2010. New perspectives on ancient warfare. Leiden.

Gabrielsen, V. 1994. Financing the Athenian fleet: Public taxation and social relations. Baltimore.

———. 1997. The naval aristocracy of Hellenistic Rhodes. Aarhus.

———. 2001a. “Economic activity, maritime trade and piracy in the Hellenistic Aegean.” REA 103: 219–40.

———. 2001b. “Naval warfare: Its economic and social impact in ancient Greek cities,” in T. Bekker-Nielsen and L. Hannestad (eds.), War as a cultural force: Essays on warfare in antiquity. Copenhagen, 72–89.

———. 2007. “Warfare and the state (archaic and classical Greek),” in CHGRW 1: 248–72.

Gardiner, R. (ed.). 1995. The age of the galley: Mediterranean oared vessels since pre-classical times. London.

Garland, R. J. 1987. The Piraeus. London.

Hale, J. R. 2009. Lords of the sea: The epic story of the Athenian navy and the birth of democracy. New York.

Hanson, V. D. 1996. “Hoplites into democrats: The changing ideology of Athenian infantry,” in Hedrick and Ober, 1996., 289–312.

Harris, W. V. 1979. War and imperialism in republican Rome 327–70 B.C. Oxford.

Hattendorf, J. B. (ed.). 2007. The Oxford encyclopedia of maritime history. Vol. 1. New York.

Hedrick, C., and J. Ober (eds.). 1996. Demokratia: A conversation on democracies, ancient and modern. Princeton.

Hornblower, S. 1996. “Old Oligarch.” OCD³, 1063–4.

Hurst, H. 1994. Excavations at Carthage. The British mission. Vol. 2: The circular harbour, north side. Oxford.

Jameson, M. H. 1962. “A revised text of the decree of Themistocles from Troizen.” Hesperia 31: 310–15.

Jameson, M. H. 1963. “The provisions for mobilization in the decree of Themistocles.” Historia 12: 385–404.

Kallet, L. 2001. Money and the corrosion of power in Thucydides: The Sicilian expedition and its aftermath. Berkeley.

Kallet-Marx, L. 1993. Money, expense and naval power in Thucydides’ History, 1–5.24. Berkeley.

Konen, H. C. 2000. Classis Germanica: Die römische Rheinflotte im 1.-3. Jahrhundert n. Chr. St Katharinen.

Krentz, P. 2007. “Archaic and classical Greek war,” in CHGRW 1: 147–85.

Lancel, S. 1995. Carthage: A history. Oxford.

Lazenby, J. F. 1987. “The diekplous.” G&R 34: 169–77.

Lloyd, A. B. 1972. “Triremes and the Saite navy.” JEA 58: 268ff.

———. 1975. “Were Necho’s triremes Phoenician?” JHS 95: 45–61.

———. 2000. “The Saite Navy,” in G. Oliver, S. Brock, T. Cornell, and S. Hodkinson (eds.), The sea in antiquity. Oxford, 81–91.

Mann, J. C. 2002. “Name forms on recipients of diplomas.” ZPE 139: 227–34.

Mark, S. 2005. Homeric seafaring. College Station.

Marr, J. L., and P. J. Rhodes (eds.) 2008. The ‘Old Oligarch:’ The Constitution of the Athenians attributed to Xenophon. Oxford.

Mason, D. J. P. 2003. Roman Britain and the Roman navy. Stroud.

Miles, R. 2010. Carthage must be destroyed: The rise and fall of an ancient Mediterranean civilization. London.

Mitchell, R. E. 1971. “Roman Carthaginian treaties: 306 and 279/8 B.C.” Historia 20: 633–55.

Momigliano, A. D. 1944. “Sea power in Greek thought.” CR 58: 1–7.

Morrison, J. S., with J. F. Coates. 1996. Greek and Roman oared ships, 399–30 B.C. Oxford.

Morrison, J.S., and R. T. Williams. 1968. Greek oared ships 900–322 B.C. Cambridge.

Morrison, J.S., J. F. Coates, and N. B. Rankov. 2000. The Athenian trireme: The history and reconstruction of an ancient Greek warship. 2nd ed. Cambridge.

Murray, W. M. 2008. “The development of a naval siege unit under Philip II and Alexander III,” in T. Howe and J. Reames (eds.), Macedonian legacies: Studies in ancient Macedonian history and culture in honour of Eugene N. Borza. Claremont, 31–55.

———. 2012. The age of titans: The rise and fall of the great Hellenistic navies. Oxford.

Paget, R. F. 1968. “The ports of ancient Cumae.” JRS 58: 148–69.

Pugliese Carratelli, G. (ed.). 1996. The western Greeks: Classical civilization in the western Mediterranean. London.

Rankov, B. 1995. “Fleets of the early Roman empire, 31 B.C.–A.D. 324,” in Gardiner 1995: 78–85.

———. 1996. “The second Punic war at sea,” in T. Cornell, B. Rankov and P. Sabin (eds.), The second Punic war: A reappraisal. London, 49–57.

———. 2008. “Roman shipsheds and Roman ships,” in R. Hohlfelder (ed.), The maritime world of ancient Rome. Ann Arbor, 51–65.

Rauh, N. K. 2003. Merchants, sailors and pirates in the Roman empire. Stroud.

Rawlings, L. 2010. “The Carthaginian navy: questions and assumptions,” in Fagan and Trundle 2010: 253–87.

Reddé, M. 1986. Mare Nostrum: Les infrastructures, le dispositif et l’histoire de la marine militaire sous l’ Empire romain. Rome.

Rice, E. E. 1991. “The Rhodian navy in the Hellenistic age,” in W. R. Roberts and J. Sweetman (eds.), New interpretations in naval history. Annapolis, 29–50.

Rich, J. W. 1993. “Fear, greed and glory: The causes of Roman war-making in the middle Republic,” in Rich and Shipley, 38–68.

Saddington, D. B. 2007. “Classes: the evolution of the Roman fleets,” in Erdkamp 2007: 201–17.

Schäfer, C. 2008. Lusoria—Ein Römerschiff im Experiment: Rekonstruktion-Tests-Ergebnisse. Hamburg.

Serrati, J. 2007. “Warfare and the state,” in CHGRW 1, 461–97.

Spalinger, A. J. 2005. War in ancient Egypt: The New Kingdom. Oxford.

Spaul, J. 2002. Classes Imperii Romani. Andover.

Starr, C. G. 1989. The influence of sea power on ancient history. Oxford.

Steinby, C. 2007. The Roman republican navy. From the sixth century to 167 B.C. Helsinki.

Strauss, B. 1996. “The Athenian trireme, school of democracy,” in Hedrick and Ober 1996: 313–25.

———. 2000. “Democracy, Kimon and the evolution of Athenian naval tactics in the fifth century B.C.,” in P. Flensted-Jensen, et al. (eds.). 2000. Polis and politics: Studies in ancient Greek history. Copenhagen, 315–26.

Thiel, J. H. 1946. Studies on the history of Roman sea-power in republican times. Amsterdam.

———. 1954. A history of Roman sea-power before the second Punic war. Amsterdam.

Trundle, M. 2010. “Coinage and the transformation of Greek warfare,” in Fagan and Trundle 2010: 227–52.

Turfa, J. M. and A. G. Steinmayer. 2001. “Sewn hulls and self-defence.” IJNA 30.1: 122–7.

Tusa, S., J. Royal, and C. Buccellato. 2011. “Rams, warships and sea-battles in Sicily according to new archaeological evidences.” Conference abstract published at http://www.deguwa.org/data/File/IPR-XVI/Abstracts%20IPR%20XVI.pdf.

van Wees, H. 2004. Greek warfare: Myths and realities. London.

———. 2010. “‘Those who sail are to receive a wage’: Naval warfare and finance in Archaic Eretria,” in Fagan and Trundle 2010: 205–26.

Wachsmann, S. 1998. Seagoing ships and seamanship in the Bronze Age Levant. College Station.

Wallinga, H. T. 1993. Ships and sea-power before the great Persian war: The ancestry of the ancient Trireme. Leiden.

———. 1995. “The ancestry of the trireme 1200–525 B.C.,” in Gardiner 1995: 36–48.

———. 2005. Xerxes’ Greek adventure: the naval perspective. Leiden.

Whitehead, I. 1987. “The periplous.” G&R 34: 178–85.