Wednesday, February 29, 2012

Dreaming of Cavalry, III: Seeing Death

Cavalry, again. Specifically, what the Encyclopedia Britannica's designated expert thought it was going to be doing in the next war, as of 1909. Interestingly, it's not (operational level) reconnaissance. Today I'm going to pursue the point, and lay some groundwork for talking about Jutland next week.

A professor emeritus at Toronto reported conceived an interesting, if self-indulgent project when I was a graduate student there. He assessed the Eleventh Edition of the Encyclopaedia Britannica, once considered  an epitome of knowledge and latterly, if I may put it politely, an artefact of its times. His conclusion was that the latter reputation was somewhat overstated. The past was not a safety school.

 As an owner of a copy of the Eleventh Edition, I basically agree, but then there's Frederic Natusch Maude, editor of an awful translation of Clausewitz, as well as tracts more fully explicating his social Darwinist creed, and, of course, the Eleventh Edition's article on "Cavalry." (Lifted from here, corrected, without too much guilt after I noted that the author didn't edit out Maude's explanation of why Catholic cavalry is inherently inferior to Protestant)*:

Imagine an army of 300,000 men advancing by five parallel roads on a front of 50 m., each column (60,000 men, 2 army corps) being covered by a strong advance guard, coming in contact with a similarly constituted army moving in an opposite direction. A series of engagements will ensue, in each of which the object of the local commander will be to paralyse his opponent's will-power by a most vigorous attack, so that his superior officer following him on the same road will be free to act as he chooses. The front of the two armies will now be defined by a line of combats localized- each about a comparatively small area, and between them will be wide gaps which it will be the chief business of the directing minds on either side to close by other troops as soon as possible. Generally the call will be made upon the artillery for this purpose, since they can cover the required distances far more rapidly than infantry. Now, as artillery is powerless when limbered up and always very vulnerable on the flanks of the long lines, a strong cavalry escort will have to be assigned to them which, trotting forward to screen the march will either come in contact with the enemy's cavalry advancing with a similar object, or themselves find an opportunity to catch the enemy's guns at a disadvantage. These are opportunities for the cavalry, and if necessary it must sacrifice itself to turn them to the best account. The whole course of the battle depends on success or failure in the early formation of great lines of guns, for ultimately the victor in the artillery duel finds himself in command of the necessary balance of guns which are needed to prepare the way for his final decisive infantry attack. If this latter succeeds, then any mounted men who can gallop and shoot will suffice for pursuit. If it fails, no cavalry, however gallant, has any hope of definitely restoring the combat, for against victorious infantry, cavalry, now as in the past, can but gain a little time. This time may indeed be worth the price at which it can be bought, but it will always be more economical to concentrate all efforts to prevent the emergency arising. After the Franco-German War 'much was written about the possibility of vast cavalry encounters to be fought far in advance of the main armies, for the purpose of obtaining information, and ideas were freely mooted of wide-flung raids traversing the enemy's communications, breaking up his depots, reserve formations, &c. But riper consideration has relegated these suggestions to the background, for it is now evident that such expeditions involve the dissemination of force, not its concentration. Austria and France for example would scarcely throw their numerically inferior cavalry against the Germans, and nothing would suit them better than that the latter should hurl their squadrons against the frontier guards, advanced posts, and, generally, against unbeaten infantry; nor indeed would the Germans stultify their whole strategic teaching by weakening themselves for the decisive struggle. It follows therefore that cavalry reconnaissance duties will be strictly local and tactical, and that arrangements will be made for procuring strategical information by wireless telegraphy, balloons, motor cars, bicycles, &c.,

So you thought that cavalry was for reconnaissance? That's silly. That technology stuff will rise to its steampunk occasion. Or, at least, it had better rise. The cavalry is needed for its real work: charging, if necessary, "great lines of artillery."

This is, I think that we can agree, bugnuts.
Photograph by Dan Alex, hosted at
Again, this is the M.1897, the French field gun, adopted by the United States Army, thanks to which rootwebs, of all places, hosts the service manual. Also, Craig Swain's discussion.

This is transformational technology. If I could just communicate the importance I sense in the incredible range of technological developments from the 1870-1895 generation that you are seeing here in field operational condition, you'd probably mistake me for Ray Kurzweil's grandpa. You've got your nitrated cellulose, ancestor of all modern plastics, albeit still a little more flammable than might prefer; nitroglycerine, the first true explosive, by the technical definition of having the brisance needed to shatter rock and thus make the Suez and Panama Canals possible; the first forged-steel made in industrial processes on Siemens hearths; a little revolution in precision engineering (in the recoil system) that will lead in short order to the machine age. The result?  A fully recoil-compensated gun that fires a 16lb round charged with 290 shrapnel balls (actual shrapnell, not shell fragments) out to 7500 yards. A good crew can get off two rounds a second. But I'm not going to argue that the "pre-Singularity" (work with me here, I'm playing with a trope) happened because of the incredible culmination of multiple strands of technological development that is the Soixante-Quinze. I'm arguing that it was the reaction to it that made some vital bit of modernity real.

The sheer magnitude of technological change inherent in this weapon is, however, Colonel Maude's excuse. He might be a bit of a second-rater (search the name if you want to find out why he's "already accepted"), but if he'd really assimilated what this gun could do, he would surely have realised that he'd just come up with a tactical solution to the problem of there being too many horses in the world.

Why? I'm going to focus on two other aspects of the Modele 1897 that rise to my attention: the aiming telescope, and wooden wheels. I'm not going to be able to put them together as smoothly as I'd like below, but they both represent, in different ways, a revolution in our relationship with the landscape, our understanding of what information  might be. It's all at once, a great leap into the dark. We'll do it in the context of the state's full-throated preparation for a great power war a little more than a decade away, where we will test our understanding of the synthesis by throwing a few million lives away. And then we'll stand back and wonder what we've done, and how we've changed the world, when it's already changed and we don't even quite know how.

Notice why Maude thinks that the fronts of the opposing armies will thicken up along great lines of artillery. It's because the horsedrawn artillery is so much faster than the infantry. This is vital, and, in its own way, consequential. Notice how the discussion of the British QF, 18 Pounder starts with the design issue of its trail. This is the part of the gun that has to be hooked up (quickly) to a team of 6 horses. Incidentally, the resulting pole trail restricts the range of the gun to an absolute maximum of 9,300 yards, which no-one cared about in 1905, because they thought that even less range was perfectly acceptable. A new 18-pounder with a splayed-out trail to allow the breech to depress further didn't come into service until well after the Battle of The Somme. But don't think of that as an oversight. Think of it as a transformation of our relationship with the landscape, and it will make more sense: wooden wheels and telescopes.

The first part is the wheels. This goes to roads, which, when I started on this line of thought led me to an old civil engineering textbook by the simple expedient of pulling the most useful volume off the shelf in the general neighbourhood where Civil Engineering was shelved. Those halcyon days are gone, but Wikipedia on John Macadam will do well enough for these purposes. Cut through the stuff about camber and drainage and foundation. What matters here is the concept of "metalling" the road by paving it with a layer of crushed rock. That's why the chain gang breaks big rocks into little rocks, and where we get the "Russian steamroller." Crush rock, once it has had the heck rolled out of it and been washed down with water, forms a level, flat, water-impermeable, smooth surface. We call it "pavement,"  just as we call modern asphalt-on-concrete roads "macadamised." It's not true. The roads that people were building at a flat out rate in the last half of the Nineteenth Century with steam tractors and steam rollers were much harder than modern roads.

This is where Whig history serves us badly. This wasn't a new invention of the Eighteenth Century, on the contrary, roadbuilders had been working towards this kind of surface for a long time, restricted mainly by the limits of labour. As a result, wagon wheels, at least on the main roads had been getting narrower and narrower, eventually acquiring first an iron "tyre" and eventually coming to be made of solid steel. The thinner the surface in contact with the road, the less the friction, and the greater the freight that the horses could haul. I've recommended Dorian Gerhold's books on the Transportation Revolution before, and I'll do it again. The rapid pace of economic development in the Eighteenth Century is over-explained by one revolution too many in my mind, but as long as we're taking cognisance of dubious monocausal explanations for a global phenomena that's clearly in fact caused by the Columbian Exchange, we should notice this rather important one. New and better roads, narrower, harder wheels, and more stuff transported overland at less cost.

So why the change to asphalt? Because of those nasty bicyclists, of course. Long before our modern culture wars, the first pneumatic-tyred bicycles took to the road and proceeded to start vacuuming up the water-repellent rock-dust pavements that made these roads possible. The problem really took off with the coming of the automobile, and the First World War was fought in the midst of the self-unravelling of our land transport infrastructure.

That being said, the M. 1897 had neither a steel tyre nor a rubber one, but rather wood. That was to reduce the ground pressure and allow the guns to run freely off road. It also made them lighter, so that horses could haul them and men could manhandle them, and it get the riggers in a job. We live in a world where pavements are perfect. It's easy to write, say, Roman history, and forget that in the winter, the Po Valley and the Pannonian plain between Vienna and the Iron Gates, or the Fens that cut Anglia off from the rest of the Low Countries, or so many other places, were inland seas. But they were. That's a world that we're cut off from, and we moved from that world to the one where the ground is always firm beneath our feet, literally under fire, as we sought to bring ammunition up to the guns in 1914--18.

That's the wheels: now the telescope. See, it's not enough to be able to fire a "storm" of shrapnel at the enemy, whether cavalry obligingly charging in the open, or, more plausibly, enemy infantry hiding in holes in the ground and taking pot shots at you with their notoriously smokeless rifles. You need to be able to aim at them, and very quickly, too. The M.1897 had a gun-free sighting mechanism that allowed the gun to be aimed without disturbing the view of the gun-aimer through his telescope. That was a good start on matters, but you also need to have a range so that you can elevate the gun. And, in fact, you even need to be able to take into account the gun's variation from level across the axle. It's all quite complicated. Actually, it's more complicated than is really feasible under the Oh-My-God-People-Are-Shooting-At-Me conditions.

Fortunately, this had already been taken into account. Not my artillerists, mind you, as they were still of the opinion that the gun was the best ranging tool, and not, science notwithstanding, entirely without reason. It was, rather, the professional field of land surveying that had been thinking about this problem for about a century now, and come up with a wide range of gadgets to speed up, and thereby encheapen (because imitation is the sincerest form of-), surveying. It's right there in my Eleventh Edition, (v. 26: 341ff of the compact edition): Tacheometry: the art of rapid surveying. The Wikipedia article even lifts the first few paragraphs of the article. Unfortunately, it thereby misses the key point in the rather beside-the-point technical issues of, you know, actually surveying real estate. Which is that, far more than the other origins of the concept of which I've been made aware by the fine folks at the IEEE Annals of the History of Computing, the surveyors were the first guys to really, seriously try to solve and simplify computing problems by designing machines to do it for them.

Consider this beautiful little piece in the Smithsonian's collection of surveying and geodesy instruments.

It's an example of Eckhold's patent omnimeter, explained by the Smithsonian curators here (I'm trying to drive traffic to the Institution because I'm guilty about image leeching off them). It has a little level, rangefinder-type gizmo, and actual theodolite, allowing the surveyor to "accomplish the work of theodolite, level, and Chain," without actually having those. Or paying guys to do the Chain readings, of course. Cheapskates.

One more thing: this particular example of a computing/surveying instrument is by a firm that, if you've read your computing history, you may have heard of: Elliott Bros. You will not, unfortunately, read much about the company's prehistory as an optical surveying instrument maker in the official company history. That stuff is before our ken, back in the misty prehistory that so often claims history of technology before we realise that it's important. Suffice it to say here that the transition is no accident. As John Brooks documents, Elliott Bros. was brought into the naval fire control problem precisely the need for optical instrument makers in the field, and gradually transitioned from instrument-maker to computer pioneer because of the close connection between computing optical instruments and computing fire control solutions.

But that's for next week. This week I want to talk about moving guns and seeing death. If a gun can fire 30 shells a minute, each containing 280 balls to a reliable point in space anywhere within 7500 yards and within the admittedly limited traverse of the gun, or, with a little more delay, after moving the gun, then being visible to any point within that range is no longer an option.

Now, you can cheat here, and use a gunshield. In fact, everyone cheated. Gunshields allowed guns to approach enemy rifles, and even cannons, much more closely. Fit a gunshield-and-gun with some kind of auto-propulsion, and you might even have a new tactical concept. Actually, it's not really  new. People have been barnstorming tanks since the beginning of time. The truly novel concept is hiding the gun.

After struggling for years to produce a gun that could move faster and shoot straighter and more devastatingly, people are suddenly confronted with the serious proposition that they have to fire from out of line of sight. It's not a new idea, in that the siege, coastal, and even howitzer batteries had been doing it for years. The difference is that field guns are intended to bring massive and rapid fire on moving targets at unknown range. How do you even do that? Someone can stand out in the open and tell you where you should shoot; but how do you convert that information, quickly and accurately, into gunlaying instructions? More importantly, what do you do when the observer gets shot, or, at least, has to move out of earshot?

With tacheometry. Here's a discussion of the rapidly evolving technology of what I'm going to call computed firing solutions. All kinds of factors are taken into account with remarkably complicated instruments, considering that they're meant to be used in the field and that some of them are mounted on a gun barrel. 

And it's all predicated on a landscape that the eye can no longer view. What can see, can be shot. We need to be able to build a synthetic landscape, based on information acquired by other means --Maude's "&c." And we need to communicate that information to the gun battery, and to turn that information into gun-aiming  instructions. We're going to need aeroplanes, and electrical communications, and cameras, all of which seem a great deal more important, and, to be fair, are more important (for now) than the tacheometric devices that compute the aiming instructions. 

I'm going to come right out and call this a virtual landscape. It basically consists of data that has to be captured and processed tacheometrically. It has to be computed. It is an information-technology landscape, and it has been created in less than twenty years since the first  M. 1897 soixante-quinze rolled past the reviewing stand at the Bastille Day parade. We've gone from a world encompassed and imagined by hunting-and-shooting folks riding by in mad pursuit of a fox to a virtual world. We haven't even managed to invent the heuristic that will explain what we're working with, and won't for another three generations. And you think that the pace of technological change is fast today.

*"It is to the growth of Protestantism that cavalry owes its next great forward leap. To sweep the battlefield, it was absolutely essential that men should be ready to subordinate selfish con siderations to the triumph of their cause. The Roman Catholicism of the day gave many loopholes for the evasion of clear duty, but from these the reformed faith was free, and it is to the reawakened sense of duty that Gustavus Adoiphus appealed." I told you that the Nineteenth Century was nuts about this religion thing. I'm just happy that we're past it now.

No comments:

Post a Comment