Saturday, 23 November 2013

The Downside of Models

I have written quite a lot about how much we use models, and how useful they are. Models, I have claimed, are basic to how we wargame, and, indeed, how we can wargame. Without models there would be no rules, no process of playing the game and, if I stretch the use of the term model to the scale models of soldiers that we use, there would be no figures to wargame with.

Now, of course, in a true spirit of equitability, I am going to have a go at our use of models. I am going to contradict myself, even though I will claim, I imagine, before the end of the post that I am not really doing so. We shall see.

What possible problem could there be with using a model for a wargame?

Firstly, of course, I have already said that we actually have a large number of models interacting within our rules. We try to ignore the fact, and treat a set of rules as a single model, but the clue is in the name: a ‘set’ of rules. Each rule within a rule set constitutes a model, which is part of an overall process, a dynamic model, of a battle. Thus, the rule models which  we have have to be made to work together.

But that is not, in many cases, the issue. What is the real issue here is the fact that while models are useful, and perhaps inevitable, in life as in wargaming, they do have a tendency to limit as well as assist our thinking. A model is an intellectual construct for thinking, for making things intelligible, but it is always limited, in the sense that a model is not the complete thing modelled, but an abstraction. A model, we hope, picks out the important bits of the thing modelled and allows us to concentrate our intellect upon it.

Now, as I said, this is all well and good and inevitable. We cannot hope to understand real life in any depth as a whole. The complexity would simply overwhelm us. A quick look at the history of science and philosophy (the two were, after all, only separated in the 17th century, or so. Newton wrote far more theology than he did physics (plus the fact that he was fairly close to being a heretic)) shows that attempting to understand the world via its details is more or less impossible. The big breakthroughs have come about when abstractions have become available.

For example, and to stick with Sir Isaac, gravity was a puzzle. It was thought, for example, that there was a force pushing down on a rock in the air which made it fall to earth. Alternatively, a rock fell because its natural place was at the centre of the universe, which was the centre of the Earth. The only thing that stopped the rock achieving its final goal was other bits of rock attempting to do the same thing.

This sort of issue, and an apparently unrelated one due to the motion of the planets, was resolved by Newton with a new abstraction, that of gravity, as a universal force. The fact that the details could be worked out mathematically and checked against empirical measurements was a bonus, of course, and launched all sorts of research and some fairly wild speculation. But the fact is that newton had proposed a new model of gravitational attraction, and it worked. The older, more concrete models were swept away in a matter of a century or so.

The thing is, however, that Newton’s models for the universe became a constraint on how people could think about how the universe worked. While the evidence mounted up, for example, that atoms were not billiard balls, and that the perihelion of Mercury did advance, physics was a bit constrained by the very success of the Newtonian system. It took a lot of work, brilliant thinking outside the box and some careful experiments before modern physics, of relativity and quantum mechanics, was accepted. Even then (and even now) people find these parts of physics hard to grapple with; the Newtonian model, because it refers to the scale of people and our perception, dies hard.

So, how does this affect our wargaming. As I have said, wargaming is about models, and systems of interacting models creating the whole rule set and dynamics of a battle. But if you look at most rule sets, including the ones I have written, then fall into distinct bits.

These models are fairly consistent across rule sets. We have a section for movement, a section for fighting, a section for morale. These are the basic models, and they are, excluding details, fairly similar. These rules constrain how we think about wargaming and, in all probability, have a normalising effect on how we think about battles. We can, and probably do, back project our views of, say, morale, onto an account of a battle.

Thus, the models which we have evolved in wargaming, in modelling a real battle, then constrain out thinking about that real battle. We find it more difficult to think outside the constructs of our intellect to find a different view of what happened, of how the world in fact works. The anomalies of battles which are not described by our rules and models are shelved, much as the anomalies of Victorian physics were shelved, until someone came along and incorporated them into a new scheme of models.

Now, of course, history is not physics. Models in physics are everywhere, and new experiments can offer new perspectives, while we only have one set of fairly dodgy data for our battles. But I do not think that that excuses us from looking for new models, new ways of understanding battles and gaining insights into them.

However, I will have to confess that I am not really sure how to go about this. But I do think that one thing is clear. We have to try to avoid basing our models on secondary sources, and go back to original descriptions of battles in the hunt for new explanations, new models. Secondary sources are, often, I think, based around an understanding of the battle which is similar to those we already have, and so will only reinforce our prejudices.


  1. On the topic of the pursuit of "realism" in wargames by the building of mathematical models, I did a blog post a couple of years ago which I would struggle to add anything to now. If anyone is interested, and can bear it, check out

    I claim no great merit or originality, but the cast includes Peter Young, BP Hughes and NN Taleb - no expense spared there, even if the script was a bit iffy!

    Maybe we do need new models - I think what needs to be new is the approach - some of the models we have already inherited suffer from the fact that the originators tried to stick too closely to reality, or their idea of reality, which is significantly different.

    1. Interesting. I recall Firepower, and wondering how anyone ever managed to win a battle if they didn't hit something. I think Bert Hall's Renaissance Arms and Armour does a similar thing for earlier firearms, his point being, in part, that muzzle loaders were inherently inaccurate, due to the bullet bouncing down the barrel.

      On the other hand, muskets and artillery going off and possibly killing you is quite scary enough. I guess we could count casualties in figures as reductions in effectiveness, but it is confusing one model with another. Plus, when you remove figures, the units get small (something very odd with WRG rules).

      As for reality, my guess is that we'll never get there, but we might obtaining an increasing grasp on it as we innovate. Possibly...

  2. I think one of the most refreshing things about wargaming as a hobby is the host of different models that are out there and the readiness of many wargamers to experiment with new ones. Yes, a lot of the models use the same basic scaffolding of command, movement, shooting, morale, but in very different ways. The perfect set that fulfils everyone's requirements for the perfect game is a long way off yet, but it's fun trying to find it.

    Realism, of course, depends on perspective. We do sometimes base our model on a single dodgy source and a lot of best guesses. Good primary sources can be a bit of a luxury sometimes.

    And yes, I am that soldier. I used Arthur Taylor's rules. What can I say - I was twelve.

    1. Many (though not all) gamers like to try things out, and that is a good thing. I do wonder if we get a little too focussed on some of the issues, and, as I think I said before, morale is an ill-defined concept anyway, but usually agreed to be the most important.

      I used Taylors rules too, very briefly, before rewriting the ECW set to include dice. I think they are still on my 'dead rules' shelf. And i think I was less than twelve at that point. either endearing precocious or an annoying little imp, I suppose.

    2. You can't ignore the effect of our prejudices though and I always find it funny that people can read the same sources and come to completely different conclusions. I can only think it's a case of seeing what you want to see sometimes.
      I'm thinking of the number of times I've had the rules query "The French Imperial Guard aren't invincible under your rules. Is it a mistake?"

    3. I understand from some figure manufacturers that French Imperial Guard and French line sell in roughly equal quantities...
      Even 200 years later we still believe the propaganda.

      Anyway, to take as an example, Cromwell's 'God made them as stubble to our swords...'. Is this evidence that it was an easy victory, an over the top metaphor, Puritan religious fervour, or ample evidence that Cromwell needed help from some kind people? Some of the answers to this question impact on writing rules, of course.

  3. Just a couple of random thoughts but I'm not sure exactly how it would help. If I read more primary sources, fair enough, I'm in a better position to critique existing uses of those primary sources. But the primary sources may not be a much better guide to 'what happened at Arbela' because surely the actual primary source may be in possession of 'models' which are even less good than my own.
    For example, some ancient/medieval sources seem to stress a religious element, either from entrail reading or from religious portents. But I wouldn't include them my model of those battles, as I assume them to have no actual significance. On the other hand, the secondary source writer has all kinds of tools of comparison and analysis unavailable to the primary source writer.

    1. Yes indeed - I am nervous about suggesting that I could do with more information about entail reading, but I'm not sure how this worked. Were the religious procedures intended to influence events, or did they just avoid fighting on days when the entrails didn't look good? Or both?

    2. Possibly such religious elements did have a role, because the people who saw them were encouraged or discouraged by the particular entrails or whatever. The question then arises how do we incorporate them into a model or rule set? I don't have an answer, but it is perhaps our own secular age which makes us dismiss such things from our models so readily.

      As to reading primary sources, my view at present it that we do need to do so, but not just the battle reports. Everyone has a view of how the world works, and this can only be discerned by wider reading of the author and their times.

      So we have to develop a model of how, say, Herodotus writes before we can start to say anything about his reports, reliability and so on. and that, of course, is only our model. Sooner or later, that way madness lies, of course.

  4. This comment has been removed by the author.

  5. Is there any relevant military knowledge that can be used to examine the models we might derive from reading primary sources? In other fields (science or medicine or maths, even economics perhaps) then one might use one's current knowledge as a basis for developing a model for judging the effectiveness of the approaches of ancient philosophers/doctors/mathemiticians/politicians etc. Is there a similar basis in modern understanding of conflict to help form models of ancient conflicts? Such work as I have seen focuses on logistics - in the form of 'how large an army could be supported in "Dark Age" Britain' questions - but can anything be said more generally about fighting itself?


    1. I'm not sure. some things, like human physiology would transfer, but a lot of other things, such as world view, would have changed massively.

      For example, Dark Ager-s were probably more used to death and gore than modern Western liberals. On the other hand they would be much less used to loud noises. Trying to decide which of these is important, and how, is really difficult, I imagine.

      Plus we would also project assumptions from our own time back onto theirs, which can mislead as well.

      it is a necessary but difficult thing to do, I think.