Sound-to-Print vs Print-to-Sound: Why the Starting Point Changes Everything

By Eyal Rav-Noy, Co-Founder, Capit Learning

Almost every child in the world is taught to read the same way: point to a letter, and say what sound it makes. It feels so obvious that few people stop to ask whether it's the right starting point. But the direction you start from—letters first, or sounds first—turns out to decide whether English looks like a chaotic mess of exceptions or a logical, learnable code. This is the difference between Print-to-Sound and Sound-to-Print, and it is the single most important choice in early reading instruction.

The hidden assumption: "letters make sounds"

Traditional phonics rests on a phrase teachers repeat thousands of times: "This letter makes the sound…" It sounds harmless. But look closely, and it contains a strange idea—that a letter can talk.

Letters don't talk. People do. When a person speaks, they make sounds with their mouth, and we invented letters to write those sounds down. The letter is the record of the sound, not the source of it. When we tell a child that the letter is doing the talking, we hand them a problem with no clean solution—because a single letter represents many different sounds.

Consider the letter "a." It represents the sound /a/ in apple, /ay/ in acorn, /aw/ in ball, /o/ in father, and /ə/ in above. If we start at the letter and ask "what sound does this make?", the honest answer is: it depends. That uncertainty is exactly what makes English feel confusing, inconsistent, and random. And notice where the confusion enters—it enters at the starting point.

English is a reversible code

To see the alternative, imagine inventing writing from scratch. You have a word you can say—"cat"—and you want to record it. You listen, and you break it into its sounds: /k/ /a/ /t/. Then you assign each sound a shape: /k/ becomes c, /a/ becomes a, /t/ becomes t. You write cat. You've just invented spelling. The technical word is encoding—turning speech into marks on paper.

Now run it backwards. Start with the marks c-a-t, turn each shape back into its sound /k/ /a/ /t/, blend them, and the word hands itself back to you: cat. That's decoding—reading. Reading and spelling are the same code running in two directions. This is what it means to call English a reversible code.

And a reversible code gives you a choice about where to begin:

• Print-to-Sound: start with the letters on the page and work up to the sound. This is how nearly every child is taught.

• Sound-to-Print: start with the sound and work toward how it's written. Almost no one teaches this way—and it changes everything.

Why the direction matters

The two directions do not lead to the same place. Starting from print, English looks unpredictable: one letter, many possible sounds, a fog of rules and exceptions. Starting from sound, English becomes orderly: each of the language's 40+ sounds has a defined set of ways it can be spelled. The sound is one stable concept; the spellings are simply its different written forms.

An analogy helps. Once you know what a triangle is, you can recognize an equilateral, right, isosceles, scalene, acute, or obtuse triangle—they're all instances of one concept. The sound /ee/ works the same way: it can be spelled many ways (see, Pete, eat, donkey, taxi, etc.), but those aren't separate facts to memorize—they're instances of a single sound. Sound-to-Print instruction teaches the concept first, then its written forms.