The Benefits of a Sounds to Letters Approach
- Dr. Steve Underwood
- Nov 19
- 6 min read
Updated: Nov 20
I had a great week supporting schools in NW New Mexico this week. It's such a beautiful place!
I spent the time observing teachers and coaching them on the Science of Reading.
Something I was reflecting on during my school visits is the importance of a Sounds to Letters instructional approach to phonics.
Two Phonics Approaches
There are basically two ways people teach phonics. One is best thought of as letters to sounds. The other is sounds to letters (or as some refer to it, speech to print).
I spent a fair bit of my coaching this week talking about the differences.
While the research doesn't absolutely come down on either side, according to Dr. Tim Shanahan (2024), many leading thinkers, including Louisa Moats, are with me in seeing sounds to letters (aka, speech to print) as the more logical and more efficient way to teach. Dr. Shanahan hedges in his discussion saying that the debate may not be 100% settled yet, but he goes on to say that "the preponderance of current data" does lean toward that conclusion.
The Problem with Letters to Sounds
Letters to sounds looks something like this.
Teacher: Class, what sound does S make? (Teacher then shows a flash card with S or writes S on the board.)
Students say the sound /s/.
What's wrong with that you might ask?
Let me re-frame it for you. If a student was reading the word "shirt", what sound does the letter S make?
See the problem?
Essentially the problem is that the letter S doesn't "say" anything. It is a symbol. Its meaning depends on a few different things.
Here's an analogy. The digit 7 doesn't mean anything by itself. It is a symbol that has to have context to understand its meaning. For example, a 7 in the ones place means something very different than a 7 in the millions place. In addition to place value, you have to consider 7 what? With coins, 7 pennies is very different than 7 quarters, for instance.
In the same way, letters are symbols. What they represent is defined by the code that is behind them.
Why Sounds to Letters?
English is considered an opaque language.
Unlike a transparent language, such as Spanish, in which each speech sound only has one letter symbol that represents each sound, the English phonics system has two layers – (1) a simple sound-to-letters relationship plus (2) a sound-to-etymology-to-letters relationship – because our language was made through the merger of Old English (a Germanic Language) and Old French (because of the Norman Invasion). Old French was a derivative of Latin. English then evolved and later we added new words for science, history, and other fields of study that were created from Latin and Greek roots. This history impacts the correct spelling (orthography) of English.
Spanish, however is more directly descended from one language: Latin. In Spanish, for example, you can teach letters to sounds because the relationship is predictable due to the simplicity of the language's history. The code works and makes sense either way: sounds to letters or letters to sounds.
English, on the other hand, isn't predictable in both directions.
Because of the merger of the languages, English reading and spelling makes more sense if you go in the direction that the code was created.
How was the code created, you might ask?
Every human language started as a spoken language of sounds before it ever became a printed language. Humans are innately created to learn spoken language from the time we're born, so the spoken side of language learning comes naturally to us.
Printed languages, though, are all human invented codes. Some break spoken language down to a word or idea level. For example, Japanese uses symbols that may represent entire words or ideas. Other cultures break spoken language down into speech sounds (phonemes) that are represented by phonic symbols (graphemes). English is the latter.
This means learning to read and spell involves two complex codes – an innate one and a human created one – that have been mashed together.
With English, in particular, our spoken language has approximately 44 phonemes (speech sounds), depending on dialect. Then, every spelling pattern, or grapheme, is connected to one of those 44 phonemes based on the history of how the word came into English. Words that are more basic typically come from Germanic roots. Words that are more complicated typically come from French, Latin, and Greek origins.
Due to the history of English, we only have 44 phonemes, but we have about 120 graphemes (with some estimates significantly higher).
The English spelling system starts with sounds and then is impacted by word history, using different graphemes to represent sounds to maintain the historical record and meaning of our words.
According to researchers (Hanna et al., 1966, in Moats, 2005), the code for English orthography can be understood this way:
50% of English is completely regular with basic phonics (like Spanish)
34% is mostly regular, though these words may contain one or two complex phonics patterns
12% is mostly regular, but with some sort of etymology that explains any perceived variation
Only 4% is truly irregular for reading or spelling
Because of all of this, it's a lot easier, in my experience, for students to learn how to read and spell when teachers base all of their phonics instruction in this fact. It's so much simpler to teach students that there are only 44 sounds as a foundation than to try to teach them 120+ graphemes as the foundation.
Using a Sound Wall
To put into practice a Sounds to Letters (or Speech to Print) instructional approach, you have to start with what is commonly called a Sound Wall. All of the best core elementary literacy programs have approximately 44 Sound Spelling Cards, like the three examples I have in the attached picture below. Notice that each card represents a single sound (phoneme) and has all of the primary spellings (graphemes) that students need to learn for that sound. The Bee card, for instance, only represents Long /Ē/ and not short /E/.
With Sound Walls like these, students only have to learn 44 categories for reading and spelling, compared to the Letters to Sounds approach which forces them to learn approximately 120 graphemes separately without any overlying mental model.
Wrapping Up
With this in mind, let's return to the example of the letter S from earlier. The letter S shows up with the following three phonemes:
/s/ as in salad
/sh/ as in shut
/z/ as in cherries
If students are taught a Sounds to Letters approach, they come at their reading and spelling understanding that speech sounds are the foundation of the code and that letters are just symbols that depend on context for interpretation.
When they have this mindset, they don't get confused and think that the letter S has meaning by itself or that the letter S should "say /s/" all the time. A Sounds to Letters approach helps set a foundation in Kinder, for example, that doesn't force a 1st grade teacher to reteach children when they come to more complex phonics where the letter S shows up in the sounds /sh/ and /z/. The teachers can just go straight into explaining how we represent the sound /sh/ with the letters "sh" and we represent the sound /z/ with the letter "z" most of the time, but at the end of words, /z/ is often represented with the letter "s".
If you're an educator or school leader, what approach does your school use? If you are using a letters to sounds approach, what might be the benefit of switching to a Sounds to Letters approach for your students?
For Further Reading
Shanahan, Tim. (August 2024). Print-to-Speech or Speech-to-Print? That is the Question. Blog: Shanahan On Literacy. https://www.shanahanonliteracy.com/blog/print-to-speech-or-speech-to-print-that-is-the-question-1
Hanna, P.R., Hanna, J.S., Hodges, R.E., and Rudorf, E. H., Jr. (1966). Phoneme-grapheme correspondences as cues to spelling improvement. As found in: Moats, Louisa. (Winter 2005/06). How Spelling Supports Reading. American Educator.
