What the other ancient writers say
Outside Polybius and Livy, the ancient evidence for the pass comes to six sentences. Each is given below in the original, with what it does and does not establish.
τέτταρας δ᾽ ὑπερβάσεις ὀνομάζει μόνον· διὰ Λιγύων μὲν τὴν ἔγγιστα τῷ Τυρρηνικῷ πελάγει, εἶτα τὴν διὰ Ταυρίνων, ἣν Ἀννίβας διῆλθεν, εἶτα τὴν διὰ Σαλασσῶν, τετάρτην δὲ τὴν διὰ Ῥαιτῶν, ἁπάσας κρημνώδεις.He [Polybius] names only four passes: the one through the Ligurians, nearest the Tyrrhenian sea; then the one through the Taurini, which Hannibal crossed; then the one through the Salassi; and fourth the one through the Raeti; all of them precipitous.Polybius 34.10.18, quoted by Strabo 4.6.12 (Meineke). This is the only statement by Polybius that names the pass, and it comes from his geographical book, written after his own Alpine journey. It places Hannibal's pass in the Cottian Alps, between the coast road and the Aosta valley, and so excludes the Petit and Grand St Bernard. Strabo's own geography (4.6.6) puts the Taurini, "a Ligurian people", on the Italian face of the range held by the Medulli, that is, at the mouth of the Susa valley, and the Salassi in the valley of the Dora Baltea (4.6.7). Pliny locates the source of the Po, where the Traversette comes down, "in the territory of the Ligurian Bagienni" (Padus e gremio Vesuli montis ... finibus Ligurum Bagiennorum visendo fonte profluens, 3.117), not of the Taurini, though the tribal map of the first century AD need not be that of 218 BC. The Susa valley passes fit "through the Taurini" without argument. The Traversette fits it with a fifty-kilometre march to Turin, which Polybius's narrative allows (3.60.8).
quas quinque viis Varro dicit transiri posse: una, quae est iuxta mare per Ligures; altera, qua Hannibal transiit; tertia, qua Pompeius ad Hispaniense bellum profectus est; quarta, qua Hasdrubal de Gallia in Italiam venit; quinta, quae quondam a Graecis possessa est, quae exinde Alpes Graiae appellantur.Varro says the Alps can be crossed by five ways: one, which is by the sea through the Ligurians; a second, by which Hannibal crossed; a third, by which Pompey set out for the Spanish war; a fourth, by which Hasdrubal came from Gaul into Italy; a fifth, which was once held by the Greeks, whence they are called the Graian Alps.Servius on Aeneid 10.13 (Thilo). Varro's five and Polybius's four both begin at the coast and run away from it, Varro's ending at the Graian Alps and Polybius's at the Raeti, so both lists are probably ordered from south to north. If so, Hannibal's pass lies between the coast road and Pompey's. Pompey wrote to the Senate in 75 BC that he had "opened through the Alps a route other than Hannibal's, and more convenient for us" (per eas iter aliud atque Hannibal, nobis opportunius, patefeci, Sallust, Histories 2.98.4 Maurenbrecher), and Pompey's pass is usually taken to be the Montgenèvre, the line of the later Via Domitia. On that reading Hannibal's pass is south of the Montgenèvre: the Traversette, the Agnel or the Larche, and de Beer built on this. Two things weaken that reading. Which pass Pompey meant is not known: Ammianus credits King Cottius under Augustus with building the road over the Cottian pass from Susa (15.10.2-3), so Pompey may have found no made road on the Montgenèvre, and his pass has been placed at the Mont-Cenis and at the Larche as well. And Varro's fourth way, Hasdrubal's, contradicts Livy 27.39, where Hasdrubal in 207 BC "led his army through places mostly made passable by his brother's crossing" (per munita pleraque transitu fratris), that is, by the same pass. One of the two is wrong. The list therefore shows that Hannibal's pass was distinct from Pompey's and from the Graian pass, and suggests, without proving, that it lay south of Pompey's.
ad Alpes posteaquam venit, qua Italiam ab Gallia seiungunt, quas nemo umquam cum exercitu ante eum praeter Herculem Graium transierat (quo facto is hodie saltus Graius appellatur), Alpicos conantes prohibere transitu concidit, loca patefecit, itinera muniit, effecit ut ea elephantus ornatus ire posset, qua antea unus homo inermis vix poterat repere.When he came to the Alps, which divide Italy from Gaul, which no one before him had ever crossed with an army except the Greek Hercules (whence the pass is today called the Graian pass), he cut down the Alpine people who tried to stop his crossing, opened up the ground, built roads, and made it possible for a harnessed elephant to go where before a single unarmed man could scarcely crawl.Cornelius Nepos, Hannibal 3.4 (Fleckeisen and Halm). Nepos puts Hannibal on the Graian pass, the Petit St Bernard by its Roman name, and Pliny records the same popular tradition for the Pennine pass: Aosta lies "next to the twin gates of the Alps, the Graian and the Pennine; by the latter, they record, the Carthaginians crossed, by the Graian Hercules" (iuxta geminas Alpium fores, Graias atque Poeninas: his Poenos, Grais Herculem transisse memorant, 3.123). Livy attacks this tradition at 21.38.6-9, together with the Cremonis iugum that Coelius named, on the ground that both passes bring an army out among the Salassi and not among the Taurini, and he adds that the name Poeninus comes from the god worshipped on the summit, not from the Poeni. Livy's argument is sound, and the votive inscriptions to Iuppiter Poeninus found on the Grand St Bernard confirm the god's cult there. Polybius's "through the Taurini" settles the same point independently. Nepos's claim that no army had crossed before is rhetoric (Polybius says the Gauls had crossed with armies "not once or twice but many times", 3.48.6), but his picture of a pass unfit for a laden animal until Hannibal built a road agrees with Polybius's broken path and tells against the great roads, the Montgenèvre and the Mont-Cenis, though the sentence is rhetorical and carries little weight.
Taurinis ducentibus accolis, per Tricasinos et oram Vocontiorum extremam, ad saltus Tricorios venit. Indeque exorsus, aliud iter antehac insuperabile fecit; excisaque rupe in immensum elata, quam cremando vi magna flammarum acetoque infuso dissolvit, per Druentiam flumen, gurgitibus vagis intutum, regiones occupavit Etruscas.Guided by natives from among the Taurini, he came through the Tricasini and the far edge of the Vocontii to the passes of the Tricorii. Starting from there he made another road, until then impassable; he cut through a cliff rising to a vast height, which he broke up by burning it with a great force of flame and pouring on vinegar; and by way of the river Druentia, unsafe with its wandering eddies, he seized the Etruscan regions.Ammianus Marcellinus 15.10.11 (Rolfe's Loeb text). Ammianus follows Livy's itinerary and Livy's vinegar and adds the Taurine guides, which is what Polybius implies when he says the guides who met Hannibal were "those living around the pass" (3.52.3). He is a witness to the Livian tradition as it stood in the fourth century and adds no independent knowledge of the ground.
ἐλθὼν δὲ ἐπὶ τὰ Ἄλπεια ὄρη, καὶ μηδεμίαν μήτε δίοδον μήτε ἄνοδον εὑρών ... τὴν πέτραν ἐκ τοῦδε ψαφαρὰν γιγνομένην σφύραις σιδηραῖς θραύων, καὶ ὁδὸν ποιῶν ἣ καὶ νῦν ἐστιν ἐπὶ τῶν ὀρῶν ἐντριβὴς καὶ καλεῖται δίοδος Ἀννίβου.Coming to the Alpine mountains and finding no way through or up ... breaking the rock, made crumbly by this means, with iron hammers, and making a road which even now is a worn track on the mountains and is called Hannibal's pass.Appian, Hannibalic War 4 (Mendelssohn). Appian names no place, gives the march six months from Spain against Polybius's five, and records that in his day, the second century AD, a track was shown as "Hannibal's pass". Where that track ran he does not say. Zonaras's epitome of Cassius Dio (8.23) has nothing on the pass. Silius Italicus (3.446-476) versifies Livy's peoples and his Druentia.
Two further scraps are sometimes cited. Strabo's description of the mountain of the Medulli (4.6.5), with "a great lake in hollow ground above, and two springs not far apart" from which the Durance and the Dora flow, has been claimed for the Mont-Cenis plateau and its lake and for the Savine basin below the Clapier. It is Strabo's geography of the Maurienne, garbled, and it does not mention Hannibal. Pliny's remark that the Po's source at Monte Viso is "a sight worth seeing" (3.117) shows only that the head of the Po valley was known and visited in the first century AD.
These passages, taken together, settle only a negative. The Petit St Bernard and the Grand St Bernard rest on a Roman tradition that Livy and Polybius both contradict, and on a folk etymology. They leave open the whole Cottian range from the Larche to the Mont-Cenis, with a lean towards the Susa valley from "through the Taurini" and a lean away from the Montgenèvre from Pompey's letter and Varro's order.