Sounds first, and say them in the ordinary run of the day rather than as a lesson. When a letter turns up on a cereal box, say /m/, not "em" and not "muh". The letter names arrive later, mostly on their own. What matters more than the order is how you say the sound, and most adults, without noticing, say it wrong.
Say the sound, and drop the vowel
Try saying the letter b out loud right now. If what came out was "buh", you added a vowel that is not in the word. The real sound is smaller than that: lips together, a puff of air, and then you stop. It is barely a sound, which is why adults pad it.
The extra "uh" has a name, the schwa, and phonics trainers spend real time getting adults to stop producing it. Parent sites almost never mention it.
There are two kinds of consonant, and each has its own trick.
- The short ones you clip. b, d, g, k, p, t. Say them as short as you can and stop before any vowel arrives. Keep the voice in b, d and g; a whispered b comes out as p.
- The long ones you stretch. m, s, f, l, n, r, v, z. Say "mmmm" and "ssss" and hold them. A child can hear these and copy them.
- Vowels get their short sound first. The a in cat, the e in bed, the i in pin. The letter name ("ay") is the long sound, and it can wait.
It takes about a day of catching yourself.
Why a child who knows all the sounds still cannot read cat
Picture a four-year-old at the kitchen table with three magnetic letters in front of her. She points at each one and says "cuh", "a", "tuh". Then she tries to push them together and gets "cutter", or "cutta", or gives up and says "cat" because she saw the picture on the fridge.
Nothing is wrong with her. She was taught five sounds and told they were three. Cuh-a-tuh has two extra vowels in it, and there is no way to blend them out. The first time she hears the sounds said clean, /k/ /a/ /t/, the word appears, and it looks like magic to her. To you it should look like three sounds with nothing in between.
This is the most useful thing I know about early reading, and I would put it above every toy on our shelf. Say the sound short, and the blending later takes care of itself. Say it padded, and you will spend a year wondering why the flashcards did not work.
What the research supports, and where it stops
Children who arrive at kindergarten knowing their letters tend to become earlier readers, which is how alphabet drilling turned into a preschool industry. The question is whether teaching letters harder produces the reading, or whether both come from the same home.
The best synthesis I know of is a 2010 meta-analysis by Shayne Piasta and Richard Wagner in Reading Research Quarterly. They pooled the alphabet instruction studies of the time and asked what the teaching moved. Alphabet instruction reliably improved alphabet knowledge, and they found minimal evidence that it transferred to early phonological, reading or spelling skills.
Be careful what that shows. The literature has grown since 2010. It tested alphabet instruction as programs and lessons, rather than a parent saying /s/ at the cereal box. Nobody has run that trial, and we will not pretend the meta-analysis covers it.
What it does undercut is the flashcard set. In the studies pooled, drilling letters as a standalone program produced children who knew their letters, with little measurable gain on the phonological, reading and spelling outcomes tracked. That is an argument against running a curriculum at three, and none of it touches saying the sound correctly when a b comes up. The wider Montessori evidence is in Are Montessori toys actually better?
What it looks like on an ordinary Tuesday
The whole method fits inside ordinary talk. Under a minute, several times a day, and only when the child is already looking.
- Start with their name. The first letter of a child's own name is the one they care about. Say the sound when it appears on a cup, a coat peg, a birthday card.
- Play with first sounds before letters. "What starts with /s/ ... sock, soup, sandwich." No letter is in the room. At three this is the whole game.
- Rhyme badly and often. Songs, nonsense words, a made-up verse about the dog. Hearing where words split is the skill underneath all of it.
- Answer the letter that gets pointed at. If a child asks what a sign says, say the sound and the word and move on.
- Skip the test. "What sound does this make?" ends the game. Children keep playing games that nobody is grading.
At three most of this is sound play and rhyme, and the shelf of toys for 3 year olds leans that way on purpose. Around four the link between a sound and a shape starts to stick, which is where the movable letters on the 4 to 5 year olds shelf come in.
It will be found that the muscular memory is in the young child the most tenacious and, at the same time, the most ready. Indeed, he sometimes recognises the letters by touching them, when he cannot do so by looking at them.
Maria MontessoriThe Montessori Method, trans. Anne E. George (New York: Frederick A. Stokes Company, 1912), Chapter XVII, "Description of the Method and Didactic Material Used," p. 277
Lowercase and cursive, and what Montessori gets from tracing
Montessori classrooms do this in a particular way, and nearly all of it is tradition rather than evidence.
The classic material is a set of sandpaper letters: lowercase, often cursive, one letter per board. The child traces the rough shape with two fingers while saying the sound rather than the name. The tracing and the saying happen together, which is the point.
The lowercase-first argument is practical. Open any picture book and a child meets lowercase on nearly every line; capitals show up at the start of a sentence and in names. A toy that teaches only uppercase teaches the rarer shape first. The cursive-first argument is a practitioner's one: the letters flow into each other, and b and d are harder to confuse when each begins with a different stroke.
Those are reasons, and I find them convincing, but I could not find a trial that sets lowercase-first against uppercase-first, or cursive against print, in preschoolers. Montessori guides do it because Montessori did it and because it seems to work in their rooms. A tray of sand does the finger half for nothing, if you want to try the tracing before buying a single board.
If you are going to buy an alphabet toy anyway
Nothing above requires buying anything. Most families end up with an alphabet toy anyway, so here is the filter I use for our shelf. It is a filter, not a boost. The right toy mostly stays out of the way.
Most of the alphabet toys I come across are uppercase, print, and picture-cued. "A is for apple." A child looks at the apple and says "apple", which is reading the picture, and the a never gets its sound.
- Lowercase available. Uppercase only is a pass.
- No picture glued to the letter. A plain a is harder to sell and easier to learn from.
- No voice. A battery that announces "ay!" every time a tile is touched drills the letter name, cheerfully, all day.
- Letters you can pick up and move. Blending is a physical act at four. Three wooden letters pushed together on a table do more than a printed row.
- Nothing that keeps score. Sorting, matching and building words are play, and points turn all three into a test.
Our own language and literacy shelf is small and I would rather it stayed small than filled up with singing alphabets. If a toy there fails this filter, tell me and I will take it down.
The one thing to change this week
Clip the consonants. Say /t/ instead of "tuh" from now on and the next year of letters goes better whether or not you buy a single thing.
Say the sound when a letter turns up, play the first-sound game in the car, rhyme at bath time, and do not test. Let the letter names come from the alphabet song, which they will. A three-year-old who is not interested is a three-year-old, and a four-year-old who suddenly is will move faster than any program can. The work on your side is smaller than it looks, and most of it is unlearning a vowel.
Where this came from
- Piasta, S. B., & Wagner, R. K. (2010). Developing early literacy skills: A meta-analysis of alphabet learning and instruction. Reading Research Quarterly, 45(1), 8-38. A 2010 synthesis of classroom and program studies; it did not test a parent saying sounds at home, and the literature has grown since.
- Clipping consonants short without a trailing vowel (/b/, not "buh"), and keeping b, d and g voiced, is standard phonics teaching practice and the author's judgement, not a trial finding. We found no study that compares it against the padded version in parents at home, so treat it as sensible practice rather than evidence.
- Lowercase-first and cursive-first are Montessori classroom convention, from the sandpaper-letter presentation, backed by practitioner reasoning. We could not find a trial that sets lowercase against uppercase, or cursive against print, in preschoolers.
