Decoding Multisyllable Words by Chunking

[Rod Everson]

Okay, I'm on a mission. All, and I mean all, of the ways that we currently teach kids to decode an unfamiliar multisyllable word are too difficult for a child learning to read to understand. In fact, a lot of parents don't understand them. Not more than one parent in one hundred ever learned Orton-Gillingham's six syllable structure, for example, and I'm talking about now, as adults. They certainly didn't use them to learn to identify unfamiliar words in second, third, and fourth grade.

 

Similarly, all of the prefix, suffix, root word nonsense is useless for a language like English. Sure, it helps (sometimes) for deriving meaning, but for decoding, they just don't address a high enough percentage of longer words to encourage a child, and especially a 9 or 10 year old, to adopt it as a primary decoding strategy.

 

But for a decade I taught struggling readers and used one easy method to teach them how to tackle multisyllable words. And it works reliably enough that they would quickly adopt it as a primary decoding strategy.

 

I'm going to lay it out, in full, here and hopefully people will try to pick it apart. In doing so, I think most of you will come to the conclusion that it will work for your child and it's so easy to teach that anyone can do it.

 

It consists of one main rule and three exceptions.

 

The Main Rule: Stop each chunk after the vowel sound and use the first vowel sound in each chunk.

 

Terminology explanation: Vowel sound is the spelling of the vowel, like the "er" in "herself", not just the letter "e"

More terminology: Each vowel spelling has a first option, and may have two, three, or even four. The letter "e" has /e/ for the First Vowel Sound and /ee/ for the Second Vowel Sound, for example.

 

Example of words decoded on the first pass:

 

ta--blet, de--cade, to--pic, show--er, ligh--ter, far--mer, bor--der, nee--dle, ca--ta--stro--phic (of course, this assumes they know the code, i.e., that "igh" is the /ie/ sound, that "ar" is the /ar/ sound, that "le" is /ul/, etc., but that's another issue.

 

Now, of course there are a lot of words that won't be decoded correctly on the first pass, but first, here are the exceptions:

 

Exception 1: If the next chunk starts with a doubled consonant, add it to the previous chunk.

 

Example: ra--bbit becomes rabb--it (decoded correctly because doubled consonants almost always are preceded by a First Vowel Sound)

 

Exception 2: If the next chunk begins with one of the four markers (ck, tch, dg, and x, all taught previously), add the marker to the previous chunk.

 

Example: lo--cker becomes lock--er, ha--tchet becomes hatch--et, ba--dger becomes badg--er, and bo--xing becomes box--ing, all correctly decoded on the first pass. (Markers nearly always "mark" the preceding vowel sound as a First Vowel Sound.)

 

Exception 3: If the next chunk is "Hard to Say", add the first letter to the preceding chunk. (In adult terms, if a chunk is "hard to say" it is starting with an illegal blend in English, such as nt, or bp, or bt, etc.

 

Example: co--ntact becomes con--tact (because ntact is "hard to say") and is decoded on the first pass. (Note that this exception has the effect of automatically forming the "closed syllable" of Orton-Gillingham fame.

 

I have taught the Main Rule and those Three Exceptions to nearly 200 kids, and this system works. There is one added element to it, however, that is explained not as a rule or an exception, but as the overall strategy.

 

Strategy: If the first pass through the word doesn't result in a word you recognize, then start systematically trying Second and Third Vowel Sounds in all the chunks that haven't had an exception applied. If you go back and look, each of the exceptions share two important characteristics:

 

1) They tend to follow First (short) Vowel Sounds, and since the Main Rule states that the First Vowel Sound is to be applied first (hence their name), those chunks are almost always correctly pronounced on the first pass through the word. Thus, the child can ignore them when testing options.

 

2) The second common characteristic is subtle, but important. All of them create situations that "look normal" to the child. Thus, words don't start with doubled consonants like ss or ll, but they do end with them, so moving them to the previous chunk looks right to a child. Similarly words don't start with the markers ck, tch, dg(e), and x, they tend to end with them, so they too look better at the end of a chunk. And finally, nt, or bt, or mp, also don't start words, so it's easy for a child to learn to move one of the letters to the previous chunk. In other words, they all look strange at the beginning of a chunk, so it's easy for a child to learn to move them.

 

Now, if you play with some pretty long words, it's surprising how many of them decode relatively easily with this approach. Of course, this is English, and you'll find thousands of challenges, but I maintain, and have observed personally, that most kids will take to this like the proverbial duck to water, assuming they've been properly prepared. By that I mean they need to have a reasonably secure grasp of the code and understand the concept of markers and first, second, and possibly third options for vowel sounds.

 

Some examples from the previous paragraph: "surprising" decodes as sur--pri--sing. The "ur" has only one option, so the child tries the /ie/ sound in "pri" on the second pass and is very likely to get the word even though they will be changing the /s/ sound of "sing" to "zing" at the same time.

 

"decode" decodes as de--code (with a short "e") and is easily recognized on the second pass when they switch to the Second Vowel Sound (/ee/).

 

"easily" decodes as ea--si--ly, with the correct pronunciation of each vowel on the first pass, because /ee/, /i/ and /ee/ are the first vowel sounds of "ea," "i," and "y" respectively, and most kids will easily switch the /s/ sound of the letter "s" to a /z/ sound, particularly if they've been taught both sounds for "s".

 

"possibly" decodes as poss--i--bly and is correct on the first pass.

 

"secure" decodes as se--cure and is probably recognized on the first pass (the "e" is a schwa, but /e/ will usually work)

 

"second" decodes as se--cond and is again probably recognized on the first pass.

 

"understand" decodes as un--der--stand and is easily recognized on the first pass.

 

In addition, what this method of multisyllable decoding does is continually emphasize the pronunciation options for the vowel sounds. Children are continually rehearsing the first, second, and third sounds of the letter a, and the two options for "ow", or the three options for "ea", etc. Using this method cements the code knowledge in their heads. They learn to systematically approach long words, knowing that they can successfully tackle most of them, while realizing that some are still tricky. However, even the tricky ones eventually fall into patterns that become routine. The "a--tion" at the end of so many words, for example, always has the Second Vowel Sound for the letter "a", and they soon learn to skip trying the first one. A word like "sensation" then becomes decodable on the first pass instead of the second (sen--sa--tion).

 

As I said, I'm on a mission. I've put all of this on my website at OnTrack Reading in full detail, exactly as I teach it, with worksheets and instructions to move from 2 syllable, then to 3 syllable and finally to 4/5 syllable words in sequence, with words chosen carefully so that the concepts are learned in a useful sequence. And it's all free starting on the page Multisyllable Method Overview. I'm convinced that if every child was taught this method beginning in later first or early second grade most would easily learn to tackle multisyllable words with confidence, instead of the ridiculous guessing that many of them now do. And teaching it would be far simpler than the methods now commonly, and generally unsuccessfully, being used. Most kids do learn to decode longer unfamiliar words, but many of them are doing it entirely on their own because the strategies they're being taught are too complicated to be used by a nine-year-old and/or too unreliable to be trusted by them as good strategies.

 

Rod Everson

OnTrack Reading

[julikins]

This sounds good. Checking it out for my 10 yodd who struggles with this aspect of reading and writing. Thanks.

[Rod Everson]

You're welcome. I think you'll be pleased with her reaction to the method.

 

You mentioned writing, and I just wanted to add that most of us spell by chunks also, but that kids might need to be told that fact. For example, when spelling "concentrate" we usually say (and spell) con....cen....trate, breaking the word at the natural verbal boundaries. A child should be told to focus on the more unusual, or easy to miss, parts of each chunk. For example, "con" is easy. With "cen" they need to recall the "c" for the /s/ sound, and with "trate" they need to remember the a-e spelling of the /ae/ sound.

 

Incidentally, with a word like "rabbit" the natural verbal boundary is ra--bbit (unless you were trained as a child to say it with two /b/ sounds (rab--bit) as many of us were. So, when a child learns it (following the first exception) as rabb--it, he's not following the natural oral break. This can serve as a memory device. "Why am I saying 'rabb'?...It must have a doubled consonant." A similar thing happens with markers.

[MandMum]

Wow thanks!! This is exactly what my 8 year old struggles with so will try this with him

[Rod Everson]

ETC4 has it as rab|bit. Their rule is when 2 consonants stand between 2 vowels, the word is usually divided between the consonants: vc|cv.

Also, when a word has 2 consonants that come between 2 vowels, the word is divided into syllables between the 2 consonants - vc|cv, for example, Nap|kin.

 

Will I fall into trouble moving from ETCs method to the op above?

 

I don't think you'll have any trouble, but you will have to start treating the "bb" in rabbit as a digraph for the /b/ sound, along with most other doubled consonants.

 

The difference between the sort of syllable rules that ETC uses and the ones in the OnTrack Reading method are pretty stark. To figure out a word using ETC's rules you have to be thinking about vowel letters and consonant letters, and rules that apply to them.

 

To figure out a word the OnTrack way, your child just stops every single chunk after the vowel sound and applies the first (and usually most common) sound to each chunk, applying the three exceptions (which are very easy to learn) as he goes along. As I pointed out in the main article, since the First Vowel Sound for a, e, i, o, and u are all "locked in" by any one of the Exceptions, changing the ending of the chunk doesn't require going back and changing the vowel sound. In fact, adding a doubled consonant, or one of the four markers (tch, ck, dg, x), or breaking a consonant off the following chunk because it's "hard to say" each lock in the short sound of the vowel letters.

 

Taking a fairly complicated word like "concentrate" for example, as he's reading left to right he'll say "co" (cah--the short sound of "o" now), then see the "nce" and easily decide that's hard to say and change first chunk to "con" (locking in the short, or first, vowel sound). The second chunk starts as "ce" (which he might pronounce "se" or "ke" depending on his knowledge of the Rule of c--the one rule I do teach), see the ntra at the beginning of the next chunk and revise the second chunk to "cen" (sen or ken, depending on his choice).

 

Note that thus far, he's not even made it to the end of the word. No looking at vowels or consonants or patterns like cvvc, vccv, etc., just reacting to something that is clearly "hard to say." Finally, all that's left is "trate" which if he's ready to work on multisyllable words, he should easily read correctly. So he's got either con-sen-trate, or con-ken-trate, and the only real option left to try is to adjust the sound he chose for the "c" in "cen" if he chose /k/ first.

 

What I observed in working with a lot of kids, one-on-one, is that their confidence grew tremendously as they gained familiarity with the process. But beyond that, a huge advantage of this method is that they are continuously reviewing the various options for the vowel spellings (and some of the consonant spellings as well, like the "c" in "cen")

 

As I said, I'm on a mission. This is the way kids should learn to tackle longer words, because it's easy, intuitive, and reinforces code knowledge in a significant way.

 

Incidentally, your example "napkin" starts out as na...pkin, which gets revised to nap-kin due to the third exception (pk in is hard to say) which has the effect of closing the syllable naturally, and locking in the short vowel sound. Almost all two syllable words of the vccv structure like that are easily decoded on the first pass. It's just that instead of focusing of a vccv rule, they're reacting to something they don't normally see, an illegal blend that is "hard-to-say".

 

And if the cc in the vccv is doubled consonant, like bb, tt, zz, etc., they're both moved to the preceding chunk in accordance with the first exception, with the same result, locking in the first (short) vowel sound. Note that it's vital that they always try the first vowel sound "first", which is why it's named that.

 

I apologize for the long delay in responding, by the way.

 

Rod

 

P.S. There are always tricky words. This is English after all. But I'd be happy to discuss any examples people have if you mention them in a comment.