by Dexter Hoyos
Justin’s showing, he was sensible and well informed.22
Orosius, a Spanish church presbyter oppressed by the disasters befalling
the now Christianized empire early in the fifth century, put together a seven-
book world history to show carping pagans that these were nothing
compared to the calamities of past times. He deals with the period of the
first two Punic wars in 20 pages, much of them devoted to pious lesson-
drawing. His account follows standard Roman tradition, for instance on
Hannibal’s oath, and goes back, probably indirectly, to Livy.
Briefer sources need mention too. Anecdotal compilations include Valerius
Maximus’ nine books (from the reign of Tiberius) of famous deeds and say-
ings by Roman and foreign leaders, in which Hamilcar, his sons and even
Hasdrubal son of Gisco contribute mostly well-known items. Julius Fronti-
nus, ex-governor of Britain and City water-services commissioner, late in the
first century AD put together four books of anecdotes on military stratagems,
which for Roman history seem to draw much from Livy and naturally offer
some items from the Barcid era. A generation or so later a skilful rhetorical
writer, one Florus, boiled down Livy’s and others’ accounts of Roman wars
to Augustan times into a fluent two-book compendium, occasionally useful if
also prone to error and exclamations. In the 160s AD a Greek rhetorician,
Polyaenus, tried his hand too at stratagems-collection, of minimal use to
Barcid history.
At some period, probably the late empire, an anonymous person compiled
a series of one-paragraph biographies of notable men in Roman history (and
three women including Cleopatra) down to the end of the Roman Republic.
221
S O U RC E S
Hannibal and some of his Roman contemporaries figure among them, but
the entries are merely skeleton résumés—which does not save them from
various careless mistakes. Unlike Aurelius Victor’s similarly compacted lives
of the Roman emperors, very little of independent value can be found in De
Viris Illustribus. It stands as a sobering illustration of how not to compress.23
A few other writers supply miscellaneous information. The Augustan-age
writer Strabo in his world geography describes not only the topography but
also the ethnography of Spain, Italy and Africa as well as other lands; Pliny
the Elder’s encyclopaedia contributes a great variety of individual items
(it mentions the rich silver mine at Baebelo, for instance); a hundred years
later the Greek Pausanias wrote a travel account of Greece with one or
two bits of relevant information, and useful too is Ptolemy’s Geography, a
catalogue of countries, peoples and cities. And late Roman itineraries—
catalogues of major routes with their towns and distances—have their uses,
as does the epitome of Stephanus of Byzantium’s sixth-century geographical
encyclopaedia.
222
A P P E N D I X
Special notes
1 Hamilcar’s daughters, and other family questions
(chapter II, note 2)
Appian, Hann. 20.90, mentions a military commander Hanno as Hannibal’s
nephew (kinship doubted by Lenschau, RE 7.2357); he was son of Bomilcar
‘the king’ (Pol. 3.42.6), so if Appian is right about the kinship it would mean
that his mother was a sister of the general. A daughter promised by Hamilcar
to Naravas in about 240 (the original of Flaubert’s Salammbô): Pol. 1.78.8.
Livy 29.29.12 mentions ‘a Carthaginian noblewoman, daughter of Hannibal’s
sister’, who by 206 was widow of another Numidian prince, Oezalces: she
may have been a sister to Hanno son of Bomilcar (cf. chapter XIII).
Seibert suggests that Hannibal may not only have been reared, but even
have been born, in his father’s Sicilian camp ( Hann. 9 note 12). But this was
hardly the place for a baby and its mother—still less so after the move to Eryx
in 244, when a second son probably arrived too. Hasdrubal’s birth-date: Diod.
25.19, lines 9–10 (actually from the Byzantine versifier John Tzetzes (chapter
V note 15)) describes him as 12 and Hannibal as 15 in late 229, which is wrong
arithmetically but may be right on the difference between their ages.
Seibert, Hann. 20, notes too that, according to Val. Max. 9.3 ext. 2 and Cas-
siodorus, Chronica, Anno Urbis Conditae 524, Hamilcar had four sons and—as
a fourth never appears in history—infers that the newborn fourth formed
part of a major infant-sacrifice which, he further infers, took place during the
Mercenaries’ War. But Valerius describes all four as Hamilcar’s lion-cubs
being raised to destroy Rome. It is likelier that the numeral is due to a fuzzy
awareness that Barca had more children than just three sons. Or the numeral
in these texts may be mistaken: in Valerius’ text the preceding word is
‘odium’, which could have corrupted a ‘iii’ to a ‘iiii’ or ‘iv’—and Cassiodorus’
item seems to come from Valerius.
2 The artificial ports at Carthage (chapter II, note 5)
Nearly all the datable remains are from the second century BC, hence the
excavators’ view that the ports were built in Carthage’s final years. But those
223
A P P E N D I X
years may mark refurbishments and repairs (H. Hurst, Antiquaries’ Journal
(1979) 27; and in Atti del I Congresso Internazionale di Studi Fenici e Punici . . . 1979
(Rome 1983) 2.609). To start so colossal a project in an era when the
Carthaginians were forbidden by treaty to have a navy, and were terrified at
any prospect of clashing with the Romans (then the only major power with
one), was pointless expense. True, Livy reported Roman complaints about
Punic warship-building then ( Epit. 47, 48, 49), but ships alone do not a port
make, and even the ships may have been anti-Punic propaganda.
True again, there is an almost complete lack of late fourth- and third-
century material—only a third-century coin of Tarentum and one struck at
Carthage in the period 241–221 or just after (Hurst, Antiq. J. (1979) 27)—but
a late third-century date is possible enough, most probably the first decade of
the Second Punic War (cf. chapter VIII §IV). It does make more sense for the
ports to date to an era of Punic naval power (so too Seibert, FzH 111–13).
Seibert suggests the interwar years of 241 to 218; but when the Second Punic
War began Punic naval strength was woefully low (chapter VIII §II). By con-
trast, Romans raiding the Punic coast in 210 learned from prisoners that the
Carthaginians were readying a massive fleet (Livy 27.5.13). True, Punic naval
forces after 210 never exceeded 70 to 100 ships (cf. Lazenby (1978) 197), but
if the naval port was being or had recently been constructed the Carthagini-
ans may yet have had hopes. Refurbishments one or two generations later
could point to both ports then being used for Carthage’s prosperous marine
commerce.
3 Carthaginian revenues (chapter II, notes 10–11)
(i) Tribute of ‘Lepcis’ in 193 (Livy 34.62.3): Lepcis, later Lepcis Magna, was a
big and flourishing Phoenician colony, but 360-plus talents a year would be a
colossal sum to pay
in tribute. Yet Lepcis certainly lay in an area called Empo-
ria as Livy states (rightly, despite W. V. Harris, CAH 2 8.145, and map, ibid. 144;
see chapter II note 10) and had prosperous neighbours, Oea and Sabratha.
Leptis Minor in its turn was not by the Lesser Syrtes but at the southern
end of the gulf of Hammamet further north; nor was it as important as
Lepcis, so a daily talent of tribute is unthinkable from it alone. But though
Livy clearly singles out ‘Lepcis’ ( or ‘Leptis’) itself as paying the talent—‘una
civitas eius [sc. regionis] Lepcis; ea singula in dies talenta vectigal
Carthaginiensibus dedit’ (34.62.3, where ea can hardly refer back to the
understood regionis)—this may be a misreading of his source: cf. Briscoe
(1981) 143–4; Mattingly (1994) 50. Leptis Minor did stand close to the fertile
Emporia region south-east of Carthage, the later Byzacium (Pliny, NH 5.24).
Moreover Masinissa in 193 coveted Emporia—this is in fact why Livy men-
tions the region and ‘Lepcis’—a lust that fits the area adjoining his kingdom
better than territory far away to the east, even if that region counted as part
of Emporia.
224
A P P E N D I X
Briscoe (1981) 143–4 is probably right then to suggest that Leptis Minor
served as the administrative centre for Emporia (Hadrumetum and Thapsus
were more important cities, but were independent allies of Carthage). Cau-
tiously then the daily talent can be taken as paid by Emporia as a
whole—even perhaps including Lepcis Magna and its neighbours far to the
east.
(ii) Carthage’s Spanish revenues: We have only one indicator, the output of
a silver mine at Baebelo—site unknown—which yielded Hannibal 300
pounds daily according to the Elder Pliny (33.97). In Roman denarii this
would be about 25,000 a day, or just over 4 talents (cf. Frank (1933) 47). But
that would make Baebelo by itself produce nearly 1500 talents a year; more
likely Pliny’s figure is exaggerated, optimistic or flawed. Perhaps the whole of
the Spanish empire by 218 paid 1,500 a year—much of it no doubt spent in
the province. Incidentally the silver mines near New Carthage in the second
century BC likewise yielded 25,000 drachmas/ denarii a day ‘to the Roman
people’, according to Polybius (cited by Strabo 3.2.11, 148C).
Whatever the wealth of Spain, Kahrstedt’s estimate ((1913) 135–7) of
Carthaginian income in 218 of up to 6,000 talents seems wildly optimistic,
and not much less so the 2,800 a year he supposes for 200–191.
4 Carthaginian population in the third century
(chapter II, note 12)
Strabo 17.3.15, 833C, reports ‘seventy myriads of men [i.e. persons] in the city’
in 149. K. J. Beloch estimated 200,000–300,000 in the city ( Die Bevölkerung der
griechisch-roemischen Welt (Leipzig (1889) 467); Kahrstedt 125,000–130,000 in 218,
including 20,000 non-Carthaginians ((1913) 23–4, 133). Warmington (1964)
150 reckons some 400,000, including slaves and foreign residents; Picard (1961)
61 about 100,000 for the city proper and 100,000 for Megara. Tlatli reckons
243,000 all told in the city and Megara, plus some 400,000 in a territory of
18,000 square kilometres ((1978) 107–9, 117–18, 124), perhaps rightly, but that
area is too large—it includes cities like Utica, Hippo and Mactar which did not
belong to Carthage’s own city-territory. Huss (1985) 51 accepts Strabo’s figure,
whereas Ameling (1993) 205–6 holds to a broad range of 90,000–225,000.
Two hundred thousand citizens, with their families, for city and territory
together can only be a rough estimate, but the total of men, women and chil-
dren would then range between 571,000 and 714,000, depending on what
percentage was male in the total population (probably between 28 and 35 per
cent: cf. Brunt (1971) 59, 116–17). Additional to these would be any Libyans
dwelling in Carthage’s own territory, plus resident aliens, and slaves—surely
another 100,000 at least.
The Libyans outside the city’s territory, the North African allies like Utica
and Hippou Acra, and their slaves, are a separate matter. For the city and
empire in 218 Kahrstedt’s detailed estimates come to about 2.1 million apart
225
A P P E N D I X
from the recently won Spanish territories ((1913) 133), but include implausi-
bly low estimates or guesstimates for the city of Carthage (above) and for the
Libyan subjects (650,000); Ameling (1993) 225 likewise estimates the Libyans
at under a million. If instead there were 700,000–800,000 people in Carthage
and its own territory, plus 1.5–2 million Punic allies, subject Libyans and their
slaves, and 1.5 million in Hasdrubal’s Spain—quite possibly too low a guess,
Kahrstedt thinks 2 million—then the total population under Barcid leader-
ship around 221 would be 3.7 to 4.3 million. For comparison, Rome and Italy
together Brunt estimates at 3.5 million in 225 ((1971) 59–60).
5Naravas’ family (chapter III, note 7)
Naravas seems to be the Nrwt mentioned on a Libyan inscription set up in
129 or 128 BC near Mactar by Nrwt’s grandson Wlbh (Picard (1966) 1257–65;
Huss (1985) 60 note 65). It names Nrwt’s father as Zililsan, known from an
inscription of 140 or 139 at Thugga as father of Gaia (G. Camps, Masinissa ou
les débuts de l’histoire [= Libyca 8 (1960)] 283). This is the royal family of the
Massyli in eastern Numidia. If the identification is correct, Naravas was
brother of Gaia, whose famous and long-lived son Masinissa was born
around 240. Another brother, Oezalces, late in the Second Punic War mar-
ried a daughter of one of Hannibal’s sisters (Livy 29.29.12): see chapter XIII.
6 Massiliot colonies in south-eastern Spain
(chapter V, note 12)
Strabo (3.4.6, 159C) mentions the three Massiliot ‘little towns’ (πολ´ιχνια) in
the present tense (’εστ´ιν), which itself rules out Acra Leuce, a ‘very large city’
and a Punic one from Hamilcar’s day, being one of them. He sites them ‘not
far from’ the Sucro river (modern Júcar) and names one: Hemeroscopeion,
the Roman Dianium, which apparently was at modern Denia by Cape de la
Nao, 36 miles/60 kilometres south of the Sucro (Rouillard (1982) 427).
Stephanus lists an Alonis (s.v.) as a ‘Massiliot city’, and other evidence places
this near Alicante (Mela 2.93; Itin. Rav. 304), but it does not count very obvi-
ously as one of Strabo’s three. If all three were near to or south of Cape de la
Nao, Strabo’s phrase loses all meaning; why not describe them instead as ‘not
far from’ the cape?
7 The Saguntines’ fractious neighbours
(chapter VII, note 9)
Livy calls them Turdetani and Turduli, Appian Torboletae or Torboletes, but
none of these suits the region. By contrast the town of Tyris or Turis, at the
mouth of the nearby river still called the Turia, existed until the first century
BC and is altogether apt (Hoyos [1998] 188–9). In Latin its people would be
226
A P P E N D I X
called Turitani; in Greek ‘Tyrieis’,‘Turietae/-tes’, ‘Tyristae’ or the like. The
‘Thersitae’
who, along with the Olcades and others, supplied troops to North
Africa early in 218 (Pol. 3.33.9) may again be these people in yet another Gre-
cized form; the Thersitae are otherwise unknown and efforts (Walbank
1.362; Huss (1985) 297) to identify them as Tartessii or Turdetani are less
plausible.
8 The grand army at the start of 218
(chapter VIII, note 15)
On the expeditionary army, recent discussions include Walbank 1.366;
Lazenby (1978) 33–4; Scullard, CAH 2 8.40; and Lancel, Hann. 103–4, who all
doubt Polybius’ starting total; so does Seibert, FzH 179–83, suggesting
70,000 infantry and 10,000 cavalry. By contrast Barreca (1983–4) 44–5
defends Polybius’ total, and Goldsworthy (2000) 154, 158–9 accepts it with-
out discussion. No one suggests that Polybius’ figure may include the 15,000
men Hannibal assigned to his brother Hasdrubal.
Livy’s details about the disillusioned Spanish troops (21.23.4–6) are fuller
than Polybius’ and look well based, perhaps from a Hannibalic source like
Silenus—directly or via Coelius. If we suppose a fighting loss of 7,000 in
north-eastern Spain, which seems plausible if losses were serious (8 per cent
of an expeditionary total of 87,000), this plus the dismissed troops and
Hanno’s corps would add up to 28,000. Hannibal would then indeed have
had 59,000 to lead over the Pyrenees.
Interestingly the Roman prisoner of Hannibal’s, L. Cincius Alimentus (cf.
Hoyos (2001a) 78), who conversed with the general around 206, later wrote a
history of Rome and estimated the Punic army at 80,000 foot and 10,000
horse on its arrival in Italy (Livy 21.38.3–4). Perhaps he misremembered a
rounded-up estimate of Hannibal’s of the original numbers departing for
Italy. The figure certainly does not refer to Hannibal’s strength after the Boii
and others joined his army in north Italy, for he still had only some 40,000
men at the Trebia: see §9 below.
9 Hannibal’s route to Italy and numbers on arrival: some
views (chapter VIII, notes 23–4)
(i) On the general’s constantly discussed route recent studies include Proctor
(1971); Lazenby (1978) chapter II; Connolly (1981) 153–66; Huss (1985)
298–306 with 298 note 35; Seibert (1989), also FzH 191–200 (with a thor-