words that can't be decoded?

Hi Robin,

Instead of looking for a list, try this.

Whenever you and your son run into a word with unusual “sound pictures” in it, draw a double line under what each of you think is the unusual part. You might not even agree. Then, if the unusual part is an extra letter or letters, in your opinion, stop there. But if it’s unusual because it represents what you think is a very uncommon sound, write that sound under the double underlines.

The reason for the double underline, by the way, is to differenciate it from the way the curriculum tends to indicate a sound, by underlining each sound, so use the double underline when you do this.

Here’s a couple of examples:

broad: (Double-underline the “oa” and write “aw” under it.) You both know that “oa” is almost always the /oe/ sound, but can it be /aw/ to in several words? Well, if it can, it’s just got an /oe/aw/ overlap, right? But if “broad” is the only one you ever come across, then it’s a very unusual spelling of the /aw/ sound which needs to be noticed, but not generalized. If, six months from now, you’ve only got one word with the “oa” underlined, it should probably be on the list you’re talking about. Or should it? Isn’t it decodable, or at least 75 percent decodable?

sword: (One of you chooses to double-underline the “w” as an “extra letter” and the other double-underlines the “sw” as a sound pic for the /s/ sound, writing “s” under the “sw.”) Who’s right? Depends on whether you can come up with more words where “sw”=/s/, doesn’t it? “Answer” is the only one I’ve found so far, so I tend to think of the “w” as an extra letter in both words. But are they decodable?…sure…pretty much anyway. And will your son become more aware of the unusual spellings this way…definitely.

After you do this for a while, you’ll end up with a list of 40 to 80 words and will be convinced that there are very few English words that aren’t decodable to a very large extent. “Eye” is probably the worst. What would you do with that one?…See “lye,” “bye,” “rye” and “dye” for clues, but let him find them himself…..Rod

Rod , We’ll try this method but I know I saw a list posted here or on the READ NOW board.The reason I’m more inclined to use a list is because sometimes I’m really not sure myself if a word can be decoded or not.So far I have just been giving him the word and telling him that it is undecodable,but if a word really isn’t decodable wouldn’t that have to be more like a sight word list?Are you saying we should be adding these new spellings we’re seeing to our list and telling the child it is another way of showing that sound?Thanks and please bear with me I’m new to PG but am willing to learn.

Where are you in the program? Are you in the advanced code, chapter 5? If you are, look at the charts in that chapter for the different choices of letter combinations for the sounds and that should help you. There are very few words that can’t be decoded but words that are derived from other countries do not always follow the sounds for the letter combinationns. Also, don’t forget that there are suffixes and prefixes that he should be taught that are always certain sounds. Email me directly if you want and I will help you with PG.

Hi Robin,

Bear in mind that the philosophy underlying the program you’re using is that given enough information as to how English is constructed, kids can figure out the rest of the code on their own. By implication, so can you, once you accept the fact that it’s a code.

For example, the words “bye,” “dye,” “lye” and “rye” that I mentioned in the earlier posting clearly indicate that “ye” is a digraph for the /ie/ sound in at least four words. To my mind, this is enough to make these words completely regular and decodable, even though my earlier inclination was to think that the word “bye” just had an extra “e” attached to it (like “were” does, for instance.) The main point is that the curriculum you’re using never bothers to dig that deep into the code. There is no need to, once you know how it works….you train yourself to look for recurring patterns, and for the inevitable exceptions.

Basically there are hardly any true sight words, in the sense that the code in them makes no sense at all. Now, if we spelled, say, “kisses” as “XXX” then “XXX” would be a sight word, which of course it is….. (although even here, the sounds would be /ks/ks/ks/, so who knows…..maybe it’s not a sight word after all….*s*…..Rod

Hi Robin; The answers you’ve gotten have been interesting, and I’m sure helpful, but perhaps you are like me and still would like to have a simple list to start with. I wanted the same thing 3 yrs. ago when I became a PG therapist; I called and the girls there gave me a pathetic list that they’d made up (some were soooo decodable!), so I used the good ones and thought up the rest on my own. I add to it now and then; good use of my brain! Here are just a few; I truly believe there are only a couple hundred, which is nothing if you consider that the avg. person has a working vocab of 20,000 t0 50,000 words. (by the way….have you read the book by Dianne McGuinness …the one who figured out the code…the mom of Geoffrey McG…her book is “why our children can’t read and what we can do about it”…fascinating stuff! Here’s a start for you:
aisle, ancient, answer (Rod can add SWORD to his list of “sw” words)…asthma, ballet, bisquit, business, bury, busy , buoy, build, beautiful, colonel, cupboard, czar, clothes, cushion, depot, debris, dungeon, dinghy, exhaust, half, hearth, island, jeopardy, leopard, ocean, one, once, of, people, pizza, psalm, plaid, sergeant, says, sword, sew, soldier, surgeon, two, the.
That should give you food for thought! If I’ve misled you on any of these, I’m sorry; I know the pg code inside and out, but even then, you can decode a few things differently, but that’s a subject for another time. If you need any personal help, email me at my address. good luck!
ps…if you want to make your kid think he’s a super-star, give him this assignment; how many ways are there to spell the sound “sh?” After he shows you the “sh” way, show him the “s” (sugar, sure….my mind is tired; there are more examples of that), then show him “ch”..machine is a good ex….I follow up with a list of French words, such as champagne, chagrin, chandelier…there are several more. Anyway, he’ll think it’s cool and I’d then ask him to take the challenge to his mom or dad, or older sibling. It’s a real boost for most kids to see they can know something a good reader doesn’t(as long as it’s not done in a haughty way!

Leslie, Thank you for your reply.Concerning the list of words you gave me,these are more for my use than anything else.I’m not sure if I would just out and out hand him a list of words and tell him that they are not decodable.Ofcourse I’m thinking that instead of trying to get him to go back over all the sounds and try to find a place for these spellings that are unusual it might be easier and less confusing to him just to tell him that “yes,there are some words that just can’t be decoded.” I’m not sure but I think in time,after encountering these words in reading it will come naturally.I have to believe this because he knows
and those are not decodable.On the other hand if he comes to a word that he doesn’t already know and can’t be decoded he will have no idea of what it is or how to decode it.Probably should try and give him a way of dealing with these words other than memorizing them.Guess I’m just looking for an easy way to get this to him.
I’ll try the “sound seeking”game you suggested and no I haven’t read the Dianne M. book but I will.I do have Reading Reflex and the other one How to Raise Your Child Verbal Intelligence both very good in my opinion.Sorry so long and again thanks for all your suggestions.

Rod, Hello and thanks for your reply.I’m thinking maybe I’m making too much out of this decodable word list.I mean if a kid learns to read using PG there are not going to be a whole lot of words he comes to that he doesn’t know or if he can’t figure it out there’s always the dictionary right?I just don’t like coming to a word while were reading and have to say oh sorry that’s another one of those words that we can’t use the code for.Ofcourse I had to learn to read the word somehow and there wasn’t PG or Phonics when I was growing up.I will trust that you are right and that a kid will come to notice that yes there is a pattern in a lot of these words and it will be learned naturally.Thanks so much.

Hi Robin,

Yes, if you assume that the vast majority of words are decodable, you will be on good ground.

It’s important to build a foundation for matching up the sounds and symbols so that even if one does have to ask for, or look up, the pronunciation of a word, he can then match what he hears with what he sees and figure out why the word was hard to read. By doing this, words like “collage” which almost anyone will mispronounce the first time they see it in print, will start to make more sense as other words are added which have the “ge” representing the /zh/ sound, such as “beige,” “mirage” and “barrage.”

Now, you can waste time teaching every possibility, or you can build the foundation and then let the child explore the possibilities while reading good literature. One way is mind-numbing and the other mind-expanding….your choice…Rod

PS: I’m not saying it’s a waste of time to worry about or be interested in this stuff….I love it….but it is probably bad strategy to try to pound it into every kid’s head, just because we find it interesting.

Here’s a start for you:
aisle, ancient, answer (Rod can add SWORD to his list of “sw” words)…asthma, ballet, bisquit, business, bury, busy , buoy, build, beautiful, colonel, cupboard, czar, clothes, cushion, depot, debris, dungeon, dinghy, exhaust, half, hearth, island, jeopardy, leopard, ocean, one, once, of, people, pizza, psalm, plaid, sergeant, says, sword, sew, soldier, surgeon, two, the

many of the above words do follow English phonetic code,

the ones that don’t such as yacht, were not phonetically changed for English,

not sure why /clothes/ cannot be decoded, the /th/ is in there albeit, very quiet

ancient can be decoded,

/a/ /ng/ /ci/ (representing the sh sound which does repeat many many times in our code) /e/ schwa /n/ /t/
the word ancient is on the Burns and Roe IRI sight word test for fourth grade,

/ci/ is first encountered in the common word /special/, so kids need to learn the /ci/ /ti/ and sometimes just the /t/ can represent the sh sound or they will struggle with many MS words and many times the /t/ represents the ch sound such as in statue

ocean is also decodable

/o/ /ce/ /a/ schwa /n/

most words do have code that represents sounds, the word surgeon is also decoded

/s/ /ur/ /ge/ representing j /o/ schwa and /n/

jeopardy and leopard both have the /eo/ which represents the ‘e’ sound,

so most words are easy to decode, few as a i said before, just do not follow our phonetic sounds such as ballet,

the language of origin is i guess french and was not changed when brought into our english code

so the code is fun and fascinating, and by all means read Diane McGuinesses book Why our children cannot read, she talks about the code and it makes for fascinating reading,

it is prefaced by a guy i think his name is Pinkerton, and he is a linguist from Princeton University

but she did know our English code and wrote a fantastic book

i wish she would write another one, libby

Hi Robin, about non-decodables; I just tell my students that this a word that you can’t use the code for and luckily, there aren’t many of them. The truth is, it’s a piece of cake to memorize the 200 or so, if a big deal is not made of it. I never give lists; just use text that uses the new word a few times and I find they have it. When I tell them they will be able to decode about 50,000 words and can’t do only about 200, it puts it in perspective. How cool is it when a child knows the spelling possibilities for a sound? A student of mine (age 7) came in last week telling me he can’t believe his teacher told another student that she’d just have to memorize that “ai” in “said” makes the sound “e” and it’s an exception to the rule. His exact words, “first, she thinks letters MAKE SOUNDS, and then, she doesn’t even know that there are 3 ways to spell “e” in English!” (he went on to write “e,” “ea,” and “ai.” I explained that for a long time, I didn’t know the code, either, but together we can get everyone to learn it, teachers included. But, we have to do it in a nice way!

This approach can be the slippery slope to you-know-where.

You start to teach a system based on decoding.
Then you make a list of SO-CALLED ( I disagree strongly with the concept) “non-decodable” words. Then you try to have him memorize these words by sight. Then every time the student meets a new word that he doesn’t know, he has to find some magical way to tell if this is a word he can decode or not — has to be magic, because if he doesn’t know the word by sight already, there is no way he can tell if it’s on your special list or not. So he tries to run through a list of memorized words by visual memory, guess from pictures and context, and if that fails, use decoding skills. And guess what? You’re right back in sight memorization or “whole-language” approaches. And he is dependent on asking you whether this is a word he can sound out or not, so he cannot ever be an independent reader.
This method is extremely slow and inefficient and frustrating, and leads to poor comprehension and other problems because the student has so much mental space and energy tied up in searching through the memory banks for words, testing guesses in contexts, and so on, that there is no space left for the content of the reading.
It is simply not effective to try to read using two separate and mutually contradictory sets of rules: Rule A says IF it’s a decodable word, sound it out — but there are no rules to say whether it is decodable or not; Rule B says IF it’s on our list memorize it by sight — but there is no explanation of *how* you memorize by sight, or what you do with thousands of similar-looking words — and there is no system to know if it’s on our list or not. Imagine yourself learning a foreign language written in another alphabet, say Russian for example — would you be a good learner if you were told that you should sound out sometimes but memorize sometimes? Or would you ask people to explain the rules, please?

You mention that you were not taught phonics in your youth and this is new to you — it’s easy to slip back to old habits, and having been taught yourself by a system of memorizing lists, you tend to do this. Please try to avoid it — or at least if you use lists, use lists of regular patterns to teach what works most of the time, and *not* lists of exceptions to make things seem even more difficult.

There are NO, repeat NO completely “non-phonetic” words in English. The worst I know are “one” and “once”, but even there the n and ce = s and silent e are good clues. “eye” is bad but not that bad; y = long i is a common pattern in little words such as my, try, sky, etc, and silent final e is very common, so the only odd thing is the initial e. Most of our consonants/consonant groups are very dependable in sound. Our vowels have two or three main sounds and a couple of less common variants. Yes, there are irregular words; but even these have degrees of irregularity. There are silent consonants — some of these can be learned in patterns such as mb in lamb and comb, and some are relatively unique such as sw in sword and answer. There are vowels that have an unusual variant such as ai in said. There is the French soft g in collage and ch = sh in chef and machine. This is a nuisance, but it is systematic and teachable and learnable — it doesn’t require a magic wand and a whole separate system of reading.

A much more productive approach is to sound out words as far as you can, and then as Rod suggests, note the parts that break the rules; these parts will be a *small portion* of the whole word, most often a weird vowel pattern in 70 to 80% of cases, a silent consonant in perhaps 20% of cases, and an unusual consonant pronunciation in a few words. Then in order to read, you use the clues you have from the consonants and try vowel variants until something makes sense. You never tell the student that a word “can’t” be sounded out — that totally contradicts the effort you have put into teaching him that yes, he *can* read by a logical system. Rather, you tell him to **sound out as best he can** (which will be totally correct for 80 to 90% of words, and halfway for the others) and use context to pull together *small* irregularities. To learn to spell (preferably at the same time you first meet the word in reading) you study the odd *parts* of the words and make special note of them, and spell the regular parts by sound. This drops the memory load in both writing and spelling by 90% and makes both easier and more accurate, as well as leaving energy to understand and enjoy what you are reading.

As an adult reader I have come upon more words like this than I care to count! How is this decoded?

Hi SAR,

You may not “care to count” them, but it wouldn’t be that big a job. You’d have a hard time getting past 100 if you sat down right now and tried, without a dictionary to help you out, I’d wager.

And, even then, as victoria says, 75 percent or so of each word would be decodable.

The problem is usually extra letters, not missing letters. If you read my post right under Robin’s, “victuals” (pronounced “vittles”) would simply have the letters “c” an “u” double-underlined, since “ct” is not a digraph for the /t/ sound and “ua” is not a normal vowel digraph. So, “c” and “u” are extra letters, probably due to the corruption of English pronunciation of a foreign word.

As for spelling it, I always thing “vic-tu-als” turning it into a three syllable word, like we all do with Wed-nes-day, even though its Wens-day……Rod

PS: I agree with victoria about the swamp you create when you try to impress a child with the existence of the mysterious “sight-word” list.

Victoria, I now see after reading all the replies from everyone that I am creating a monster by even suggesting my son memorize these difficult words as sight words.I mean the idea is to keep remediation simple.I’m really pleased with the progress he’s made so far and don’t want to make it difficult for him.That’s why I posted to this board,I have been reading the posts for months and knew I would get some really good suggestions.Thanks

Well, originally, and I believe still in England, the word *was*/*is* a three-syllable regular word “vic - tu - als”. Not being from the southern US, I never hear or use the word in speech, so reading to myself I pronounce it that way and it is regular.
Having read a number of books which try to replicate regional pronunciations, I have seen the “vittles” pronunciation written, though honestly I can’t say I’ve heard the word spoken more than five or ten times in my life (maybe in a few TV comedies, but not sure.). I’ve always considered this a dialectical variant.

Adult version of explanation:
As far as what happens to the pronunciation, this is logical but a little hard to explain.
The “hard c” or “k” sound, and the “t” sound, are both what are called voiceless stops. They are quite hard to distinguish; you have probably had the experience of a staticky phone line, trying to spell something and mistaking a k or p or t sound — which is why we have the habit of saying “No, it’s T as in Tom”. So when a hard c = k sound and a t sound are next to each other, it’s very easy to combine them into a single sound. That’s why the c has disappeared in the common southern pronunciation.
As for the vowel, the “schwa” or UNstressed “uh” sound (phonetic symbol upside-down e) is the most common vowel in spoken English: the second a in Santa, the second e in elephant, the i in orbit, the o in second, the second u in beautiful, the sound before the r in her, bird, fur, work, and Earth, and the sound with the l in bottle. *Most* unstressed vowels fall to the schwa in quick normal speech. In the “vic - tu - als” pronunciation, the a is already dropped or reduced to a schwa. So when the c gets subsumed into the t and the middle syllable needs to go somewhere, it is normal for the u to disappear into the schwa.

Student version of explanation: sometimes we drop sounds when we talk. We all know about contractions such as he’s for he is, and we all know about saying “gonna” (with the schwa pronunciation of a) for going to. Well, in the word victuals, many people drop the c and slur the vowels. Just one of those things.

Hi victoria,

Just curious. Do linguists consider the “ur” in “fur” to be two sounds (schwa + r)? Don’t they acknowledge an /er/ sound as standing alone as a single sound (bird, work, etc.)? As I said, just curious…..Rod

PS: It’s funny, but I’ve always thought of “victuals” the same way you have. I even checked the dictionary, where I found “vittles” as the accepted pronunciation. “Vic-tu-als” isn’t even one of the options. I’ve both heard “vittles” and seen “victuals” many times, but never heard and read it at the same time. When I saw “victuals” my thought was, oh,oh, wonder if that’s “vittles.” Another one of those learning experiences.

In linguistics, as in any field, there are vicious internecine battles over the tiniest details. The group which I find most reasonable consider the ur, ir, er sound to be schwa plus r; r by itself is just a growl (general american dialect, please let’s not get into r-less speech!).
Many reading programs like what I call “phonics made difficult” — they teach every pattern as a separate item, complicating the system until you might as well just do word memorization (which I believe is often the unwritten goal of the authors). This kid of program teaches, for example, blends as separate items, not as being simply fast combination f already-known consonant sounds. In these programs a lot of fuss is made over ur as a separate sound. But it doesn’t need to be.

As far as looking up “victuals” in “THE” dictionary — I’ll make a bet that your dictionary is Webster’s. A perfectly OK dictionary, but often rather parochial; it generally only has US usages, and often only East Coast US at that. Check with Oxford. And please let me know what you find.
(I can’t check with my Oxford at the moment because it’s in one of thirty boxes somewhere under twenty sheets of Gyprock waiting for the builder to return, sorry.)

Hi victoria,

You’re correct about the Websters….*s*

I think though, that I’ve finally found an area where we would part company as to approach. I can’t imagine teaching a young child to sound out “bird” with four sounds. I would teach “cow” with three (as one curriculum I’ve seen actually does) before I would attempt “bird” as four. I trust, though, that you manage to make it work. It’s just at odds with the way I teach, as is the schwa generally. I try to go more with a perfect pronunciation…….Rod

No, ypu’re right, I do *teach* “bird” as three sounds, because the “er” combination is a common pattern in English.

There are at least three issues here: scientific phonetics/linguistics, phonetic representations, and pedagogy.

For linguistics and phonetic representations, you need a system that is as accurate as possible but not insanely complex. Most languages can be represented by somewhere between twenty and fifty distinct phonemes (independent distinguishable sounds). English is on the high side with forty-four to forty-six, depending on who’s counting (dialect issues). I’ve been looking at my French materials and have come up with thirty-six or thirty-seven; will have to check with the formal text (pedantic French - gag) when it comes out of storage; I’m surprised it seems so few.
In actual use, every phoneme is affected by accent and dialect and by the other phonemes around it; the l in lemon is very different from the l in bull, for example. But if we started making a different symbol for every variant, there would be no end to it, so we restrict ourselves to the minimum set of phonemes that can represent every word in the language and distinguish auditorially distinguishable words.
Thus we prefer to show the “er” sound as schwa plus r, because there is no *need* to invent a forty-seventh phoneme to distinguish it

Pedagogically, on the other hand, “er” is very common, as in her, bird, fur, and sometimes or as in work, so it’s a good thing to teach it as a sound pattern.

There is a tendency which I nickname “phonics made difficult”. Some teachers and authors like to elaborate and elaborate their systems until there are dozens of signs and several hundred patterns to learn. At that point you lose the forest for the trees and the kids learn that phonics is difficult and annoying and pointless — which unfortunately that kind of program is. So it is important to think before teaching whether things are getting over-complicated. I teach *pronunciation* of er as schwa plus r, and I teach it as a standard spelling pattern er/ir/ur sometimes or, memorize only which vowel.