I find the debate really interesting because I have 2 sons who were early readers and they taught themselves by more or less guess / check. I watched them bootstrap by recognising whole words and memorising them. They start noticing "EXIT" signs and words in environment, asking what words on the juice box say, see common words in books etc. Once they've seen enough of them (it seems to require a LOT) they seem to naturally figure out how to decode. I have seen the neuroimaging studies that claim ALL word recognition is phonetic "decoding" but I just don't believe they mean what phonics proponents claim. Some kids will recognise whole words before they've learned the alphabet. How are they "decoding" the sounds?? I mean, the whole existence of logographic writing systems contradicts these ideas. Chinese readers could not exist if what they say is true.

My daughter's brain does not work in this way and she absolutely benefited from explicit and careful phonics instruction in school. Whole word recognition does not come naturally to her.

All of them were read to for on average an hour a day up to age 5, so thousands of hours, nothing happens "magically".

I think I can understand why phonics would be the better method of instruction for schools. It's slower, but has less variance. The "old way" of teaching was more or less telling all kids to try to copy what kids who learn to read easily do. I've seen first-hand that this doesn't work well for everyone.

One thing I do wonder about is changes in culture and home environment, alongside demographic shifts. If children are not having as many books read to them or there are fewer kids with the exposure required to bootstrap "whole word" reading, then the old "YOLO methods" (do $whatever reading activities that mainly serve to focus attention on the topic) would appear to steadily decline in effectiveness.

My observation has been that phonics as a method of instruction doesn't do much for kids who learn to read by themselves or at home. BUT - those kids don't need any help to hit educational targets. So in a "phonics only" setting the method "gets the credit" for their success despite it being tons of classroom hours that don't teach them anything except that instructional content "doesn't apply to them".

It's an interesting tradeoff because you could extend this to all areas. You could target ALL educational content towards kids who are going to naturally struggle, using "remedial" methods. I think there are strong equity arguments for this. You could argue that the purpose of school is to help everybody meet minimum standards.

An unfortunate side-effect is that schools run on this philosophy become basically daycare from the perspective of children who don't need this. They are left to do ad-hoc activities (because there is no formal content targeted to their needs) or goof off, while teachers focus all attention on kids who are struggling. I don't think this is "wrong" but I do think it's a tradeoff.