In the small corner of the blogosphere that I am able to observe, the D&D morale system has been the source of recent rumination based upon events in the bloggers' games.
First, I read noism's posts on morale as a world-building tool (here and here). The proposal was that the BECMI morale rules work acceptably well but could be improved (read: overhauled) by creating a morale table for each monster detailing specific behaviors/encounters for each of the morale results.
One of his points, made more explicit with the slow worm example, is that there are a number of creatures for whom the standard morale system makes absolutely no sense - any prey-status animal (rabbits, deer, etc.), barring some extraordinary circumstance, would not attack or be hostile to the party. They would flee. And this extraordinary circumstance would not happen once every 36 times such an animal was encountered. This is a good observation - the morale system, as written in BECMI, does an acceptable job of producing 'reasonable' behavior in most of the enemies present in a standard D&D game.
This morale system generates enemy behavior based upon a mostly arbitrary die roll (one might apply modifiers, but the 2d6 roll is highly variable), and noisms' modifications use it as an encounter generator, shaded by the attitudes of the original enemy behavior.
I think the idea of a morale table for each creature one might encounter is an excellent idea, in theory. I am opposed to the idea not because of the work such a project would entail but in that I already spend too much time consulting tables during play. I have barely begun to flesh out my world, and already there are at least 20 or 30 creatures they can encounter in the areas for which I have data, based upon the time of day at which the encounter is determined. To then give each creature their own table means that I am adding a another sizeable step to my encounter procedure.
It seems unnecessary, since what is perhaps more helpful is understanding what each creature's motivations are, and what might drive them to encounter the party.
Most creatures will see a group of well-armed, battle-seasoned individuals and steer clear. With the exception of super-apex predators, the kind that only exist in D&D games (trolls, manticores, etc.), most regular predators will not go after a group of humanoids. A lone humanoid, or maybe a pair? Absolutely. But 4+ large creatures traveling together? There are deer and rabbits aplenty to eat. And if they do engage, unless the first round of combat is fairly successful (weakest prey incapacitated), they will break and run, and the worst they will do is run in and harry the party over time - it's how most predators (those that don't ambush or swoop in for a single kill and swiftly depart) get their food. And so an encounter with such a creature or creatures will likely be the characters becoming aware of (or failing to perceive) their presence nearby and then choosing to pursue or ignore it.
Especially since this understanding of a creature's motivations is a necessary part of using that creature in your world - and if you haven't, I'd suggest you consider Maslow's hierarchy of needs. So if I choose goblins for my encounter, I roll the group's size and then let that number, combined with the location for the encounter and the time of day, determine the goals of that group which immediately tells me how this group might interact with the characters.
What is useful to me is a tool that will help me generate realistic combat behavior for my creatures - to determine if they rally when things are going poorly or flee even when they hold the advantage - and I'll continue the brainstorm next post.