This is the fifth and last part of our series (I (https://acoup.blog/2026/04/10/collections-raising-carthaginian-armies-part-i-finding-carthaginians/), II (https://acoup.blog/2026/04/24/collections-raising-carthaginian-armies-part-ii-the-african-backbone/), III (https://acoup.blog/2026/05/01/collections-raising-carthaginian-armies-part-iii-generals-warlords-and-vassals/), IV (https://acoup.blog/2026/05/08/collections-raising-carthaginian-armies-part-iv-allies-and-mercenaries/), V (https://acoup.blog/2026/05/22/collections-raising-carthaginian-armies-part-v-how-a-carthaginian-army-fights/)) looking at how Carthaginian armies were raised and constituted. Over the last four parts, we’ve looked at the larger components of Carthaginian armies: the relatively small role of Carthaginian citizens, the more prominent role of North African conscripts, of Numidian and Iberian vassals, and of mercenaries and allies from Italy and Gaul. As we’ve noted, the place of many of these troops within Carthage’s armies changed over time, particularly in the third century as Carthage exercised a more direct presence in Spain, Gaul and Italy, thereby transforming mercenaries into vassals and allies.

To close off this week, I want to briefly discuss some of the ‘odds and ends’ of Carthaginian military forces that we haven’t gotten to yet, most notably the role of light infantry slingers from the Balearic islands and of war elephants. But I want to spend most of the time here discussing how these composite armies fight.

Now that, in and of itself, is a tricky proposition. For one, the composition of these armies quite evidently changed over time. Worse yet, for most of Carthaginian history, our sources provide us few battle accounts in which the dispositions and tactics are both detailed and reliable. We have a few early battle descriptions (mostly in Diodorus), but to put this effort at analysis on somewhat firmer ground, I propose to focus on the third century in particular.

But first, as always, raising large armies of mercenaries, subject conscripts, vassal warlords and allies is expensive! If you too want to help me invade Italy with a multi-ethnic army of diverse origins in a doomed effort to stop the Roman Republic, you can help by supporting this project over at Patreon (https://www.patreon.com/user?u=20122096). If you want updates whenever a new post appears or want to hear my more bite-sized musings on history, security affairs and current events, you can follow me on Bluesky (@bretdevereaux.bsky.social). I am also active on Threads (bretdevereaux) and maintain a de minimis presence on Twitter (@bretdevereaux).

						Email Address						
					
				

				
					
					
					
					
											
						Subscribe!						

Slings and Elephants

Before we get into the battles themselves though, we have a few more odds and ends to add to our armies, most notably Balearic slingers and war elephants. In both cases our sources don’t give us a ton of information to go on, but these were both regular parts of Carthaginian armies.

The residents of the Balearic islands had a well-earned reputation as exceptional slingers in antiquity; Phoenician settlement on the island of Ibiza in the seventh century steadily led the islands to drift into Carthaginian influence, making Balearic slingers available for Carthaginian armies, though as far as I know it remains very unclear how much actual governance Carthage exerted on the islands.

That said, while Balearic slingers are a recurring ‘specialist’ unit in Carthaginian armies, their numbers remain few. When we hear about their detachments, they’re really quite small. Hannibal, for instance, when disposing his forces in 218 sends 870 of his slingers to Africa and leaves 500 in Spain (Polyb. 3.33.8-16). While he is also taking some with him (so that is not a total count of his Balearic slingers) I think it is worth contrasting the scale of other troop movements in the dispositions: the force left in Spain is fifty-two warships, 450 African cavalry, 300 Spanish cavalry (Ilergetes), 1,800 Numidian cavalry, 11,850 Libyan infantry, and 21 elephants. 500 slingers seems a small detachment, in comparison. Likewise, the force heading to Africa was composed overall of 1,200 cavalry and 13,850 infantry alongside the 870 slingers. These are thus quite small detachments: small units of specialists rather than major contingents of an army.

From the excellent https://www.trajans-column.org/ (https://www.trajans-column.org/), a detail of scene 66 on Trajan’s Column showing a slinger, likely Balearian, with his sling and a set of slingstones; he wields a small shield as well, but is otherwise unarmored.

That said our sources (Polybius, mainly) often keeps track of them, so they’re a distinctive unit. In addition to the 1,370 slingers left behind, Hannibal took some number with him when crossing the Alps, but we can’t really track how many because in Hannibal’s army they always appear brigaded together with his lonchophoroi (who as you will recall, are also light infantry skirmishers, using javelins), in a combined unit of 8,000. I suspect that, at least by the time Hannibal is in Italy, the Balearians represent a distinct minority in that formation too. In that 218 disposition above, Hannibal (advancing into Spain reportedly with 90,000 infantry and 12,000 cavalry (Polyb. 3.35.1), a figure I suspect is inflated, but it is what we have), has essentially advanced on Italy with three-quarters of his force, splitting the remaining quarter to act as the core of armies to be formed up (if necessary) in Spain and Africa. That might imply something like 5,500 total Balearian slingers, of whom about 4,000 are with Hannibal. The problem, of course, is accounting for casualties: Hannibal loses half of his cavalry and three-quarters of his infantry getting to Italy (he drops into the Po River Valley with 20,000 infantry and 6,000 cavalry, Polyb. 3.56.4).

Even assuming a specialist unit like his slingers might have been spared the worst of the casualties, we might reasonably expect their numbers to at least be reduced by half, meaning that light infantry ‘brigade’ we see at Trebia might only have 2,000 (or even just 1,000) Balearians in it, with the rest made up of North African – and also perhaps Spanish or Gallic – javelin troops. Of course Hannibal’s army would subsequently expand back upwards with a fresh infusion of Gauls, such that by Cannae he had 50,000 troops (40,000 infantry, 10,000 cavalry), of which as noted above, at most we might expect a couple thousand to have been Balearian slingers.

In terms of equipment and fighting style, Balearic slingers fought unarmored, using slings and carrying small round shields, with spears for close-combat – though given their lack of armor, that must have been something of a weapon of last resort.1 So these were very light troops: very mobile, but not able to really stand up to anything. Hannibal never puts this mixed brigade into his battle line (that we’re told) – instead it is deployed as a screening force (Polyb. 3.72) or in rough ground (Polyb. 3.83) and withdrawing them before the main clash of infantry lines. Slings are really effective weapons in skilled hands – sling bullets can arrive with a lot of punch and be quite accurate at relatively long range – so even a small force of slingers mixed into a larger force of skirmishers would certainly make their presence known.

As for Carthage’s war elephants, we actually discussed war elephants at length way back in 2019. We may add a few notes here on Carthage’s elephants in particular. First, Carthage used war elephants, fairly regularly. The size of Carthage’s elephant corps seems to have been primarily limited by logistics: elephants were hard to move overseas (though it could clearly be done and the Carthaginians do it) and hard to keep supplied. Carthage had “nearly a hundred” elephants at Bagradas (255), supposedly 200 in Spain in 228 under Hasdrubal the Fair (Diod. Sic. 25.12; I suspect this number is quite inflated), but Hannibal marches out of Spain with just 37 elephants (Polyb. 3.42), leaving – as noted above – only 21 elephants behind, suggesting he only had 58 to start with.

Still my favorite Etruscan elephant plate (c. 275-270 BCE), from the National Etruscan Museum at the Villa Giulia in Rome. Given the Etruscan context of this famous plate, the elephant here might be Carthaginian, rather than Hellenistic. Interestingly, it is depicted with its calf in tow, which would make the grown elephant female, though generally only bulls were used as war elephants.

Carthaginian elephants, like Ptolemaic elephants, were drawn from the now-extinct North African elephant (Loxodonta africana pharaohensis), likely a relative of the smaller African forest elephant (Loxodonta cyclotis), not the bigger African Bush elephant (Loxodonta africana) you are likely more familiar with but which has never been domesticated. These North African elephants were smaller than the Asian elephant (Elephas maximus), which was a problem for the Ptolemies (who regularly faced Indian elephants in Seleucid service) but not really for Carthage. It is possible Carthage may have at times obtained Indian elephants in small numbers: Hannibal’s personal elephant was named Surus, ‘the Syrian;’ if he was from Syria, that would mean he was an Indian elephant imported by the Seleucids, although it seems equally likely to me that someone might name an uncommonly big North African elephant ‘Syrian’ because in its large size it resembled the larger Asian elephant (which a Carthaginian might associate with Syria – where the Seleucid elephant program was – rather than with India, where the elephants were actually from).

Of course ‘smaller’ doesn’t mean ‘small:’ male African forest elephants ‘only’ get to be about 7ft tall at the shoulders (compared to 9ft for Asian elephants and 10ft for African bush elephants), which is still a mighty big animal. As we discussed back in the original series on war elephants, the logistics demands of keeping elephants were substantial: they cannot be effectively bred in captivity so they must be captured and tamed and once domesticated, you have to feed them and they eat a lot. Nevertheless, they could be a clear military asset.

Alas, we know very little about how the Carthaginian elephant corps was organized: it’s unclear who the mahouts (the skilled elephant drivers) would have been or how they fit into Carthage’s mobilization system. It was clearly an important component of Carthaginian power – Carthage mints coins with elephants on them (https://www.ngccoin.com/news/article/14333/ancient-coins-elephants/), likely as a symbol and expression of Carthaginian power (especially in places where elephants were not native, so the only elephants around would be Carthaginian war elephants).

Via Wikimedia Commons (https://commons.wikimedia.org/wiki/Category:Punic_coins_of_Hispania#/media/File:Carthage,_quarter_shekel,_237-209_BC,_SNG_BM_Spain_102.jpg), a Carthaginian quarter-shekel from Spain, showing on elephant on the reverse. It’s striking to me that Punic coins with elephants on them seem more common in Spain and Sicily – where the elephants would have to have been imported – than in Africa itself, where they were native.

How a Carthaginian Army Fights

Unlike Roman, Macedonian (https://acoup.blog/2024/01/19/collections-phalanxs-twilight-legions-triumph-part-ia-heirs-of-alexander/) or Greek (https://acoup.blog/2025/11/14/collections-hoplite-wars-part-i-the-othismos-over-othismos/) armies, we don’t have any discussions of Carthaginian tactics from the Carthaginian point of view, or even artwork from Carthaginian contexts showing things like battles. The closest the sources come are some general comments by Polybius, which I think have to be treated with a great deal of caution. We’ve already seen that Polybius’ depiction of Carthage’s armies as wholly mercenary is at best deceptive. Likewise, his comment that Carthage “entirely neglects its infantry” and merely “pays some slight attention” to its cavalry (Polyb. 6.52.3) doesn’t stack up against the performance of Carthaginian arms in the third century: Carthaginian infantry appears, if less capable than the Romans, more capable than Greek or Macedonians, while Carthaginian cavalry appears flatly superior. So we have to be careful simply taking Polybius’ word for things when it comes to Carthage’s military ability.

That leaves us reliant on Carthage’s battles to understand how Carthaginian armies tend to fight. There are a few things to note here. First, in the third century, Carthage loses battles with Rome somewhat more often that it wins them – never nearly so lopsided as the Hellenistic record against Rome, mind you – even with Hannibal considered in the record. I want to note that out front because our sources tend to focus more on the occasions where Carthage wins against Rome (because it is to some degree surprising) and so that is when we tend to get something like a complete order of battle that lets us assess Carthaginian tactics. But I don’t want to give a distorted impression of effectiveness here: Carthaginian armies are a real threat that can beat Roman armies in the field, but the Romans win more than they lose.

All that said, I think there are some things we can say about how Carthaginian armies fight. These armies tend to be somewhat more cavalry heavy than Hellenistic or Roman armies, although they maintain a strong infantry ‘backbone.’ Tactically, because Carthaginian armies are so varied in composition (given the variable numbers of mercenaries, allies and vassals they may have), they tend to have much more varied dispositions than Roman or Macedonian armies (which both have a fairly ‘standard’ battle plan), but there is a preference towards envelopment using cavalry (as distinct, I’d argue, from Macedonian ‘breakthrough’ using cavalry). Finally, there is also a clear preference in placing mercenaries and auxiliaries in high-casualty positions that is remarkable, especially compared to Roman or Macedonian armies which tend to place their most reliably troops – who tend to be the highest status (citizen or Macedonian) troops in the center.

So let’s look at a few Carthaginian armies in action to see how this plays out.

The pitched battle from the First Punic War (264-241) we get the most details for is the Battle of Bagradas River (255), which is where Marcus Atilius Regulus’ (cos 267, 256) expedition to North Africa – which had been making gains – falls apart, necessitating a naval rescue mission to extract what is left of his army later that year. Polybius (1.33-34) is our best account of this battle and he doesn’t give us a ton of detail, but what he does give us, I think, is indicative of how Carthage expects to fight.2

The Carthaginian army, led in part by the Greek mercenary general Xanthippus, arrives with 12,000 infantry, 4,000 cavalry and nearly 100 elephants, which is a very cavalry (and elephant) heavy force, though it seems like nearly all of the infantry here is heavy African infantry (Libyans and perhaps also Carthaginian citizens). The Carthaginians draw up with the ‘phalanx of the Carthaginians’ in the center, with the elephants strung in a single line in front of them. Xanthippus puts his own mercenaries on the right wing – right where a Macedonian commander would have put his elite infantry in an Alexander-Battle formation (https://acoup.blog/2024/01/19/collections-phalanxs-twilight-legions-triumph-part-ia-heirs-of-alexander/) – a but splits the remaining cavalry across both wings. The Romans seem to respond to the threat of the elephants by forming up unusually deep and thus also narrow (Polyb. 1.33.9-10), which turns out to have been the catastrophic mistake of the battle.

What ends up happening is that in the center, the Romans are able to push past the elephants, but have their lines disordered by it and as a result are thrown back by the Carthaginian heavy infantry. Meanwhile on the flanks, the numerically superior and more capable Carthaginian cavalry quickly routs the Roman cavalry and begins what we’ll see is a standard Carthaginian tactic – double-envelopment – wrapping around the Romans on the flanks. The one spot where the Romans perform well is, ironically, against Xanthippus’ mercenaries on his right (the Roman left), where the Romans are able to get around the elephants (and presumably inside the cavalry) to engage the Carthaginian right-wing and send it reeling back to camp (Polyb. 1.34.4). The envelopment proves fatal: no army can fight effectively if beset on all sides and the Romans are no exception (unless Julius Caesar is leading (https://en.wikipedia.org/wiki/Battle_of_Bibracte)): repulsed in the center and then encircled, the Roman army falls apart, with the Carthaginians able to inflict heavy losses – all but 2,000 out of a force of 15,500 – in the rout.

What I think is most striking is that here already we see the Carthaginians doing what is going to become a standard approach in Hannibal’s battles, double-envelopment with cavalry. This is, I think, quite distinct from the Macedonian practice of ‘Alexander-battle’ (https://acoup.blog/2024/01/26/collections-phalanxs-twilight-legions-triumph-part-ib-subjects-of-the-successors/); somewhat ironically the Carthaginians essentially are doing the ‘Total War tactic’ that I spent some much time insisting the Macedonians do not do. Whereas the Macedonian approach is generally to try to ‘breakthrough’ with their cavalry often at the point where an army’s center joins one of its wings – and generally only on one flank, with the other flank merely buying time – the Carthaginians really are looking to ‘flank.’ Put another way, Macedonian cavalry goes through one side, but Carthaginian cavalry goes around both sides, aiming to disperse the enemy cavalry screen at the flanks and then loop around the flanks and rear of the enemy force rather than smashing through.

Vassals and Allies on the Battlefield

Moving forward chronologically, we can look at some of the dispositions of the Second Punic War to see some of the same patterns as at Bagradas, but also – because our sources (mostly Polybius, now with some Livy) provide more detail – some additional details.

After a major skirmish at Ticinus, Hannibal’s first major pitched battle in Italy is the Battle of the Trebia (218; Polyb. 3.71-4; Livy 21.54-56). Hannibal’s plan here is clearly another envelopment battle, similar in conception to Bagradas. Hannibal sets up a single line of ‘line infantry’ (both his African ‘heavies’ and his Spanish and Gallic ‘mediums’) in the center and his cavalry on the flanks. But he then does two things to reinforce his flanks: he deploys his elephants there, rather than along the center, and he has his skirmishers – initially deployed in front of his army as a screening force – retreat to the flanks once they had bested the Roman velites. The result is that Hannibal ends up stacking up his cavalry and elephants and skirmishers against the Roman cavalry on the flanks. That must have left his main infantry line somewhat thin: he has less ‘line infantry’ than the Roman force (probably around 25,000 Roman and socii heavies against c. 21,000 Carthaginian heavies and mediums, when you subtract out the velites and Carthaginian ‘lights’ (Balearians and lonchophoroi) and what he has is meaningfully lighter, but he has to match the same width.

Via Wikipedia (https://en.wikipedia.org/wiki/Battle_of_the_Trebia#Battle), a decent map of the Battle of the Trebia after the Carthaginian lights had fallen back behind their main force and the Roman velites had been driven back. Livy and Polybius actually disagree slightly about where the elphants went: Polybius puts them in front of the cavalry, Livy out on the wings outside the cavalry (for some reason this map does neither).

But we get another interesting note here that, as we’ll see, is a trend. Polybius reports of Hannibal’s losses, “for they all were very glad about the battle, [thinking it] as a great accomplishment, for it happened that the losses of the Iberians and Libyans were few, most of the losses being of the Celts” (Polyb. 3.74.10). Now given what Polybius has told us – that the Gauls, Iberians and Africans are all in one line – it’s not quite clear how that outcome happens (the Iberians, at least, are no more heavily armored than the Gauls!), but as we’re going to see, it is something of a pattern.

Hannibal’s next major battle is at Lake Trasimene (217) but this is something of a rarity: an actual ambush at battle-scale rather than a pitched battle. While ambushes are common in small actions, it is actually quite rare for one field army to ambush another: field armies are so big they are quite hard to hide and tend to have a lot of scouting. That makes this kind of ‘true’ large-scale ambush quite rare, but it also means the dispositions and tactics aren’t really applicable to the more common pitched battles. There is, however, one detail that is worth noting, which were the casualties among the Carthaginians, of which Polybius says (Polyb. 3.85.5), “He [Hannibal] now rested his own [troops] and honored the dead of the highest ranks, thirty in number; the overall losses were fifteen hundred, of which most were Celts.”

Which gets us to Cannae (216), which I have analyzed elsewhere (https://warontherocks.com/the-importance-of-the-battle-of-cannae/) and so needn’t do in detail here. The plan is once again double-envelopment using cavalry deployed on the flanks and a relatively weak center, with Hannibal’s innovation here coming in two parts: first the center is arced forward to invite the Romans to attack it and second Hannibal pulls his North African troops – his heaviest and most reliable – into two formations that sit on the flanks of the combined Spanish and Gallic ‘medium’ infantry main line. The result, famously, is that the Romans, when they push back the Gallic-and-Iberian center, will put the Africans around their flanks, while Hannibal’s cavalry first disperses the Roman cavalry and then completes the encirclement by striking the advancing Roman infantry force in the rear.

Via Wikipedia (https://en.wikipedia.org/wiki/Battle_of_Cannae#/media/File:Battle_of_Cannae,_215_BC_-_Initial_Roman_attack.svg), a decent enough diagram of the Battle of Cannae’s opening positions, although the map on my WotR article (https://warontherocks.com/the-importance-of-the-battle-of-cannae/) is somewhat better.

Now there are solid tactical reasons to arrange the army this way – Hannibal does, after all, win the battle – but it is hard not to see another consideration at work: almost all of the heavy losses are guaranteed to be sustained by the Iberian and Gallic troops. By contrast, the North African troops, who are by this point every bit as heavy as Roman troops (Polyb. 3.114.1; Livy 22.46.4) are not placed into the thick of things but held off onto the side, where they mostly avoid the brunt of the Roman assault. The Numidian cavalry seems to have been given orders to skirmish, to ‘tie up’ Rome’s socii cavalry, while it is the Iberian and Gallic cavalry that has to punch through and attack to create the encirclement.

There is a consistent pattern here of risking Iberian and Gallic troops in order to preserve Carthaginian, African and Numidian troops. And the result is predictable. Polybius (3.117.6) gives Hannibal’s losses at Cannae as, “of Hannibal, the Celts lost 4,000, the Iberians and Africans 1,500 and the cavalry 200.” One wonders, given that division of losses, if – when Polybius says that the Gauls and Iberians are in the center – Hannibal has in fact put the Gauls in the absolute center (furthest forward), with the Iberians on their wings and the Africans on the Iberian’s wings, essentially creating a ‘spectrum of peril,’ with the Gauls in the most dangerous spot and the North Africans in the safest.

Hannibal’s deployment at Zama (Polyb. 15.11; Livy 30.33) echoes this concern: he puts the elephants out in front (like Xanthippus at Bagradas fifty years prior) and then in his first line he puts Ligurians, Gauls, Balearians and Mauritanians: his skirmish specialists (the Balearians) and then all of his expendable auxiliaries. He may have intended these fellows to retreat to the flanks like at Trebia, but in any case he made no preparations for them to withdraw down the center – when they did so they were cut down by the next line (Polyb. 15.13.3-10). Then behind that line he places his North African and Carthaginian citizen troops – fresh levies from North Africa; the Romans would have put these greenest troops in the front, but Hannibal shelters them in his second line. Finally, perhaps having learned something from the value of the Roman triarii, Hannibal puts his own veterans in a third, final line in the rear.

Had Hannibal won at Zama, rather than lost, we’d presumably have had another line of Polybius about how the great majority of his losses were taken by the Gauls and Ligurians (and Mauritanians) that he threw forward at the outset of the battle.

That said, we get quite a different approach at the Battle of Metaurus (207). The sources for this battle are, I should note, something of a mess (with Polyb. 11.1-3 and Livy 27.48 not quite agreeing), but a mess that is untangled quite capably by J.F. Lazenby, Hannibal’s War: A Military History (https://amzn.to/3PlBRXu) (1998), 189-90. Hasdrubal’s army is a mix of Iberians, Gauls and Ligurians, with thirty elephants. Hasdrubal’s problem is that his army is quite outmatched and he knows it, fighting the battle because he has no other choice, having not yet had time to fortify a camp. The battlefield had hilly terrain on Hasdrubal’s left, so he seems to have settled on a gambit of trying to concentrate all of his combat power on the right so as to smash the Roman left.

What is striking then is that Hasdrubal stacks his right – the ‘hammer’ arm – with his elephants and his Iberians, places his Ligurians in the center (where they need to hold) and then puts his Gallic troops up on the hill on his left. It suggests that he doesn’t have a whole lot of faith in those Gauls, because in the event the hill is sufficiently steep that the Romans can’t even really approach their position (Polyb. 11.1.5). In the event, the Romans win when C. Claudius Nero, commanding the Roman right (making no headway up that hill and realizing it), detaches part of his force to extend the Roman left (while leaving a pinning force), wrapping around the flank of Hasdrubal’s right-flank-hammer. But what I think is notable is that where Hannibal – confident and expecting victory – exposes his Gallic troops to let them take the brunt of the losses, Hasdrubal – panicked and merely trying to avoid defeat – puts his Gauls where they can do the least harm.

Both sentiments seem to suggest that the Barcids, at least, held their Gallic allies in relatively low esteem: expending them when convenient but avoiding relying on them whenever possible.

Conclusions

Carthaginian armies were complex creatures – far more so than something like a typical polis army. What I find perhaps most interesting is that for the most part, expansive Carthaginian recruiting was more about broadening Carthage’s base of military resources than it was about acquiring specific capabilities. Carthaginian generals do not seem to use Gallic, Ligurian, Iberian, Greek or Italian troops in dramatically different roles than their own North African troops. Indeed, Gallic and Iberian ‘mediums’ are deployed as line infantry the same as Carthaginian citizen or North African ‘heavies.’ We do not get, for instance, the fairly clear contrast in positioning and usage between ‘heavies’ and ‘mediums’ that we see in Hellenistic armies (https://acoup.blog/2024/01/26/collections-phalanxs-twilight-legions-triumph-part-ib-subjects-of-the-successors/), though of course they have a pike-phalanx to consider. Instead, when Carthage recruits in Gaul and Spain, they seem to want more troops rather than different troops. After all, they already have capable javelin light-infantry (North African lonchophoroi) and heavy line infantry, they just want more of those roles.

The exceptions are clear: Balearic slingers and Numidian cavalry. These are specialist troops that supply new capabilities – longer-range skirmishing (a factor, for instance, at Trebia, where they outrange and outshoot the Roman velites) from the slingers and highly capable, fast-moving skirmish cavalry from the Numidians. Carthage’s heavy cavalry in turn, is a mix of Carthaginian citizen cavalry, North African cavalry, and Iberian or Gallic cavalry – depending on what is available given the time period and location.

The result was not necessarily a more tactically complex army – Carthaginian armies seem to have had fewer moving parts than Roman ones – but the challenge of leading such a polyglot, multicultural army must have been considerable, as Polybius himself alludes to (Polyb. 1.67.4-9), especially when the very recruiting principles of these troops were different: some citizens, some conscripts, some mercenaries raised with money, some allies raised with promises, some vassals raised through very particular personal relationships with the generals themselves. That complexity may serve to explain to some degree why Carthage preferred long-serving generals over a regular rotation: the relationships generals established and their personal knowledge of their armies would have been difficult to pass on. By contrast, Roman armies, while more tactically complex, where organizationally much more ‘plug-and-play,’ each army working more or less like the next.

There is a frustrating tendency in the scholarship to denigrate Carthaginian war-making and I suspect the rather ‘motley’ nature of these armies – which do not look very much like the western ‘ideal’ of an army (uniform and almost mechanical in its function, a ‘war machine’) – contributes quite a lot to this. But Carthage was a military over-performer, especially in the third century: Carthage withstood Pyrrhus and was able to go two full rounds (and one more) with Rome, albeit losing in the end.

I’ve mentioned this before, but the contrast with the Hellenistic kingdoms of the East is so striking: Carthage spends a combined forty-years at war with Rome in the third century, peaking at more than 160,000 men in the field, matching Rome on land and at sea,3 matching the Roman capability of fighting in multiple theaters simultaneously and not-infrequently defeating Roman armies. By contrast, in the second century, the Seleucids and Antigonids manage to fight Rome for just fourteen years combined (including the Fourth Macedonian War, which isn’t even an Antigonid war!), lose every major battle and never manage to put more than 80,000 soldiers or so in the field at any one time (the Antigonids don’t even get close to that).

In short, if we understand the complex Carthaginian mobilization system as an effort to reach more broadly for military resources, we ought to understand it as a success. Carthage, from 254 to 201, deploys massively more military resources than comparable large (larger in the case of the Seleucids) Hellenistic states.

That said, the system was not without flaws. The largest was that it was quite obviously more fragile than its Roman equivalent. Hannibal, despite stunning victories, struggles to get a critical mass of Rome’s socii to revolt. By contrast, Carthaginian control in both North Africa and Spain was relatively more easily disrupted, as shown by the Mercenary War (241-237), the collapse of the Barcid system in Spain after the Fall of Carthago Nova (209) – although Carthage continued to maintain large armies in the area for another five years – and the ability of Rome to draw the Numidians away from Roman service through Masinissa’s defection in 203.

Via Wikipedia (https://en.wikipedia.org/wiki/Masinissa#/media/File:MASSINISSA_-_MAA_23_-_87000716.jpg), a Numidian coin showing a Numidian king, possibly Masinissa (r.203-148) or perhaps Mcipsa (r. 148-118). Interestingly, that Numidian king is represented in Hellenistic style, complete with a Hellenistic diadem, an intentional employment of Greek symbols of kingship in a very non-Greek context.

It is also worth noting that while Carthage’s strategy of recruiting non-state warriors from Spain and Gaul enabled it to field a lot of raw manpower, the warriors they got in the bargain were not as heavily or expensively equipped as either the Romans or Carthage’s own North African troops. The Carthaginian system was thus one that, by the Second Punic War, if not earlier, was forced to seek quantity over quality in order to match the staggering effectiveness with which the Romans had turned Italy into a machine for the generation of military power.

I also suspect, had the Carthaginians not been defeated by Rome, that their system of long-serving generals setting up veritable fiefdoms abroad would have eventually spelled disaster for the Carthaginian Republic. In a sense, we watch this same development play out in the Late Roman Republic, but the Barcid private empire in Spain was if anything even more of a private fiefdom than anything enjoyed by the Late Republic’s ‘rogue generals.’ One imagines, had Carthage continued with an empire that other Carthaginian figures would feel compelled (as rival Roman dynasts felt so compelled in the first century) to establish their own bases of power, leading to predictable results.

All of that said, Carthage’s military system deserves better than to simply be treated as a failure or – even more inaccurately – as the product of an ‘unwarlike’ people. Certainly, the Carthaginians were not able to overcome the Roman Republic – but no one else, not the ‘warlike’ Gauls or the ostensibly more ‘western’ (despite being more eastern) Hellenistic kingdoms – no one else was able to either.

Carthage got the closest, by far, for which the Romans would never forgive them. Ironically, had the Carthaginians been worse at war, Carthage might well have lasted longer.4