A taxonomy of cavalry

[Patrick Waterson]

Maybe we should not think about think about this as categories of troop types, but categories of troop interactions?

Won't this be even more complicated?   The number of troop types that could interact and the number of situations they could react in?

It might actually be simpler in concept, e.g. when a mass of infantry meets a mass of infantry, you get (I think) one of three basic types of interaction:
1) Both stand in place and hew away (or shoot, if not keen on contact)
2) One pushes the other back, leading to eventual collapse if not remedied
3) One charges rapidly to be met by a steadier opponent with one of the two outcomes above.

The next step is to decide on one's level of detail and fine differentiation.

[Erpingham]

Assuming we have two types of infantry only "mass" and "dispersed", for your mass category :

*Both units advance to spear fencing distance/missile exchange , which may lead to one of the other outcomes after time, or one just slinks off. Or, following certain models of combat, small units cross the gap and engage until enemy is worn down , or line relief happens.
*One advances in a tight formation and attempts to break through the other.  It might break through, grind to a halt and do 1 or 2 in the Patrick's list or become enveloped and swamped.
*One charges the other and the charged party loses its nerve and runs off, pursued by triumphant charger.


I'm sure there are more options, even if we ignore tactical circumstances like direction of attack, disorder, terrain.

Interactions are more complex than taxonomy, IMO.

[Patrick Waterson]

It may be that some of these suggested possibilities exist in the minds of theorists but not on the historical battlefield, allowing a bit of re-simplification.

Both units advance to spear fencing distance/missile exchange , which may lead to one of the other outcomes after time, or one just slinks off.

Mmm ... any historical examples?

Or, following certain models of combat, small units cross the gap and engage until enemy is worn down , or line relief happens.

Again, can we point to any of this actually happening?  (We can with line relief, but this is basically perpetuating a fight-in-place interaction.)

One advances in a tight formation and attempts to break through the other.  It might break through, grind to a halt and do 1 or 2 in the Patrick's list or become enveloped and swamped.

If it fails to match its opponent's frontage, it introduces another dimension into the equation, which is probably why most classical armies made an attempt to match frontages unless they were very confident about having sufficient momentum to do something decisive before the enemy's extra frontage did something serious.

One charges the other and the charged party loses its nerve and runs off, pursued by triumphant charger.

This did historically occur quite a few times, although in essence it gives us the task of modelling infantry pursuit of infantry as opposed to another form of combat interaction.

I'm sure there are more options, even if we ignore tactical circumstances like direction of attack, disorder, terrain.

There may be, but I think they are along the lines of one side attempting a stratagem or unusual deployment which ultimately leads to one of the basic types of infantry interaction.  A flank attack, for example, is essentially one side pushing the other back, but from a different direction.

Interactions are more complex than taxonomy, IMO.

I think interactions may be simpler, because for infantry there seem to be essentially two types: 1) both sides stand and 2) one side shifts the other.  The complexity lies in judging which troop types will bring about which interaction, and when, and what the consequences of such interaction will be for both sides.  For example, a host of unarmoured Britons close with a thin red line of armoured Romans and both sides stand for some time.  The result is a lot of dead Britons and a loss of confidence among the remainder, so the action changes to Romans shifting Britons.  Boudicca cannot stop the rot and after a further time interval loses the battle.

Another example: Caesar and Pompey face off at Pharsalus.  The main infantry forces stand and inflict casualties and fatigue on each other, while on Caesar's right we get a cavalry interaction and an interesting infantry-cavalry interaction which sees off Pompey's cavalry and allows Caesar's spare infantry to do quite a bit of the second type of infantry interaction, namely shifting whatever gets in their way.

Actually this suggests the third basic type of infantry interaction which earlier eluded me: one side massacres the other, advancing while covering the ground with bodies.  Not too common, but probably essential to distinguish from Type 2 for certain actions.

[Erpingham]

Both units advance to spear fencing distance/missile exchange , which may lead to one of the other outcomes after time, or one just slinks off.

Mmm ... any historical examples?

Its a common suggestion for shieldwall battles.   Also, followers of Hanson may use this to explain their belief in a formal phase of doratismos in hoplite battle.

Or, following certain models of combat, small units cross the gap and engage until enemy is worn down , or line relief happens.

Again, can we point to any of this actually happening?  (We can with line relief, but this is basically perpetuating a fight-in-place interaction.)

As I'm sure you recognise, this a model of Roman combat much favoured by some modern scholars who provide considerable evidence.  However, I also recognise that previous debates here have shown others utterly reject this evidence

Overall, I think the idea of only two or perhaps three outcomes and one or maybe two classes of infantry suggests very high level abstraction.  To return to our original topic of cavalry taxonomy, we could postulate the same high level outcomes and perhaps two classes of cavalry; contact and not contact.  All else is mere detail   But don't we want some detail - some idea that the toy soldiers are representing historical warfare in some way?

[aligern]

So right Anthony. When the level that a set of rules is operating at becomes such that they could be used for almost any period where there is near contact between the forces then perhaps it has gone too far.  The ideal is to build in the flavour of a period. The more this can be done by making the rules such that the general naturally takes decisions that fit with the style of the period or by legislating within the rules so that armies are restricted or enabled in the direction of 'realism' .
So I would look for a set of Ancients rules to be based upon an analysis if how troops actually perform and then encourage or enforce that fighting style.  That isn't easy because we don't find easy agreement on even the well recorded Romans. I suggest that either rules should operate at the level in which there are three lines of Romans that can replace/ reinforce each other, or, on a higher level, that the characteristic that Roman legionary formations have is represented. I would suggest that this is a matter of outlasting opponents and gradually degrading them?
For cavalry, If Roman and other cavalry operate a Cantabrian or other such tactic then they ought to be able to stand off from an infantry opponent to their front and cause it casualties and yet charge effectively when sufficient disorder had been caused. We would have to agree on why Persian cavalry formed so deep and allow Macedonians and Thracians to manoeuvre more speedily. I suggest we have to get to that level of depiction, or we could use the same rules to fight a Napoleonic battle with no perceivable difference.

[Patrick Waterson]

Which leads to the question of why we dispensed with WRG 6th Edition.  Took too long.  Too much record-keeping.  Players wanted something quick, not something realistic.

Are we about to reverse that trend?

[aligern]

I am not at all sure that building in more of those  elements that are  period or army specific need add much to record keeping, though it is likely to add to time and runs a bit counter to the desire to hold all the game information in your head or on the tabletop.
Let us imagine a simple interaction, Vikings versus Saxons.
The two bodies close.
There is an opportunity for one or other to quit before contact.
There is a mutual exchange of missiles. One or neither side wins this.
Fighting occurs along the line. There is an outcome, stasis, pushback or rout.
One or both sides may launch an attack on the enemy leader , successful or not.
Fighting recommences along the line.
This is repeated until.
Someone wins.
There is a free hack and flight.
Morale tests might intervene at each stage or posthumously.

All this can be rationalsed down to each side having a fighting factor and then a die throw added and an outcone decided. The outcome can be rout/ pursue or mutual withdrawal. If one wanted one stage up from an instant resolution then the die roll can be repeated.

In a large battle the likelihood would be that the sevond method holds, in a smaller, or one with plenty of time to play it out then the first method would be practical.
I remember once playing a game with Tony Bath in which he was using Minifigs 2mm blocks. There were lots of little blocks, painted mainly red or blue. Combat resolution was at the level of fighting factor and die throw for each block to block conflict. It worked quite well because there were so many blocks and so many dice thrown  that it avoided the single catastrophic throw that cracks a unit and then wrecks a flank.  However it could have been any period, though Tony, famously, did not jib at Pharaonic Egyptians fighting Burgundian knights.
Given the launch of so many Ancient rule sets recently I think it fair to ask their promoters just how they deliver period flavour and still manage to resolve the action in an acceptable timescale.
As you can see I am concerned if the period element in a set of rules is simply having different weapon and armour factors. In the example I cited above there was nothing to say that the Saxons were different from the Vikings, perhaps because there wasn't, but say it was Late Maceconians versus Republican Romans , could we capture the essence of the difference between the two?
Roy

[Erpingham]

Which leads to the question of why we dispensed with WRG 6th Edition.  Took too long.  Too much record-keeping.  Players wanted something quick, not something realistic.

Are we about to reverse that trend?

There is an assumption in your post that granularity and realism are the same.  WRG gave detail but were its outcomes more realistic than other sets?  Even at the time I remember questioning whether a game with a careful ground/time scale that ensured battles were brought to a point of conclusion after three or four scale minutes could be considered realistic.  In order to get this to work, it also had a very artificial casualty rate (although I will confess I only realised this later - I thought at the time that the system was pretty neat).  And some things to do with scales were fudged mightily, like how quickly an order change could happen or how long it took to dismount.  So WRG 6th were good in their day but lets not pretend they were some acme of realism.

I actually think concentrating on the outcomes of interaction is an interesting approach to combat, as opposed to counting swords.  I think, though, to get it to work you need a taxonomy of troop types and an understanding of how tactical circumstances (formation, order, ground etc.) play their part .  Otherwise there really are too many possible outcomes.

[Patrick Waterson]

There is an assumption in your post that granularity and realism are the same.  WRG gave detail but were its outcomes more realistic than other sets?

I think yes: not necessarily more realistic than some of the detail-intensive sets around at the time, but more so than many we have seen since.  If by 'granularity' detail is meant, I would agree only insofar as the detail is relevant.

Even at the time I remember questioning whether a game with a careful ground/time scale that ensured battles were brought to a point of conclusion after three or four scale minutes could be considered realistic.

This was a frequent, or at least persistent, criticism which led to a number of excuses and abstractions and finally different time scale in WRG 7th.  In fact the problem was not with the system so much as the players, who often zoomed into action in a way that no historical general ever did (where I can find advance rates, armies with the exception of hoplites generally did so at half speed, usually following an extended wait while the skirmishers or archers did their stuff).  A tendency to cover the entire table with shallow formations also made for rapid conclusions - which is just what players want for a competition, but is not really faithful to historical practice or outcomes.

What WRG 6th lacked in my opinion was effective command rules.  These were the major constraints on historical armies, but replicating them is much more of a challenge, particularly the way a brilliant general/great captain operates inside his opponent's 'command loop'.

In order to get this to work, it also had a very artificial casualty rate (although I will confess I only realised this later - I thought at the time that the system was pretty neat).

Paradoxically, yours truly was not too impressed at the time, but came to appreciate the virtues of the system later.  I was eventually surprised how well the casualty rate actually worked out when attempting historical actions.  An added consideration was that the whole system, or at least its modelled interactions, tended to be driven by the casualty rate, so it was already modelling more aspects than just body count.

And some things to do with scales were fudged mightily, like how quickly an order change could happen or how long it took to dismount.  So WRG 6th were good in their day but lets not pretend they were some acme of realism.

I think the order change was about right, at least where signalling was concerned; dismounting, yes, that did happen a bit quickly, but funnily enough I never played against anyone who dismounted anything during a battle.

I actually think concentrating on the outcomes of interaction is an interesting approach to combat, as opposed to counting swords.  I think, though, to get it to work you need a taxonomy of troop types and an understanding of how tactical circumstances (formation, order, ground etc.) play their part .  Otherwise there really are too many possible outcomes.

I am inclined to agree: interactions without troops would resemble a play without actors.  However if we begin with basic types of interactions and fine-tune these based on troop types, national practice, morale situation, terrain etc. then we may have a more sustainable system than if we start by classifying our troop types and then find ourselves nonplussed by the question of what happens when they meet each other.