Warfare 218 BC

Novelists writing about the Hannibalic Wars

What for archeologists, classical scholars and historians is a continual barrier to academic understanding, is for the historical novelist like me, our greatest strength. Imagination and invention isn’t just making it up. Any historical novelist worth their salt, implies that making it up only takes place after all the written sources of evidence, archaeology, experimental evidence, practical and logical implications have been borne in mind, sifted through and intellectually exhausted. Only then can the imagination be fired up and take flight. At its very best, making it up ought to be higher up the food chain, because it’s founded on a pyramid of solid fact and research. Historical novelists inhabit the hinterland between evidence and probability, not truth and fantasy. This liminal ground between these two worlds is what fires and excites us as writers.   

When L. Lewis Johnson published his history of flint knapping experimentation in 1978, archaeologists were outraged. How could a humble workman, albeit a skilled stone toolmaker, write authoritatively about the Neolithic period? Quite rightly, toolmakers turned this argument on its head and asked of archeologists how they could possibly write about flint knapping unless they were skilled toolmakers and had tried it themselves? Thankfully, we now live in a more enlightened age where replication experiments are now a respectable and accepted branch of archeology. 

There’s been an interesting debate recently about the boundaries between imagined (novelistic) history, and evidential/documented history. Hilary Mantel’s recent BBC Reith Lectures stresses that history is always imperfect. While there may be some facts and figures from the past that we know to be indisputably accurate, there will always be many gaps, errors and unreliable witnesses. Mantel says history is “what’s left in the sieve when the centuries have run through it – a few stones, scraps of writing, scraps of cloth.” Writers of historical fiction “need to know ten times as much as they will ever tell.”

Bearing in mind Johnson’s practical archaeology and Mantel’s warnings about errors and unreliable witnesses, let us apply this to a simple problem: how did the famed Numidian cavalry in Hannibal’s army fight, and what tactics did they employ?

Let’s start with the facts. We know reasonably accurately what they looked like and what weapons they carried. There is clear evidence from artefacts and illustrations (e.g. Trajan’s column) they wore a simple tunic, carried a small round shield and were armed only with a clutch of javelins. They were not armoured in any way (helmets, mail, leather cuirass etc.), rode small horses (almost certainly mares) without saddles, and wore their hair in a style we today would describe as ‘dreadlocks.’ The lack of saddle and reins is already providing evidence.

Numidians must have been skilled horseman. It is thought they controlled their horses using the string around their horses’ necks. Every horseman I’ve spoken to about this says it would be of no use, and that it’s perfectly possible to direct a horse using knee and leg movements. But when it comes to Numidian battle tactics we enter the realm of probability and theory almost immediately. Polybius gives us some tantalising glimpses.

This sounds like classic hit and run tactics. So if we were to reconstruct this tactic, it might possibly happen like this. A line of horseman riding abreast would charge the enemy (with great daring), raising the speed of their mounts until they reached a gallop, release their javelins then wheel around and retreat. This form of attack could be maintained until each horseman ran out of javelins! 

There’s a great deal of guesswork going on here. But also a great deal of reasoning, logic and common sense. Consider a Numidian horseman as a weapons system; the most effective way to deliver that weapon would be at close range and maximum speed. Numidian javelins must have been light, otherwise they would not have been able to carry a clutch of them (say five or six – any less than this would mean fewer charges). Light javelins would only be effective if launched as a volley, hence it’s logical to suppose Numidians charged together in line. 

How close could they approach their enemies at full gallop and still turn around their horses and retreat? The answer to this is astonishing; they could get remarkably close. Perhaps as close as five or six metres. The evidence for this can be seen at any gymkhana event.

In barrel racing, small agile ponies are trained to turn 180 degrees in two strides! From a headlong gallop, it’s possible to turn a horse completely around in two strides, head off in the opposite direction (retreat), almost immediately reaching a gallop. It requires two things; skilled horsemen, and a light, agile pony. Large stallions would not be able to accomplish this feat. 

This gives us another clue about a Numidian cavalry line. The horseman would have had to space themselves out a little if they were to turn around without colliding with their neighbours on either side. It would not be possible to charge stirrup-to-stirrup, as it were. This suggests another question; were Numidians trained to turn their hoses around in the same direction? If they were, this simple solution would avoid accidents or mishaps.

The next question is how effective was a Numidian light javelin? In combination with a leather throwing thong and hurled from a galloping horse, the answer would appear to be very effective, if not deadly. Let’s look at the throwing thong first.

German reenactors, experimenting with the longer Greek spear (Akontion) at Olympus before the 2004 Athens Olympics, discovered they could throw the weapon forty metres further if they used a thong (Greek; ankyle/amentum). Bear in mind the reenactors were not trained athletes. It has been suggested that Greek Olympians, using a throwing thong and a run-up, might be able to reach distances of 100 metres. The lighter Numidian javelin would easily approach these distances, especially when thrown from a galloping horse adding a further 20 – 25 mph to the javelins velocity. 

Such a weapon, thrown say ten metres from the enemy, would have significant penetrative  qualities.

There is another significant advantage to using a throwing thong. When the leather unfurls, it sends the missile into a spin. Like the rifling effect on a bullet, a spinning missile is far more accurate.

The use of the throwing thong might also clear up another mystery. There is some evidence that Numidian cavalry sometimes fought on foot in melee situations. Why would they abandon their advantages of speed and mobility? In melee situations where all coherence had been lost and mobility no longer needed, with a throwing thong even a light javelin would still be a highly effective weapon on foot.

For the historical novelist this is where the imagination can now take off. What’s left in the sieve when the centuries have run through it has been evaluated and assessed for its likely accuracy. Now as a write, all I want is for my Numidians to thunder forward and wreak havoc.

Robert M. Kidd 

October, 2020      

Buy Kindle edition of The Walls of Rome from Amazon

Paperback coming very soon.