Greg, thanks for the analysis of differentiation in this article. The analysis raises savvy points, especially that "differentiation" is a very elastic, adaptable, and flexible concept and that the evidence of benefits in outcomes for differentiating is a bit like like the new clothing worn by the unnamed emperor in the Hans Christian Anderson fairly tale.
The discussion provides an opportunity to ask what is, to me, a fundamental question: What is *un*differentiated instruction? I suspect some would answer by suggesting that Big DI is an example, but most reasonable folks know that Big DI includes low-stakes ability grouping (low, middle, and high groups for math that might differ in membership for language arts) and different numbers of practice opportunities according to whether the learners are "firm."
It's important that you mentioned special education, too, because the topic is particularly relevant there. That's because "differentiation" has application at two (or more) levels in special education. First, at a broad policy level, educators differentiate education simply by having special education: These students get general education but these students get special education. Second, at the level of individual students, we differentiate instruction by providing an Individualized Education Program or an Education, Health, and Care Plan or an Individual Learning Plan. I know these terms are not strictly aligned, but they all represent efforts to differentiate instruction.
There are important (to me, at least) concerts about associating special education with differentiation. Those considerations at too numerous and detailed to present in a comment on your post, but let me please point to those general and individual aspects of differentiation when applied to special education and note that I plan to present them in greater detail elsewhere.
And please let you thank you once again for broaching this topic!