RC, in response to my comment that if I mention a barkeep, my players will assume that s/he is an evil cultist (or a similarly interesting potential antagonist), you say:
Yes, they would.
Whether or not you find it problematical, this is exactly what I described upthread. If you only mention the things you believe to be important, the players automatically assume (1) that anything mentioned is important, and (2) that anything not mentioned is not important.
I don't disagree with this. In fact, I've expressly said that I only like to mention what's important (or, in my preferred terminology, what's interesting). And I'm happy for my players to assume that if I haven't mentioned it, it's not important. (Of course, they may not always make that assumption. If they would like a given game element to exist - such as a friendly priest who was once a cultist of Orcus but then converted to the good side - they may have their PCs set off to look for it.)
The point I am making is that such an assumption being made by the players doesn't preclude meaningful choices.
The meaningful choices, once the players have had their attention drawn to the barkeep, are:
(i) What steps, if any, are to be taken by their PCs take ingame so that the ingame state of knowledge can be brought into alignment with the players' metagame hypothesis. By the standards of many posters here my table is pretty liberal on metagaming, but we do have a mild aesthetic preference that there be at least an ingame figleaf. This choice is meaningful, because the range of options - from surveillance to interrogation to mind probing to mind control to killing the barkeep and then coercing his/her dead spirit - is non-trivial, and which choice is made makes a potentially significant difference to the unfolding ingame situation, as well as the unfolding metagame situation (is this a game about nice guys, or about vicious torturers, for example? or even a game about how nice guys sometimes get corrupted into vicious torturers?)
(ii) What, if anything, is to be done to the barkeep. Is s/he to be killed? Converted? Or are the PCs to join the cult and use the barkeep as their ticket to do so? These are obviously non-trivial choices. The choice to join the evil cult, in particular, is a hugely game-changing one (at least from my experience of having GMed games in which those sorts of choices have been made).
Neither of these is a "choice" on which nothing of significance for the game turns. It therefore satisfies your criterion (1):
(1) In order for a choice to be meaningful, the result of a given option cannot be the same as that of all other possible options.
You also say:
(2) Assuming any goal, whether GM-presented or player-driven, if the result of an option moves the PCs forward toward that goal, it is (in general) a better option than one that does not. Likewise, if resources are to be taken into account, an option that reduces expenditure whil moving toward a goal is (in general) superior to one that includes heavy resource attrition.
As far as resource efficiency is concerned, in the game I run that tends to be relevant mostly (although not exclusively, as skill challenges sometimes consume powers and surges) in combat, both within combats and across combats that happen prior to a given extended rest. In 4e these sorts of "optimum because efficient" choices are shaped by the interaction between the mechanics and the geography. The best way to make meaningful choices possible is therefore to make the geography fairly apparent - via a battlemap - and let the players go to town with the mechanics. (There are some exceptions, like hidden traps or invisible opponents, which reward some character build and skill deployment choices to do with perception - I don't draw hidden things on the battlemap until they become unhidden.)
As to goals - I think it's a bit more complicated than your description of it. Suppose a player's goal is for his PC to flirt with the darkside but in the end be redeemed. This is the sort of goal at least some of my players adopt for their PCs from time to time. Relative to this goal, is it optimal to kill the barkeep, or to invite the barkeep to explain in more detail the theology of Orcus? I don't know. I expect my player doesn't know either. Either choice could lead to some degree of corruption of a character. Neither seems likely to lead to iredeemable evil. My player is, I think, more likely to go one way or the other based on what strikes him at the time as fitting with his feel of his PC, the scene, and the current mood of the game.
In any event, to the extent that there is any optimality here, it's completely orthogonal to any issue of detail in description or not describing only interesting details. It's not as if my player's choice would be
more meaningful if I'd made it harder for him to work out that the barkeep is a cultist. That would only have made it harder for the player to actually reach the point at which the meaningful choice is to be made.
(3) Since we are discussing the results of meaningful choices, there must be both context (sufficient information to at least make the choice something other than the result of random chance) and sufficient consequence (i.e., all choices cannot be equally optimal based upon the goals of the players, whether self-generated or GM-generated) to make the choice meaningful.
A choice where no option is more optimal than any other is, essentially, meaningless.
I don't fully agree with this.
First, as to context - your examples of Wolves in Sheep's Clothing, and other cases of narrative detail, suggest that you are assuming the context for choices is provided overwhelmingly by information about the ingame state. To the extent that you are making such an assumption, I don't accept it. Information about the ingame state (eg that the barkeep is a cultist, that in the crypt the PCs find a statue of Orcus) provides some of the context. But a lot of context is provided by external considerations, such as a player's knowledge that if his PC just walks into the bar and kills the barkeep, the PC'll be judged as a pretty merciless individual by onlookers, and perhaps also by the player's fellow players and/or their PCs.
Second, as to consequences - it's simply not true that differences in optimality are a necessary condition of meangingfulness. This probably isn't a place to rehash the writings of Hume, Nietzsche, Joseph Raz, John Gray etc, but there is a strong body of argument that suggests that some values are incommensurable in principle - and hence that some choices are radical, in that they are meaningful but no consideration of reason tells in favour of one or the other. Even if one rejects this as a philosophical account of value, in practice there are many choices where optimality can't be measured. For example, a decade ago my job as a public servant came to an end and I had to decide whether to look for another public service job or instead to try for an academic career. I took the latter choice. I believe that it was a meaningful choice. Did I make the choice because I believed it was superior to staying in the public service? Obviously I thought it was an attractive choice. But I had also enjoyed being a public servant. Being an academic would let me do things that I couldn't do as a public servant, but obvioulsy the converse is also true. In the end, I had to choose - one can't deliberate forever - and I believe that I made a sensible choice. But I'm pretty confident that had I stayed in the public service I would also be able to reflect back on my choice and my (alternative, hypothetical) range of achievements and regard them as sensible and more than satisfactory. (A complicating factor is that a choice doesn't only deliver outcomes to be evaluated relevant to a preference set, but also shapes the preference set by which the value of the outcome is measured - this means that anticipatory valuation is not necessarily the best guide to post hoc satisfaction.)
As I said above in relation to the pursuit of player goals, when my players choose what to do with the Orcus statue or the Orcus cultist, they aren't engaged - certainly not explicitly and in my view not implicitly either - in a cost-benefit analysis. (They save these analyses for combat planning.) They are responding to a choice situation in a way that they think is sensible and interesting. The meaning is provided not by the opportunity to rationally maximise, but by the opportunity to determine the character - aesthetic, moral, thematic, . . . (insert other appropriate evaluative categories here) - of the unfolding game, and their PCs as integral elements of that game.
Counter-examples always, perforce, contain some element of optimization, even if what is optimal is the level of fun the participants enjoy.
For all sorts of reasons, including some of those given above, but also others to do with the dependence of certain sorts of emotional responses upon not just the outcome that they are a response to, but the means whereby that outcome is achieved, I don't find it helpful to analyse playing an RPG by reference to rational maximisation of a desired emotion.
To put it another way - as a general rule, I don't think that the best way to pleasure (or at least the sort of pleasure that an RPG provides) is to set out to maximise that pleasure, any more than the best way to get an aesthetically satisfying (and therefore, in some sense, entertaining) movie is to set out to make an entertaining movie.
Typically, then, when a player chooses to responsd in a certain way to the statue or the barkeep, I don't think they are following the imperative "Maximise fun!" I think they are responding to much more complex aesthetic and other emotional cues, which hopefully will have as an upshot the generation of fun.
multiple courses of action allow for choices that optimize what game you and your players want to play (i.e., they allow the players to drive the game in a direction they find more enjoyable), and the excitement arises from the tension between what is known and unknown while driving the game in that direction.
The excitement in my game comes primarily from action resolution - especially combat, which in D&D 4e is very decision and consequence heavy with a healthy dose of randomisation thrown in. The overall game I would say is interesting but not always exciting. To the extent that it is exciting, I think that excitement arises from doubts about how it is going to resolve. But those doubts don't arise from unceratainties in expectation about GM-described ingame elements (eg Is that really a rabbit on the stump, or in fact a disguised carnivorous plant?). They arise from uncertainty as to how the action of the game will resolve (eg What is going to happen to me, now that I've decided to become a supplicant of Orcus?).
Thus, you say
<snip pemerton's presentation of mechanical-driven approach to Wolves in Sheep's Clothing>
which is, of course, a perfectly valid way to play, but one which (to some degree at least) minimizes player choices because the PCs never have a "full view" of their environment.
<snip>
In fact, by limiting what you describe of the world, you perforce constrain player options.
This is consistent with my earlier suggestion that you see "context" as shaped overwhelmingly by ingame elements. It is also consistent with my remark upthread that your views on description seem to be influenced by a more Gygaxian/exploration-based approach to play.
If you take away these two assumptions - that is, if you assume that metagame evaluations and emotional responses are an important part of context, and if you assume that the game is not primarily about exploration (in the literal sense) of the gameworld - then the effect of non-description of rabbits on stumps on player choices becomes close to zero. Because failing to describe rabbits on stumps has a neglible impact on the possible range of metagame evaluations and responses that shape the metagame context for choice, and has a negligible impact on the players' view of the ingame environment that is relevant to generating such evaluations and responses - because the ingame environment that generates those evaluations and responses is an ingame environment of PCs killing or sparing cultist barkeeps, cleansing or fleeing in fear from statues of Orcus, and so on. Rabbits on stumps are irrelevant to this.
And as I said earlier, if my player would like an ingame element to exist that speaks to their concerns and the way they want to develop their PCs, and I haven't mentioned it yet, then they will have their PCs look for it. Thus, one player wanted to know whether or not, among the elves the players were camping with, there were any members of the secret society he had written into his renegade drow's backstory. He therefore had his PC flash a secret handsign and see if any elf responded. As this was the first I'd heard of the secret society, and as I hadn't thought much about the elves other than what the module told me about their need for an idol to be recovered from a crypt, I made a quick decision - that one of the elves was a member, as the player hoped - but it was not the leader, as the player hoped. Instead it was the crafter to whom the player's PC had given a dragon tooth to be shaped into a magic dagger. This was the first bit of personality that that crafter had manifested as an NPC.
I was subsequently able to develop this new gameworld element, by having the crafter taken prisoner in the course of bringing the dagger to the PC - whom the crafter now had an extra special reason to help. (This also made my life easier as GM - I wanted to get the item to the PC, but not until it was a level-appropriate treasure, and having the crafter be kidnapped while trying urgently to deliver the dagger to his fellow secret society member gave a perfect rationale for the dagger to end up in the PC's enemies' hands). As the rescue of the prisoners is still ongoing in the game, and continues to provide the backdrop against which a lot of player decisions have been made, this has been a development that has produced a lot of payoff for a small outlay by the player and me.
I would regard this as a paradigmatic example of (i) only describing what is relevant, (ii) following the players and their interests in designing encounters, and thereby (iii) facilitating meaningful choices. Choosing only to focus on what is interesting has not cut off options for the player in question - it has made meaningful options available to him.
And, yes, I do understand that there are benefits to this approach. For one, you do not need to worry about the players going off on a tangent, simply because no tangents are presented.
I don't understand the notion of "tangent" here. It's up to the players to decide the direction in which they take their PCs. There is no predetermined direction, in relation to which some other chosen direction threatens to be tangential.
My point - and it's really just rehearsing the point that Mal Malenkirk made way upthread - is that descriptions of rabbits on stumps aren't relevant to any direction my players are interested in going, because they don't engage the relevant emotional and evaluative responses.
if you decided that you were the sort of GM who wanted to lead the players by the nose. It is easy enough to do. All you have to do is decide what the players should choose aforehand, and make ever other potential choice result in maximum suckage.
I don't see how this touches upon the question of whether or not describing only relevant things is leading the players by the nose.
I believe I've said enough about my game and more genrally about how I see 4e as playing, in this and other threads, to make clear that mine is not a game in which the GM leads the players by the nose. I believe I've also made it eminently clear that I only describe those gameworld elements that are relevant/interesting/important. I've done my best over a few posts now to explain how these two approaches to play are consistent.
If I wanted to lead my players by the nose I wouldn't change my practice in describing ingame elements. Instead I would dictate their PC backstories to them. I would instruct them (either directly or implicitly) who the PCs' ingame enemies are to be. I would impose my own moral judgement on my players via a combination of mechanical penalties to and ingame lynching of their PCs for doing things I disapprove of (traditionally in D&D the alignment mechanic has been the vehicle for achieving this). I would make sure that whatever decisions the players try to make about who their PCs fight and who they talk to, the ones I want them to fight always fight, and the ones I want them to talk to are always 20th level gold dragons or Elminsters who are (relative to the PCs) invulnerable. I would make sure that the king always hates them however nice the PCs are to him, and that the same patrons present the same plothooks in the same taverns with the same likelihood of turning out to be a turncoat regardless of the moral, political and mythical universe the players are trying to create in the gameworld via the actions of their PCs.
In my view, what I have described above is the main way in which D&D play, over the years (and especially but by no means exclusively in the 2nd ed AD&D period), has encouraged a style of play in which GMs lead their players by the nose. The range of approaches to ingame description has, in my view, played comparatively little role.
