It's a classic "can't see the forest for the trees."
Maintaining a process leadership means squat if chip design and production efficiency/capacity falls by the wayside. Its pretty obvious the process R&D team is having too much sway inside Intel and they are undermining the overall interests in the company.
I think it's more along the lines of the requirements of maintaining a coordinated cadence across such gigantic and very different but simultaneously interdependent business sectors being pretty much impossible to fulfill, and Intel making some overly conservative decisions based on the continued belief that they'd get out of it "soon".
You might be right that the process R&D team has pushed for keeping an impossible goal rather than lowering requirements for the 10nm process, but considering how good 14nm has grown over its revisions, they don't really have a choice. Even best-case roadmaps show 10nm to be mostly inferior to 14nm except for density. Stepping back goals would likely mean scrapping the node entirely.
In a broader perspective, Intel (the CPU/chip maker, not so much the fab owner) has coped with this crisis as well as could be reasonably expected, but they've also pretty much hit the worst case scenario: fab capacity running out long before an upgrade is available, and chip sizes increasing at the same time. The only thing they could have done to make it better would be unplanned large-scale fab upgrades to 14nm, but that would both be a very expensive short term investment (as the entire purpose of it would be to alleviate pressure until 10nm arrives) and cause investors to run for the hills (as it would be a very clear admission that 10nm was nowhere close). Their current plan has two major issues: it pisses off customers (both large and small), and it is
still reliant on 10nm arriving soon. The longer 10nm stays out of reach, the more precarious Intel's situation becomes both in terms of product competitiveness, production capacity, and customer goodwill. If this keeps up for another year, I wouldn't be surprised if AMD's market share came close to overtaking them. For now, they can get out of this "unscathed" (only having lost their 3+ year process advantage and a lot of customer/partner goodwill), but in a year? They'd survive (they're Intel, after all), but it would be a bloodbath.
If they'd set fab upgrades in motion in late 2016 or early 2017, they'd be in volume production already - but Intel's stocks would also have taken a significant dive, and the arrival of Ryzen would likely have been much worse for them. Intel gambled (big) on this being a short-term predicament, but they're running dangerously close to losing that bet outright. If they start fab upgrades now they won't be in volume production for at least a year, which won't help them much.
).
