I always enjoy your writing, John. On first reading of this piece, it appears you’ve set this up to appear as if hierarchy is the cause of dysfunction in goal-oriented, human cooperation. I don’t know if you meant to do that, but it’s misleading.
To be really basic, humans cooperate when there is repeat exposure, mutual upside for cooperation, and significant cost for defecting. These conditions are necessary for cooperation, but they aren’t sufficient for effective and timely achievement of a shared goal. For that, you need clarity on who is supposed to do what, the authority of that person to make decisions that enable them to do the things they are responsible for, and the capability (talents/skills/interests) to do the thing responsible for.
In the less structured (positive) example you provide, those three things can be worked out ad hoc and as the project emerges — so long as it’s not a terribly complex project. In the dysfunctional example, you have poorly designed hierarchy. You describe a situation in which accountabilities are parsed smaller than the capability of the people and a severe mismatch between authority those same people to make decisions that impact the quality of their work. Odds are good there is a leadership capability problem — because I’ve seen those symptoms emerge before.
I talk about the importance of syncing up accountability, authority, and capability in my blog post The Trust Algorithm. At first read, what I say may be counter-intuitive to some because it doesn’t match many pet theories about trust based on anecdotes and correlations-not-causes. I urge those folks to think through what pre-conditions enable the correlates of trust to exist. The Trust Algorithm gets you 90% of the way there.