Phonology is not about spelling. If you don’t believe me, look for the word “spelling” in the index of a phonology textbook.
If you try to derive phonology from spelling, you run into all sorts of nonsense—like the idea that the [ks] of the “x” of “box” stands for one phoneme, while the [ks] of the “cks” of “socks” stands for two. (Or the idea that [ks] is an affricate in any language).
Linguists are not, qua linguists, any more qualified than anyone else is to tell people how to teach reading.
Neither are psychology professors, unless they know something about automaticity—and about how and why it’s important in reading.
The fastest route to automaticity involves reducing cognitive load, presenting material systematically, and lots of patterned drills, a la “cat-mat-sat”; “cat-cab-can.” This is true even for material in which there are—dare I say it?—exceptions to the patterns. “What”????
Some people think that the following is an example of explicit, systematic instruction for beginning readers
Let’s pick a word in this story. What do you think it sounds like? Can you think of some related words? Let’s formulate some hypotheses and investigate. Now let’s take what we’ve learned and write some word sums and create a word matrix.
They also think that the cognitive load here is totally worth its weight in learning.
And they think that the material is so inherently interesting that five- and six-year-olds will be totally riveted. (Much much more riveted, surely, than they would be to all that “cat-mat-sat”; “cat-cab-can” nonsense).
Two more thoughts: Teaching systematic phonics has a small effect size on reading ability only if you insist on including comprehension, which you shouldn’t.
But if comprehension is to be included, all those other variables (vocabulary, background knowledge, inferencing skills…) need to be eliminated, and kids should be measured only on their comprehension of stuff they’d have no trouble understanding were it spoken.