In 2018, Michael Coughlin shared an essay with the Boston Mensa Writing Group about “fonetics”. It was Michael’s dream to revamp the English language (and all languages, really) by using new symbols to represent sounds. Part of his obsession was to present his case in a very specific format. He wanted the lines to wrap in exactly the right place. He wanted each character to be presented just the way he saw them.

We helped Michael again and again with formatting ideas, but he just never was happy with the end result. The only format which he was even remotely satisfied with involved a JPG image of his essay, so he could wrap each line exactly where he wanted. Having a JPG-only solution then posed problems for a number of reasons, such as how a visually-impaired person could “read” it.

Michael put this project off until he figured out a solution. Sadly, he never got that opportunity.

Here is what Michael was working on. We can share in his dream that the English language become easier to use. First I will show the JPG versions which he preferred. After that I will have the raw text, which might be harder for some display systems to show properly.

Image version of Michael Coughlin’s essay on fonetics:

Next, here’s the text version of this same essay by Michael Coughlin. Depending on your browser / device, you might not be able to see all the characters properly. For example, Michael used as a description the symbol less-than (<) with I and greater than (>). This combination means “italics” to most web browsers. So work had to be done to prevent this type of issue. This is exactly the issue Michael was having with his project.

* * *

When I was ten years old I discovered a disturbing aspect of the study of the English language. The alphabet had three extra unneeded letters (C, Q, X) that could be replaced with other letters. Worse yet many letters needed by the English language are missing. We say the alphabet has five vowels -- A E I O U -- but the English language uses three times as many.

It took a long time for me to realize that we did not have an English alphabet, but used the Roman alphabet for medieval Latin instead. How many letters were missing?  The more I looked, the more I found, and the more confusing the subject became.

The answer was not to be found in any of my schoolbooks, or the books in the children's library, or the books in the adult public library. It was only when I searched in the largest textbook store in New York City that I found a good answer. I needed a book on the topic of phonetics. I did not need to study this subject to graduate from high school or college, I just wanted it to answer a simple question that any child might ask.

How many letters does a real English alphabet need?

This is something a ten year old can understand, but adults will have trouble with.

So can the computer help with this puzzle? Now we can write all the main languages of the world with the Unicode characters we have on our current operating systems, a big change from the days of green screen CRT monitors. We even have the International Phonetic Alphabet (IPA) characters available with some type fonts. These show up on web pages about languages between / / (or [ ] when being more fussy). Regular spellings can be quoted with < >. With the IPA we can write down any language even if it has never been written before.

Lets try that with English.


The first missing letter I always think of is <TH> which is neither T nor H. We need ð from Icelandic or Old English. But wait, ðere are two <TH>s. So we need θ from Greek. 

English uses stressed or accented syllables. ðis just means important syllables are longer and louder ðan oðers. An unaccented or unstressed syllable frequently uses one of two special vowels, ə or ɪ. ðese are ðe most common letters in English but don’t have ðeir own symbols.


Now we have enough new letters to show a true fact about English ðat everybody knows but nobody can write. ðə definite article <the> is almost always unstressed and pronounced ðə before words ðat start wiθ a consonant but as ðɪ before words ðat start wiθ vowels. It’s a similar change as ðɪ indefinite article <a an> = /ə  ən/. 

Anoðer missing letter is needed for words like ship  -- ʃip so we can distinguish it from words like sip and use ðə right number of letters. ðere is anoðer letter ðat goes wiθ it, ʒ, to distinguish bays from <beige> = /beiʒ/.

Now we have what we need to describe one of ðə strangest parts of conventional English spellings. Words like church and judge. Church starts wiθ a t. Judge starts wiθ a d. Sometimes ðə t or d gets put in at ðɪ end of syllables but not at ðə beginnings of words. Church is really tʃurtʃ and judge is really dʒudʒ. 

But wait, it gets worse. ðese pairs of letters in speech actually only take up ðə space of one letter. So we need to have symbols showing ðey are squashed togeðer. ʧurʧ ʤuʤ. Now we don’t need to use j for ʤ, we can use it as it was originally intended a few centuries ago, a variation of i which was pronounced <ee> in Latin.

Here’s more of ðə story of <the>. It's ðe before consonants, ðɪ before vowels and <thee> /ði/ when its accented, someθing ðat doesn’t get written but does get spoken.
 ðə letters M and N are nearly ðə same. M goes wiθ P and B and N goes wiθ T and D. But ðere is anoðer letter ðat needs to be used wiθ K and G. So we need ŋ for words like <ink> = /ɪŋk/.


Finding ðə new letters we need for all ðɪ English vowels is very difficult. We changed all ðə names for ðə letters a few centuries ago and didn’t change all ðɪ words and spellings. ðə first letter in ðɪ alphabet is ðɪ worst. In Latin it is called “ah”. ðə English name of A actually goes wiθ ðə Latin letter e. ðɪ English name of "I" starts wiθ “ah”. ðə main ways of pronouncing A vary wɪθ where people live. ðə IPA uses different type forms of letters for different sounding vowels. <Fool> becomes /ful/ and <full> becomes /fʊl/. ðose two different vowels need two different symbols. Likewise for O. ðə word <so> = /so/ but <saw> becomes /sɔ/. A word wɪθ only two letters should be written ðat way.

Dealing wiθ A E and I is quite a problem. Would you believe e should always be pronounced <eh>? No? Neiðer did I at first. It's a foreign conspiracy started by ðə Romans and Etruscans. ðɪ oðer version of E gets ðə symbol ɛ so we can write <bait> as /bet/ and <bet> as /bɛt/.

<I> is more of a problem. Its pronounced <ee> in languages ðat don’t fool around changing θings like English. We have a lot of words like <machine> ðat use it ðə old way. A small capital I is used for <pit> = /pɪt/ to distinguish it from <pete> = /pit/. An interesting case is <pity> which could be /pɪti/ or maybe /pɪtɪ/ depending where ðə stress falls. Unlike ə, ɪ can be used in boθ stressed and unstressed syllables.


I’ve saved ðə most confusing vowel for last. <A> has more variants ðan you can shake a stick at. So ðə IPA uses a whole bunch of different A type faces for completely different vowels.  ðɪ old sound from Latin is just ɑ as in  <bar> = /bɑr/, which is a different vowel ðan <bat> = /bæt/, and not ðə same vowel as <ask> = /ask/, ðə IPA doesn’t change ðat word. ðə symbols ɑ, a and æ are always getting mixed up boθ in speech and writing IPA depending where you are from.


One vowel ðat is in a class by itself is ʌ. You use it in words like <but> = /bʌt/.

Two oðer leftover letters are j and w. ðey are related to ðə vowels i and u but get stuck onto someθing else and don’t make whole syllables.


If single vowels are confusing, ðen θink of what to do wiθ diphθongs wich are vowels ðat start off one way and change into anoðer in ðə same syllable. ðɪ English name for I and ðə first personal pronoun is just such a case. It might start off wiθ ɑ, a or æ and finish off wiθ i or ɪ. I like to use ðə leftover j and write I as aj. Its ðə way ðey do it in Polish. Some people use ɑɪ instead. Two different ways of writing ðə same sound. Maybe ɑɪ means <ah> <ee> in two syllables instead. IPA transcriptions are somewat a matter of opinion.

Oðer English diphθongs are ɔj aʊ and oʊ. 


Anoðer troublesome letter is R. It is pronounced or left out in different ways all over ðɪ English speaking world. Only people from Scotland can pronounce R. Everybody else just mumbles it. A word like <nurse> and a word like <word> doesn’t use ðə separate sounds of u o or r and needs anoðer symbol /nɝs/ /wɝd/. For unaccented symbols we need /ɚ/. Where I’m from we almost lose R at ðɪ end of words, but not quite, and not always. It depends on wheðer ðə next word begins wiθ a vowel.


wi naw kæn mɛʒɚ ænd kaʊnt ɔl ðə ɪŋglɪʃ lɛtɚz. 
hir ðe ɑr. 


p b t d f v s z θ ð ʧ ʤ ʃ ʒ k g m n ŋ r l 
 
h j w ɪ i e ɛ ə ɚ ɝ o ɔ ʌ u ʊ æ a ɑ 
  
aj ɔj aʊ oʊ



ðə lɝnəd ʧajnis ɛmpɝɝ kāŋ ʃī dì roʊt -- θɪŋgz ju stʌdid æz ə ʧajəd ɑ ðə lajt əv ðə rajzɪŋ sʌn; ðə stʌdiz ɪn jɔ mæturɪti, ə kandəl.

Michael Coughlin Obituary – Boston Writing Group