Saturday, 22 October 2022

Information Flow

One of the things which have slowly come to my attention in wargaming, perhaps particularly in campaigns but not exclusively, is the question of information flow. That is, who knows what and when, and the movement of information between participants.

I suppose that this is particularly acute in pre-modern times before telecommunications came into existence. How often do you read a campaign report where the orders were inflexible because they had to be, and the subordinate commanders struggled or failed to adjust to a changed situation? I think it happened quite often. A decent commander might be one who trusted and was trusted by their subordinates sufficiently to adapt to circumstances and one whose network of information gathering and order dispatch was at least adequate to the task.

As you might guess this has been reinforced for me by the Jersey campaign, where at least three companies of Jersey militia are busy guarding their villages and beaches in blissful ignorance that the enemy is already ashore and their colleagues could really use their presence. The orders to march failed to get through. No one is being insubordinate or dense, they just do not know the situation and therefore cannot react. At the current rate of progress, the first they will know of the successful landing is when the Parliamentarian army marches into their parish.

The wargaming problem here is manifold, of course. The wargamer, even if not a solo player, has a somewhat omnipotent view of what is going on. If Colonel Bright on the left wing can see that the enemy has failed to deploy before him, then the commander in chief and Colonel Dime on the right are also going to know. We can introduce rules which try to simulate the latter two not knowing but it is very difficult to remove knowledge once it is in the wargamer’s head. It is hard not to react to the knowledge even if the ‘men on the ground’ would not know the information.

There are various ways around this problem. We can assume instant telepathy between our commanders (perhaps some science-fiction or fantasy games do so). That seems a little too much for a historical scenario, of course. In Twentieth-Century games, we can assume radio communications (although there were not necessarily reliable) and so more control is available to the commanders. On the other hand, there can be too much information. One of the problems at the Three Mile Island nuclear plant (which nearly went bang) was that so many alarms went off the operators could not see what was important. Similarly, when there is an emergency event there are so many mobile phone calls about it that switchboards are overwhelmed. Too much information can be as difficult to handle as too little.

Assuming that telepathy and radios are unavailable in our historical period, most news travelled by people. These include couriers and aides dashing around on horseback delivering messages from the commander or a scouting party to good old rumours, which may travel less quickly than a man on a horse, but may not. This is information flow, which we can conceptualise, should we feel so inclined, as a network of nodes, some of which light up when a new piece of information appears.

As a node is illuminated, we have to decide what happens next. The node should be programmed to pass the information on to the next node. Hence the scouting party reports back to the unit commander ‘We have seen a troop of cavalry’. That node then also has to decide what to do. Is the report useful and reliable? Does it fit in with the overall picture emerging from other scouting parties and other knowledge received from headquarters? Do we pass it on or seek clarification?

If the decision is to pass the information on, then it goes up the chain of command (hopefully). If the force is fairly small it can go straight to the commander-in-chief, possibly (in ECW armies) via the Scoutmaster (which must have been an unenviable job, it has to be said). But then the commander has to wonder: so what? What does this mean, and how should I react? So the light cavalry has seen a troop of horse. Is that the enemy scouting screen? Is that where they are advancing from? And so on.

We can of course replicate some of this in a campaign game, or even a tabletop wargame. But the thing is that we know, as the wargamer, what is going on. That troop of horse consists of the third troop of the 6th Lancers and they are the lead for the enemy’s fourth division. I know that, but my little lead commander does not. We have to try to persuade ourselves to react to the limited information disclosed rather than the fuller picture we have at our disposal.

I have found it rather hard to control this. Courier cards and their equivalents help. I can record the information sent ‘Enemy in sight’ and to whom it is sent, and when it arrives. I can even control how the recipient reacts. Often that is obvious, but using personalities and an initiative roll helps. Some people will ignore the blindingly obvious, while some will overreact to the slightest news.

In a larger campaign, this does get complicated. I recall one campaign set in Ancient Greece where couriers and ambassadors were zipping around the map in their multitudes. It started to get difficult to record when they arrived where and what information they were bearing. I am sure I missed a few crucial treaties or declarations of war along the way. I cannot be sure. As this was not really a map-based campaign I could not resort to pins for the couriers, although that would have solved part of the problem, at least.

I do not think there is an easy answer to all of this. I suppose that if I had the time and inclination I could write a bit of code that would ping up a message each time a courier arrived somewhere, but I am not sure I want to tie my wargaming to an electronic device. I shall have to stick to map pins and a campaign diary, I suppose.


  1. Great post, very thought-provoking. I wish I could write code because I know exactly what I would like to do to tackle this - at least the 'low information' environment. I would write a code simulation to get an indication overall of how it worked, then I would design a manual "rough best fit" replication for actual gameplay (I wish I could do this for lots of aspects of gaming, TBH).

    1. Thank you; I can write code but I am really not sure if I want to tackle this one. As a solo player, it depends on what I want to happen automatically, or by assumption, and what I want an overview of, I suppose. How much detail do I want to delve into?

  2. In my home-brewed sets, played solo, I have adopted the following approach:
    Units act according to the last orders given. The orders can only change when the unit passes a test (bonus for having a sub-commander with high initiative rating). Orders are only very simple, like march in that direction or stand there. If you want them to stop at a particular feature you still have to pass a test - if they fail they carry on (“oh I thought you meant THAT farm!”). I reason that people on the ground can’t see what the general sees and vice versa (shades of the Light Brigade), or don’t recognise it’s significance, or orders don’t get through. Chaotic but fun. You have to apply your preferred narrative reason after the event.
    ‘The Twilight of the XY’ series of rules have something a little similar.

    1. Yes, I think that sort of approach works. In PM: SPQR you set units moving, and then can only stop them by issuing additional orders. The time delay is ignored, admittedly, but it does lead to some interesting situations with columns marching straight past the enemy. The upshot is that being a general or other commander is probably a lot harder than we imagine.

  3. Always an interesting problem. I get around it by making up what amuses me most or what fits into how I want it all to play out. This does of course miss out on those serendipitous situations that arise because of dice rolls and cascading actions.
    When I have more time I think I will move to something like what you are doing.

    1. I think it depends on what you want from your game. So long as it fits the story and is interesting and acceptable I don't see a problem. The serendipitous situations don't arise all that often in my experience. Usually the outcome is realistically silly.

  4. The example of your current campaign has me thinking about the locals. Surely on such a small place as Jersey, the local grapevine and/or screaming locals would have passed on 'information'? Then, of course, it depends on what that is, how exaggerated and how much store the commander puts in said info.
    So, one could even have a matrix for each event. What comes from 'within' (e.g. couriers), what from 'without' (friendly/unfriendly locals) and, perhaps any mis-information spread by the enemy.
    In my own attempts at campaigns I have had a table with a modified die roll to dictate what information a player/side reveals to the other when forces are within a given distance or closer. This ranges from general to more specific in terms of the force/composition and strength ±x%. Modifiers include cavalry size of force, quality of command and no. of days in 'contact' (closer or equal to the specific distance).
    Regards, James

    1. I think that you make a very good point. I confess, I tend to ignore the civilians, because it saves atrocities. But you suggest a two-speed flow of information, I think. One of reasonably reliable 'official' information via courier, and one of rumour, refugee and sensationalism. I am sure that could be incorporated with enough thinking about, and be extended to cover friendly and hostile locals as well.
      Hm. Now my imagination is full of expanding circles of information. I think I might need a lie down...