Maybe a better way to put it would be that agreement on meta-level principles more reliably forces agreement on simple object-level issues?
In cases I can think of, this is the reverse of what happens. In fact, agreement on (simple, often too simple) object-level issues allows abstractions which can be agreed on.
Physics is a great example. Intro physics lectures include a massive number of demonstrations. These are simple (too simple to use to build a car), but are clearly and incontrovertibly real-world behaviors. There is object-level proof that the abstractions have at least some tie to reality.
In cases I can think of, this is the reverse of what happens. In fact, agreement on (simple, often too simple) object-level issues allows abstractions which can be agreed on.
Physics is a great example. Intro physics lectures include a massive number of demonstrations. These are simple (too simple to use to build a car), but are clearly and incontrovertibly real-world behaviors. There is object-level proof that the abstractions have at least some tie to reality.