Here's a Straussian reading of Tyler Cowen's comment that, perhaps, intellectual flexibility is a form of cowardice, and it'd exhibit better intellectualism to have rigid belief structures that can be completely demolished instead of flexible ones that endure: that he, as a voracious and well-read infovore, he is attempting to see the virtue in people who (essentially) don't need to read the news to know what they think about it. Seeking empathy isn't a natural task in general, and I think he's maybe only 40% right about what's to admire about that way of thinking.
First off, although he hasn't said it directly, it's worth mentioning that making specific predictions and having them tested is the only way we can be sure we understand anything. Sure, we can know a whole lot, but if we can't imagine the way something could be, we don't understand its context. But of course, sometimes we're wrong, and that should be humiliating, because we want to train our inner pigeons to hate being wrong. So we should make specific statements and back them up until we do some really specific humiliation ritual – like saying "I was wrong" – to remove their burden on us. Next time we won't have as strong a burden. We will predict and imagine better, to push away the terrible day when our beliefs turn into falsehoods.
But mostly, having a specific model of the world just saves time. Not everyone is as well read as Tyler Cowen. I spend a lot of time reading, and I'm not even close. Maybe Charlie Munger has enough explicit feedback about his world model, and already spends enough time reading, that it makes sense for him to use a flexible and adaptive model. For the rest of us, there are rules of thumb and apologizing for our errors.
This shouldn't just be written off as a detail! There's far too much information in the world today for a human mind. We all use shortcuts, and even Tyler Cowen is explicit about using his. But permanently committing to a lower level of curiosity because you want to focus on solid rules of thumb... that strikes me as tough, I'm just too curious. But I know it's the right choice.
The real clincher is that I want to use my knowledge for things. Just understanding isn't enough. Even just making accurate predictions isn't enough. Predictions and feedback are the whetstone. I am here to practice my art. It is a strange and boring art, where I find myself making spreadsheets, for instance. But it is just as profound, and demands more of me than slack can deliver. Every step must be optimized, otherwise I won't have any time to pet a dog or smile at a girl, ya know, the real important stuff. So to master my art even my mind must be made lean. These are not times for indulgence, and you can overindulge in learning.