Quantizable and meaningfully quantizable are both beside the points of usefully quantizable, and useful to whom.
Case in point: one of my wife's middle school students in humanities (basically English + history) was getting quite competitive and was obsessing over her grades in specific, narrow areas, to the point that her overall performance in class was deteriorating -- her scores on individual tests and assignments were good, but her actual comprehension was lacking. After talking with the parents, my wife floated the notion of not providing the child with a grade, i.e. not quantizing her performance, in an effort to get the child to stop obsessing over the number. The student calmed down, stopped obsessing, and her understanding of the material increased. And, in not being so competitive about the number she was assigned, she became friendlier and socialized more.
Part of the dynamic in this case is something that gets lost by any test-centric approach. Specifically, there's more to school than just the subject matter, particularly at the younger grades. How does one quantize a student's sociability? Friendliness? Cooperativeness? Etc. Many of these different aspects certainly can be quantized, but without any objective measure for doing so, these numbers are meaningless outside of the subjective context of whomever is assigning them. Sure, 1 + 1 = 2. But how does one objectively work out the math for "my pet hamster died and I feel sad and don't know how to talk about it, and don't want to"? Or, "I don't get along well with this teacher because our communication styles are too different, and she reminds me of that horrible Aunt Edith who spits when she talks and always gives me scratchy wool for Christmas, and I'm allergic to wool"?
Humans are deeply contextual. Math isn't. Trying to apply math to human contexts doesn't always work very well, and often has unintended consequences. One of the biggest issues is when a number score ostensibly represents a particular metric, but a deeper inspection of the scoring algorithm reveals that the metric doesn't actually measure what it's supposedly measuring. Quantization represents a gross kind of summarization, and in extreme cases, the baby does get thrown out with the bathwater (that is, all of the detail that's been summarized away). Sometimes the numbers do effectively lie.