Reposting here from chat following @LongForWisdom’s recommendation:
I have a worry that as governance processes become more rigid, they also accrue risk in the sense that they will need to be “broken” in more and more situations, where before there would have been no rigid rule, or the rule would have been vague enough to just look “bent”. Possible examples: sudden Dai peg break; hostile slate supported by a surprise whale.
I don’t like the idea of a big legitimacy loss one day where a vote bypasses all processes, even for good reasons. I don’t either like the prospect of us having to go for suboptimal decisions more&more often just so it doesn’t look like we’re breaking our own rules. Possible examples: arguably the small vote delay we have for debt ceiling increase is suboptimal (I don’t think it’s a big deal here and I’m interested in the general case).
This ties in with a more general line of thinking: the optimality/process tradeoff tradeoff (in favour of process) is perfectly reasonable in the physical world because all people have there is process. There is no ultimate backstop to the political conversation, just a series of gradually more senior&abstract guidelines. Compare with our situation where the actual decisionmaking rules are enforced by “god”. Since we have that backstop, do we need to codify the softer decisionmaking processes as lawyerly as it happens in the physical world? Clearly the optimality costs of strict processes are worth it in legacy politics but I could be convinced that they are not in our case when compared to a) efficiency gains, b) the fact that the advisory nature of our governance processes may become invisible, until the day when it becomes super visible.
Addendum: after sleeping on it, my current opinion is that there’s no choice but to do the hard work of 1) following a very structured schema in normal conditions, 2) having reasonable escape hatches, and 3) messaging around those escape hatches to make clear that they don’t create serious abuse of power risks since actual power was always in the votes anyway.