The “Games are Conversation” point has been made many times, but here’s one more note in that symphony. I’ll be approaching it from an angle that presumes a discourse between a game designer and player.
The first exchange in this model occurs at character creation. Trad systems, at least the vast majority of them, create a very specific kind of mindset in players. The thought when making characters is almost always “how much can I get away with?”. You want a character who is maximally good at any cost. But right out of the door you have a mindset problem, where you have an antagonism between the players and the system. As players create characters, they feel they must “beat” the system by making a crazily effective character. Anyone who has played Shadowrun knows this feeling intimately. This can cause amusing problems if your system is complex enough to break, which is to say, if your system contains numbers. Though some games can beat this, it often comes at a steep cost. Take D&D 4E, where all class powers are some variation of “roll a die for damage, placement effect”. Fun if you put work into it, but incredibly bland as a base. Then look at the previous editions, and their impossibly powerful wizards.
This antagonism in character generation, and its underlying system message, in turn encourage the mindset of killing orcs to take their pie as distinct from killing orcs because they burned your family alive or stole your dog or whatever. Mechanical goals completely replace personal ones. Exalted, a game about the incredible power of god-kings, often boils down to looting and minmaxing because the most important parts of the system all encourage that style of play implicitly. They do this by making it advantageous, and by making all the hard work of character advancement take place in that sphere. This that Target Fixation that John was talking about. If there is no mechanical benefit to making your character grow, and no attention that needs to be paid to their personality, then it feels ridiculous to have a monologue in which your ranger reveals a deep-seated guilt for his genocide of the orc peoples.
So in the conversation between player and designer, System is one avenue of interaction in the conversation of gaming. It’s like when you’re on a date. If your date talks about their dogs for ages and ages, the content of that conversation only encourages a few avenues of interaction. Maybe you can talk about your dogs, or about how cats are better. But without being vulgar or truly creative, continuing the conversation is going to involve talking about animals. But unlike real conversations, a complete game is a static terrain. It’s going to be dogs forever, unless there’s some built-in space for other conversations in there. So in the conversation with a designer, as expressed in the medium of the game, if they only talk about the truly numerous ways in which they killed some dudes and took their pie then the only avenues of interaction you’re going to get are violent pie-taking ones. This is why you simply cannot play a game of 4E as a socio-political thriller. The tone of the conversation is dictated completely by the overwhelming presence of killing mechanics, and the near-complete lack of social ones.
Now, what would interest me is this: tacking a totally new, indie-inspired mechanic onto a trad game, and seeing where that took you. How it changes the landscape and tone of the game. Maybe putting a Guilt mechanic onto D&D. Or a Humanity mechanic into Shadowrun. Fun ideas, anyone?