Do you remember the old 'unanswerable' question, "Which came first, the chicken or the egg?" Actually, it was answerable, but few people have the patience to go through a long biological discourse on the subject. In brief terms, it seems that birds evolved from reptilian stock. Chickens, as we know them, came in later. Ergo, it appears that oviparous chickens were part of a long evolutionary process and neither came first, in the true sense of the word. In plain language, it is evident to me that chickens and eggs were a natural and parallel continuing result in the development of various avian fauna over perhaps millions of years.

When we talk about language, on both the historical and the individual level the answer seems simpler. It is apparent that sounds precede writing, every time. The first axiom must be that all languages were spoken before they were written. There are still quite a few languages on our planet that are not written. The second axiom is that all fully functional humans first hear and later speak their native languages, before they learn to read and write them. 

I'm not a scientific linguist, obliged to prove to my academic peers statistically and through empirical investigation that what I say is true. Instead, I'm a right-hemisphere dominant teacher, who has spent many years in ESL classrooms, figuring out, through trial and error, what seems to work quickly and effectively and what doesn't work as well. One conclusion that seems to have been held for decades by some of the better teachers, despite some "scientific" arguments to the contrary, is that adults seem to acquire new languages most quickly and effectively, the same way that they acquired their native languages. 1.LISTENING 2.SPEAKING later 3.READING 4.WRITING. The problem is that few classroom teachers disciplined themselves to the point that this is how they really taught, even when the LS/RW sequence was fashionable. There is a certain convenience and tendency to pop books open on every occasion. This may be part of the problem, rather than the solution and is certainly easier to do.

Adult second language learners have disadvantages and advantages over native speaking youngsters in the language acquisition process. 1.DISADVANTAGE: Adult second language learners always have another language in the way, which tends to interfere with the new language, often predictably. 

2. ADVANTAGE: Adults are able to process informational input and use deductive and inductive logical processes more quickly than children, simply because they (theoretically) know more.

So what am I getting at? Last year I attended a workshop in which the presenter went over the processes that she used to teach pronunciation from the written language. She had numerous examples from phonics, which gave all of the rules to convert written English to spoken English. She felt that she was getting excellent results from her students who were learning to use the numerous formulas to come up with auditory results, expressed in terms of "long I, short A," etc. There seemed to be potentially hundreds of letter combinations to work from. She was obviously getting some results, or I don't think that she would have been there. I was uncomfortable with this approach, because I had always thought of phonics as being a system to use with young native speakers of English to teach them how to read. Children who are native speakers of English are already programmed, by default, with the phonology of their own language. I will not join the controversy on using this system with young Americans, as I have always worked in ESL for adults. There is quite a bit on the web on this subject, both pro and con.

I checked around and found that quite a few other teachers were using this same system in ESL, and it made me uneasy, particularly because we are talking about English, which is, to my knowledge, graphically farther from the spoken language than any language I have dealt with. Written English expresses its history, much more clearly than its phonology. If somebody is teaching Turkish, the gaps between the written and spoken versions of the language are minimal. This is because modern written Turkish was designed in the 1920's, after the Arabic alphabet was tossed out. This was after the Ottoman Empire period. Using the newly-adapted Roman alphabet some linguistically sophisticated persons were able to get a very accurate rendition of their language in written form, close to their phonemic system. What you see is mostly what you get. So if a student is learning Turkish as a second language, the writing system will interfere minimally. Even Spanish, which is certainly closer to its written language than is English, has quite a few graphic versus spoken-language anomalies, caused by word-environment differences (allophones). For example, voiced stop consonants in Spanish convert to fricative sounds between vowel sounds. Dedo (finger) is pronounced "day though," except with much shorter syllables than would occur in English. English speakers, reading this word, invariably pronounce it "day dough," and extend the syllables, as would occur in English. An English speaker hearing it pronounced correctly, without written input, usually doesn't have this problem.

Following is a reminder of what we are up against, when we learn to read our own language. The good news is that we bring an enormous amount of language skills, non-conscious and conscious phonological skills, and information at the time we learn to read our native language. We learn to recognize certain words from experience. Are we to believe that ESL learners could under any circumstances figure this one out, even with all of the phonic rules available?


When the English tongue we speak,
Why is "break" not rhymed with "freak"?
And the maker of a verse
Cannot cap his "horse" with "worse"
"Beard" sounds not the same as " heard",
"Cord" is different from "word",
"Cow" is cow, but "low" is low,
"Shoe" is never rhymed with "foe".
Think of "hose" and "dose" and "lose",
And of "goose," and yet of "choose".
Think of "comb" and "tomb" and "bomb",
"Doll" and "roll" and "home" and "some",
And since "pay" is rhymed with "say",
Why not "paid" with "said", I pray?
We have "blood" and "food" and "good",
"Mould" is not pronounced like "could".
Wherefore "done" but "gone" and "lone"?
Is there any reason known?
And, in short, it seems to me
sounds and letters disagree.

author unknown (and obviously British, which is fine.)

Here is what I know. Under the phonological system that I find most useful in teaching American English, we have eleven vowel phonemes, three diphthong phonemes and 24 consonant phonemes. This makes a total of 38 distinct sounds in the average American's phonological inventory. I am not a mathematician, and my wife won't even let me touch our checkbook. However, I have managed to figure that there are many fewer phonemes in American English, than there are phonics letters-to-sounds conversions. I believe that this is obvious, ergo, why not start with the smaller numbers of things to "get" and move on later into the more complex, if that is necessary.

Following is some of what I do in my ESL classes at various levels, usually more near the bottom than the top. It seems to work for most students, even some at more advanced levels, and they enjoy it. Also I must modestly add that my students, on the average, are able to discriminate words and pronounce them better than some other peoples' students, who learn from emphasis on the written word. My students also seem to be able to read at a later time, faster and better than some others.

We don't have space and time here to go through all of the methodologies and techniques that I suggest. These are all covered in my workshops, "Applied Phonology" and "Teaching English from Day One." Descriptions of what we do, in two four-hour sessions can be found at Ted Klein's ESL.

The use of applied phonology in the classroom seems to be missing in many ESL teacher-training programs. Most ESL instructors don't seem to be very familiar with the American English sound system, or even if they have had that training, how to use it in the classroom. That is a necessary component, in order to put the ears before the eyes. Many ESL students are in a confused state, caused by the fact that they aren't hearing and discriminating English words, even if they are seeing them in the written language.

I always start with vowel sounds and concentrate entirely on listening and identification skills, before we work on trying to actually make them. This is with the goal of achieving "phonemic awareness," which is one of the stated goals of programs, which put the written language first. However, it is my opinion that one of the major problems is that adults in particular, strongly favor visual input and often do not hear what they conceptualize from visual input. If their ears are properly trained first, they can bring some of the assets to the reading process that native speakers bring.

The word "clotheys," which I have heard from many non-native speakers of English, when they say "clothes," certainly doesn't come from the spoken language! "Who's that kuhnocking" on my door?" Ought/ rough/ plough/ dough? Rules??

With beginning students, I use vowel hatches, simple schematic tic-tac-toe types of designs that roughly represent the interior or the human mouth, facing to the left. Each vowel is assigned a permanent number, which is used for all future references to vowel sounds. A word is assigned to each hatch, with the vowel sound being the only difference between the words, except in a couple of cases where a real word that fits doesn't exist.

Students first listen to the vowel sounds and numbers, while watching the hatch on the marker board. Next I move to the back of the room, where students can't lip read and call out words at random from the hatch. Students chorally or individually identify these words and others numerically. Average students are able to identify these words numerically more than 80% of the time within a few weeks, with some practice for a few minutes every day. I also issue students cassettes or CD's to listen, identify and later to practice orally at home or while driving. This accelerates the whole process.

Diphthongs, which are gliding mixtures of two known vowel sounds, are given double number designators; the word "buy" is 6+2, the word "bough" is 6+9, and the word "boy" is 7+2. This is much more accurate than the so-called "long I" of "buy" and the "short I" misnomers. One teacher who was upset with me for that observation said that the "long I" is much longer. She then said "fiiiiiiiiiiiiiiiiiiiiight," to prove her point. I was whelmed. I then demonstrated, using normal speech, that the "long I" in "fight" was, in reality, shorter than the "short I" in "fizz." If these designators had been the more neutral "blue I" and "green I," perhaps there would be less confusion with acoustic realities.

The second vowel hatch includes; 1 heed, 2 hid, 3 hayed, 4 head, 5 had, 6 hod, 7 hawed, 8 hoed, 9 hood, 10 who'd, 11 Hud. These are more difficult for most students to identify, even after they have become reasonably competent with the first hatch. This is because of lengthening, caused by the change to final voiced stop consonants. As the students get better at identification, emphasis begins on production of these sounds in a word context. Early production is discouraged, to avoid reinforcement of existing problems from hearing one's own voice or those of classroom peers. It's easier to sing after one has heard the song. This is particularly true of vowel sounds, because explanation of vowel formation is much more difficult than explanation of consonant production. Once they are perceived and recognized, they can be imitated.

Whenever new words are introduced, in the context of vocabulary lessons, realia identification or picture flashcards, the students are challenged to identify the central vowel sounds numerically. This gives them the opportunity to hear and see new words in a less contrived environment, associate the words with well-programmed numerical insights, and receive constant reinforcement, which is critical to language acquisition. They also begin to acquire familiarity with the various anomalies in English spelling, plus begin to acquire reading ability through recognition, rather than attempting to "decode" written English.

At the same time, students are being equally familiarized with consonant sounds, both systematically and freely, and getting to recognize constant input from the written language as they move on in listening and identification skills. I have observed students becoming more competent readers, using these processes on a consistent basis.

My conclusion is to give students ear training first, oral production training next, and actual emphasis on reading and writing last. I'm not suggesting that the written language shouldn't be at least a part of aural and oral familiarity, throughout their training. However I'm convinced that a delay in serious reading is helpful and that the written language shouldn't be the source of pronunciation training. It works. This egg will hatch!

The preceding article was published in a slightly condensed version in the TexTESOL III Newsletter, August 2004.