The Elephant in the Room: Teaching Phonics
 

When people talk about teaching phonics, the conversation usually centres on programmes, fidelity, and pupil outcomes. Schools discuss which systematic synthetic phonics programme they use, how closely it is followed, and whether children pass the phonics screening check. Researchers often examine the effectiveness of programmes or compare instructional approaches.

Yet one critical issue is rarely discussed.

Phonics instruction relies on teachers helping children map spoken language onto print. That involves more than “knowing the sounds”. It involves analysing words at the point where speech meets spelling, and making decisions about how the sound structure of a word is represented in writing.

In real teaching, the work is not simply identifying phonemes in isolation. It is word analysis:

• deciding how the word should be segmented for teaching and spelling

• deciding which graphemes represent each sound in that word

• deciding what to say when there are multiple plausible grapheme choices

• deciding how to justify a mapping decision so it is learnable, not just asserted

This is the part of phonics that is often treated as obvious, but it is not.
Teachers are already skilled readers. This means they map words without conscious effort. They don't think about it. This is the 'orthographic mapping phase' they are tasked to guide their pupils towards and yet may not be fully aware of what happens at this stage. 
 

The hidden linguistic system behind phonics
 

In linguistics, speech sounds can be represented using the International Phonetic Alphabet (IPA). The IPA provides a shared reference system that supports consistent discussion of speech sounds across contexts and accents. Linguists, speech and language therapists, and phoneticians use it to describe speech and to support analysis.

However, most teachers in England are not trained to analyse speech and print relationships using the IPA or any equivalent phonetic framework.

Instead, teachers are expected to teach a limited set of phoneme–grapheme correspondences through structured phonics programmes. These programmes introduce correspondences in a planned sequence, including around one hundred core grapheme–phoneme correspondences. Within this structure, the analytical work of mapping words can appear settled in advance, as if every word’s structure will be transparently handled by the programme sequence. But developing orthographic knowledge requires far more than that. 
 

When the programme doesn’t tell you
 

Phonics programmes specify which correspondences to teach and in what order. They provide word lists and examples aligned with correspondences already introduced.

However, children encounter thousands of words beyond those lists, including hundreds of untaught correspondences, and teachers regularly have to support reading and spelling with words that sit outside programme content. Children cannot figure out those new words using only the core set of correspondences taught within phonics lessons.

When a child asks how to read or spell a word that has not been explicitly covered, the teacher has to make a chain of decisions, for that specific word, in that moment. For example:

• how the spoken form should be segmented for teaching and spelling

• which sounds are relevant for the mapping task

• which graphemes represent those sounds in this word

• which option to choose when more than one grapheme could fit

• how to explain the decision in a way the child can apply again

These are not only phoneme identification decisions. They are phoneme–grapheme mapping decisions, and they include judgement about spellings, alternatives, and explanatory clarity. Teachers need to know how to ensure that the new learning is secured in the orthographic lexicon, to not only be retrieved later but used for future statistical learning

Without an explicit shared reference system, teachers may segment and map the same word in different ways. Accent variation can influence how a word is heard and segmented. Some sound boundaries are not obvious. Some graphemes represent different sounds across words, and some sounds have multiple possible spellings. All of this affects what a teacher underlines, groups, labels, and teaches as the mapping for that word.

Yet phonics instruction often proceeds as if the mapping is self-evident and logical. It is not, unless you are already a skilled reader. Even then, most aren't made aware of what their brains are doing when reading. How can they read words with a fir degree of accuracy if they were never taugt those correspondences?
 

Why this matters for spelling
 

Spelling makes the issue particularly visible.

To spell a word, a learner must attend to the sounds in the spoken word and represent them in writing, selecting graphemes that correspond to those sounds (Treiman, 1993; Moats, 2005; Berninger et al., 2010). If word analysis varies, mapping becomes unstable. Two adults can guide a child towards different segmentations and different grapheme choices for the same word, especially when the spelling pattern has not been explicitly taught.

So the challenge is not simply “choosing the correct spelling”. It is being able to analyse and explain the word’s phoneme–grapheme structure in a consistent, learnable way that supports self-teaching (Share, 1995, 2014).

This step is rarely made explicit in phonics training. It is often implied, assumed, or outsourced to programme materials. Many teachers do not understand it. 
 

What research has examined so far
 

Earlier research explored trainee teachers’ graphophonemic awareness through segmentation tasks that required participants to identify letters representing phonemes in words (Scarborough et al., 1998; Stainthorp, 1999). These studies suggested limitations in teachers’ ability to analyse the sound–letter structure of words.

However, these studies took place before systematic synthetic phonics became statutory in England.

Since the policy shift towards mandated phonics instruction, research in England has focused largely on programme implementation and pupil outcomes. Much less attention has been given to how teachers themselves analyse words and articulate phoneme–grapheme mapping decisions through enacted word-level tasks, particularly when they must work beyond programme-specified correspondences.

In other words, we know a great deal about phonics programmes.

We know far less about the decisions teachers make when the programme does not specify the mapping, and the correspondences have not been explicitly taught.
 

The question rarely asked
 

If phonics depends on phoneme–grapheme mapping, what do teachers actually do when they must analyse a word beyond programme examples?

How do they decide how to segment it for teaching and spelling?

How do they choose which graphemes to assign, and how do they justify that choice?

How confident are they in those decisions, and how consistent are they across teachers?

These questions point to what might be considered the elephant in the room.

Phonics policy and practice assume that teachers can carry out word-level mapping analysis reliably. Yet the knowledge, reasoning, and decision-making involved in doing this work remain largely implicit and comparatively underexamined. Emma Hartnell-Baker, who has developed the first system to show which letters are graphemes and their sound value (and which can be adapted by the user according to accent), seeks to change that, in part by launching Phonics Reform England. The goal is to raise awareness of the huge gap between the code that is taught and how to ensure that every child can read, if most learning to read is acquired through statistical implicit learning, not explicit instruction.
 

References
 

Berninger, V. W., Abbott, R. D., Nagy, W., & Carlisle, J. (2010). Growth in phonological, orthographic, and morphological awareness in grades 1 to 6. Journal of Psycholinguistic Research, 39(2), 141–163. https://doi.org/10.1007/s10936-009-9130-6

Moats, L. C. (2005). How spelling supports reading: And why it is more regular and predictable than you may think. American Educator, 29(4), 12–22, 42–43.

Scarborough, H. S., Ehri, L. C., Olson, R. K., & Fowler, A. E. (1998). The fate of phonemic awareness beyond the elementary school years. Scientific Studies of Reading, 2(2), 115–142. https://doi.org/10.1207/s1532799xssr0202_2
Share, D. L. (1995). Phonological recoding and self-teaching: Sine qua non of reading acquisition. Cognition, 55(2), 151–218. https://doi.org/10.1016/0010-0277(94)00645-2

Share, D. L. (2014). Self-teaching hypothesis. In P. Brooks & V. Kempe (Eds.), Encyclopedia of language development (pp. 579–581). Sage.

Stainthorp, R. (1999). Phonological awareness: Development and intervention. In I. Lundberg, F. E. Tønnessen, & I. Austad (Eds.), Dyslexia: Advances in theory and practice (pp. 103–119). Dordrecht: Kluwer Academic Publishers.

Treiman, R. (1993). Beginning to spell: A study of first-grade children. Oxford University Press.