A History of the Roman World (74 page)

2
C
ROSSING THE RHÔNE
. For this picturesque incident see Pol., iii, 42, Livy, xxi, 28. The crossing cannot be located with certainty. Napoleon (
Commentaires
, vi, 159) was probably right in setting the limits between the Rhône’s tributaries, the Durance and the Ardèche. For discussion and bibliography see Jullian,
Histoire de la Gaule
, i (1908–9), 464, who, however, favours a crossing further south at Tarascon. Cf. also De Sanctis, III, ii, 70, Walbank,
Polybius
, i, 378. On Hannibal’s elephants here and elsewhere see H. H. Scullard,
The Elephant in the Greek and Roman World
(1974), 154 ff. There is no doubt that the main herd comprised African elephants, though there is a possibility that Hannibal may have had one or two Indian beasts in addition.

3
T
HE ALPS
. The topographical evidence given by Polybius (iii, 50–6) and Livy (xxi, 31–7) does not allow any conclusive identification of the pass used by Hannibal. For older discussions it will suffice here to mention the bibliography in
CAH
, viii, 725 and to list the more probable passes and some of their advocates: (a) Little St Bernard: Niebuhr, Mommsen, Lehmann, Viedebandt; (b) Mt Cenis: Napoleon, Osiander; with the variant of Col du Clapier: Azan, Colin, Wilkinson; (c) Mt Genèvre: Neumann, Fuchs, Marindin, De Sanctis; with the variant of Col de Malaure: Bonus; (d) Monte Viso (Monviso): Col d’ Argentière and variants: Freshfield. See further De Sanctis,
SR
, III, ii, 65 ff.; Walbank,
Polybius
, i, 382 ff. The views of Sir Gavin de Beer,
Alps and Elephants
(1955), more ingenious than convincing, have been refuted by Walbank,
JRS
, 1956, 37 ff., and A. H. McDonald,
Alpine Journal
1956, 93 ff. A balanced survey of the problem is given in D. Procter’s excellent
Hannibal’s March in History
(1971). If any trend can be detected, it perhaps leans towards the Col du Clapier (cf. also E. Meyer,
Museum Helveticum
, 1958, 227 ff.).

4
T
REBIA
. Scipio’s movements after Ticinus have caused much difficulty, partly owing to Livy’s inferior account which places the battle on the right of the river. The account
given in the text follows De Sanctis and Kromayer. When Scipio had recrossed the Po (presumably at Placentia; according to Lehmann,
Hist. Zeitschr.
, 1916, 101 ff., a little further west and according to Fuchs near Pavia) he camped
πεϱὶ πόλιν Πλακεντίαν
(Pol., iii, 66, 9): on the east or west of the Trebia? Topography and strategy suggest on the west; i.e. the left bank, some way off at Stradella; Lehmann and Fuchs place the camp nearer Placentia at Rottofreno, but this is in the open plain. The old view of Grundy (
Journ. Phil.
, 1896, 83;
Cl. Rev.
, 1896, 284) though rejected by later critics (e.g. Kromayer) has been revived by Beloch (
Hist. Zeitschr.
, 1915), who places Scipio’s camp on the east bank of the Trebia. This view alters the whole topography of the battle which followed; according to it, after the desertion of his allies Scipio then withdrew and camped on the left of the river, while Hannibal encamped on the right and the battle was fought on the right bank. Scipio’s retreat over the Trebia becomes, on this theory, an advance, while Beloch even regards it as a doublet of his retreat after the battle of Ticinus. The view has been rejected by De Sanctis, Lehmann, and Kromayer, and the old view that Scipio camped with Sempronius on the right, Hannibal on the left bank and that the battle was fought on the left bank also may be retained. Laqueur’s theories (
Polybius
, 99) are not happy; see De Sanctis’ criticism (p. 98). Tenney Frank (
JRS
, 1919, 202 ff.) believes that Placentia before it was refounded in 190 was near Stradella: a theory which solves many difficulties, but is supported by no evidence. It may be added that Livy (xxi, 57–59) gives a long account of campaigning during the winter, but this may safely be rejected as reduplication. See also Walbank,
Polybius
, i, 404 ff.

5
H
ANNIBAL CROSSES THE APENNINES
. The route suggested in the text is supported by Kromayer (
Schlachtfelder
, iii, 104–47) and De Sanctis (III, ii, 104–9). It involves the supposition that there were then marshes between Pistoia and Florence, which is highly probable. The four days mentioned by Polybius present a difficulty, which is not insuperable. Routes further north and west are too long and too near the sea and would lead through marshes which were impassable for ancient armies before the land was drained; routes further east were not marshy and would bring Hannibal too near Flaminius at Arretium. The marshes were probably subject to flood, and Veith refers to similar conditions suffered by the Austrians at Muzakja in Albania in the winters of 1916–18.

6
F
LAMINIUS
. The extant tradition, which is aristocratic in outlook, depreciates such popular leaders as Flaminius and Varro, but an impartial estimate of their careers goes far to modify such criticism. Tradition asserts that Flaminius refused to co-operate with Servilius, which the facts contradict. He acted rashly but did not fall into every trap that Hannibal set. He was probably hastening south, not to fight, but to join Servilius further south, instead of awaiting him in the north.

7
T
RASIMENE
. The various hypotheses suggested for the actual site of the battle all agree that it was on the north shore of Trasimene; but various positions are chosen: (1) Borghetto-Passignano. This view has been adopted in the above text: see Fuchs (
Wien Stud.
, 1904, 134), Pareti (
Riv. di Fil.
, 1912, 383), De Sanctis (109) and Hallward (
CAH
). (2) Passignano-Monte Colognola: here the mountains are much nearer the sea. The view of Henderson (
J. Phil.
, 1897; 1899), Kromayer and Walbank; criticized by De Sanctis. (3) Borghetto-Tuoro; a very confined space – see Grundy (
J. Phil.
, 1896, 83; 1897, 273), Reuss (
Klio
, 1906, 226), M. Caspari (
Eng. Hist. Rev.
, 1910, 417). (4) Up the river Sanguineto. This improbable view presupposes that Flaminius was not surprised on the march by Hannibal, but saw the enemy above the Sanguineto and advanced against him in battle formation – see Sadée (Klio, 1909, 48) and Lehmann (
Jahresber. d. phil. Vereins Berlin
, 1915, 81). A recent attempt to identify the site from the finding of
what were alleged to be the ashes of the dead by G. Susini is criticized by Walbank,
JRS
, 1961, 232 ff (cf.
Polybius
, ii, 638).

8
S
ERVILIUS
. See De Sanctis,
SR
, III, ii, 122 ff. Appian,
Iber.
, 9, 11, who follows an inferior tradition, places the defeat at the lake of Plestia (Pistia) east of the Via Flaminia.

9
C
ALLICULA
. The pass probably lay between Cales and Teanum (cf. De Sanctis,
SR
, III, ii, 124 ff; Hallward,
CAH
, viii, 50); less probably further north at Mte Caievola, as Kromayer.

10
M
INUCIUS RUFUS
. On his office see T. A. Dorey,
JRS
, 1955, 92 ff. On the topography see De Sanctis,
SR
, III, ii, 54; Kromayer,
Schlachtfelder
, iii, 248.

11
C
ANNAE
.
Numbers
. The Roman forces considerably outnumbered those of Hannibal, though they probably did not amount to the 80,000–90,000 men that Polybius and Livy give (Livy also knew of other assessments). Cf. De Sanctis,
SR
, III, ii, 131, B. H. Hallward,
CAH
, viii, 52. The basic reliability of the number of legions in the field in the years 218–167 given by Livy is defended by De Sanctis and by P. A. Brunt,
Manpower
(1971), 416 ff., 645 ff. against the criticism of M. Gelzer,
Kleine Schriften
, iii, 220 ff.
Site.
Two main solutions of the difficulties in fixing the site are offered: (1) It lay on the north, i.e. left bank of the Aufidus, with the Romans facing roughly east and the Carthaginians west. This is the view of Delbrück, De Sanctis and Hallward. An improbable variant, with the armies almost north and south, was proposed by Lehmann (
Klio
, 1917, 162), though later retracted (
Klio
, 1930, 71). Judeich (
Hist. Zeitschr.
, 1927, 1) places the armies facing north-west and south-east; criticized by Kromayer (
Schlachtfeld.
, iv, 611). (2) It lay on the south bank, i.e. the right. This school falls into two divisions: (
a
) those who place the battle to the west of Cannae, with the armies facing very roughly north and south; this view is very improbable. It is proposed with individual variations by Arnold, Hesselbarth and Reusch (who alters the course of Aufidus) and recently by Lehmann in his recantation (
Klio
, 1930). (
b
) those who place the battle east of Cannae with the Romans backing the sea some three miles distant. This view, adopted in the text above, is that of Kromayer (proposed in
Schlachtfeld.
, iii, and defended in iv, and
Atlas
), Kahrstedt, Cornelius, etc. The discovery of a large cemetery south of the Aufidus (
Arch. Anzeiger
, 1938, 717; M. Gervasio,
Iapigia
, ix) seemed to have established that the battle was fought on the south bank, but these burials appear to be medieval (see H. H. Scullard,
Historia
, 1955, 474 f.; F. Bertocchi,
Rendic. Ac. d. Lincei
, xv, 337). For full discussions of Cannae, see especially Kromayer, De Sanctis and Walbank
Polybius
, i, 435 ff. Cornelius’ ‘Cannae; das militärische u. das literarische Problem’ (
Klio
, xxvi, 1932) contains much of interest, but his main contention is untenable: namely, that in the Polybian account the Romans broke through the Gallic-Spanish line and were caught behind that line by the Africans held in reserve. The Gallic line bent but did not break; nor were the Africans a reserve.
    Six deep grain depositories, which may have formed part of the rich granaries captured by Hannibal, have been found: see
The Times
, 2 August 1930.

12
N
AVAL BATTLE OFF EBRO
. For the naval aspect of the war see J. H. Thiel,
Studies on the History of Roman Sea-Power in Republican Times
(1946). A description of the battle of the Ebro is probably found in a fragment of the Greek and pro-Carthaginian historian Sosylus: see F. Jacoby,
FGrH
, 176 F. Our five sources all give different reasons for the victory. Polybius (iii, 95–6) gives the nearness of the Carthaginians to their infantry on the shore; for Livy (xxii, 19–20) it is surprise; Frontinus (
Strat.
iv, 7, 9) tells of the throwing of burning projectiles; Zonaras, of the destruction of the Carthaginian sails; while Sosylus says the skill of the Massiliotes foiled the enemy’s manoeuvre. The reasons given by Polybius, Livy and Sosylus are not mutually exclusive.

13
T
HE SCIPIOS IN 217
. Traces of their camps have been found at Almenara, five miles north of Saguntum; the accuracy of Polybius’ topographical description is thus strikingly
confirmed: see A. Schulten,
Arch. Anz., 1933, 622 ff.
, and, for photograph, Scullard,
Scipio Africanus
(1970), plate 24. In tibili lay near Benicarlo, and Iliturgi in Catalonia at Cabanes west of Oropesa: see A. Schulten,
Hermes
, 1928, 288 ff. The site of the more famous southern Iliturgi in Andalusia is now shown by an inscription (in honour of its
deductor
Ti. Sempronius Gracchus) to have been near Mengibar in the province of Jean:
A. Espan. Arch.
, 1960, 193 ff.

14
T
HE SCIPIOS’ ADVANCE
. It is very improbable that the Scipios also penetrated into Andalusia and captured Castulo, as Livy, xxiii, 49; xxiv, 41 f. Such annalistic accounts, which Livy derived from Valerius Antias, may contain elements of truth, but it is safer to reject them. Strategic considerations make an advance and victories in southern Spain very improbable. However, this idea has been revived by R. C. Sanchez,
Habis
, 1975, 213 ff., who placed the defeat of Publius Scipio near the river Genil, south-west of Corduba.

15
P
HILIP’S TREATY WITH HANNIBAL
. On the terms and Philip’s subsequent war with Rome see F. W. Walbank,
Philip V of Macedon
(1940), 70 ff. The terms of the treaty are given by Polybius, vii, 9, who provides a Greek translation of the Punic document which fell into Roman hands. E. Bickerman,
TAPA
, 1952, 1 ff. equates the oath with a Hebrew covenanted treaty (
berit
). See also A. H. Chroust,
Classica et Med.
, 1954, 60 ff.; Walbank,
Polybius
, ii, 42 ff. It is noteworthy that the terms imply that Hannibal’s war aims were limited and that he does not appear to have aimed at the complete destruction of Rome.

16
THE AETOLIAN TREATY. The terms are summarized by Livy, xxvi, 24. Part of a copy of the text, in Greek, was found in 1949 inscribed on a stone in Acarnania. See G. Klaffenbach,
S.-B. Berlin
, 1941, 13 ff.; A. Momigliano,
Quinto Contrib.
977 ff.; A. H. McDonald,
JRS
, 1956, 153 ff.; E. Badian,
Latomus
, 1958, 197 ff.; Walbank,
Polybius
, ii, 162, 179 ff., 599 ff.; and G. A. Lehmann,
Untersuchungen zur hist. Glaubwürdigkeit des Polybios
(1967), who deals with the problems at length. See also E. Badian,
Titus Quinctius Flamininus
(1970), 49 ff.; J. Briscoe,
Commentary on Livy xxxi–xxxiii
(1973), 273 f.; D. Musti,
Aufstieg NRW
, I, ii (1974), 1146 ff. The terms were probably agreed by Laevinus and the Aetolians in the autumn of 211 and ratified in Rome two years later.
    Livy records that if the Romans took any cities they were to have the movable booty while the cities and their territories fell to the Aetolians. But the inscription adds two further clauses (the second fragmentary): booty from cities taken jointly by Romans and Aetolians should be shared, and any cities that went over voluntarily to the Romans or Aetolians could be received into the Aetolian League. This last clause causes problems in regard to Flamininus’ settlement of Greece in 197 when he denied the Aetolians any right to four Thessalian cities on the ground that they had surrendered to him. Questions of Flamininus’ honour or Polybius’ accuracy may be involved, but possibly in the lost part of the inscription there was some further qualification about the cities that surrendered voluntarily under which Flamininus’ demand might be justified (cf. Walbank, op. cit, 600).