On Artillery

I have, on and off, been pondering early modern artillery and its usefulness, effectiveness, and, above all, why armies bothered dragging the pieces over the countryside. After all, I think it was King Charles I who commented that the artillery train was a sponge that soaked up all the money.

By the War of the Three Kingdoms cannon had been around for a while. I think the first illustration of one dates from the early 14^th Century, and they were certainly deployed by the English in the 1327 campaign, not that they were particularly useful on that occasion.

Still, by the 17^th Century, things had improved. Powder, for example, was more powerful, and improved construction techniques meant that the pieces could take the strain, although a full charge of corned powder could well be too much for many gun tubes. That, I imagine, comes down to the skill of the individual gunner and his knowledge of his piece.

Gunpowder weapons of all sorts were horribly inaccurate. Bert Hall’s excellent book on early firearms reports a test of 325 discharges of smooth-bore muskets (dating from the 16^th—18^th Centuries). Surprisingly high muzzle velocities were reported (averaging 454 m/s, or 1490 ft/s), although that depends, of course, on the charge weights. There are some similar results for ACW artillery.

Spheres are the worst regular shape possible for generating drag, and the bullets decelerated at about 2.5 m/s for every meter of travel. About half the kinetic energy is lost at 100 m from the muzzle. Hall reckons that early modern weapons only have a possibility of being lethal at 100 – 120 m. At Blenheim and Fontenoy the high casualty rates were generated by shooting at about 40 meters. Drag meant that under normal circumstances at best 10-20% of shots hit the target, and, frequently, it is more like 5%.

Of course, artillery kept its velocity better, or at least, because the bullet was heavier they maintained lethal velocity for longer. But both firearms and artillery were notoriously inaccurate. Frederick the Great of Prussia calculated that 650000 rounds had been fired at Chotusitz (1742) killing about 2500 Austrians and wounding a similar number. That is, 1 in 130 discharges hurt someone. Other numbers bandied about were even more pessimistic.

Inaccuracy was also measured. At 100 meters the size of a square enclosing half the bullets is about 1 foot on each side. Actually, rifled muskets from the same period did not do much better. So the chance of hitting someone at 100 m was about 50%. Some weapons were better than others.

The problem is that spherical balls with windage – the difference between the diameter of the ball and the inner diameter of the barrel – bounce along the barrel. Ward reports that the standard windage was about a quarter of an inch in cannon. So the ball strikes the sides of the barrel after the gun is discharged, and picks up spin from such collisions.

As any cricketer or baseball player will know, balls with a spin on can behave in unusual, counterintuitive, ways. In a controlled spin, such as in a sport, this can be helpful. When aiming a weapon with the spn in an unknown random direction, it will move in the air. Hall reports an amusing experiment with a slightly left-bent barrel where the ball actually struck the target to the right of the centreline.

Artillery, although with longer ranges, suffers from the same issue. The ACW artillery deviated by 3 feet at 600 yards and 12 feet at 1200 yards for a 12-pounder. Lighter pieces seem to have deviated more. The aim of gunners, of course, was also to keep their shot below the height of a man. There is also the consideration of ‘point blank’ range, where the shot starts to drop away from its expected trajectory. Ward records point blank to a saker (5 ½ pounder) as 300 paces and a culverin (19 pounder) as 420 paces, which is the furthest in the list. Ultimate random ranges were 1500 paces and 2100 paces respectively, although the deviations Hall records for smooth bore artillery indicate that you were unlikely to actually hit what you aimed at at those ranges.

I have long suspected that part of the impact of artillery is simply the noise. A cannon discharging is, from my experience of attending re-enactments, very, very loud. Musket fire is less so, but these noises were probably the loudest ones available to most people. It would, at least until you got used to it, be frightening. Along with the noise the smoke and the possibility of being hit by a cannonball (which could damage you if it hit you at even extreme ranges), there was quite enough to perturb most people.

In Polemos: ECW we have a maximum range of artillery as 4 base widths and a factor of 3. At over 1 base width the artillery loses 1 from its factor. This, as JWH has pointed out in the past, seems a bit too swingeing. I suggested an alteration, making the artillery fire as normal over 4 base widths and then losing 1 per base width from its factor. That seems to make artillery too effective, as my experiment with Beatrice last week might suggest. I also added a second dice roll at over 6 base widths – the cannon needed to roll a 6 to hit.

In my homebrew Wars of the Counter-Reformation rules, I have adopted the above model. It attempts to depict the problems with the deviation of cannon shots, as well as the disruption artillery fire can cause. Beatrice disrupted the attacking Parliamentarians but did not cause any problems other than command and control. That seems about right to me, although I still suspect that the factor is a little high – perhaps it should be 3 rather than 4, although firing at dense pike blocks might give a bit of an advantage. But I am trying not to have too many factors in the rules.

So, in summary, I think early modern artillery should be a disruptor in battles, rather than a critical element. The hope of early modern generals must have been to shake the enemy and disrupt their deployment and advance rather than anything else unless there were specific tactical situations, such as narrow bridges or flanking fire, which would give the cannon an advantage. Sieges, of course, are a different thing again.