Teaching Letter Names First: A Critical Response

I generally have no desire to argue with people on the internet. It’s an unproductive and stressful experience that rarely yields worthwhile results. However, this Instagram post was sent to me by several people asking my opinion, so, it seemed like there was enough interest for me to write a full response.

Fair Use
Copyright Disclaimer under section 107 of the Copyright Act of 1976, allowance is made for “fair use” for purposes such as criticism, comment, news reporting, teaching, scholarship, education and research.
Fair use is a use permitted by copyright statute that might otherwise be infringing.
Fair Use Definition
Fair use is a doctrine in United States copyright law that allows limited use of copyrighted material without requiring permission from the rights holders, such as commentary, criticism, news reporting, research, teaching or scholarship. It provides for the legal, non-licensed citation or incorporation of copyrighted material in another author’s work under a four-factor balancing test.

Here’s the post:

Post Caption:
Period. There’s no debate to be had.
Research-backed, best practice instructional strategies are the ONLY strategies that should be used to teach.
Saying things like: “letter names aren’t that important anyway” or “kids will just pick it up” or “they only need the sounds to read” is a massive 🚩.
A red flag rooted in pride. A red flag rooted in privilege. A red flag rooted in ignorance. Because that individual doesn’t recognize the child that is struggling.
Did you know that in any given kindergarten class of 25 students, at least 1 child (if not more) has dyslexia? Often undiagnosed at that age. Saying “most of the kids will figure it out anyway” leaves out that 1. That 1 has the right to quality phonics instruction.
When teaching phonics you MUST teach letter names, letter sounds, and letter symbols upper and lowercase TOGETHER. I’m not making this up, it’s scientifically proven.
Swipe through for some of the research, and don’t forget that learning should be designed for ALL students. Not just the majority.

This carousel feels really aggressive to me and I get why so many people sent it my way. My reading program, a linguistic phonics program modeled after the original Montessori Method of reading instruction, very explicitly says to wait on teaching letter names and capital letters until about halfway through the program.

The post asserts a method of early reading instruction that is allegedly “science-backed,” but the claims it’s making are not supported by the “paper” it cites. I’ll explain in more detail below, but the practices this instagram post is promoting are, to my knowledge, not supported by research.

The caption of this post also seems to indicate that you are “problematic” if you disagree with this post. That you are “privileged” or “ignorant” if you think this post is wrong. All I have to say to that is that science is not a fixed body of knowledge, it’s a way of thinking. A way of systematically interrogating the universe to discover what is true. If anyone ever tells you that “the science is settled” or “scientific debate isn’t welcome” you’re dealing with someone who is intellectually dishonest and doesn’t understand the philosophy or mechanism of science.

First Things First

What is a writing system?

Simply, a writing system is a human invention that transmits language over space and time. Well before audio or visual communication technology, we could share our words through writing.

A writing system is a code with a systematic relationship between units of sound and visual symbols. All writing systems are a code, whether you’re reading English, Chinese, or Hebrew.

When we say we are “teaching children to read,” what that really means is that we are teaching children how this code works. We are teaching children to be codebreakers.

Spelling is encoding, reading is decoding.

All codes are reversible, meaning you can encode and decode them. For children to be successful codebreakers, they should understand the relationship between sounds and symbols both from the direction of sound-to-symbol and then symbol-to-sound.

This is the task at hand. What myself and the other author disagree about is how to approach this task.

I believe that reading instruction should isolate the difficulty on these two endpoints, sounds and symbols. They believe that you should teach many more things all at the same time.

The case FOR letter-name first instruction

The Instagram slideshow was made from this “paper,” which we can dissect. I’m choosing to address the “research” instead of the Instagram post itself, because anyone can say anything on the internet (including me!). This document comes through the Colorado Department of Education website and it is not a research paper, it is a Word Document.

TheImportanceofTeachingLetterNames Download

In Colorado public schools, the same department that published these “best practices,” 62% of their students read “Below Basic” which means two out of three students cannot read age-appropriate texts for understanding. Maybe it’s snobbish, but I’m not sure I trust these folks as experts in the field of evidence-based reading instruction.

Line by line, lets discuss:

The Document Says	My Rebuttal
The basis of English literacy instruction is knowledge of the alphabet.	The basis of English literacy is the spoken English language, not the visual symbols. This is the fundamental premise that we disagree on.
While it is true that decoding instruction is based on letter-sound instruction,	Agreed.
the knowledge of letter names provides a common language to discuss letters.	Why are we discussing letters? The name of the letter isn’t important. At the beginning, you don’t need to mention a symbol’s name. This is a problem with letter-first or visual-phonics strategies that can be completely avoided when you teach from speech-to-print.
Many letters have more than one sound that they produce and many sounds are produced by more than one letter.	Letters do not make sounds, people do. What we agree on here, is that each sound can be written in many different ways (main, pay, weigh), and that the same symbols can represent more than one sound (head, bead, break). This orientation however isn’t me just being knit-picky. Our visual symbols never make sounds, that’s nonsense, or it’s a confusing metaphor. Children that are literal-thinkers can actually believe that just by looking at a letter, the sound will be made in their head, kind of like watching TV. When they don’t “hear the sounds letters are making,” they think something is wrong.
Without a consistent language surrounding the symbols of the alphabet, it is more difficult for students and teachers to discuss these letters consistently.	Do you really need to discuss letters? Check your premises. What I think the author is trying to indicate is that letter names are the only stable part of the writing system. Letters are capricious and “make” a lot of different sounds, but the name is constant. “C is see even if it says /k/ or /s/.” This is a misidentification, the shape is what stays the same. But neither the name nor the shape are the important anchor point. The stable part of the writing system is the 40+ phonemes. “/k/ is /k/ even if it’s spelled C, K, or CK.”
This is particularly important when considering spelling instruction because sounds have to be mapped to letters.	Agreed. Our difference is that, as a proponent of the Montessori method of reading instruction, I believe you should teach children to read through spelling instruction.
Letter names also provide a connection between upper and lowercase letters of the alphabet. Teaching the letter name “A” provides a bridge between the symbols of A and a. These connections are critical to build memory networks in the brain for the alphabet.	At the stage that I teach uppercase letters, letter-names have been taught. Where we disagree is that I believe uppercase letters and letter names are taught after the child can phonetically spell each of the 40+ sounds of English (I call this the Basic Code.) Whereas this document proposes teaching uppercase, lowercase, names, and sounds all at the same time.
Many letter names also provide clues to the most common sound that the letter makes. For instance, the letter name “B” includes the sound /b/ within the name of this letter. In these instances, the letter name provides important background information for children to learn letter-sound correspondences more rapidly.	My classroom experience completely refutes this claim and so does reliable evidence. B is pronounced “bee,” this is two sounds /b/ and /ee/. The emergent reader needs to have sophisticated enough phonemic awareness and conceptual knowledge to know that they need to segment off the first sound /b/. Students that can do this are likely not children who are at risk for reading difficulty.
Unfortunately, this does not apply to all 26 letters of the alphabet.	This isn’t “unfortunate,” it is a critical flaw in teaching letter names first. This pattern of segmenting off a single sound from a name doesn’t apply to a mere 26 individual letters, and there are between 280-500 spelling patterns in English! For a spelling pattern like TH, what part of “tee- aitch” has to do with /th/ as in thin? For OI, what part of “oh-eye” has to do with /oi/ as in boil? Repeat for 280+ spelling patterns.
Letters such as “W” do not include their most common sound. In fact, young children sometimes confuse letters because the name does not include the sound from a different letter such as the letter name “Y” beginning with the /w/ sound. This potential confusion is often sighted as justification for not teaching letter names.	This isn’t a minor point! And it’s not a “potential confusion,” this actually confusing to children. Imagine going to a party and meeting bunch of new folks: “This is Betty, but she prefers to be called /b/, this is Kevin, but he goes by /j/.” Later, when you see these folks again, which name will you remember? Betty or /b/? Kevin or /j/? The extra cognitive load of remembering two pieces of information, and simultaneously remembering that one of those pieces is irrelevant to your task can be very overwhelming. And yes, it is a major justification I have for not teaching letter names first.
However, even with these confusions possible, a study by Piasta and Wagner (2009) found that students who were taught letter names and letter sounds together actually had better letter sound knowledge than students who were only taught letter sounds. This finding indicates that potential confusions of names and sounds is not a valid reason to not teach letter names, but rather that letter names actually support letter sound acquisition.	Let’s take a look at this study: Fostering Alphabet Knowledge Development: A Comparison of Two Instructional Approaches The study was conducted on a small group of preschool age children, 58 children with the average age of 3.7 years. This study taught 26 sounds and 26 uppercase letters (so no lowercase letters, SH, AY, etc). This is a problem; there are 40+ sounds of English and 280+ spellings. Even a highly constrained set should include all 40+ sounds and at least 40+ spellings. In the letter-name group, children were taught “kay makes the /k/ sound.” Letters do not make sounds. In the sound-only group, letters were referred to as “the letter that makes the /k/ sound.” This means they still gave the letter a name, they replaced a short name, kay, with a new longer name the-letter-that-makes-the-/k/-sound. This instruction still placed the visual symbol as the primary, not the sounds of the spoken language. This is not an accurate representation of a sound-first instructional method. You should say: “/k/ is spelled like this [write or point to K.]” This study was not a comparison of two instructional approaches (such as a linguistic phonics program like mine or Montessori’s vs. a visual phonics program), it is a comparison of two visual phonics programs that both teach from print-to-speech. The results of the study were that children who got letter-name instruction performed better in alphabet knowledge and letter-sound connection. How much better? Children receiving letter-name instruction learned an average of 6 letter-sounds while those in the control group and the sound-only group (or should we call it the “long name” group?) learned an average of 4 letter-sounds. But there was no difference in outcomes for vocabulary, phonological awareness, emergent reading, or developmental spelling between any group. So even though the children in the letter-name group learned two additional letter sounds, it did not translate into better reading outcomes. This paper itself has a nice summary: “The lack of [alphabet knowledge] transfer to emergent literacy skills also deserves further attention. As the ultimate intention of providing effective early alphabet instruction is to impact reading and spelling abilities.” “Hence, although alphabet knowledge and reading/spelling development *may* be causal related, the current study and its emergent literacy measures *may not have been powerful enough to detect these relations*.” [Emphasis added.] I suspect the Colorado Department of Education only read the abstract, because the paper itself is rather clear: Learning letter names does not improve reading outcomes.
It is clear that letter name instruction should not be overlooked in favor of letter sound knowledge, but it is also important to remember that the basis of decoding instruction relies on letter sound knowledge.	I do believe children need to learn letter names, they are important. We spell outloud and refer to letter-symbols by their name quite often. I do not believe they important for emergent readers. Do not teach letter names first. Teach them later, after the child has a reasonable grasp of our complicated English code. We agree: The basis of decoding (reading) relies on letter sound knowledge.
With this in mind, letter naming instruction should not be belabored to the detriment of letter sound instruction.	Agreed! I propose that all letter name instruction, for the emergent reader, is at the detriment of letter sound instruction. Let’s wait until the child has a fundamental grasp of reading and spelling before introducing the symbol names. With a linguistic phonics program, this is only a delay of a few months. It makes almost no difference to the instructor but it makes all the difference in the world to the child.
Jones and Reutzel (2012) developed an instructional technique called Enhanced Alphabet Knowledge (EAK) Instruction. This instructional format teaches letter names and sounds along with the written form for each letter in a quick and efficient format with consistent distributed review cycles. Within an EAK format, all 26 letter and sounds are taught within the first 26 days of school and then reinforcement lessons are taught consistently for students who need more time or on frequently confused letters. This study demonstrated significant increases in students performing at benchmark on measures of letter name fluency and a dramatic decrease in at-risk students after receiving EAK instruction. An example EAK lesson can be found within the article by Jones and Reutzel (2012) and can be accessed through your local library.	Enhancing Alphabet Knowledge Instruction: Research Implications and Practical Strategies for Early Childhood Educators “EAK emphasizes identifying the letter name and sound, recognizing the letter in text, and producing the letter form.” This is what each of these activities looks like: Identifying the letter name and sound: Point to a letter, A. What is the name of this letter? What sound does it make? (I’m a broken record: Letters don’t make sounds.) Recognizing the letter in text: Find all of the A’s on the page. (Possibly useful, but I’m doubtful. What about digraphs like AI, AY, or AW? The second letter will not be “found” even though its an important part of the spelling pattern.) Producing the letter form: Write an “ay.” (This activity is similar, but critically different than the Sandpaper Letters. With the Sandpaper letters you write an /a/. The sound, not the name.) None of those tasks are decoding (reading) or encoding (spelling), why are you doing them? The program only teaches 26 letters and 26 sounds. This is a problem; there are 40+ sounds of English and 280+ spellings.

And that’s it. A short document with weak evidence showing that if you teach children letter names then they learn letter names. With the level of aggression that the Instagram post put forward, I was expecting something a bit stiffer.

Correlation and Causation

Letter-name knowledge is a strong predictor of reading outcomes. But that’s just it, it’s a strong predictor of reading outcomes, we do not know that it causes improved reading outcomes.

Letter-name knowledge, number-name knowledge, color-name knowledge, shape-name knowledge, and cartoon-character name knowledge all strongly and positively correlate with (predict) reading achievement. But this doesn’t mean we should teach the rainbow to help children learn to read!

What the evidence seems to indicate is that children who have an aptitude for paired-associate learning or have good memories in general are good readers. If that is true, then students with lower aptitudes in these areas will struggle more if they are given too much information to wrangle- so why would you introduce more complexity on their memory and paired-associative memory?

If a child has more difficulty with paired-associative learning, then adding in additional unnecessary information will make this process harder. For these students, our students at highest risk for illiteracy, we must isolate the difficulty and only present the essential information. You do not need to know that “double-you, eye, enn” says win; you need to know /w-i-n/ is spelled win.

We are burdening them with initially useless information.

What’s in a name?

The names of letters are just a way for us to talk about these symbols. They are not related to the core skills of decoding (reading) or encoding (spelling), they are only used for talking about letters in isolation.

Would you be as keen to teach letter names if cat was spelled out loud as, “the-letter-that-makes-the-/k/-sound, the-letter-that-makes-the-/a/-sound, the-letter-that-makes-the-/t/-sound” or “Charlie, alpha, tango” or “stick, ox, sign?” Those are just as arbitrary as “see, ay, tee.”

Put yourself in the emergent reader’s shoes: For our Letter of the Week Lesson, we were going to learn all about “soo.” “Soo” makes the sounds /k/ and /s/ as in cup and cent. This is different from “kiff” that makes the sound /k/ as in kangaroo and “iss” that makes the sound /s/ as in silly. Sometimes “soo” works with “ith” to say /ch/ as in chat.

Objecting to this “best practice” is what prideful, privileged, ignorant people do? Thinking this is a bad way to teach reading means I don’t “recognize the child that is struggling”?

Agree to disagree.

The problem is bigger than this

This “debate” between teaching letter names first or later isn’t the lynchpin issue holding our children back from reading. (Though the original Instagram post charmingly asserts, “There is no debate to be had. Period.”) We have bigger fish to fry.

Reading is a multifaceted skill, we all know this. There is a complex dance between kinesthetic, visual, and auditory memory at play when we read and write words. There is a child’s entire life experience, vocabulary exposure, and spoken language environment that affects reading outcomes, and we as educators play a very small role in that.

A child needs the prerequisite phonemic awareness skills of segmenting and blending, they need explicit instruction in letter-formation, they need a robust spoken language to comprehend what they’re reading. Children need background knowledge, cultural knowledge, and social knowledge. They need a grasp of grammar, they need to see clearly, they need to develop focus and attend to the words on the page.

In other words, decoding is the easy part on the path literacy.

Let’s put aside our pride and do what’s best for all learners.