Those who do not study history, and do not learn from it (as is ironically often the case with history majors) are doomed to repeat it, and re-invent it, poorly.
Still, engineering mindset aside, how do you explain that so many people with college degrees are poor, while the engineers don't feel that pain?
Having studied both in arts and sciences I would say that both approaches have advantages when applied in the real-world, and limitations. I don't believe (or know of any evidence) that "Arts" people are more likely to suffer the Peter principle. It's a function of seniority that you lose information versus specialists in an area - so if anything it's connected with bombastically believing you're right rather than what degree you did.
If you're very intelligent, you can do very well with a social science or humanities degree. In fact, you will do better than with a STEM degree—the highest positions in society are dominated by people with non-STEM degrees.
On the other hand, it's very easy to end up un(der)employed if you're not particularly intelligent and got a humanities degree. Meanwhile, your equally intelligent peer who got an engineering degree is doing just fine as developer or technician.