… the basic formula for what would become gunpowder (saltpeter, charcoal, and sulfur in a roughly 75%, 15%, 10% mixture) was clearly in use in China by 1040 when we have our first attested formula, though saltpeter had been being refined and used as an incendiary since at least 808. The first guns appear to be extrapolations from Chinese incendiary “fire lances” (just add rocks!) and the first Chinese cannon appear in 1128. Guns arrive in Europe around 1300; our first representation of a cannon is from 1326, while we hear about them used in sieges beginning in the 1330s; the Mongol conquest and the sudden unification of the Eurasian Steppe probably provided the route for gunpowder and guns to move from China to Europe. Notably, guns seem to have arrived as a complete technology: chemistry, ignition system, tube and projectile.

There was still a long “shaking out” period for the new technology: figuring out how to get enough saltpeter for gunpowder (now that’s a story we’ll come back to some day), how to build a large enough and strong enough metal barrel, and how to actually use the weapons (in sieges? against infantry? big guns? little guns?) and so on. By 1453, the Ottomans have a capable siege-train of gunpowder artillery. Mehmed II (r. 1444-1481) pummeled the walls of Constantinople with some 5,000 shots using some 55,000 pounds of gunpowder; at last Theodosius’ engineers had met their match.

And then, in 1494, Charles VIII invaded Italy – in a dispute over the throne of Naples – with the first proper mobile siege train in Christendom (not in Europe, mind you, because Mehmed had beat our boy Chuck here by a solid four decades). A lot of changes had been happening to make these guns more effective: longer barrels allowed for more power and accuracy, wheeled carriages made them more mobile, trunnions made elevation control easier and some limited degree of caliber standardization reduced windage and simplified supply (though standardization at this point remains quite limited).

The various Italian states, exactly none of whom were excited to see Charles attempting to claim the Kingdom of Naples, could have figured that the many castles and fortified cities of northern Italy were likely to slow Charles down, giving them plenty of time to finish up their own holidays before this obnoxious French tourist showed up. On the 19th of October, 1494, Charles showed up to besiege the fortress at Mordano – a fortress which might well have been expected to hold him up for weeks or even months; on October 20th, 1494, Charles sacked Mordano and massacred the inhabitants, after having blasted a breach with his guns. Florence promptly surrendered and Charles marched to Naples, taking it in 1495 (it surrendered too). Francesco Guicciardini phrased it thusly (trans. via Lee, Waging War, 228),

They [Charles’ artillery] were planted against the Walls of a Town with such speed, the Space between the Shots was so little, and the Balls flew so quick, and were impelled with such Force, that as much Execution was done in a few Hours, as formerly, in Italy, in the like Number of Days.

The impacts of the sudden apparent obsolescence of European castles were considerable. The period from 1450 to 1550 sees a remarkable degree of state-consolidation in Europe (broadly construed) castles, and the power they gave the local nobility to resist the crown, had been one of the drivers of European fragmentation, though we should be careful not to overstate the gunpowder impact here: there are other reasons for a burst of state consolidation at this juncture. Of course that creates a run-away effect of its own, as states that consolidate have the resources to employ larger and more effective siege trains.

Now this strong reaction doesn’t happen everywhere or really anywhere outside of Europe. Part of that has to do with the way that castles were built. Because castles were designed to resist escalade, the walls needed to be built as high as possible, since that was the best way to resist attacks by ladders or towers. But of course, given a fixed amount of building resources, building high also means building thin (and European masonry techniques enabled tall-and-thin construction with walls essentially being constructed with a thick layer of fill material sandwiched between courses of stone). But “tall and thin”, while good against ladders, was a huge liability against cannon.

By contrast, city walls in China were often constructed using a rammed earth core. In essence, earth was piled up in courses and packed very tightly, and then sheathed in stone. This was a labor-intensive building style (but large cities and lots of state capacity meant that labor was available), and it meant the walls had to be made thick in order to be made tall since even rammed earth can only be piled up at an angle substantially less than vertical. But against cannon, the result was walls which were already massively thick, impossible to topple over and the earth-fill, unlike European stone-fill, could absorb some of the energy of the impact without cracking or shattering. Even if the stone shell was broken, the earth wouldn’t tumble out (because it was rammed), but would instead self-seal small gaps. And no attacker could hope that a few lucky hits to the base of a wall built like this would cause it to topple over, given how wide it is at the base. Consequently, European castle walls were vulnerable to cannon in a way that contemporary walls in many other places, such as China, were not. Again, path dependence in fortification matters, because of that antagonistic co-evolution.

In the event, in Italy, Charles’ Italian Vacation started to go badly almost immediately after Naples was taken. A united front against him, the League of Venice, formed in 1495 and fought Charles to a bloody draw at Fornovo in July, 1495. In the long-term, French involvement would draw in the Habsburgs, whose involvement would prevent the French from making permanent gains in a series of wars in Italy lasting well into the 1550s.

But more relevant for our topic was the tremendous shock of that first campaign and the sudden failure of defenses which had long been considered strong. The reader can, I’d argue, detect the continued light tremors of that shock as late as Machiavelli’s The Prince (1532, but perhaps written in some form by 1513). Meanwhile, Italian fortress designers were already at work retrofitting old castles and fortifications (and building new ones) to more effectively resist artillery. Their secret weapon? Geometry.

Bret Devereaux, “Collections: Fortification, Part IV: French Guns and Italian Lines”, A Collection of Unmitigated Pedantry, 2021-12-17.