by Dexter Hoyos
of faults—and despite laying responsibility for the Second Punic War on
both of them. But he is not interested enough in the Punic state or commu-
nity to offer even an outline of Carthage’s political system (a contrast to
Book 6 on Rome’s) or name any notable Carthaginian except military com-
manders: that Gisco was the name of the senator whom Hannibal summarily
shut up in 203 we learn from Livy, while the sufete who spoke for Carthage in
218, in the fateful confrontation with the Romans’ war-embassy, remains
anonymous. Again, his comment that by 218 the Punic political system was
tilted towards democracy is better than no comment at all, but still leaves us
to make what we can out of Aristotle’s sketch in the Politics plus occasional
items like Hasdrubal’s and Hannibal’s election as generals, and Livy’s brief
mention of the postwar ‘order of judges’.11
All that being said, Polybius has unique strengths. He consulted pro-Punic
accounts like Hannibal’s and Sosylus’ as well as Roman ones, and was not
over-trusting of either side (though at times hypercritical). In spite of his
Roman connexions he is not prepared to give the Romans the benefit of the
doubt every time: certainly not, for instance, over the rape of Sardinia in 237,
while—cautiously but unmistakably—he sets out his opinion that the
Carthaginians were justified in going to war in 218. In a different vein, his
description of the special horrors of a Roman city-sacking is the more chill-
ing for being matter of fact.
The Hellenistic equivalent of a Renaissance man, Polybius not only was a
leader, diplomat and organizer in his own career but also contributed to mili-
tary science (improved fire-signalling methods, for example), travelled widely,
inspected sites and terrain, looked up documentary sources like Hannibal’s
inscribed memoir and the treaties between Rome and Carthage, and inter-
viewed eyewitnesses (King Masinissa for instance). All these aspects show up
in his history. On military, geographic and chronological matters he is a great
deal more knowledgeable than our other sources, and he takes time out to
describe and discuss them. Not that this does guarantees clarity every time—
most famously over Hannibal’s route from the Rhône to Cisalpine Gaul, but
also for instance over the chronology and topography of the Mercenaries’
War—but it is important that he realizes the need. In the same way his
interest in explaining psychology and motivation is valuable even though not
216
S O U RC E S
every explanation may convince: thus, famously, his theory of Barcid hostility
and war-planning against Rome.12
III
Well after Polybius came the next two surviving writers, Diodorus and Livy,
both of the later first century BC. Diodorus’ Historical Library, a 40-book
world history, offers only excerpts from its second half, mostly selected by
medieval Byzantine copyists. A Sicilian himself, he devoted substantial space
to the island’s history and for the First Punic War drew on his fellow-Sicilian
Philinus as well as on Roman sources. In turn, his few excerpts on the Merce-
naries’ War match Polybius’ account so closely that efforts from time to time
to thrust some intermediary fount between them remain unconvincing. By
contrast, his invaluable few paragraphs on Hamilcar’s and Hasdrubal’s doings
in Spain draw on some other writer, well informed, detailed and Barcid-
friendly: it is tempting to think of either Silenus or Sosylus.
Diodorus’ modus operandi, much of the time anyway, was to rewrite or
abbreviate his source of the moment with few substantial changes; in other
places (like his account of the fateful year 264) he uses more than one source
but seems to be faithful to each in turn. This not very original way of writing
history makes the lack of his full text the more deplorable, for that might
have given a better view of what some of the pro-Punic sources were like.13
T. Livius, his younger contemporary, started his history of Rome around
30 BC and outdid all his predecessors in length (142 books—90 per cent of
them on the centuries from 272 on) and in artistry. Of the 35 books surviv-
ing, Books 21 to 30 record the Second Punic War; then Books 31 to 39
provide most of the details we have of Hannibal’s postwar fortunes, includ-
ing his year as sufete. Like Diodorus, Livy is heavily dependent on his
sources, but with much more reshaping and recasting of his finished narra-
tive. As mentioned earlier he often draws on at least two accounts to create
his own blend, something readily seen when one account is Polybius’. For the
later third century and the first half of the second, besides Polybius he draws
chiefly on Roman predecessors, with the already-noted partiality for Claudius
and Valerius; but—at least by his own claim—makes use from time to time of
a wider selection, including Fabius and Cincius. Whether he directly con-
sulted all such works or merely reproduced citations made by a narrower
group (Coelius and the late annalists have been urged) we do not know.
Livy follows Polybius in his praises of Hannibal’s leadership and continues
to be interested in—and sympathetic to—the general after 201. Nevertheless
he makes it clear that the Romans were provoked into the war and had the
moral fibre to win it. Literary grandeur apart, his value lies not just in basing
his narrative partly on Polybius but in supplying detailed accounts of Roman
affairs of every kind, materials that his industrious predecessors amassed
from older records like the annals of the pontifices. Industrious himself (the
217
S O U RC E S
history took him about four and a half decades), he lacked direct experience
of politics or warfare or the world outside Italy—rather like Polybius’
favourite bête noire Timaeus—and could apply no great independence of
judgement to issues. Therefore he reworks what his sources tell him: and so
preserves a wealth of unique information, along with rather numerous mis-
takes and confusions. Now and then, he pauses to summarize the differing
opinions among his sources that are causing him trouble (Hannibal’s Alpine
route and army losses, Fabius the Delayer’s exact legal position vis-à-vis his
deputy Minucius late in 217, Hannibal’s route to Rome in 211, Scipio’s booty
from New Carthage). Patriotism plays its part too, for instance to distort the
Ebro accord with Hasdrubal and import romantic tales about Scipio
Africanus in Spain and Africa.
One result of Livy’s methods and limitations is extensive scholarly scepti-
cism about many items that lack Polybian corroboration and smell instead of
annalistic inventiveness, or seem to. Minucius’ self-abasement after being res-
cued by Fabius the Delayer from annihilation is one uncontroversial example,
and so too is (or should be) the Senate’s rejection of the peace negotiated in
203 by Scipio in Africa. The non-Polybian details reported for the elder Scip-
ios’ operations in Spain down to 211 are, some or all, dismissed as annalistic
fictions too, which (on the other hand) m
ay be oversceptical; likewise, on a
smaller scale, the episode of Centenius Paenula in 212. Such controversies
make it necessary to read Livy, even more than Polybius, with care.14
IV
Sizeable narratives of operations in both Punic Wars are provided by the late
second-century AD Appian of Alexandria, in his history of Roman wars laid
out by geographical regions: in particular the books Hannibalica, Iberica and
Libyca. They are very uneven work. On second- and first-century wars
Appian is generally sound: his telling of the Third Punic War in Libyca, for
instance, is at least partly based on Polybius and puts no strain on a reader’s
credulity, nor does the lengthy account of the wars against Mithridates of
Pontus. In glaring contrast, his accounts of the various theatres in the
Second Punic War (those for Italy, Spain and Africa survive) call for constant
caution.
It is not only that Appian fails to tell apart the Ebro and Tagus rivers, sites
Saguntum north of the former, and still manages to identify it with New
Carthage. His narratives teem with items missing from any other source and
often incompatible with those in Polybius or Livy. Events in Africa between
204 and 202 are thus enlivened, not to mention a good many of Hannibal’s in
Italy. The Appianic battle of Cannae, for instance, is nothing like the standard
version (among other things, he transfers to it Hannibal’s stratagem at the
Trebia—troops hidden in a ravine to ambush the enemy’s rear). Hasdrubal’s
Ebro accord in Appian has provisos more plentiful and less believable than in
218
S O U RC E S
any other source. Again, the elder Scipios’ disastrous last campaign as Appian
tells it is at odds with Livy’s more thorough account.
Sometimes Appian does it the other way round, for example telling us
nothing of the important operations (including at least one big victory over
Hasdrubal brother of Hannibal) that Cn. Scipio carried out in 218 and 217
before his brother joined him. The younger Scipio’s operations in southern
Spain from 208 to 206 are run together and one battle fills in for both Baecula
and Ilipa. Highly rhetorical speeches and occasional grotesque dramatizations
(notably the hand-to-hand combat between Hannibal and Scipio at Zama)
add extra colour and disbelief.15
All the same Appian cannot be ignored, even if he has to be used with care
and some scepticism. Not all the details that he alone offers are incredible.
Once the Tagus is substituted for the Ebro, for instance, his description of
Hasdrubal’s sway in Spain becomes plausible. Like other later writers he
makes a Roman propaganda mess of the Italian traders episode during the
Mercenaries’ War and then the seizure of Sardinia; but he may well be right
that the Romans allowed the Carthaginians to recruit soldiers in Italy during
that war. A good deal of his narrative does, moreover, coincide with other
sources’ versions. Hannibal’s oath is in the standard Roman version, as
against Polybius’ (and Nepos’); his army-strength on leaving New Carthage
matches Polybius’ figures; and the Appianic battle of Trasimene is largely the
same as in Livy. Later on, like Livy and Diodorus, Appian too makes Hanni-
bal massacre his reluctant Italian soldiery before embarking for Africa, not
that this adds any probability to the tale.16
His treatment of the general, the only Barcid dealt with at length, is censo-
rious in places. Political calculation and glory-hunting prompt Hannibal to
start the war, he slaughters prisoners at times (plus the recalcitrant Italians),
victimizes the family of a treacherous Italian ally, and as his fortunes start
their decline he takes a winter holiday in Lucania with a mistress. Yet Appian
gives him his due as well—mild treatment of Italian captives after Trasimene,
Cannae lauded as a splendid victory, honours for the fallen commanders
Gracchus and Marcellus, later on in Africa a (quite improbable) effort to
make a treaty with Scipio which his fellow-citizens abort, and after Zama
more wise counsel to the ungrateful Carthaginians about accepting Scipio’s
terms. This varied depiction again mirrors the ambiguities of the standard
Roman tradition towards their most admired enemy.17
That Coelius was his basic source, at least for the Second Punic War, has
been suggested. But Appian’s way of so often combining factual material
with fanciful, and reporting items that Livy—who did use Coelius—contra-
dicts or ignores, points to a far more eclectic and uncritical blending of
borrowings, not to mention a strong dash of carelessness. This means that
Appianic items on their own cannot be automatically treated as sound evi-
dence merely omitted by other writers: scrutinized carefully, some may pass
muster while others have to be left in doubt or be rejected. Overall his
219
S O U RC E S
treatment of the period suggests that he preferred annalistic sources (a great
contrast to his account of the final Punic war) and put them together fairly
uncritically—not to mention ignorantly at times.18
V
Other sources are less extensive. In the later first century BC Cornelius Nepos
included a three-paragraph biographical sketch of Hamilcar and a lengthier
one (13 chapters) on Hannibal among his collection of lives of distinguished
foreign generals. As far as they go these are useful at least for various
details—we owe him our knowledge of Silenus and Sosylus being in Hanni-
bal’s entourage—and, interestingly, he gives as much space to Hannibal’s life
after the war as to all his earlier career. Except for the general’s suicide, these
later details are not matched in Livy or other major sources, but one or more
turn up in other lesser writers like Justin. At the same time, compression (like
covering the years from 216 to 203 in one paragraph) and various errors or
confusions (he places the events of late 217 after Cannae, and misunder-
stands Hannibal’s postwar position) limit what he has to offer.19
The philosopher and biographer Plutarch, of the late first and early second
centuries AD, wrote no biography of a Carthaginian and his life of Scipio
does not survive. Those of Fabius the Delayer and the pugnacious Marcellus
treat largely of their subjects’ doings in the Second Punic War. They con-
tribute to the record of events, usually from Roman sources (Livy included),
but the occasional sayings and anecdotes of Hannibal that Plutarch
includes—like the jest to Gisco before Cannae—must derive directly or indi-
rectly from a Punic source, perhaps Silenus or Sosylus again.
The third-century AD consular historian L. Cassius Dio treated the Punic
Wars period at some length in his 80-book general history of Rome. But all
that remains of Dio before his Book 36 (the late Republic) is a collection of
excerpts, mostly short, and the sizeable epitome of his work that the retired
Byzantine administrator John Zonaras made in the eleventh century. Dio’s
history rests almost entirely on Roman and pro-Roman sources (though he
claims he read every work relevant to his theme, which for this
period ought
to include Sosylus and Silenus at least). He relays, for example, the story that
Roman envoys went to Hannibal during the siege of Saguntum; has Hannibal
after Trasimene advance down the Tiber valley towards Rome; reports
Maharbal’s advice to him after Cannae as Livy does and narrates the Metau-
rus campaign much again as in Livy. Likewise he makes the Romans refuse to
ratify Scipio’s peace-terms in 203 until Hannibal has departed Italy.20
There is also Silius Italicus. An eminent ex-consul in the later first century
AD, he composed a lengthy epic poem, Punica, on Hannibal’s war—the
longest epic poem in Latin and the least inspired. A collection of episodes
told in standard epic form and heavily imitative of Vergil, it offers plenty of
sonorous names and similes, Homeric-style combats (including between
220
S O U RC E S
gods, not to mention Scipio versus Hannibal again), and Scipio—who we
learn is really the son of Jove—at one stage summoning up the ghosts of
past Roman heroes and heroines for consultations. The poem uses Roman
historical sources, notably Livy, and adds no special information of its own
apart from naming Hannibal’s wife and giving them a son, both of which
items may or may not be accurate. An effort has certainly been made to see a
pre-Livian source behind Silius’ telling of the war’s preliminaries, so as to use
him as a foil to the supposed bias and distortions in Polybius and Livy, but it
is hard to find this plausible.21
Late Roman historical writers add only incidentally to the register.
Eutropius, a retired senior administrator in the later fourth century who com-
posed the most concise of Roman histories for the presumably easily
distracted emperor Valens, summarizes the entire period from 264 to 201 in
ten pages. His most useful contribution (not always believed) is the date of the
battle of the Aegates islands in 241. The much more useful Justin, probably
again fourth century in date, epitomized a lengthy history of the non-Roman
world by the Augustan-age writer Pompeius Trogus, with valuable if concise
information about Carthage’s earlier times and equally abbreviated references
to Barcid doings—mostly on their Spanish conquests, Hannibal’s exile and his
dealings with Antiochus III. That we lack Trogus’ full work is a pity since, on
