Righting the Mother Tongue: From Olde English to Email, the Tangled Story of English Spelling

by David Wolman

  At any number of links in this fragile chain of operations, things can go awry. The occasional typo, misspelling, misspoken word, stutter, or misreading—those are just hiccups. In other cases, the wiring itself is faulty, and that system error is, in a nutshell, dyslexia. This diagnosis is rendered after alternate explanations—poor teaching, hearing, or vision problems, or perhaps an unsupportive home environment—are all ruled out.

  Dyslexia can involve an inability to synthesize text; a deficit in sounding out letters or in storing or recalling words, sounds or meanings; or some combination of these or other limitations. One person may be able to identify all the letters y-a-c-h-t, but can’t link that string of letters to the sound “yot,” even if he knows perfectly well what a yacht is and can say the word in conversation. Someone else who’s dyslexic may be unable to divvy up the sound units of a word, so that when she hears extraordinary, she can’t build syllables from sounds in the divide-and-conquer way that most of us go about learning to read and write. Other people can identify the individual letters in a series, whether it’s in a real word like altitude or something made up, like snorbitude, but can’t sound them out. And still other people can read a passage, yet can’t think along with the text to digest meaning.

  Here are a few things dyslexia isn’t. It’s not when you write down a phone number and mistakenly flip a couple of digits. It’s not when you say left turn but mean right turn, when you mix up the names of your children, or even when you write an email message riddled with misspellings. Nor is dyslexia simply a mental mirroring that results in backward pictures of letters or words. Dyslexics do sometimes transpose letters, b and d most notably, but this is a small thread in a more complicated web of difficulties. The frequent association with visual mix-ups traces back to the early days of research into impaired reading, when hypotheses about the roots of dyslexia hinged solely on the visual system.

  For the brain, English words come in two types. The first consists of words that can be pieced together by recognizing and producing units of sound. English has forty-four sounds represented by some 1,100 different letter arrangements.2 When linguists talk about these units, called phonemes, they’re not talking about syllables but contrasting sounds, so that “s”, “i” and “t” are the three phonemes that comprise the word sit. (In the word van, the sound unit represented by v is different from the sound unit represented by the f in fan because a v is pronounced along with vocal cord vibrations, while an f is ‘unvoiced.’ For other words, phonemes can be represented by more than one letter, such as the th in throw, which is distinct from the th in these. And down the rabbit hole we go.) When it comes to reading tens of thousands of English words, the ability to identify and produce these units will get you where you need to go. If you can sound them out, you can write them. Trunk, basket, reformer, lift, glom, hospital, kernel, sunset, editor.

  This “phonemic awareness” is what young children use to build their early reading and writing skills. Look at the writing of a child between, say, four and eight, and it’s peppered with constructions that make perfect phonetic sense to a mind that only builds words by sounding out constituents. Kik the bal. I luv mi momi. Saly haz appls—cute stuff that proud parents often post on kitchen refrigerators. When my niece was first beginning to write, she asked me how to spell Dave. I told her to give it a try, repeating it back to her extra clearly before I realized that I was hindering more than helping.

  For schoolchildren nowadays, education has in many ways caught up with linguistics and brain science. Stern spelling instruction, penalization, and ego-injuring spelling books are, for the most part, a thing of the past. The talk in education today is about “creative” or “invented” spelling. The idea is that during early development of writing skills, correct spelling doesn’t matter. Children just need to get comfortable with the manual task of writing out letters and then words. Gradually, the child grows more familiar with the relationships between letters and sounds; that those relationships can be temperamental and inconsistent is a later (and lifelong) lesson. The rest of the theory goes that most children, after encountering a word’s correct spelling enough times, will naturally begin using that form of the word, so there’s little need to worry about younger years’ misspellings. (That is, unless a child is gravely off track, in which case there’s a more obvious barrier to learning that needs to be dealt with).

  Language skills teachers are, and should be, taking a more evolutionary approach to teaching spelling conventions. One frequently used strategy is to give students a list of new spelling words each week. The words are incorporated into the curriculum in various ways, and then there’s a spelling test on Friday. By this time each child has had practice reading, writing and saying the words in varying contexts, so that the quiz at week’s end isn’t so much a memorization test as it is a way to synthesize recently acquired knowledge. The emphasis is on the word and its meanings, and less so on its physical dimensions.

  All of this, however, assumes no major roadblocks to learning. At some point, schoolchildren must face more than sound out-able words because English is also endowed with so many words that fall outside the neat and tidy phonetic orbit. This brings us to the other category of words. English is unlike languages such as Spanish, Dutch, Finnish, Italian, Greek, or German, in which, by and large, what you see is what you get. Yacht is a favorite example among researchers. There’s no way you can sound it out, and although most people don’t remember that day, chances are good that the first time you saw yacht you were confused. In this sense, one scientist told me, “English is the most Chinese of the European languages” because our less regular orthography has so many words that are stored as visual data. (Languages like Chinese use symbols that link to associated sound and meaning, as opposed to using symbols to represent sounds that build words. Japanese is a mashup of the two.)

  Brains process irregular words differently, and they do so beautifully. The trick is in making an image in our minds of what the irregular word looks like, and tagging that stored image to the pronunciation that goes with it. To read English, every time we encounter words like yacht, lieutenant, phalanx, mediaeval, and quiche, there’s a nanosecond of calculation that takes place, as the brain rifles through a database of stored words and corresponding sounds. Reverse this order for a spelling task, and the process is the same; writing and reading are essentially two sides of the same cognitive coin. Think of the last time you paused to sort out a spelling before typing the word. Perhaps you looked skyward to try to picture the word in your mind, or scribbled possible spellings on scrap paper. The same thing is happening in the heads of those spelling bee whiz kids when they look up at the ceiling or trace imaginary letters on their hands. In this sense, spelling is very much about seeing mental pictures.

  The seat of dyslexia is the brain, but it starts with genes. Scientists have known this for decades, thanks in part to studies that reveal strong heritability of reading and writing difficulties, as well as rare cases in which family members show the same unusual deficits, such as the ability to understand the same word when written but not when spoken, or the other way around. By studying dyslexia in different cultures and in twins, researchers are narrowing in on the genetic architecture of the reading system, implicating genes with names like ROBO1 and DYX1C1. As one psychologist explained to me, by identifying the suite of genes needed for proficient reading, we could soon have a way to predict what people will need to help them learn, because we would know precisely which genetic deficiencies trigger which learning disorders.3

  Still, to say that genes are at the root of something like dyslexia doesn’t mean there’s a single gene for it that either you inherit it or you don’t. That is the case for specific genetic syndromes in which a clearly identified letter error among the billions of letter pairs comprising the genome results in a seriously debilitating disease like hemophilia or sickle-cell anemia. When it comes to behavioral conditions like dyslexia, though, genes are implicated as part of
the puzzle, not the single piece that explains the whole. Nevertheless, scientists are confident that genes play a major role predisposing people to a weak link, or links, in the chain of processes required for fluid reading and spelling.4

  All this talk of biology would suggest that the vicissitudes of a particular orthography don’t really matter. One of Uta Frith’s recent studies compared the brains of people reading English and Italian.5 She found that English speakers took more time to begin reading real words and nonsense words—plig, wost, splonk, flumstery—than Italians did, presumably because Italians’ brains can rely almost exclusively on pronunciation rules, whereas English speakers’ brains have to do a bit of additional legwork, determining if a word can be sounded out or needs input from the stored lexicon of irregulars.*

  Frith’s research suggests that whether or not someone is reading or writing in a language replete with orthographic ambiguities, the patterns of brain function in dyslexics appear to be consistent.†6 But dyslexia is more prevalent, or at least more easily detectable, in countries where languages like English and French are spoken. These languages act like magnifying glasses, making the condition more noticeable by way of orthographic pitfalls. On the other hand, a trickier orthographic code means milder forms of dyslexia are easier to spot, which is good news for parents and educators. In countries like Italy, Germany, or Finland, trouble with reading can be more easily masked by sticking close to stable phonemic rules.

  English orthography doesn’t cause or exacerbate dyslexia, yet it does pose challenges where some other orthographies do not. To modern-day spelling reformers, these findings are delicious. Frith and her colleagues did not, and do not, advocate spelling reform. They don’t mind, however, saying that their results support, or at least encourage sympathy for, the theoretical argument that altering complex orthographies might help improve literacy.7

  But before becoming a member of the Simplified Spelling Society or sending your kids to school in Finland, it’s worth considering some strong arguments in support of the English spelling code, arguments that contradict the spelling reformers’ tenet that English orthography is unnecessarily difficult and that learning it wastes valuable time in school. None other than the high priest of linguistics, Noam Chomsky, has said that English spelling has evolved into “a near optimal system.” That sounds bizarre, especially after so many miles and pages spent exploring English spelling’s warts and barbs, but the idea is that an orthography that is perfectly reflective of pronunciation may not be ideal. In isolation, words with silent or extra letters may strike people as inefficient, and at times they are. But in other cases, they help our brains draw dotted lines between words with related meanings, such as sign and signature, condemn and condemnation, dough and doughnut, or bomb and bombard. Another benefit to spellings that aren’t exclusively geared for pronunciation is that, in the case of the suffix-ed for instance, meaning is instantly and uniformly clear, even if the sound of the suffix in the spoken words vary (the d sound at the end of the word showed, versus the t sound at the end of hiked).8

  There is indeed much method to the madness. Upwards of 84 percent of English words are spelled with a predictable lineup of letters—that is, they have logical sound-to-letter correspondence.9 Add to that tools like a mastery of Latin roots and detailed understanding of parts of speech and etymology, and you’re on your way to center stage at the Scripps National Spelling Bee. OK, maybe not, but you’re at least dealing with an orthography in which a great many spellings can be easily deduced.

  On the other hand, there’s no denying that English has traps where some other languages do not, and this can make it extra tough on dyslexics. For someone with a severe form of dyslexia, a seemingly simple spoonerism can be a huge obstacle. With “Helen Mirren,” for example, Frith said she sees people who might respond by saying only melon. “That’s as much of a mountain as they can climb, if they even understand what’s being done with the sounds in the first place.” The same is often true with other tests. Try this. In the following words, identify the emphasized vowel: Predicament. Emphasis. Illuminate. Most people correctly come up with predica-ment, emphasis, and illuminate. But some subjects with dyslexia often have trouble with this task. Others may do fine on this type of sound-based exercise, but have trouble with visual-or memory-dependent tasks, like recalling that the written letter string yacht is pronounced “yot.”

  Compensated dyslexia is essentially a low-grade dyslexia that a person, or rather a person’s brain, has found ways to work around, often without the individual knowing it and with little noticeable effect on everyday life. One scientist told me about a lawyer who was a terribly slow reader and probably a compensated dyslexic. How did he ever get through the huge volumes of reading required in law school? By sharpening his skills as a skimmer, not in the sense of cutting corners, but skimming so that he could more rapidly hone in on the essential components of a text and then be sure to read—and process—those portions with added care.

  When children are first learning to read and spell, a deficit may be more pronounced, hurting their ability to sound out or piece together words. But with age comes more vocabulary, context, and experience, especially if the home environment is one in which reading is encouraged, all of which can help propel children past what may be early-years-only difficulty. Gradually, a picture of a motorcycle is matched to the letter string that’s close to that of motorcycle. Gradually, they come to know that a pint of ice cream is written like a mint candy, even though pint and mint sound different. Gradually, they come to know through conversation what a yacht is. Months or years later, when they encounter a sentence about rich people doing maritime things, their brains match that word with the proper pronunciation. This is how learning plays out over the course of an orthography education.

  A few days after meeting with Frith, I had dinner with an English couple. They are both doctors and one of them teaches at an über-prominent university. (He asked that I not use his name because he worried that his comments were too politically incorrect.) I asked what they thought about compensated dyslexia, before sharing the fact that I may be a case in point. The husband wasted no time: “I think that’s basically an excuse of affluence—a diagnosis of convenience for getting more time on exams,” he said. Dyslexia, although mysterious, is without question a real thing. But ever-more inclusive diagnoses of mild forms of a condition like dyslexia can, in his view, end up obscuring the real cases. If we’re all claiming to be a little OCD, ADHD, dyslexic, and Asperger-y, the people who really need special attention will be drowned out by the overdiagnosed masses.

  His wife added that she sees more upper-class parents expressing concern that their children might have a condition like dyslexia or ADHD. Then again, that could be because wealthier families simply have the resources to look into such things, and not necessarily because they’re desperate for a diagnosis that will explain away their non-valedictory children. The truth is that the line between disorder and natural variation of brain function is always going to be somewhat blurry. After our dinner, though, I was determined to verify or nullify Frith’s back-of-the-envelope diagnosis of my brain with a more detailed analysis of my word-making inner self.

  THE EXAMINER AND I sat at a round wooden table in a beige meeting room. I was in San Luis Obispo, California, at the headquarters of Lindamood-Bell Learning Processes. When teachers or parents see that a child is having difficulty learning, whether it’s a problem with attention, writing, numbers, comprehension, or what have you, they can turn to a resource like Lindamood-Bell for a more thorough assessment of what’s going on in the child’s head.

  My first task was to read aloud a series of nonsense words: raff, gat, twem, adjex. They slowly got tougher: quiles, cyr, phomocher—gobbledygook constructions built for the sole purpose, it seemed, of tripping up someone trying to calmly read aloud, which of course is impossible when you know you’re being tested for a learning disability.

  In the next exercise, I had to
hold letter sequences in my mind. The examiner showed me a flashcard that read “htjlf” for four seconds, then hid it and asked me to repeat what I saw. The drill was significantly easier once we moved from consonant-only cards to nonwords that had vowels, constructions like pregreply and recanciously. Without thinking, I kept muttering the words aloud, as if the action of utterance would give the nonword an in-stereo permanence inside my head.

  Next came the colored blocks. After the tester said a series of sounds, I lined up blocks to represent each one, using the same color if I heard the same sound twice. When she said, “Show me ‘p’, ‘i’, ‘p,’” I lined up three blocks: red, yellow, red. At times it was hard to keep distracting thoughts out of my head, like about how the row of black, yellow, and red blocks made a German flag, but for the most part I did OK on this one.

  I finished the day with two spelling tests. The first was only three words long; I had to write out the following nonsense words: spreft, spligrity, yetterswipper. I got all three correct. The fifty-word spelling test was next, and I figured this larger sample would, finally, expose my deficit. Some of the toughest words were: exaggerate, cacophony, camouflage, malfeasance, belligerent, and pusillanimous. I nailed exaggerate but botched the others, even though I’d scribbled belligerent correctly in the margin when trying to determine what looked right.

  I had to wait an hour for my scores to be processed, so for lunch I walked down the hilled driveway from the Lindamood-Bell offices and crossed Higuara Street to Ben Franklin’s Sandwiches. Awaiting my order, I flipped open the Los Angeles Times and came to a story about researchers who had discovered that the rate at which irregular verbs become regular is mathematically predictable. The past tense of help was once holp, and the past tense of chide was once chode. Like so many once-wackier verbs, they have since become regular, forming the past tense by simply tacking on the-ed suffix (helped and chided).10 Some workhorse verbs, namely be, take, have, and go aren’t going to be regularized anytime soon. But other irregular verbs, like wed, which are used less frequently, might want to get their affairs in order. Irregular verbs are the ancient bones of linguistic rules or trends long since passed, and the only thing compelling us to hold on to them is convention. The same is true of spelling.11