What is it?
Why is it needed?
Can English be unspelled?
Can English be written phonetically?
Is this a sneaky attempt at spelling reform?
Why doesn't it use the Latin alphabet?
How was it designed?
What was the initial motivation behind it?
How does one type it?
What is its status?
Who would benefit from it?
How do I learn it?
What is it?
English is a relatively simple language to learn: a simple, analytical grammar with no declension or conjugation tables to memorize, a largely international vocabulary mainly derived from French and Latin, and a sound system that features just a few sounds that are exotic.
In spite of that, around 50% of all native English speakers struggle with learning to read and write, and the levels of functional illiteracy in English-speaking countries are often many times those of other developed nations. Of the billion or so students around the world who are studying English at any given time, only a very small percentage go on to achieve any sort of competency in it.
The reason for this is perfectly simple: English spelling is a nightmare. It was haphazard to begin with, but then it was simply frozen in time sometime in the 17th century. Since then, the way English sounds has evolved almost beyond recognition. English has no orthographic rules, just an assortment of patterns. Which of the many patterns applies in any given case depends on what word it is. Because of this, to learn English one has to separately memorize how each word is written and how it is pronounced. It is not a system that can be taught or learned—only memorized.
Unspell provides an alternative orthography of the English language in which each symbol is equated with exactly one speech sound. To read Unspelled English, all one has to do is drag the finger along, sounding out each symbol. It takes several hours to learn.
Unspell uses a unique symbol set designed to work around dyslexia and various visual and learning impairments. It is very fast to learn, easy to read, and very fast to write. Most importantly, because it is distinct from the Latin alphabet, learning Unspell does not cause confusion with spelled, regular English.
Unspelled English directly embodies the alphabetic principle of phoneme-grapheme correspondence. It builds phonological awareness—understanding of the sound structure of English—in those who learn it. According to research findings from cognitive science, these are the two vital ingredients to learning to read successfully.
Why is it needed?
English spelling poses a huge impediment to learning. Functional illiteracy rates in English-speaking countries run as high as 40%. English orthography was haphazard to start with; then, in the 18th century, the spellings of English words were fixed for all time, mistakes included. Since then, spoken English has evolved, but written English hasn't.
In the U.S., children spend eight years memorizing the spellings of words to achieve basic competence in written English. But eighth-grade-level reading and writing skills are too limited for most practical uses, such as understanding law, science, medicine, technology or commerce. In contrast, schoolchildren in countries where the national language has a regular, consistent orthography achieve adequate literacy in just a year or two, by memorizing a small set of rules, and are then free to learn other things. It is little wonder that many of these countries are surging ahead while English-speaking countries are falling behind. Unspell can level the playing field.
To learn English spelling is to memorize thousands of obsolete spellings of words: “whale” still has an “h” in it; “gnat” still starts with a “g”. Children are forced to cram such non-information into their heads in order to pass tests that allow them to get on in life. Is this really necessary? No, not really!
Can English be unspelled?
Much of the trouble comes from the confusion that reigns in the minds of English speakers as to what is a word. They think that a word is some sequence of Latin letters, plus some way of pronouncing it that, more often than not, has to be learned separately. So, “there,” “they're” and “their” are all different words, even though they all sound the same.
But another way to look at it—one favored by linguists—is that a word is a sequence of phonemes. A phoneme is the smallest phonetic unit in a language that is capable of conveying a distinction in meaning. In a well-designed, rational writing system, a small set of rules determines how a word is written down by mapping phonemes to graphemes, which are letters or combinations of letters. The mapping should be unambiguous: there should be exactly one way to write a word, and there should be exactly one way to pronounce a word. This makes a language as easy to read and write as it is to speak. Many languages follow this prescription quite closely, providing a short and relatively simple path to literacy for nearly everyone who speaks them.
English does not follow this prescription. Take the grapheme “th”: it corresponds to two phonemes: [θ] (the sound in “thing”) and [ð] (the sound in “this”). Therefore, it is not possible to determine how the grapheme “th” is pronounced. Going the other way, take the phoneme [i], which is the sound in “keen,” “bean,” “people,” “fierce” and “creme.” There are many graphemes that correspond to it. Therefore, it is not possible to determine how the phoneme [i] is written. Unfortunately for those who seek to learn to read and write English, these two examples are typical cases rather than exceptions. Not a single letter in the English alphabet is pronounced unambiguously, and not a single sound of the English language is written unambiguously.
Some people are apt to say that having multiple spellings for many words is somehow useful or efficient. They are yet to present any evidence in support of this claim. Indeed, their case is hard to make: there are many more English words that are spelled the same in spite of having multiple meanings. Take the word “date”: it can be either a romantic get-together, a calendar day, or the fruit of the date palm. Nobody is confused by this, because our brains pick out the right meaning automatically and unconsciously, and the sentence “I ate a date” does not give rise to suspicions of cannibalism. The fact that many words have multiple meanings is not a problem to be solved. Even if it were, assigning a different arbitrary sequence of letters to represent their different meanings would not qualify as a solution.
If it were a problem, then books on tape wouldn't exist because nobody would understand them. Either that, or the reader would have to constantly pause to spell out words: “There was a ewe (that's Ee-Double-Yoo-Ee) grazing out in the field” (as opposed to a yew or a you). The fact that English prose, when it is read aloud, can be understood without reference to how words are spelled shows that it is not necessary to have different spellings for different meanings of the same word.
Can English be written phonetically?
There are some languages where the writing system is so simple that all one has to do is learn which symbols (or combinations of symbols) indicate which sounds. In these languages, a spelling bee would instantly bore everyone to distraction, and learning to read and write amounts to learning to listen and to speak while using symbols in place of sounds. These languages include Turkish, Hungarian, Finnish, Estonian, Hawaiian and quite a few others.
There is another group of languages where there is more than one way to write down the same sound, but there is almost always a consistent way to pronounce it. This group includes languages such as Spanish, French and German. This is usually an accident of history: once upon a time, there were more sounds than there are now, and they remain in the orthography as vestiges. They make the language a tiny bit harder to learn to read, and quite a lot harder to learn to write because memorization is required to learn which letter to write down in which case.
There is a third group of languages that has a phonological process called “vowel reduction”: stressed vowels are pronounced fully, but most unstressed vowels are reduced. The reduction rules vary by language and even by dialect. This group of languages includes Portuguese, Russian and English. In English most unstressed vowels decay to something called a “schwa”: an indistinct middle vowel. The question is, how does one write down a schwa? It does have the clumsy symbol 'ə' in the International Phonetic Alphabet, but it is used for phonetic transcription, not any actual writing. Giving it an orthographic identity by assigning an orthographic symbol to it means promoting an automatic phonological process (something everyone does unconsciously as a matter of habit) to the status of an orthographic rule, and that is a very strange thing to do. More importantly, words definitely do contain the non-reduced versions of the vowels: they just aren't being pronounced fully when they aren't stressed, but they are pronounced fully when they are. Take the word “syllable”: we know that the second vowel is an [æ] (the vowel in "cat") because when the stress is shifted, as in the word “syllabic,” we do hear it.
For this reason, Unspell does not directly represent vowel reduction, and unstressed vowels are written down based on how they would sound if they were stressed: the "a" in "syllable" is written as the "a" in "cat" because that's how it sounds when you stress it "syllAble".
Vowel reduction introduces some unavoidable complexity into Unspell, which does not create too much difficulty with reading (after all, vowel reduction is automatic and unconscious) but it does make it harder to write. Luckily, it isn't hard to come up with an interactive unspell-checker that will prompt you to make a choice based on hints.
Is this a sneaky attempt at spelling reform?
Unspell does not change English spelling; it changes how people work with it. English spelling stays the same; it’s simply not used. The goals are: to save time, to eliminate aggravation, to improve spelling accuracy, and to provide access to written English for people who would otherwise not have any.
Why doesn't it use the Latin alphabet?
There are quite a few reasons why Unspell doesn't use the Latin alphabet, but here are the top three:
1. Unspell uses a unique symbol to represent each English phoneme, but the Latin alphabet doesn't have enough letters to accomplish that. Various Latin-based languages use a number of additional characters, such as ç, ñ, å, š, ï and ø, but that approach would make English look like a foreign language, and not in a flattering way.
2. Reusing the Latin alphabet for Unspell would cause interference effects with spelled English (which isn't going away any time soon) by making it difficult to remember which is which. Because the symbols Unspell uses look so different from the Latin alphabet, your English spelling will not deteriorate no matter how much you are exposed to Unspell.
3. Inventing an entirely new set of symbols has allowed Unspell to solve a range of additional problems. Unspell is designed to accommodate special needs students who have trouble with the complicated curved shapes of Latin letters. English language students whose native language is not Latin-based (especially if it is Chinese, Japanese or Korean) find the stroke-based graphics easy to learn. Unspell is easy to write quickly, but, because the shapes are so simple, very hard to write illegibly. It can be written calligraphically using a brush or a pen and stenciled without modification. When embossed, the visually impaired can read it using a fingertip. It is easy to carve and embroider. It can be entered using a touchpad using stroke recognition software. It can be processed by OCR software even when hand-written. Because the shapes are rectilinear, it scales down to very small bitmap sizes without loss of legibility.
How was it designed?
The design process took an entire year. A great deal of experimentation and testing went into creating a system that is easy to read, easy to write, and, most importantly, quick and easy to learn. Each iteration went like this:
The first step was to select a minimum set of phonemes that captures all the key distinctions of spoken English across all the major dialects. The goal was to be neither overly precise (that would make it more difficult to use) nor overly general (leading to incorrect pronunciation and confusion). The number and significance of minimal pairs was taken into account; thus, Unspell distinguishes pin/pen and pull/pool, but not cot/caught.
The second step was to map the phonemes to symbols. This was done in a way that renders the most common phonemes using the fewest strokes. As many symbols as possible were made to resemble the shapes of corresponding Latin letters (making them easier to learn). Vowels and consonants were made to look different at a glance, using the same set of symbols for both but distinguished by their height, thus halving the number of symbols that have to be learned. Paired voiced and unvoiced consonants, such as p/b, f/v, k/g, etc., were distinguished using a single stroke (a voicing mark), further reducing the number of symbols that need to be learned by eight. Overall, similar sounds were assigned to similar symbols.
The overarching principle that was applied throughout the design is the Principle of Least Astonishment: there are no surprises, except for the initial shock of encountering something radically new.
How does one type it?
The Unspell symbols fit perfectly onto the QWERTY keyboard that is standard throughout the English-speaking world, taking up both upper- and lower-case registers. Stressed vowels (which are wider than unstressed ones) and the eight paired voiced consonants are accessed via the shift key. Those who can already touch-type English on a QWERTY keyboard will find that they have very little to learn.
What is its status?
The Unspell primer, Unspeller, is available in North American and World English. Also available is a number of unspelled editions of popular English children's books, such as Alice's Adventures in Wonderland, The Wind in the Willows, Tom Sawyer and Treasure Island.
Who would benefit from it?
• The many adult illiterates and semi-literates who speak English fluently but who, for one reason or another, never learned to read and write well enough to be able to enjoy reading. They can learn to read Unspell without too much trouble. This will allow them to start reading any materials which they can understand when they are read to them.
• Special needs students who, because of various learning difficulties, are unable to learn spelled English. Unspell uses simple, easily distinguished, easy to draw shapes which were specifically designed with them in mind. Unspell avoids the use of flipped, mirrored or rotated shapes that give so much trouble to dyslexics, such as p/q/d/b, u/n, y/h, w/m and s/z.
• Home-schooled kids whose parents are pressed for time and want to get them reading on their own quickly and without a lot of coaching.
• ESOL students who currently have to face the hurdle of learning English spelling even though everything they need to read is available in their native language and the English skills they need are primarily oral.
• Professionals around the world who already know how to work with written English. Many of them have mostly been exposed to written English and have never learned how to pronounce many of the words they use. They then find themselves having to participate in discussions or give talks in English, and embarrass themselves by mispronouncing numerous words. Unspell can make them sound more polished.
• English native speakers who would like to improve how they sound and lose their class or regional accent in order to better integrate into society. By reading in Unspell, they will be absorbing one of the prestige dialects: Standard American English.
• People with impaired vision whose eyes are strained by reading regular English text. The symbols of Unspell are specifically designed to reduce eyestrain.
How do I learn it?
The quickest way to learn Unspell is by playing with an online tool, which is located under resources.
Why is it needed?
Can English be unspelled?
Can English be written phonetically?
Is this a sneaky attempt at spelling reform?
Why doesn't it use the Latin alphabet?
How was it designed?
What was the initial motivation behind it?
How does one type it?
What is its status?
Who would benefit from it?
How do I learn it?
What is it?
English is a relatively simple language to learn: a simple, analytical grammar with no declension or conjugation tables to memorize, a largely international vocabulary mainly derived from French and Latin, and a sound system that features just a few sounds that are exotic.
In spite of that, around 50% of all native English speakers struggle with learning to read and write, and the levels of functional illiteracy in English-speaking countries are often many times those of other developed nations. Of the billion or so students around the world who are studying English at any given time, only a very small percentage go on to achieve any sort of competency in it.
The reason for this is perfectly simple: English spelling is a nightmare. It was haphazard to begin with, but then it was simply frozen in time sometime in the 17th century. Since then, the way English sounds has evolved almost beyond recognition. English has no orthographic rules, just an assortment of patterns. Which of the many patterns applies in any given case depends on what word it is. Because of this, to learn English one has to separately memorize how each word is written and how it is pronounced. It is not a system that can be taught or learned—only memorized.
Unspell provides an alternative orthography of the English language in which each symbol is equated with exactly one speech sound. To read Unspelled English, all one has to do is drag the finger along, sounding out each symbol. It takes several hours to learn.
Unspell uses a unique symbol set designed to work around dyslexia and various visual and learning impairments. It is very fast to learn, easy to read, and very fast to write. Most importantly, because it is distinct from the Latin alphabet, learning Unspell does not cause confusion with spelled, regular English.
Unspelled English directly embodies the alphabetic principle of phoneme-grapheme correspondence. It builds phonological awareness—understanding of the sound structure of English—in those who learn it. According to research findings from cognitive science, these are the two vital ingredients to learning to read successfully.
Why is it needed?
English spelling poses a huge impediment to learning. Functional illiteracy rates in English-speaking countries run as high as 40%. English orthography was haphazard to start with; then, in the 18th century, the spellings of English words were fixed for all time, mistakes included. Since then, spoken English has evolved, but written English hasn't.
In the U.S., children spend eight years memorizing the spellings of words to achieve basic competence in written English. But eighth-grade-level reading and writing skills are too limited for most practical uses, such as understanding law, science, medicine, technology or commerce. In contrast, schoolchildren in countries where the national language has a regular, consistent orthography achieve adequate literacy in just a year or two, by memorizing a small set of rules, and are then free to learn other things. It is little wonder that many of these countries are surging ahead while English-speaking countries are falling behind. Unspell can level the playing field.
To learn English spelling is to memorize thousands of obsolete spellings of words: “whale” still has an “h” in it; “gnat” still starts with a “g”. Children are forced to cram such non-information into their heads in order to pass tests that allow them to get on in life. Is this really necessary? No, not really!
Can English be unspelled?
Much of the trouble comes from the confusion that reigns in the minds of English speakers as to what is a word. They think that a word is some sequence of Latin letters, plus some way of pronouncing it that, more often than not, has to be learned separately. So, “there,” “they're” and “their” are all different words, even though they all sound the same.
But another way to look at it—one favored by linguists—is that a word is a sequence of phonemes. A phoneme is the smallest phonetic unit in a language that is capable of conveying a distinction in meaning. In a well-designed, rational writing system, a small set of rules determines how a word is written down by mapping phonemes to graphemes, which are letters or combinations of letters. The mapping should be unambiguous: there should be exactly one way to write a word, and there should be exactly one way to pronounce a word. This makes a language as easy to read and write as it is to speak. Many languages follow this prescription quite closely, providing a short and relatively simple path to literacy for nearly everyone who speaks them.
English does not follow this prescription. Take the grapheme “th”: it corresponds to two phonemes: [θ] (the sound in “thing”) and [ð] (the sound in “this”). Therefore, it is not possible to determine how the grapheme “th” is pronounced. Going the other way, take the phoneme [i], which is the sound in “keen,” “bean,” “people,” “fierce” and “creme.” There are many graphemes that correspond to it. Therefore, it is not possible to determine how the phoneme [i] is written. Unfortunately for those who seek to learn to read and write English, these two examples are typical cases rather than exceptions. Not a single letter in the English alphabet is pronounced unambiguously, and not a single sound of the English language is written unambiguously.
Some people are apt to say that having multiple spellings for many words is somehow useful or efficient. They are yet to present any evidence in support of this claim. Indeed, their case is hard to make: there are many more English words that are spelled the same in spite of having multiple meanings. Take the word “date”: it can be either a romantic get-together, a calendar day, or the fruit of the date palm. Nobody is confused by this, because our brains pick out the right meaning automatically and unconsciously, and the sentence “I ate a date” does not give rise to suspicions of cannibalism. The fact that many words have multiple meanings is not a problem to be solved. Even if it were, assigning a different arbitrary sequence of letters to represent their different meanings would not qualify as a solution.
If it were a problem, then books on tape wouldn't exist because nobody would understand them. Either that, or the reader would have to constantly pause to spell out words: “There was a ewe (that's Ee-Double-Yoo-Ee) grazing out in the field” (as opposed to a yew or a you). The fact that English prose, when it is read aloud, can be understood without reference to how words are spelled shows that it is not necessary to have different spellings for different meanings of the same word.
Can English be written phonetically?
There are some languages where the writing system is so simple that all one has to do is learn which symbols (or combinations of symbols) indicate which sounds. In these languages, a spelling bee would instantly bore everyone to distraction, and learning to read and write amounts to learning to listen and to speak while using symbols in place of sounds. These languages include Turkish, Hungarian, Finnish, Estonian, Hawaiian and quite a few others.
There is another group of languages where there is more than one way to write down the same sound, but there is almost always a consistent way to pronounce it. This group includes languages such as Spanish, French and German. This is usually an accident of history: once upon a time, there were more sounds than there are now, and they remain in the orthography as vestiges. They make the language a tiny bit harder to learn to read, and quite a lot harder to learn to write because memorization is required to learn which letter to write down in which case.
There is a third group of languages that has a phonological process called “vowel reduction”: stressed vowels are pronounced fully, but most unstressed vowels are reduced. The reduction rules vary by language and even by dialect. This group of languages includes Portuguese, Russian and English. In English most unstressed vowels decay to something called a “schwa”: an indistinct middle vowel. The question is, how does one write down a schwa? It does have the clumsy symbol 'ə' in the International Phonetic Alphabet, but it is used for phonetic transcription, not any actual writing. Giving it an orthographic identity by assigning an orthographic symbol to it means promoting an automatic phonological process (something everyone does unconsciously as a matter of habit) to the status of an orthographic rule, and that is a very strange thing to do. More importantly, words definitely do contain the non-reduced versions of the vowels: they just aren't being pronounced fully when they aren't stressed, but they are pronounced fully when they are. Take the word “syllable”: we know that the second vowel is an [æ] (the vowel in "cat") because when the stress is shifted, as in the word “syllabic,” we do hear it.
For this reason, Unspell does not directly represent vowel reduction, and unstressed vowels are written down based on how they would sound if they were stressed: the "a" in "syllable" is written as the "a" in "cat" because that's how it sounds when you stress it "syllAble".
Vowel reduction introduces some unavoidable complexity into Unspell, which does not create too much difficulty with reading (after all, vowel reduction is automatic and unconscious) but it does make it harder to write. Luckily, it isn't hard to come up with an interactive unspell-checker that will prompt you to make a choice based on hints.
Is this a sneaky attempt at spelling reform?
Unspell does not change English spelling; it changes how people work with it. English spelling stays the same; it’s simply not used. The goals are: to save time, to eliminate aggravation, to improve spelling accuracy, and to provide access to written English for people who would otherwise not have any.
Why doesn't it use the Latin alphabet?
There are quite a few reasons why Unspell doesn't use the Latin alphabet, but here are the top three:
1. Unspell uses a unique symbol to represent each English phoneme, but the Latin alphabet doesn't have enough letters to accomplish that. Various Latin-based languages use a number of additional characters, such as ç, ñ, å, š, ï and ø, but that approach would make English look like a foreign language, and not in a flattering way.
2. Reusing the Latin alphabet for Unspell would cause interference effects with spelled English (which isn't going away any time soon) by making it difficult to remember which is which. Because the symbols Unspell uses look so different from the Latin alphabet, your English spelling will not deteriorate no matter how much you are exposed to Unspell.
3. Inventing an entirely new set of symbols has allowed Unspell to solve a range of additional problems. Unspell is designed to accommodate special needs students who have trouble with the complicated curved shapes of Latin letters. English language students whose native language is not Latin-based (especially if it is Chinese, Japanese or Korean) find the stroke-based graphics easy to learn. Unspell is easy to write quickly, but, because the shapes are so simple, very hard to write illegibly. It can be written calligraphically using a brush or a pen and stenciled without modification. When embossed, the visually impaired can read it using a fingertip. It is easy to carve and embroider. It can be entered using a touchpad using stroke recognition software. It can be processed by OCR software even when hand-written. Because the shapes are rectilinear, it scales down to very small bitmap sizes without loss of legibility.
How was it designed?
The design process took an entire year. A great deal of experimentation and testing went into creating a system that is easy to read, easy to write, and, most importantly, quick and easy to learn. Each iteration went like this:
The first step was to select a minimum set of phonemes that captures all the key distinctions of spoken English across all the major dialects. The goal was to be neither overly precise (that would make it more difficult to use) nor overly general (leading to incorrect pronunciation and confusion). The number and significance of minimal pairs was taken into account; thus, Unspell distinguishes pin/pen and pull/pool, but not cot/caught.
The second step was to map the phonemes to symbols. This was done in a way that renders the most common phonemes using the fewest strokes. As many symbols as possible were made to resemble the shapes of corresponding Latin letters (making them easier to learn). Vowels and consonants were made to look different at a glance, using the same set of symbols for both but distinguished by their height, thus halving the number of symbols that have to be learned. Paired voiced and unvoiced consonants, such as p/b, f/v, k/g, etc., were distinguished using a single stroke (a voicing mark), further reducing the number of symbols that need to be learned by eight. Overall, similar sounds were assigned to similar symbols.
The overarching principle that was applied throughout the design is the Principle of Least Astonishment: there are no surprises, except for the initial shock of encountering something radically new.
How does one type it?
The Unspell symbols fit perfectly onto the QWERTY keyboard that is standard throughout the English-speaking world, taking up both upper- and lower-case registers. Stressed vowels (which are wider than unstressed ones) and the eight paired voiced consonants are accessed via the shift key. Those who can already touch-type English on a QWERTY keyboard will find that they have very little to learn.
What is its status?
The Unspell primer, Unspeller, is available in North American and World English. Also available is a number of unspelled editions of popular English children's books, such as Alice's Adventures in Wonderland, The Wind in the Willows, Tom Sawyer and Treasure Island.
Who would benefit from it?
• The many adult illiterates and semi-literates who speak English fluently but who, for one reason or another, never learned to read and write well enough to be able to enjoy reading. They can learn to read Unspell without too much trouble. This will allow them to start reading any materials which they can understand when they are read to them.
• Special needs students who, because of various learning difficulties, are unable to learn spelled English. Unspell uses simple, easily distinguished, easy to draw shapes which were specifically designed with them in mind. Unspell avoids the use of flipped, mirrored or rotated shapes that give so much trouble to dyslexics, such as p/q/d/b, u/n, y/h, w/m and s/z.
• Home-schooled kids whose parents are pressed for time and want to get them reading on their own quickly and without a lot of coaching.
• ESOL students who currently have to face the hurdle of learning English spelling even though everything they need to read is available in their native language and the English skills they need are primarily oral.
• Professionals around the world who already know how to work with written English. Many of them have mostly been exposed to written English and have never learned how to pronounce many of the words they use. They then find themselves having to participate in discussions or give talks in English, and embarrass themselves by mispronouncing numerous words. Unspell can make them sound more polished.
• English native speakers who would like to improve how they sound and lose their class or regional accent in order to better integrate into society. By reading in Unspell, they will be absorbing one of the prestige dialects: Standard American English.
• People with impaired vision whose eyes are strained by reading regular English text. The symbols of Unspell are specifically designed to reduce eyestrain.
How do I learn it?
The quickest way to learn Unspell is by playing with an online tool, which is located under resources.
