That doesn't really answer my question. Especially since you also play games, such as D&D, that are built around the traditional style of game play, yet you have no issue tossing that aside to run your style of game. Why expect players to be proactive with goals and desires with things other than treasure, but not treasure?
I don't think I fully get your question.
If a player has a PC goal like
I'll free my loved one from the evil duke's dungeons, then I'd expect that to drive play to some reasonable extent. And likewise if a player has a PC goal like
I will become the richest baron in all the land. But I would expect that character to drive a quite different game from typical D&D, in which the gathering of the treasure is largely an afterthought to the action, yet is treated as a major win condition. (By "the traditional style of game play" I take it that you mean looting dungeons for treasure. My sense of how treasure plays in 5e as it is presented, seems to be pretty much like Keep on the Borderlands but without the XP awards.)
I understand the "treasure as afterthought to action" approach in the context of those editions that make having treasure to spend an important element of character development, or that make it a literal win condition (in classic D&D you need to find it to earn XP, and then spend it to join the endgame; in 4e it is part of the doling out of PC abilities, either literally in the form of magic items, or in the form of the medium used to acquire magic items). I don't really get it for 5e, though. As a win condition it seems to have become detached from the actual play of the game and the develpoment of the characters. So why does it loom so large?
Maintaining a wealthy lifestyle requires around 1,500 gp a year, an Aristocratic one 3,600. Assume you want to retire from adventuring at some point means a human would need 45,000+ gold minimum. Want to build your own castle? You'll need 500,000gp just to build and another 140,000 per year to maintain. That's a CR17+ treasure hoard every year.
Of course, all of those things come with their own complications, as does spending money trying to buy magic items, as doing so could draw the attention of other powerful beings.
Spending money is all grounded in the particular story you are involved in.
I don't feel that this answers my question; it really just underpins it and sparks new, related questions.
Presumably in the imaginary world of D&D (or at least many, I'd even conjecture most, D&D campaigns) there are aristocrats who live at the "aristocratic" standard of upkeep and live in castles, but don't loot a dragon hoard every year. If a player wants his/her PC to become an aristocrat and live in a castle, why does the game suggest that killing dragons is the answer?
To be clear: I know what the reason is in classic D&D - it's a game of dungeon exploration and looting in which gp are the win condition, both in terms of achievement at looting and PC progression. But I'm wondering what's the deal with 5e - is it still a dungeon crawling game that's dropped the second (PC progression) part of the classic D&D win condition? If so, how is this tight design? If not, then why does it still connect adventuring to loot-collection in the way that classic D&D did?
And to the point of this thread and Mearl's statement, not being as granular about the rules allows DMs to better tell the story that their individual table is interested in. Could this mean that there is a disagreement at the table over how much 100,000 gp should be able to purchase? Yes, it might, but that disagreement carries a much lower consequence to it than one over whether or not the evil Orc hit you with it's sword and killed you.
That seems a pretty contentious claim, unless we accept as a premise that the stakes of play are ultimately about the PC living or dying in one-on-one combat.
Combat in D&D takes up so much design space because it is both the primary player agency mechanism and has the highest stakes. As long as the rules of life and death are seen as fair, the players can use that to impact the world with an expectation that the results will not be at DM whim.
For example, if the DM decides that no amount of money can buy a castle because the King must grant the land, the players can use the rules of combat to kill the King and build the castle anyhow. This in turn, leads back to the central conceit of the game as it was designed, for the DM and players to create an exciting story of bold adventurers who confront deadly perils.
And that to me is the whole point of the 5E design philosophy. What are the minimum amount of rules necessary to inspire and tell that exciting story? I'd argue that the extensive combat rules both inspire the imagination in exciting action packed scenes and give players a sense of agency without perfect certainty leading to careful weighing of options on how to engage the world. Extensive travel rules or shopping rules just don't, IMO, have the same rewards for added complexity.
OK, so you're prepared to embrace the contentious premise!
The bit I've bolded seems to run togehter some potentially separate things - it's the primary player agency mechanism, but that's in part because there are no other robust mechanics that generate finality of resolution without being mediated through largely open-ended GM decision-making; and even allowing that it is the primary player agency mechanism, that needn't require it to be as complex as it is.
As to the idea that travel rules or shopping rules add complexity, I don't agree with that. 4e has travel rules (skill challenge), shopping rules (skill challenge), persuading-the-king rules (skill challenge), and they're not very complex, can be used in other fictional contexts besides the ones I've mentioned, and work reasonably well. And there are plenty of other RPGs that have genuinely uniform resolution mechanics that can be used for everything from fighting to persuading to trekking to shopping. (And Classic Traveller got fairly close to this 40 years ago; closer, at least, than 5e.)
That's not to say that 5e is poorly-designed - there might be good reason to break out combat rules from a generic resolution system, give it a rather high search-and-handling time, while subordinating other possible domains of action and resoltution. But I don't think it's light, and I think to a significant extent it trades on legacy expectations and understandings.