My creation, Story Hour Academy, offers a free online phonics video course to beginning readers. I designed it to provide explicit, direct instruction in the Orton-Gillingham phonograms through a series of 100 video lessons in the style of Khan Academy. The videos are available to anyone with an Internet connection.
What I didn’t expect when I started this project was for my experience with it to make me aware of what’s wrong with phonics. If you’re teaching a beginning reader, evaluating reading curriculums, authoring reading curriculums, or involved in improving literacy outcomes for students in any way, you could benefit from what I’ve learned.
Reasons to Doubt the Effectiveness of Phonics
I’ve uncovered five good reasons to believe that there must be a better way to teach reading than to accept a currently existing phonics programs in its current form.
First, each phonics program teaches a slightly different code, none of which are free of exceptions. For example, many Story Hour Academy lessons have a section dedicated to exceptions such as how the <ai> grapheme in said does not represent the /eɪ/ phoneme like in paid, how the <oa> grapheme in broad does not represent the /oʊ/ phoneme like in toad, and how the <ey> grapheme in geyser does not represent the /eɪ/ phoneme like in they or the /i/ phoneme like in turkey. A non-universal code with a bunch of exceptions is no code at all. Consequently, phonics instruction can shut off the curiosity that teachers and students may have about why words are spelled the way they are. We all deserve more satisfactory answers about the written word than the default answer in phonics, which is that some words follow the “code” and other words don’t. We can learn things like why the words <walk> and <talk> are spelled with an <l>. This is far better than accepting that “they just are” and thinking that written English is crazy. Contrary to what phonics programs would lead you to believe, there is an understanding of the English language whereby spellings can be explained.
Second, the supposed “code” behind phonics instruction is just a curriculum developer’s opinion about what graphemes to teach, as well as which and how many of the corresponding phonemes should be taught for each grapheme. Story Hour Academy has lessons on letter groups that I once thought were graphemes, but are definitely not. It turns out there are no <ci>, <ti>, <si>, <eigh>, <ough>, <eigh>, <augh>, <ng>, or <ed> graphemes. Other phonics programs call other groups of letters graphemes that do not hold up across the system. Some use the term “split digraph” to create a grapheme out of letter strings like a_e, e_e, i_e, o_e, and u_e where the underscore represents a placeholder for a consonant. By this logic, the <a_e> “split digraph” in the word place changes to an <a> grapheme in the word placing. This is inelegant and unnecessary.
Story Hour Academy also makes the mistake of teaching how some graphemes represent just one phoneme when they actually represent more than one phoneme. For example, it teaches that <d> represents /d/ while ignoring the fact that <d> can also represent /dʒ/ as in educate, /t/ as in camped, or no pronunciation as in handkerchief. Phonics should be good at teaching grapheme and phoneme correspondences (GPCs), but how can it when every program has mistakes and omissions in its inventory of GPCs?
Third, phonics does not adequately recognize the influence of morphology on the English writing system. Instruction in grapheme and phoneme correspondences needs to coincide with instruction in morphemes because of the fact that graphemes cannot cross morpheme boundaries. For example, the word action has a base, <act>, and a suffix, <-ion>. We know <act> is a base because of the free base word act (from Latin agere) and because of the morphology of sibling words like acting and react. We know <-ion> is a suffix because there is proof of it in words like option, notion, and fusion. Knowing this, we can say the following about the word action:
- It has two morphemes: act.ion
- It has six graphemes: a.c.t.i.o.n
- It has these phonemes: æ.k.ʃ.∅.ə.n
We can learn some interesting facts from studying this word such as how the grapheme <t> can represent the phoneme /ʃ/ and how the grapheme <i> can represent no pronunciation. We can identify these patterns again in words like partial and Egyptian. If students study words in this way, with GPCs in a morphological context, it can help them master longer and more academic words.
Additionally, students can learn how each base word forms the center of a morphological family. They can study word families in order to quickly master a large number of related words in a short amount of time. I have a friend who studied the bound base <struct> with her 7-year-old son. They added prefixes and suffixes to <struct> for fun while talking about the meanings and pronunciations of each newly formed word. They had fun and ended up covering dozens of words in one short conversation.
Fourth, and subsequent to the previous point about morphology, phonics programs do not adequately recognize the influence of etymology. Etymology must be used in order to get morphology right. Words can only be in the same morphological family if they share a base with the same spelling and trace back to a common root. For instance, bases can be homonyms with the same spelling but with different meanings. The bound base <gain> in ungainly comes from Old Norse gegn meaning “straight, direct, helpful.” So, ungainly is an adjective used to describe things that are “not straight.” In contrast, the free base in gained comes from Old French gaaignier meaning “earn, gain, trade, capture, win.” So, gained is a verb used to describe having earned or captured something of value. Even though these words share a surface appearance, ungainly and gained are not in the same morphological family. The etymology of these words makes this clear.
Fifth, and most importantly, phonics programs fail to explicitly teach some of the most useful information about the English writing system. I compiled the following list of facts after getting clued into them from Peter Bowers and Gina Cooke via classes and resources such as this blog post and this forum post. Here are some facts that I now consider essential to early reading instruction:
- Words are made up of smaller parts. These smaller parts have names such as morphemes, graphemes, and orthographic markers.
- Morphology is the structure and sequence of meaning. A morpheme is a minimal distinct unit of the morphology.
- There are two main types of morphemes: affixes and bases.
- There are three types of affixes: prefixes, suffixes, and connecting vowel letters.
- Suffixing conventions such as the consonant doubling convention, the y to i convention, and the final e convention guide how suffixes are added onto words to make longer words. These conventions are thoroughly documented in “Teaching How the Written Word Works” by Peter Bowers.
- Every written word has a base.
- A base may be a content word (aka lexical word) or a function word.
- Content and function words have observable differences. Function words may have 1-2 letters whereas native content words require at least 3 letters. For example, the word I is a function word whereas the word eye is a content word.
- Graphemes and orthographic markers can only occur within morphemes. Graphemes and markers cannot cross morpheme boundaries.
- There are 82 grapheme cards in the LEX Grapheme Deck by Gina Cooke. This is the best accounting I’ve found of graphemes in the English language.
- Graphemes represent abstract units of language called phonemes. When you vocalize a phoneme, it stops being an abstract unit of language in your head and becomes human speech. The word for the smallest unit of human speech is phone.
- Most graphemes represent more than one phoneme.
- There’s a total of 49 phonemes documented by Gail Venable and Rebecca Loveless on pages 105-106 of “Dyslexia and Spelling: Making Sense of it All” by Kelli Sandman-Hurley Ed.D. This includes 32 consonant phonemes, 15 vowel phonemes, and 2 semi-vowel phonemes. Since phonemes are abstract units of language in the mind, not every person will have the same count.
- Phonemes are to phones as colors are to shades of colors. When a phoneme can be pronounced with more than one phone, the phones are called allophones. For example, the aspirated [ kʰ ] in kit and the unaspirated [ k ] in skit are allophones of the phoneme /k/. In keeping with the colors analogy, [ kʰ ] and [ k ] are shades of /k/. I mention this to suggest that the concept of a phoneme has some variability to it and that the word phoneme can not be used synonymously with the word sound.
- Most phonemes can be represented by more than one grapheme. For example, the phoneme /m/ can be represented by the grapheme <m> or the grapheme <mb>.
- A grapheme may correspond to a zero phoneme whereby it does not represent a pronunciation. For example, the second <b> in bomb represents a zero phoneme because the second <b> in its sibling, bombard, represents a non-zero phoneme, the phoneme /b/.
- English orthographic markers include e, b, d, g, h, l, o, p, s, t, u, w, z, ch, ugh, and igh.
- Orthographic markers never represent a pronunciation.
Obviously, you can’t just read this list off to a beginning reader and expect them to understand it. However, you can incorporate these facts into daily lessons in a logical, coherent, and memorable way. There’s a large and growing community of teachers using a method of study called Structured Word Inquiry (SWI) to grow their own knowledge of the English writing system and teach their students how to do the same. With SWI, a student is not confined to only the knowledge that his or her teacher has to share. The student and the teacher can expand their knowledge together.
Grading Story Hour Academy
If Story Hour Academy were graded according to its ability to convey the information shared above, it would not do well. Why?
- It has lessons with an “exceptions” section. This section covers words that do not align with the grapheme and phoneme correspondences being taught.
- It uses imprecise language such as calling graphemes “phonograms” and calling phonemes “phonogram sounds.”
- It provides almost no instruction on morpheme types or suffixing conventions.
- It never mentions content and function words.
- It never mentions orthographic markers.
- It provides lessons on some “phonograms” that are not actually graphemes.
- It does not cover every grapheme.
- It has lessons on syllable types. This obscures the morphological structure of words.
- It does not show words in the context of their morphological families.
- It does not show words in the context of their etymological families.
For now, I plan to leave the Story Hour Academy lessons on YouTube because research shows that phonics instruction, even in its unoptimized state, is better than no instruction at all. However, if you use Story Hour Academy, I recommend that you also learn and use SWI to interrogate the content of each lesson.
The Path to Better Reading Instruction
At the moment, I know of no one resource that can be used to accurately teach beginning readers the morphology, etymology, and phonology of the English language. However, plenty of teachers are learning the subject matter expertise that’s needed to teach this content in a classroom setting using a range of resources.
Most promisingly, Heather Johnson, founder of The Parenting Patch website, is working on a book similar to “Teach Your Child to Read in 100 Easy Lessons” but informed by SWI and linguistics. Heather earned a BA in English studies with a minor in creative writing from Illinois State University, an MS in library and information science from the University of Illinois at Urbana-Champaign, and an MS in English studies with an emphasis in linguistics from Illinois State University. She’s the author of ‘Structure and Meaning of Periphrastic Modal Verbs in Modern American English.’ She occasionally tweets about her progress on the reading instruction book via @LinguisticsGirl. Be sure to follow her for updates.