Round the Blend

Why an over-focus on “blends” could be slowing reading and spelling progress.

Working 1:1 with primary school children allows me to monitor teaching trends in my local area. This micro-sample is often an indicator of general tendencies in teaching on a larger scale and I try to confirm, as much as possible, the extent of what I’ve observed.

In the early years of this millennium I started noticing my students incorporating the letter y into their lists of known vowels. I also saw a rise in simplistic but no less positive sayings like ‘every word needs a vowel’ and ‘every sentence needs a verb’.

These are pleasing trends and give me much hope that linguistics is making some headway in schools. On the downside, however, there is quite a bit of false information and teaching trends which negatively impact students. Because I work with children who struggle as a result of these trends, I am especially aware of the damage they do. The children who come to me are like educational bellwethers, since they are the most affected both by good, and more frequently, bad teaching.

Recently I’ve been noticing an upswing in what I call unwelcome intruders and missing persons. To clarify, unwelcome intruders are sounds a student makes when reading a word, even though the grapheme for that sound is absent, e.g. reading ‘blend’ for bend. Missing persons is the opposite, when students omit a grapheme in spelling, even though they say it, e.g. ‘bend’ for blend.

I work with a population whose phonological awareness (PA) is typically low, so these two phenomena are relatively common, but I’m seeing it more frequently and in a wider group containing children whose PA scores are average or above.

I have a suspicion about the origins of this change. Of course, it’s a theory at this stage, but I’d like to get to the bottom of it nonetheless. It concerns the teaching of blends.

A blend can be defined as two or more adjacent consonant (C) graphemes before or after a vowel (V) in a single syllable. A blend can be represented thus:

C C V C  for example:


C V C C for example:


C C V C C for example:


The phonemes in a simple syllable are relatively easy to segment, perceive and represent. A consonant-vowel-consonant (CVC) pattern, like bat or mug presents no significant difficulty in reading or spelling, especially if the phonemes can be represented with one letter.

But as soon as a little bit of complexity is added, all hell breaks loose.

Take a CCVC pattern for instance, e.g. stop, fled, or trip. Much greater effort and skill is required to perceive and represent that pattern.

One of the reasons for this is that sounds change a little bit when they appear in clusters. Their edges adapt to the sounds around them so that they can be said in a continuous stream. This is why clusters of consonants are referred to as blends, like blended coffee beans or blends of wine. The difference between consonants, coffee and wine though, is that consonant blends can be pulled apart again.

This pulling apart, or segmenting, is as important as blending and yet seems to get short shrift. By the way, shrift is an example of a relatively uncommon CCVCC pattern.

Teaching blending in the absence of segmenting is like teaching addition without subtraction. And yet I see it all the time.

Children and parents are increasingly reporting to me that their school is placing emphasis on blend work. One boy recently told me that first thing, every morning, his teacher plays a slideshow from a commercial reading program that shows a series of blends in isolation (e.g. bl-, cl-, dr- etc.) and the class has to recite each blend rapidly as one continuous sound.

Many of my other students are seeing the same thing. I asked if they were also taught to segment and they said they weren’t.

Now, many children will learn to read and spell independent of the quality of the method used to teach them. This is fine. But many won’t. Many rely on expert, explicit, systematic teaching of the structure of words to help establish a foundation for fluency, vocabulary and comprehension. This is referred to as systematic synthetic phonics. Teaching blends as one sound is not this.

Here is what I am seeing as a result of this blend teaching:

  1. Missing persons: My students (most with some form of developmental language disorder) are increasingly prone to pronouncing a two-phoneme consonant cluster as one continuous sound, e.g. /b/ + /r/ as a rushed and indistinct ”br”. When they come to spell b-r words, they say the two phonemes but represent them with the initial grapheme only. This results in words like bring  spelled as “bing”.
  2. Unwelcome intruders: When reading CVC words, my students have started to insert consonants that aren’t there. For instance, twice last term not one but two of my students said “black” for back in a paragraph they were reading. There was a distinct increase in this behaviour last year, especially among my newer students.
  3. Teachers handing out worksheets use the term blend, diphthong and digraph (often referred to as “diagraph”) interchangeably or erroneously, e.g. “The blend ch is followed by the diphthong /oi/ and then the diagraph ‘ce’ in the word choice”.

So how do we tackle this problem?

Teacher knowledge is the key. Blend training is done with the best intentions, but when the evidence base is lacking, the vulnerable suffer. There is no research evidence, as far as I’m aware, that teaching a blend as one sound is good for students, and a study here has shown no additional increase in word-reading speed when poor readers are taught blends.

Firstly, let’s untangle the definitions commonly used interchangeably:

  • blend – two or more adjacent consonants, either written or heard before or after a vowel
  • diphthong – a vowel sound (not a symbol) made using two places of articulation but requiring only one impulse of the voice, e.g. /oy/ begins with round lips and ends with lips stretched back
  • digraph – a written symbol, which uses two letters but represents one sound, e.g. sh, ch, ph

Secondly, let’s make sure we’re giving children plenty of practice in segmenting the sounds in words as well as blending them. Presenting ready-made blends and blend worksheets is folly. There is no evidence to show that this increases reading ability or speed, and I’m seeing the fallout on an all too regular basis.

Instead, have students generate their own blend sheets by systematically taking each consonant of the alphabet and figuring out which consonant goes with which other consonants. Have you noticed the pattern? When you and your students do notice the pattern, can you figure out why this is the case? This is so much more productive and generalizable than serving random blends up on a plate and treating them as one sound.

I’d love to hear from other practitioners about their experience with this phenomenon. I’d also like to hear from teachers about what they do in the classroom concerning blends.

Leave a Comment

Your email address will not be published. Required fields are marked *

7 thoughts on “Round the Blend”

  1. Avatar

    I am extremely interested in this post, as I have observed the same issue with a 7-year old I am tutoring in regard to the omission and insertion of phonemes, indeed one being the ‘black/back’ example referenced above.


  2. Avatar

    This makes so much sense. I really like your idea of having students generating their own blend sheets. I’m going to try having the young men I work with combine consonants and then determine which combinations are legal blends and which are actually digraphs.

    1. Avatar

      Thanks Ken. When any aspect of language, written or spoken, is approached in a generative way, we make greater strides to the sweet self-teaching zone that David Share would have us understand!