When Words Don’t Fit the Model
In a recent discussion on social media thread, Dr. Pete Bowers made several claims that deserve to be read twice. They are not small claims. They challenge assumptions that are often taken for granted in early literacy instruction.
The claims were these:
- The “default” position, that words not explained by phonology alone should be memorized one-by-one as irregular, is itself an untested hypothesis.
- There is little to no direct evidence that memorizing irregular spellings in isolation is superior to explaining them through morphology and etymology.
- Avoiding meaning in early literacy may force dyslexic students to rely more heavily on their weakest areas: orthographic memory and phonological processing.
- The greatest cognitive load in classrooms is often shame from failure, not conceptual understanding.
These are not minor observations. They invite us to examine not just instructional practices, but the model underlying them.
The Untested Hypothesis
Consider the first claim.
When a word does not conform neatly to phoneme-grapheme expectations, we often assume that memorization is the natural next step. The word is labeled irregular. It is set aside. It is practiced. It is stored.
But that move rests on an assumption: that phonology alone is the primary organizing principle of the writing system. If that assumption is incomplete, then memorization is not a neutral act. It becomes compensatory. It asks students to absorb the limitations of the explanatory framework by increasing memory demands.
The question, then, is not whether students can memorize irregular words. Of course they can. The question is whether memorization should be the default response when a word does not fit the model.
What Happens Clinically
In clinical practice, this question is not theoretical. It is observable.
When memorization becomes the default response to words that do not align neatly with phonological expectations, the burden shifts to the learner’s memory systems.
For some students, this is manageable.
For others, it is destabilizing.
-Students with phonological processing weaknesses are asked to store spellings that do not feel predictable.
-Students with fragile orthographic memory are asked to retain visual forms without structural anchors.
-Students with broader language-processing difficulties are asked to hold disconnected forms without access to meaning as a stabilizing force.
When meaning is sidelined in early instruction, students are pushed toward their weakest cognitive pathways. This is not a moral failing. It is a predictable consequence of the instructional architecture.
Over time, the greatest cognitive load in these classrooms is not conceptual complexity. It is accumulated frustration. It is the quiet erosion of confidence when words must be remembered rather than understood.
Integration Is Architectural
Meaningful integration cannot take the form of simply adding prefix and suffix drills onto a phonological foundation. Morphology taught as isolated affix practice is still layering.
If the writing system encodes meaning as well as sound, then instruction must reflect that structure from the beginning. Words must be studied in relation to other words. Bases must be examined. Spelling-meaning correspondences must be made visible alongside phonological ones.
When the structure of instruction mirrors the structure of the writing system, patterns that once appeared irregular become explainable. The system stabilizes, not through increased memorization, but through coherence. The goal is not to increase complexity. It is to increase alignment.
A Structural Choice
When words don’t fit the model, we have a choice. We can increase memory demands and call the words irregular. Or we can examine whether the model itself requires refinement.
In my clinical experience, when instruction reflects the structure of the writing system, stability follows, across profiles, across ages, across levels of challenge. The goal is not more content. It is coherence.
