Monday, 9 January 2012

An Essay in Support of Teachers

The Problem

            Recently it has become popular among a number of politicians and philanthropists who consider themselves education reformers, and even the general public to blame teachers for the undeniably low reading scores and lagging performance in America’s schools.  Geoffrey Canada (CEO of The Harlem Children’s Zone, and the major figure in the much-hyped movie Waiting For Superman) claims that once teachers get tenure and are assured of reasonable working conditions and salaries, they stop exerting themselves in the classroom and just let the children teach themselves.  This view of teachers has resulted in calls to abolish teacher tenure and dismiss those teachers whose students earn the lowest reading scores.

            Anyone who has taught in an inner city urban school knows that most teachers work very hard indeed. For the great majority of teachers these negative judgments on their work are discouraging, frustrating and close to libelous.  But the most frustrating part of this scenario is that the real cause of failing schools and low reading scores is neither the teachers, the parents nor the children, but the continuing disarray in the field of American  reading instruction.  It has been many years since American teachers were free to teach children this essential academic skill as they thought best. Today teachers put their job at risk if they deviate from the currently approved method while simultaneously being blamed for that method’s poor results, which in turn influence every aspect of education.

Reading - A Classroom  History

            America’s problems with reading instruction date back to the earliest days of the republic. The first American public school system was established in Massachusetts in the eighteen thirties.  For centuries before that a child learning to read by the alphabetic system had been introduced to reading by being told to “spell out” each word by naming its letters.  The new American school system aimed to be modern and was anxious to improve on this “spelling out” system, which they considered somewhat antiquated and unscientific.

            One problem with the old spelling system was that the mostly rural students who entered these new American schools had never seen letters or print before coming to school, and were mystified by these strange shapes called “letters” and the meaningless syllables that were their “names.”  In an effort to make schooling less alien to these students Horace Mann, the Secretary of America’s first Board of Education, traveled to Europe where he was impressed by a new reading teaching method being used in Prussia.  He brought it back to his newly created public school system in Massachusetts.

            The “word method,” as this new system was called, did not require that children be taught letter names.  Instead children were asked to learn to recognize written words as wholes.  The idea was that at least whole words had some meaning for them.  After a few years teachers who tried this method began complaining that the word method worked all right “until the child tried to read new material.”  Since reciting memorized sentences could hardly be considered reading, the word method was dropped.

            Some teachers returned to the old spelling method.  Others were attracted by a new method becoming popular in Europe in the eighteen fifties.  This second method was called the “phonic method” and asked children to spell words out by their letter sounds instead of their letter names.  While this newly imported approach to reading worked well in the Continental languages of Europe, American teachers and children were having a terrible time with it.  The problem was that in English writing some letters had several different pronunciations many of which did not follow any workable rule.  The beginner spelling a new word by its sounds often had to choose among a number of possible sounds for a letter and might easily choose the wrong one, producing either a non-word or a different word than the one on the page.

            Education publishers recognized the problem and provided letter pronunciation charts, and when that didn’t work, a system of diacritical marks.  Both efforts to solve the problem of multiple possible pronunciations for the same letter proved so complex and impossible to follow that early in the twentieth century the phonic method was abandoned in its turn. Some method was needed to take its place.

            The traditional spelling out method was by now considered so old fashioned that no modern teacher would suggest it.  The only option in sight was the old “word method.”  Its failure a century earlier was remembered only dimly, if at all.  Any lingering doubts about returning to that method were erased when experiments performed by a psychologist named James McKeen Cattell proved that adult readers recognize words instantly as wholes without paying attention to individual letter sounds or names.  Why teach letter sounds or names when adult readers didn’t use them, they reasoned.  Apparently it did not occur to anyone that beginning reading might have entirely different requirements than adult reading.

            The major reading historian of the time, Edmund Burke Huey supported the return to the word method in his influential 1908 book, “The Psychology and Pedagogy of Reading.”  His endorsement apparently removed all doubts.  The word method, now also to be known as “the sight method” or “look-say” was dusted off and refurbished.  For the next fifty years the great majority of American children were taught no letter names or letter sounds but were instead expected to learn to recognize the words in their “Dick and Jane” readers by seeing them over and over again in stacks of flash cards. 

            No one provided any sensible explanation of how children were to transform these groups of nameless, soundless shapes into the twenty thousand words in a third grader’s spoken vocabulary.  In this system, the letters provided no clue to the identity of a word, not even the sound it began with.  They could have been replaced with Arabic script without making the beginning reader’s task any more difficult.

            Children were thus forced to help themselves as best they could.  One child was reputed to have explained that she knew the word on the card in front of her was “house” because of the smudge in the corner of the card.  Not surprisingly, many children still were not reading by third or fourth grade.

            Theories about why children were failing in such large numbers were fanciful.  Some experts explained that reading was basically a psychological process and could not be hurried.  The theory was that children would learn to read “when they were ready.”  Other psychologists claimed that non reading was often the fault of parents who had not “created a sense of security” for the child at home.

            Parents who asked why their children not only couldn’t read, but did not even seem to know the letter names, were told firmly to leave reading to the experts and warned that any parental interference could be dangerous to their child’s learning.  Those children, whose old fashioned parents had taught them their letter names at home, however, were often able to teach themselves to read by taking advantage of the letter pronunciation clues provided by most of the names of the letters in the written word they were trying to identify.  

            Later studies of adult illiteracy found large numbers of marginal and functional illiterates among the adult population that was in school during these years, as famously recorded by Jonathan Kozol’s 1985 book, “Illiterate America.”  This haphazard and hands-off approach to the education of children in elementary school might have gone on forever if it had not been for two things that happened in the nineteen fifties.

            In 1957, Russia sent successfully launched Sputnik, the first man-made satellite to orbit the earth. Russia, our greatest competitor, was beating us in the new frontier of space technology.  For the first time in many decades, America began paying serious attention to its schools, including the elementary years.  

            The second significant reading related happening during that period was the publication, in 1955, of a book called “Why Johnny Can’t Read.”  Its author, Rudolph Flesch, was an Austrian born writer of popular how-to books on a variety of topics.

Flesch had been appalled to find that his twelve-year-old American nephew could not read after spending seven years in school, and had no idea that written letters stood for spoken sounds.  Flesch urged in his book that America waste no time in adopting the reading teaching method that he remembered from his Austrian childhood and which made learning to read so easy for German speaking children.  This was, of course, the phonic method in which children identify written words by “spelling them out” by their letter sounds.

            The topic of Flesch’s book was timely, and the book made its way steadily up the best seller list.  Neither Flesch nor the reading teaching establishment remembered that a century earlier America had imported this same method from continental Europe, tried it and abandoned it as unworkable in English.

            As Flesch described his method it was simple, easy and logical.  All children needed to do to read words was to memorize the sound associated with each letter of the alphabet, then when they ran into a written they needed to identify, they just said the sound for each letter in the word in turn.  They could then be confident that the sounds they said were the same sounds in the same order that they would hear in the spoken word they were trying to identify.  Flesch did acknowledge that the five vowel letters had two sounds each instead of one.  “Every word in English contains a vowel,” he wrote, “so you have to start with teaching the child the letters a, e, i, o, and u, in spite of the fact that each of them spells a long and a short vowel.  The only way to solve the problem is to begin by teaching only the short vowels (which are far more common than the long ones) and postpone the long vowels to a much later stage.”

            The problem was that Flesch was writing from his memory of learning to read as a child in Austria.  The language he was describing was not English but German.  It is in German that each vowel has only two sounds and each of those ten vowel sounds is judged long or short by the reader according to the letters surrounding it in a written word.  

            The English writing system is quite different.  The English consonants serve the reader in the same way the German consonants do, but the vowels do not.  Spoken German uses ten vowel sounds.  Spoken English uses at least nineteen.  In written English, you will not find just ten vowel/letter vowel/sound associations, but over a hundred.  Our nineteen vowel pronunciations follow no rule or system.  Unlike German, our English-written vowels or vowel combinations may stand for many different sounds:  hat, call, many, was, are, etc.  And the same sound may be spelled in a variety of different ways:  fall, long, off, thought, daughter, etc.  This peculiarity of English explains why English speaking children have to study spelling.  Continental children are spared this trial, since all they have to do to spell a word correctly is listen to its spoken sounds and write the letter for each (French being the exception).  

            Teaching English as though it were German does not work.  However, in deference to Flesch, beginning reading in our kindergartens and first grades now focuses almost entirely on training the child to use the five short vowel pronunciations.  Children practice those sounds for months and even years on lists of specially chosen words and in nonsense sentences called “manipulated text,” in which the word choice is limited to those words in which the vowel is pronounced with its short sound.  Examples:  “The cat drops the gas.”  “The glass drops on the cat.”  “The dog cuts the grass.”  (“Explode the Code 2,” page 26)

            It remains unclear what beginners are expected to do with their knowledge of the five short vowel sounds.  The implication seems to be that students should use the short sound to pronounce any vowel alone in a syllable that they encounter.  However, a child who does so will mispronounce well over half of the syllables in any regular story making that story incomprehensible.

            When children have learned the short vowel sounds, the common practice is to teach a rule for when to use the long vowel sounds.  According to phonics lessons, if a word ends in a silent e, the vowel within it should be pronounced with a long sound (take, kite, rope, cute, etc).  However this information, although more explicit, is just about as useless  as the phonics advice about the short vowel sounds since the beginner will constantly  run into such words as “give,” “have,” “move,” “shove,” “none,” etc.,  as well as many cases in which a vowel has its long sound even though there is no silent “e” in sight.

            Consider, for example, the following randomly chosen passage from a Random House “Step Into Reading” book called “My New Boy.”  This story is written in normal language, not manipulated text, and is designed to be read by children in pre-school and grade one.  What happens when a phonics student tries to read it, using the commonly taught pronunciation rules? 

Sample passage:  “I am a little black puppy.  I live in a pet store.  Soon I will have a kid of my own.  Many kids come.  This one pulls my tail.  This one kisses too much.  They are not for me. Here is another kid.  He pats my head.  He says hello.  Woof.”

            In this passage, the phonics-taught child attempting to read it will encounter the following:

            Five different pronunciations for the letter “a:” am, many, are, another, says.

            Six different pronunciations for the letter “e:” kisses, they, me, here,          another, hello.

            Two different pronunciations for the letter  “i:” I, little

            Five different pronunciations for the letter “o:”  store, of, own, one, not  

            Two different pronunciations for the letter “u:”  puppy, pulls 

            Five syllables are spelled with vowel combinations, but since most vowel combinations have at least two pronunciations and some as many as half a dozen, no reliable phonics rule for pronouncing two vowels together in a word exists.  In this passage alone, there are several two-vowel combinations which have at least two different pronunciations:  tail (said); too and soon (good, blood); head (bead, great, idea, ocean); and woof (roof).

            What about the long vowels?  There are eleven vowels pronounced with their long sound in this passage.  They are:  “I,” “a,” “own,” “me,” “here,” “he,” and “hello.” However only one of these long pronunciations is signaled by a final silent e.  Additionally, there are three single syllable words in this passage that do end with a silent “e” but do not have a long vowel sound: “come,” “live” (pronounced in the sentence as rhyming with “give”) and “have.”  The vowels in “have” and “live” are pronounced with the short sound and the “o” in come is pronounced as short u.

            In summary, virtually nothing beginners learn about vowel pronunciation in their phonics lessons has proved true in this passage.  Instead of helping them to learn to read, their instruction has forced them to mispronounce so many of its words that they are left with nothing but incomprehensible gibberish.

            This will happen over and over again to phonics-taught beginners when they try to read anything but those daily doses of manipulated text and word lists that follow the rules they are taught.

            If you are tempted to believe that a few more vowel pronunciation rules would solve the problem, make up your own list of the vowel rules a child would have to keep in mind in order to read the passage quoted above.  Not only would this be difficult to do, but the child would need a whole new set of rules for the next paragraph they tried to read.  Indeed, if you think this paragraph is exceptional, pick at random your own excerpt from any text that an elementary school student might be expected to be able to read. You’ll find that this text poses the same challenges.

            The unworkability of the phonics system in English is a result of the fact that Flesch, speaking with a foreign accent, apparently overlooked the distinctive character of written English.  He recommended a reading teaching system which is no doubt excellent for a child speaking German, but will not work at all in English.  How then did it happen that even in the face of chronic reading failure in America’s schools, phonics remains the reading teaching method of choice in those schools?

            Most responsible for our unwillingness to abandon phonics and look for a better reading teaching system has been a group of reading experts appointed by Congress in 1997 to the so-called National Reading Panel.  Congress asked this group to settle an ongoing argument about reading instruction caused by the sudden popularity of a new reading teaching “philosophy” called “Whole Language” that closely resembled the old word method of the “Run  Dick, Run”  era. During the nineteen eighties many schools adopted whole language, banishing letter names and letter sounds once again from their classrooms.  The argument between partisans of the two essentially worthless methods became so bitter that it came to be known as the Great Reading Wars and was heavily politicized.

            In the year 2000 the National Reading Panel issued a four hundred and fifty page report in which its members documented research that they claimed proved scientifically that systematic intensive phonics (teaching letter pronunciation rules before reading was attempted) improved reading performance more than non-intensive phonics (teaching letter sounds while children were reading) or no phonics (whole language, the word method). 

            However, the only “reading performance” the Panel seemed to be interested in was the children’s ability to pronounce words in manipulated text or in contexts in which the vowel was pronounced with its short sound.  That is, the panel did not emphasize the reading of natural text, but only the ability of phonics instruction to enable the correct pronunciation of those words that follow phonics rules:  a somewhat circular definition.  

            Very little attention was paid to the reading of normal sentences and stories.  One small paragraph in the Panel’s final report notes that in the few studies comparing children’s ability to read and understand “connected text” (the kind that appears in books) researchers found no evidence that systematic intensive phonics was superior to the other methods tested. “Likewise, phonics programs did not produce significant growth in reading comprehension….. Substantial growth occurred in learning to decode regularly spelled words and pseudowords” (National Reading Panel Report p2-116). Phonics, in other words, was good at helping kids learn how to apply phonics rules – but this did NOT translate into helping them learn how to really read. This stunning finding was buried in page after page of statistics extolling phonics.  Congress and most of the reading establishment disregarded this paragraph and remained convinced that systematic intensive phonics had not only been proved scientifically to be superior to all other competing methods, but was the best possible way to teach American children to read.

            In the year 2001, Congress duly passed an extension of the old Elementary and Secondary Education Act that required any school receiving federal funds to teach reading using systematic intensive phonics according to federal guidelines described in publications based on the National Reading Panel Report and known collectively as “Reading First.”  This Act became known as the No Child Left Behind law. 

            Under this law children’s progress in reading is assessed by the DIBELS test in which children are asked to pronounce lists of so-called pseudowords such as, “neg” “gup” “nid” “zan” and “wob,” using the short vowel sound.  A high score is considered proof that the child has mastered decoding and other reading skills. Of course, the focus on these non-words is, in a sense, proof of the failure of phonics to effectively aid actual reading.  If it worked with natural language text, there would be no need to focus on pseudowords – and the ability to correctly pronounce pseudowords does not facilitate actual reading, since it will often lead to the incorrect pronunciation of actual words.  

            The NCLB law is scheduled to run until the year 2014, when all children are expected to be proficient in reading.  However, well before then, there were signs that the law was not working as expected.  In 2008 the U.S Department of Education released the findings of “one of the largest and most rigorous studies” ever undertaken by that agency.  The reading achievement of between 30,000 and 40,000 first, second, and third graders was tested.  Unlike the DIBELS test, measures of reading comprehension were included.  The testers commented:  “Reading achievement was low and did not improve significantly over the course of the three year study.”  The NCLB program was costing a billion dollars a year, but there has been little to show for it beyond children’s ability to pronounce pseudowords and manipulated text. We were teaching our kids to read nonsense – directly thwarting one of the most important elements of beginning reading: the notion that reading is about comprehension. 


The Solution

            Contrary to the stated conviction of Diane Ravitch, the highly respected reading historian, there is a silver bullet for fixing the education system – at least when it comes to reading instruction. No one should be surprised to hear that the solution to our reading teaching problems is ready to hand. After all, every year many thousands of American children teach themselves to read at home before they encounter school reading instruction or know what a vowel – let alone a vowel sound – is.

            As someone who has successfully taught reading to students in schools both public and private, and from a wide range of backgrounds, I can tell you that if you want your students to learn to read you will not mention vowel sounds.  Vowels are useful in our language for signaling syllables and essential for making it possible to identify written words out of context, but they have no active part to play in beginning reading. Students who are exposed to vowel rules and end up learning to read do so despite, and not because of, this instruction. Successful reading instruction should be patterned on the approaches used by speakers of written languages like Hebrew and Farsi that work perfectly well without vowel letters in their alphabet. People reading these written languages rely solely on the consonant sounds and sentence meaning to identify written words. The same approach can be followed in English, as long as students are allowed to follow their natural tendency to use only those letters that are useful to them, skipping over the vowels and relying on the consonants to pick a word that makes sense in the sentence they are reading.

            Be advised however, that this sort of reading instruction cannot begin until the children know the names of the letters of the alphabet, and have had plenty of experience listening to, and most importantly, enjoying, stories told or read aloud to them.  When children play the games that teach them the letter names, they are storing in their minds clues to the sounds of most of the letters of the alphabet, ready to be used later when they need them.  

            Enjoying stories read aloud builds children’s vocabulary, gets them used to responding to decontextualized language by creating a picture in their mind, and teaches them that “reading” must make sense and should lead to enjoyable stories (by contrast, it should be pointed out, reading pseudowords is a miserable and brain-numbing task).

            A variety of learning games that do provide children with interesting and satisfying tasks can help them learn skills that are relevant to reading.  For example, children can acquire the habit of dealing with print from left to right by playing games in which they spell words attached to the picture they name (“Who wants to spell  HALLOWEEN?” during a lesson in October, for example).  These games and many similar ones help children grasp the relationship between spoken and written words.  All these skills; learning the letter names, learning to enjoy stories read aloud (which is a skill that has to be learned and must be taught), and adopting the habit of dealing with print from left to right, are a vital foundation for learning to read.  It is the school’s responsibility to make certain that they are in place before reading instruction begins, even if it means devoting the first half of kindergarten to them.
      As soon as these skills are in place, the letter names can be put to work. Children can learn to match written words with the picture they name by matching the sound in a consonant letter name with the sound at the beginning of a spoken word (they can look at a picture of a tiger, for  instance, and tell you it begins with a "T")  This is a difficult and sophisticated skill, and mastering it marks an important milestone in children’s progress toward learning to read.  Only after all these skills are mastered in contexts that are interesting and rewarding for children, should they be introduced to reading  sentences and stories.
Once children are comfortable with these skills, which they have learned in games and activities without being aware that these had anything to do with learning to read, the teacher can guide them to succeed  in reading The children will use  the consonant sounds  to identify whole words  both by using them  to suggest a suitable word and to eliminate wrong  choices. 
A  “big book “ copy   of a simple  commercially published  children’s book is a good way to introduce the class  to written sentences  in books.  They begin to experience  the fact that when they concentrate on the meaning of the sentence a strange thing happens. Their mind , automatically and subconsciously  proposes a choice of suitable (in terms of both grammar and sense) words for the next word in their sentence.  They also begin  to understand   that  the correct word  will be the one  in which the consonant letters in the word  on the page  match the consonant sounds  in one of their word choices.  Using these  tools children who follow  the word identification method  described above  are often able to  identifiy new words  so quickly  that it seems as though   the words  are identifying themselves.  This natural word identification system  will only work  for the beginner  who approaches reading  with one goal in mind—to find out  what the sentence  is saying.

     The best way to be certain that an aspiring reader  is concentrating on what the sentence is saying is to have the child  read to him- or herself or aloud to the class from a simple commercially published children’s book. This activity is completely voluntary.   The role of the teacher  and the rest of the class  is simply  to listen  and enjoy the story being read to them. To keep the student who is reading aloud from getting stumped  on a word  and losing track  of the meaning of the sentence being read,  the listening  teacher  readily identifies  for him or her any word  that he or she cannnot figure out. The child just spells  the problem word aloud and the teacher pronounces it and the child repeats it and simply reads on. 

     Such an approach  breaks the frustrating oscillation  between  phonics and whole language approaches  that has characterized the history of reading  instruction  in the United States.  Children learning to read this way  use meaning and consonant sounds  together to identify whole words.  They are not taught to build up words from their constituent sounds, instead they use consonants as reliable cues to arrive at a whole word that fits in their sentence.
    When children using this method are deciphering a sentence in a book or a story,  the expectation  of sense and meaning  eliminates from their consideration  any word  that would be senseless or ungrammatical in their sentence, thus leaving so few choices for each succeeding word  that the consonant sounds  are usually enough  to instantly suggest  the correct word  and eliminate a wrong one. The conjecture is confirmed when the word fits sensibly into their story.  
     This teaching  method  poses  the child with no threat of  failure,  and rewards the reader and listending students with an entertaining story.  It is so easy and agreeable  that by the end of the school year  the whole class  has been swept up in it,  regardless of ethnic identity,  socio-economic  level, or previous exposure to reading.
Indeed, the students are so eager to read to the class, that the main challenge facing the teacher will be to find enough time to give everyone who wants a turn a chance to read. If this facilitating  method is continued  in the next two years, by the end of second grade virtually all children will be able to decipher in print any sentence they would  have understood  when it was spoken. They will have successfully  mastered  the tool  that will make  further education  possible  throughout  the grades  and beyond. It must be added that this description is only a thumbnail sketch of a method that has met with great success in a range of schools and with students of a variety of backrounds, and that such a sketch can only give an idea of the major issues involved in teaching according to this method. 

            It is time to abandon the fiction that we are teaching reading successfully in our public schools.  Every attempt to improve reading instruction during the last century has been a failure, including our latest attempt, the No Child Left Behind law.  According to reading scores, a third of the children in our schools are marginally or functionally illiterate and that statistic holds true from primary school to twelfth grade.

            Not surprisingly, our crippled schools have become prey to politicians and venture capitalists who see enormous sums of tax money being spent for a service they are convinced they could perform just as well and for far less money, reaping tremendous profits -- though they have not yet been successful in doing so. The school system has become vulnerable in this way because it has excluded teachers from the educational dialogue, and is completely dominated by the theories, ideas and practices imposed by writers who have never set foot in a classroom (including people like Rudolph  Flesch, Jeanne Chall, Marilyn Adams and the thirteen non-teacher members of the fourteen-member National Reading Panel). English instruction is alone among the educational systems for alphabetic languages in introducing children to reading by giving them nonsense sentences and words in lists to read.  In virtually all other alphabetic language systems children are introduced to reading by means of a simple story written in normal language.  This is impossible in English because students taught using the phonics system are only prepared to read manipulated text – an outcome that Flesch failed to foresee.  Manipulated text is our effort to find something that the phonics student can read. 

            Our school system will only be successful when teachers are returned to their rightful position as judges of which educational practices will work in the classroom.  Once the teachers are freed to teach reading in this rational manner, American schools will return to the norm.  Teachers will succeed, and their students’ achievement can be confidently expected to, at the least, rival the best in the world.

1 comment:

  1. Very interesting article! I was a 5th/6th grade teacher for 3 years, and have now been a reading specialist for 2 years, and have struggled with the amount of students entering intermediate/middle school grade lacking basic reading comprehension skills.

    I did have some questions! I have been trying to look up research regarding teaching emergent readers without vowels - can you recommend resources? Also, this prescription (As you wrote) works with a high level of exposure to spoken words. How does this translate into lower SES environments in which children are probably not even being engaged in conversation, let alone being read to? There have been countless studies indicating that students from poorer backgrounds come to school with deficiencies in vocabulary and print exposure... What does the research say in teaching without vowels for these students??

    Right now I am working with a few 2nd grade emergent readers (at about B/C reading level, up from non-readers this fall!)), and we have had great successes with high frequency words and CVC words, taught in context and in combination with comprehension strategies. I would LOVE to try scrapping the direct instruction of vowels and focus on this method, so any suggestions would be smazing!