It is an interesting fact that battles are one off events, contingent and unrepeatable. Unlike wargamers, generals were forced to give battle with what they have at the time, not some perfect balanced force. They have to put up with uncooperative allies, insubordinate subordinates, enemies who are trying very hard to kill them and their troops, and a probably lack of support from their governments who, in most ages, worry about the cost of the whole thing and cannot see why the war is not won already.
Even at a campaign level, generals tend to spend more time worrying about food and fodder than about strategy and tactics. A starving army is one which is already beaten before battle is joined. In some ages, and places in the world the most successful generals are those who win without joining battle. Indeed, it can be argued that the quest for the decisive military encounter is one of the oddities of western culture.
It can also be argued that decisive battles are not that decisive. Professional historians, even today, rather ignore the battle. They prefer, in many ages, at least, to ferret about in archives looking at pay documents and muster lists, without ever really considering what the point of all this bureaucracy is. That is not to say archival work is unimportant, far from it. We do know a great deal more than we did about the armies of the past from this work. But the work has generally been presented as giving insight into the society of the time, rather than the men who actually fought the battles.
We could, then, argue, that modern military history (with a few exceptions) is rather lopsided, and has a single angle view of warfare, mostly relating to the raising and financing of the armies, and their impacts on society and the state.
As Austin Woolrych wrote a few years ago, he thought that the important thing about wars was who won and how. In his view, explaining the English Civil war in terms of the rise of the gentry, or the fall of the gentry, or the rise of the merchant class, or the fall of the noble class, or whatever the current popular theory of the day was, it was actually more important, and more interesting, to examine the reasons why the New Model Army won.
Unfortunately, professional historians do not seem to have heeded that comment.
On the other hand, wargamers are, if anything worse. They peruse texts, usually in translated form – guilty!) , picking out occasional bits and pieces of comments about military hardware or tactics, and present them as ‘the truth’ of ‘how it happened’. I’ve mentioned before the problems over bow range which Tacitus refers to, and it is not the only such comment which has dominated our wargame rules.
The interminable threads which crop up occasionally on email fora such as Ancmed, about (to quote a recent example) whether Caesar’s legionaries used their swords or not, is a case in point. I confess, after the first dozen or so posts on the topic I rather lost the will to live, but essentially quote is being traded against quote, Livy against Caesar and so on, in an essentially unsolvable argument about whether pila or swords were decisive in battle.
The problem is that the evidence is simply insufficient to answer the question. We do not know, we cannot know in the absence of significant archaeological evidence from battle site how those participants who died met their ends. If someone were to dig up the site of Boudicca’s last battle, and most of the British bodies found had pilum wounds, then we might be able to say something a bit more useful than quoting chunks of Latin at each other.
But even if this were to happen, it would only show one thing, and that is that at that particular battle, the pilum was decisive, probably. It is not, in fact, a generalizable truth that, therefore, in all battles the pilum was king.
In this circumstance, we have to say the Hume’s inductive scepticism is correct. The observed case does not give the answer to all cases. Even if we had the evidence for Boudicca’s last stand, we would not have evidence for anything except that battle. Without finding and digging up the evidence for every chronicled battle and analysing it in such a way, we cannot say anything in general about the interaction between legionary and Celtic warrior.
The problem is, of course, as a wargame writer, we do have to say something coherent and complete. It is not enough to say ‘sorry, don’t know, so no legion vs. Briton bashes, please because we don’t have the evidence’. A rule set like that would sell even worse than the ones I write.
The problem is thus that we are stuck, and have to synthesise some model for battle without sufficient evidence. In computational language, the problem is under constrained, underspecified and thus, unsolvable. So we have to guess.
The issue with such guesswork is that it can quickly fail under any sort of scrutiny. A different interpretation of a literary source, or even just a different translation, can mean that an otherwise solid appearing rule set is regarded as a failure.
The only way to tackle this, so far as I can see, is by an internal coherence to the rules. This is what I’ve tried to do with PM: SPQR. It is not for me to say if I was successful or not, but the idea is to present something where the interpretation is clear and the assumptions made are exposed. Then, if someone disagrees with my interpretation, they can at least find the source of it and see for themselves.
On the other hand, it does seem to make the writing of rule sets rewarding, in that people can use what you think and even, if they wish, modify it to conform to their view of history. And I suppose that is why I’m starting writing the Greeks set.