Creating Instructional Casualties

A significant number of children in England become instructional casualties during Key Stage 1. These are not children with intrinsic learning difficulties. They are children failed by rigid, narrow models of instruction and a system that neither reflects how reading is learned nor prepares teachers to respond to the real needs of their pupils.
 

Synthetic phonics programmes are mandated nationally and typically delivered in a fixed sequence, often through scripted routines. These approaches standardise instruction but do so at the cost of professional flexibility and responsiveness. They operate on the assumption that all children can and should learn to read through the same method, at the same pace, using the same representations. This assumption is not supported by the evidence base on early reading development.
 

Approximately one in four children begin school without the level of phonemic awareness required to benefit from phonics instruction as it is typically delivered. The DfE has advised SSP programme developers to remove Phase 1, which offered an opportunity, even if not well designed for phonemic awareness, for teachers to identify those children and take action. Without an initial introduction to phonemic awareness that is free from the cognitive load of graphemes (which we have designed for KS1 teachers who understand why it matters), phonics instruction proceeds uniformly from day one. This makes learning phonics incredibly difficult for a large number of children. By the end of Reception, at least one in four are already struggling. The implication is that the child is at fault. They are not. The issue lies with the instruction.

The Delphi consensus definition of dyslexia identifies poor phonemic awareness, weak phonological working memory, and verbal short-term memory difficulties as key characteristics. These are the same features commonly observed in children who begin to fall behind in synthetic phonics programmes. However, the system’s response is typically to repeat or intensify the same instruction rather than adjust the approach or explore the child’s needs.

Compounding the problem is the gap in teacher knowledge. Many Key Stage 1 teachers do not have the linguistic understanding required to support children beyond the content of the programme. They are unable to identify phonemes or map phoneme–grapheme correspondences when these are not explicitly presented in their scheme. This is not a reflection of teacher capability but of systemic under-preparation. Teachers are expected to deliver phonics but are not trained in the underlying linguistic knowledge needed to support grapheme–phoneme mapping beyond the programme content.
 

This issue is reflected in the type of information shared with parents. For example, this resource from the National Literacy Trust categorises phonics into types and focuses on instructional approaches, not on how children learn to read:
https://literacytrust.org.uk/information/what-is-literacy/what-phonics/

This framing is part of the problem. Teachers are encouraged to pick an instructional approach, which typically means selecting a programme, and then follow it. But if they understood how children actually learn to read, they would not choose any of the models described in that guidance. Learning to read is not linear or uniform. It is dynamic. What children need to learn changes as they develop and as they begin to self-teach.

Visual and linguistic phonics models are not even mentioned. For example, the Speech Sound Pics Approach begins not with letters, but with speech. Children start by tuning into speech sounds through Speech Sound Play. They learn to build words using Phonemies, which simulates decoding and encoding without the cognitive load of letters. After approximately five days, they begin to connect those sounds with graphemes. These are the “pictures of the sounds”. They then practise blending and segmenting with a set of familiar grapheme–phoneme correspondences, such as those found in the initial satpin set.
 

High-frequency words are also introduced, orthographically mapped and clearly explained. The graphemes are shown alongside their sound values, which helps children understand rather than memorise. From this point, children quickly begin to map words in multiple directions. They will decode using sound–symbol mapping, and at other times begin with whole words - using meaning and context and map backwards to the phonological and orthographic structure. This is what enables self-teaching.
 

The approach is visual, linguistic, and flexible. It is also inclusive of speech variation and addresses accent differences from the outset. This is important because DfE validated phonics programmes are built on the assumption of a single accent, typically Received Pronunciation, and align rigidly with the IPA.  This does not reflect real life. Many children experience confusion early on, which is then misattributed to cognitive difficulty or poor effort, rather than to a mismatch between instruction and language background. This is why the 10 Day Plan is so important, and also that children who are already instructional casualties are offered the plan (there is an alternative for older children) as the issues identified within the definition had not been addressed in synthetic phonics instruction. This is the Dyslexia Paradox, and it is avoidable. It should be obvious to those who understand why so many children are struggling to learn phonics that more synthetic phonics will not change anything for children in KS2.
 

There is also confusion around what constitutes embedded phonics. The Literacy Trust defines it as an approach to the teaching of reading in which phonics forms one part of a whole language programme. Instruction is delivered in the context of literature rather than as separate lessons, and phonics skills are taught opportunistically rather than systematically. However, in many schools and classrooms, the term embedded phonics is often used to refer to programmes like Letterland, where pictograms or visual cues are attached to letters. These can offer limited initial support by helping children associate sounds with images, and encourage engagement. However, as Ehri (2004) points out, such visual mnemonics may result in short-term gains only. They do not support the type of orthographic mapping required for long-term word recognition. For example, if a child thinks of an ant every time they see the letter <a>, they face an unnecessary processing burden when encountering words such as was, any, water, or father. These visual anchors can interfere with flexibility in decoding, particularly in a language with an opaque orthography like English. If those teaching phonics understood what children need in order to reach the self-teaching phase, or aware of the whole alphabetic code they would not rely on pictograms or embedded mnemonics. Teacher knowledge matters. Even by displaying the whole code on walls, teachers and children are able to see it - it is too vast to be taught explicitly.