by Dexter Hoyos
probably a copyist’s slip: the name Hanno is repeated at Pol. 3.76.5 (Livy, 21.23.3
and 60.5, follows Polybius). Sound remarks on the strategic insignificance of
Spain-beyond-Ebro by de Sanctis (3.2.9–10), though Lazenby argues for its sup-
posed importance to Hannibal’s communications ((1978) 33), and Huss
implausibly sees its conquest as necessary to a successful crossing of the Pyre-
nees (298–9). Emporiae as Roman bridgehead: Pol. 3.76.1; Livy 21.60.2.
10 Spanish rivers, Hallward (1930) 36; Warmington (1964) 212; Lazenby (1978) 33;
Scullard (1980) 204. Food availability, Lazenby, ibid. Regrouping the army before
the Pyrenees: Proctor (1971) 45. Lulling the Romans: de Sanctis, 3.2.8–10;
Scullard, ibid. Consular year began on 1 March (previously in May) probably from
222: de Sanctis, 3.1.107; Broughton, MRR 2.638–9; Rich (1976) 19. Sempronius
set out before Scipio, probably in June/July: Hoyos (1998) 258–9, citing earlier
studies.
11 Eight hundred stadia in ten days: Pol. 3.50.1; treated as Hannibal’s more-or-less
average marching rate by Lazenby (1978) 35, 275; as below average by Proctor
(1971) 26–7, 29. Roman imperial armies averaged 23.7 kilometres (16 Roman
miles) a day, Proctor, 31–2; though A. K. Goldsworthy, The Roman Army at War
(Oxford 1996) 109–10, stresses such rates were under optimum conditions only.
12 Caven (1980) 98–101, suggests this less likely scenario; believed by Santosuosso
(1997) 170. Punic spy: Livy 22.33.1; Dio in Zon. 9.1.1. Hoffmann (1957/1974)
55–6, (1962) 48, stresses the danger to Carthage if the Romans invaded North
Africa.
13 Cavalry skirmish: Pol. 3.45.1–3; Livy 21.29.1–4. Hannibal’s decision: Livy
21.29.5–6; disbelieved by Lazenby (1978) 37; in fact most or all scholars suppose
he was anxious to avoid battle with Scipio entirely. His expectation of Scipio
returning, cf. Pol. 3.61.1–4 (claiming his later surprise at the consul’s swiftness).
On arbitrary grounds Walbank, 1.395–6, disbelieves this report.
257
N O T E S T O T H E T E X T
14 On dating the artificial ports see chapter II note 5 and VII §I.
15 Cape Lacinium numbers: note 6 above. 102,000 troops: Pol. 3.35.1; Livy 21.23.1;
Appian, Hann. 4.13. Hannibal suffered ‘great loss’, gave Hanno 10,000 foot and
1,000 horse, and sent home ‘the same number’ (3.35.3–6; malcontent Carpetani
and others, according to Livy 21.23.4–6). We should not assume these latter
included many horsemen. For more details about the grand army see Appendix
§8.
16 Syrian and Egyptian forces in 217 (68,000 versus perhaps 55,000 (Polybius claims
75,000)): Pol. 5.65.1–10, 79.2–13; Walbank, 1.589–92, 607; P. Green, Alexander to
Actium: The Historical Evolution of the Hellenistic Age (Berkeley and London 1990)
289–90. Roman land forces in 218 totalled 27,000 citizens and some 44,000 allies,
according to Brunt’s careful calculations ((1971) 417–19, 678).
17Consuls of 219 as envoys: Seibert, Hann. 83; Hoyos (1998) 234–5; cf. Golds-
worthy (2000) 145 (sending them to Carthage in 219 as consuls!); disbelieved by
Vollmer (1990) 137–8. It is not clear whether the Roman calendar was in accord
with the true solar year (Seibert, FzH 346–52; Hoyos, 235), but any discrepancy
looks likely to have placed it ahead—so Idus Mart. 218 might actually have been
early March or late February. Envoys demand handover, Pol. 3.20.6–8 (see chap-
ter V note 5); war let fall, 33.2–4. Visit to Melqart’s temple: Livy 21.21.9; Huss
(1985) 235, speculates about what Hannibal did there.
18 Monomachus: Pol. 9.24.5–8 (cf. Livy 23.5.12–13, Roman propaganda); cf. Wal-
bank, 2. 153; Seibert, Hann. 111 note 180. The dream: Cicero, De Divinatione 1.49,
citing Silenus via Coelius Antipater; Livy 21.22.5–9 (does not mention Hannibal
being summoned before the gods); Dio, in Zon. 8.22.9; Val. Max. 1.7. ext. 1 and
Silius 3.163–216 rephrase Livy. Cicero dates the dream simply ‘after the sack of
Saguntum’, but ( contra Seibert, FzH 184) this hardly contradicts Livy’s dating it to
the march. Dio both gives a version closer to Silenus’ than Livy’s—adding other
fabulous details perhaps again from Silenus’ propaganda (cf. Polybius’ strictures,
3.47.8, 48.7–9)—and also implies the dream occurred during the march. Barceló
(1998) 47 impossibly imagines that in Livy the dream foretells Hannibal’s ruin
and Carthage’s defeat.
19 Aerenosii, etc.: Pol. 3.35.2; Livy 21.23.2. Indibilis ‘always’ pro-Punic: Pol.
3.76.6–7; Hoyos (1998) 183. His close links to the Lacetani and others: Livy
25.34.6; 28.24.4 and 26.7; Walbank, 1.410. Ilergetes of the interior: Strabo 3.4.10,
C161; Walbank, 1.366. Note the Ilergetan cavalry squadron already in Hannibal’s
army: Pol. 3.33.15; Livy 21.22.3. Ausetani and Lacetani as Punic allies, 21.61.8—
though this episode probably followed Cn. Scipio’s victory at Cissis, 60.1–9.
Coastal Ilergetes: Pliny, NH 3.21; Jacob (1985) 252, noting the ‘Ilaraugatae’ of
Hecataeus, frg. 14; Hoyos (2001b) 69–70. Just possibly the Bargusii were around
the later Barcino (Barcelona) for, with the pro-Punic Lacetani and Ausetani hem-
ming them in, friendliness towards the Romans would be natural; though
Schulten ( FHA 3.47) prefers them well inland, on the northerly reaches of the
Llobregat around the town of Berga. Troops sent home: note 15 above.
20 Hannibal won over the Gauls, Livy 21.24; earlier conciliation, 21.20.8; some
forced to give passage, Pol. 3.41.7. According to Polybius he had been ‘very fear-
ful’ of possible resistance to the army crossing the Pyrenees (40.1), which is odd as
Polybius fails to mention any such fear when actually narrating the crossing (35.7).
21 Army size at Rhône, Pol. 3.60.5. Garrisons in southern Gaul: Picard (1967)
163–7; Seibert (1989), FzH 182–3, 193, 212, Hann. 98; contrast Goldsworthy
(2000) 167. Seibert supposes that reinforcements later used this route from Spain
to Hannibal until 207, though he concedes that the one force recorded (Bomil-
car’s in 215) went from North Africa.
258
N O T E S T O T H E T E X T
22 Napoleon’s losses in the first months of his Russian invasion: Chandler
(1965/1993) 754–5, 781–2; C. Duffy, Borodino (London 1972) 51, 62, 161. The
destruction of the army in the retreat from Moscow actually involved the loss of
fewer troops.
23 A lucid chronology of the entire march in Lazenby (1978), Appendix iii; earlier
analysis in de Sanctis, 3.2.77–81 (who dates the march from April to September
rather than May/June to October/November). The one clear indicator of time is
that the army reached the summit of the pass close to the astronomical setting of
the Pleiades (Pol. 3.54.1; Livy 21.35.6), meaning early November. Though often
treated as a loose or approximate datum (e.g. by de Sanctis and by Walbank,
1.365–6, 390) it is convincingly defended by Proctor (1971) 13–15, 40–5, 75–82;
Lazenby (1978) 29, 32–3; and Seibert, FzH 176–8. In turn its final stages, from
the Rhône on, can be worked out almost day by day from Polybius’ account, as
Lazenby shows. Losses inflicted by Allobroges et al.: Pol. 3
.51.7, 53.1–3; cf.
51.3–5 (animals lost). Losses on the descent, 3.54.4, 56.2; army’s spirits damaged,
3.54.1–2 and 7.60.3–4. Plains of Italy: 3.54.2–3; Livy 21.35.7–9. On Hannibal’s
impossible-to-identify route cf. Appendix §9.
24 Army strength after reaching Italy, Pol. 3.56.4 (‘fewer than twenty thousand’
(2.24.17) is a careless misstatement or maybe a copyist’s mistake (δισµυρ´ιων writ-
ten for τρισµυρ´ιων)). The mention of the Insubres’ territory (56.3) is generally
seen as just a loose Polybian reference (Walbank, 1.392; Huss (1985) 306 note
82), but this is not obvious. Nor, contra Seibert ( Hann. 106 note 163; cf. previous
note), need it reflect a source different from Livy’s, who has Hannibal arrive
among the Taurini (21.38.5). Dio does report desertions during the Alpine cross-
ing (Zon. 8.23.6), though he may draw this either from a source or from personal
inference. Balearic slingers: those involved in his troop-movements between
Spain and Africa constituted about 1,400 in a total of 31,000 (Pol. 3.33.12–16);
even if he himself took another 1,000 with him, their numbers too must have
thinned. On some other reckonings see Appendix §9.
25 Pol. 3.60.8–13. Bender (1997) 98–105, disbelieving in any advance soundings by
Hannibal, thinks that the Gauls had rebelled because the Romans founded the
colonies Placentia and Cremona in their midst, and came over to him only once
he showed his military strength by smashing the Taurini. But thus to dismiss
Polybius’ and Livy’s reports of the earlier soundings is unconvincing. It was
common sense for the Gauls to put off giving the invader any help until he
reached their territories—which he did after the Ticinus skirmish (3.66.7, cf.
67.6), before his big victory at the Trebia.
26 On the near-disaster of the march, Hoyos ((1983) 171–3; Shean (1996) 175–80;
Bender (1997) 98–9), improving on Hoffmann (cf. note 7 above), judges the
whole expedition as ‘the foolhardiest improvisation’. By contrast Barreca
(1983–4) 45–6, thinks that Hannibal had allowed for heavy losses. Gallic migra-
tion into north Italy: Pol. 2.17; Livy 5.34–5; Salmon (1982) 34–7; T. J. Cornell,
OCD 3 625. Gaesati: Pol. 2.22.1–6, 23.1; Walbank, 1.194–5. Losses on the descent
‘nearly as heavy’ as in the Alpine fighting, Pol. 3.54.4; quality of army at Pyrenees,
39.8.
27Seibert argues that P. Scipio sent on only about half his army with his brother
( Hann. 104–5; earlier too Errington (1971) 65; contra Pol. 3.56.6 and Livy
21.32.3–5) because Cn. Scipio in 217 had only some 35 of the consul’s original 60
warships; but for a different explanation see Lazenby (1978) 127; on Cn. Scipio’s
forces, Brunt (1971) 646–7. In any case warships, unless very numerous, did not
convey whole armies themselves but escorted transports: cf. Walbank, 1.377,
431. Romans worried about Punic reinforcements from Spain, Pol. 3.97.3.
259
N O T E S T O T H E T E X T
Importance of Africa: Hoyos (1983) 178; cf. Seibert, Hann. 488. P. Scipio’s deci-
sion is almost universally judged as sound—even as pivotal to ultimate Roman
victory: e.g. Kahrstedt (1913) 384, ‘darum hat es [Rom] den Krieg gewonnen’; de
Sanctis, 3.2.436; Hallward (1930) 57; Errington (1971) 80; Scullard (1970) 29;
Lazenby, 52; Hampl (1983–4) 26; Seibert, Hann. 490, ‘die Karthager verloren den
Krieg nicht in Italien . . . sondern in Iberien’; Lancel, Hann. 119–20; Santosuosso
(1997) 182.
28 Naval raids on North Africa: below, chapter XII §I. Hasdrubal’s planned march
to Italy in 215 and defeat at Ibera, Livy 23.27.9–29.17; chapter XI §III.
I X T H R E E G R E A T V I C T O R I E S
1 Battle of the Trebia: Pol. 3.71–4; Livy 21.54.1–56.8; Appian, Hann. 7.24–9; Zon.
8.24.4–5; Walbank, 1.404–8; Lazenby (1978) 56–8; Seibert, FzH 213–15, and
Hann. 126–9. Freedom propaganda, Pol. 3.77.3–7; doubted by Erskine (1993) on
the ground that it was a Greek concept and possibly a Polybian invention, but the
content of Hannibal’s known treaties with Italian states (chapter X §I) shows
what was meant. Brundisine garrison-commander: Pol. 3.69.1–4; Livy
21.48.9–10, naming him Dasius. Groag’s idea ((1929) 79) that Hannibal meant to
base himself in the Po region, while rousing revolt among Rome’s Italian allies,
does not convince. Gauls and Hannibal in winter 218–217: Pol. 3.78.1–6; Livy
22.1.2–4; Zon. 8.24.8 (with even more fanciful disguise-details); Polybius’ gener-
alization (23.13.1) that Hannibal was never plotted against by his men cannot be
treated on its own as literal fact. Walbank, 1.410, and Seibert, Hann. 139–40,
think the disguise-tale worthless, unlike Rawlings (1996) 89. Hannibal’s forces:
variously estimated by, e.g., Kahrstedt (1913) 406; de Sanctis, 3.2.114–15;
Lazenby, 65.
2 Livy reports a failed winter attempt to cross the Apennines (21.58), widely disbe-
lieved as a mere embroidered doublet of the successful spring crossing (e.g. de
Sanctis, 3.2.97; Hallward (1930) 8.44 note 2; Lazenby (1978) 59–60; Briscoe
(1989) 49; Lancel, Hann. 149), but accepted by Seibert, Hann. 139. Other sup-
posed operations during the same winter (21.57.5–59.9) do look very dubious: de
Sanctis, 3.2.96–9. The Arno marshes: Pol. 3.78.6–79.12; Livy 22.2.1–3.1; Zon.
8.25.3; Walbank, 1.413; Lazenby, 60–1. Hannibal blinded in one eye: Pol. 3.79.12;
Livy 22.2.11; Silius 4.751–5; Juvenal, Satires 10.158; Oros. 4.15.3. Eye badly dam-
aged: Nepos, Hann. 4.3. Seibert, Hann. 148–50, is altogether sceptical about the
supposed rigours and suggests Punic propaganda magnified Hannibal’s handicap
to link him with the one-eyed and glorious Philip II of Macedon; but the ailment
is never mentioned otherwise and its origin in the marshes is circumstantially
attested.
3 On the losses at Trasimene see Lazenby (1978) 65; Seibert, Hann. 153–4.
4 Punic fleet off Pisae: Pol. 3.96.8–10, cf. Livy 22.11.6–7; see Seibert, Hann. 156.
Less plausibly, Hoffmann (1961/1974) 339–40 supposes the fleet really meant to
harass Roman communications to Spain. Hannibal’s first message by sea to
Carthage, Pol. 3.87.4–5; Africans rearmed, 87.3, 114.1; ‘having become very con-
fident’, 86.8. Lancel, Hann. 158–9, and Shean (1996) 180–1, defend the decision
not to advance on Rome in 217; Walbank, 1.421, and Lazenby (1996) 41, hold
that he never meant to attack the city. Cf. note 8 below on Maharbal.
5 Punic army’s battered state and ensuing recovery: Pol. 3.87.2–3, 88.1–2; Shean,
ibid. Plunder and slaughter, 3.86.8–11, 88.3–6. Huss (1985) 319 note 186 dis-
misses the killings (86.10–11) as merely anti-Punic propaganda; A. Toynbee,
260
N O T E S T O T H E T E X T
Hannibal’s Legacy, 2 vols. (Oxford 1965) 2.24 note 3 imagines that Hannibal
perhaps suspended killing when he marched through allied territories (this would
require accuracy about both boundaries and individuals’ identities). Fabius’ edict
to farmers, Livy 22.11.4.
6 On the operations in the second half of 217 see (e.g.) Lazenby (1978) 66–73;
Seibert, Hann. 164–77, 182–3.
The three Campanian aristocrats, Livy 22.13.2–3.
The famous episode of the cattle with blazing horns (Pol. 3.93.3–94.6; Livy
22.16–18; Plutarch, Fabius 6–7; Appian, Hann. 13.57–15.65; minor sources in
Walbank, 1.429) is rejected by Seibert as a legend or distortion aimed against
Fabius, and he infers a more straightforward ruse ( Hann. 170–1). But this would
have served the same supposed purpose, leaving unexplained why anyone should
invent the more complicated one.
7Hannibal’s supply problems: Livy 22.32.3, 40.8–9, 43.2–4 (with rumours about
the Spaniards). Cannae-depot taken, Pol. 3.107.1; in mid-year, Walbank 1.441.
Erdkamp (1998) 163–5 argues that the army had gathered enough supplies
during 217, but it is hard to see how they could have lasted until mid-216 or how
the Gerunium district instead—with the Roman army at close range—could feed
men and horses for months. Battle at the Ebro: Pol. 3.95–6; Livy 22.19–20;
fought some time before Trasimene (despite Pol. 3.95.2 mentioning summer-
time), cf. de Sanctis, 3.2.231–2, 664. Romans sought battle in 216: Pol. 3.107.7,
108.1–2 (‘the decision of the Senate’). Joke to Gisco: Plutarch, Fabius 15. Cannae:
Seibert, FzH 228–32 (lengthy bibliography on particular issues, 227–8), and
Hann. 189–98 (detailed list of all ancient sources, 191). See also Kahrstedt (1913)
427–34; de Sanctis, 3.2.126–59; Walbank, 1.435–48; Brunt (1971) 419, 648,
671–2, estimating the Roman army at no more than 45,000, with 30,000 slain or
otherwise lost; Lazenby (1978) 75–85; Connolly (1981) 183–8; Samuels (1990);
Lancel, Hann. 169–77; Santosuosso (1997) 176–80; McKnight (1998); and cf.
Sabin (1996).
8 Romans expected Hannibal to march on city: Pol. 3.118.6; Livy 22.55.1. Troops
at Rome: Livy 22.57.7, 23.14.2; Fitton Brown (1959) 367 note 10; Brunt (1971)
648–9; Lazenby (1978) 90–1; Seibert, FzH 383–4, Hann. 207–8 with note 130;
Lazenby (1996) 41. Maharbal’s proposal: Cato, Origines frgs 86–7P; Coelius, frg.
25P (both from Gellius, 10.24.6–7); Livy 22.51.1–4 (Maharbal’s epigram, not in
Cato, was probably not in Coelius either); Val. Max. 9.5 ext. 3; minor sources in
Hoyos (2000b). Livy makes Maharbal commander of the right wing, 46.7,
whereas Hasdrubal and Hanno were the wing-commanders according to Poly-
