On Skirmishers

I have been reading a bit about ancient wars and battles, specifically, for those of you with long memories, those in the Greek and Classical worlds. One of the things that has struck me is the large number of light troops which are attached to the armies.

Numbers, in the ancient world are, of course, always rather dubious. However, the normally reliable Thucydides reports 10,000 Boeotian light troops present at Delium, although they had little influence on the battle (IV.93). Sabin, in Lost Battles, represents them with one ‘token’ levy unit.

Wargame rules (and writers of history, both ancient and modern) tend to be rather dismissive of skirmishers. I myself have fallen into this: in Polemos: SPQR a base of skirmishers is set at being about 75 people running around with javelins and so forth. Now if I transferred that to Delium, I would need over 130 bases of skirmishers alone, which would fill up most people’s wargame table without any of the battle-winning troop types getting a look in.

It seems to be the case therefore that while my representational scale might be awry, the effects could be accurate. Skirmishers had very little impact on the ancient battle. I know that there are occasional exceptions, but in the ancient world they tended to be peltasts, who appear to have been more professional and capable than your average javelin chucker.

It also has to be admitted that many sets of wargame rules do not represent skirmishers terribly well. Notoriously, in one of the version of DBM, a single base of skirmishers, set at an angle to a block of advancing heavy infantry, could delay the latter hugely. Treating skirmishers as just another troop type with compatible capabilities does not seem to represent their actual activity terribly well.

The other problem, of course, is that ancient writers were not, in fact, terribly interested in skirmishers. They were not of the right class, for one thing. Battles were won and lost by the heavy infantry (I generalise, but then, so did they). Skirmishers were the necessary but uninteresting part of the army. Exceptions do, of course, occur, but in general, light troops are not discussed or reported upon. Again, peltasts may be, but the general skirmisher is not.

So, how can we handle skirmishers within our wargame rules?

In Polemos: SPQR, as I have said, a base represents about 75 men, which, I confess, is probably incorrect. However, in my defence, I have come up with a novel (to me, anyway; I don’t imagine that it is original) method of modelling the performance of skirmishers.

If you read the few accounts that there are of skirmish type tactics in the ancient world, I do not think that you find the more modern ideas of ‘clouds of skirmishers’ or skirmishers advancing in open order. The best description of skirmishing is not, in fact, from skirmishers at all (please correct me if I’m wrong), but of the Persian cavalry at Plataea. Here, they advance in groups, throw javelins at the Greek hoplites and then retire to the main body.

There is no description of the cavalry being in open order, just advancing in smaller groups from the main body, discharging missile weapons, and then retiring.

The more I thought about this, the more likely it seemed to me to be generally the case for ancient world skirmishing. In the ancient world, there was much less emphasis, in the first place, on the individual, and so individualistic activity, like skirmishing 10 feet or so from your neighbour, is less likely. Additionally, I cannot imagine that any such troops would be trained and so, as I believe it to be the case, that untrained troops tend to huddle together, it seems to me highly likely that skirmishers would too.

There are other factors at play, as well. At Caharre, as is well known, the Parthians skirmished pretty well all day. They must have had some means of relieving their front forces, if only to resupply with arrows. This is much simpler if the whole unit was not involved at a given time. If a bunch of 50 men were dispatched from a unit of 500, for, say, ten minutes skirmishing, then each batch of fifty would have about an hour and a half to rest in between such activity, while the enemy simply saw a screen of men discharging arrows at them, constantly. The rest could be seeing to their horses, restocking with arrows, having a drink and a bite to eat, and so on.

What is very wearing for one side would be a walk in the park for the other.

Of course, the commanders would need to be careful to keep their units out of harm’s way, and we do read often of the skirmishers simply being chased off by heavier troops. However, they usually did reform and return, which suggests, in my model, that the charged skirmishers simply run back to the mother unit and reform upon it.

This then, is the model I have attempted to work with in Polemos: SPQR, and intend to implement in the Greek rules when they finally get off the drawing board. Skirmisher units represent the base unit, and their range represents the distance to which the packets of men are sent out to throw their javelins (or whatever). The whole unit is not fighting at any one time, nor are the men standing at the base and shooting. Only a small proportion of them are doing that at any one time.

Of course, it is still required that the skirmisher’s action is still, mostly, disruptive, not fatal to the enemy. That can be fixed through the relevant combat factors, of course.  And the units under attack can still advance to drive off the skirmishers, although they would have to get at the mother unit to do any real damage. A failure by the skirmishers to inflict any damage  can, of course, be interpreted as a local success by the attacked base.

Finally, of course, I can now rationalise my representative scale. A base of skirmishers can represent many more than 75 individuals, but it is that 75 who are in action at any one time. The overall base would represent, say, 750 individuals, which makes it much more viable to have the whole lot on the table at any one time.

So, there you are: a model of skirmishing in the ancient world. I wonder how much it leaks…