The World’s Language

 

Practical High-Level English Expertise for Graduate Students, Researchers, Tech Geeks and Language Lovers from All Linguistic Backgrounds. 

 

Richard Summerbell 

 

All chapters can be downloaded as free pdf files from this website.  

 

Chapter 2 

 

Inglish speling:  the nitemare beginz

 

There are headaches, splitting headaches, migraines – and then there is English spelling.  Mind you, a good headache pill has appeared in recent years – computer spell checkers.  Situations do still arise, though, where people need to spell English and then they are in truble, oops, sorry, trouble. 

 

I am lucky enough to have a photographic memory for spelling, but I still misspelled ‘accommodate’ once when I was 30 years old.  No one is immune to the terror of English spelling.  (By the way, I don’t say this about myself to be a bragging prat.  It’s just that, not being qualified to write this book as an accredited linguistics expert, my only credential lies in being a sort of linguistic idiot-savant.)  In practice, our English words function rather like Chinese characters: each one needs to be memorized as a lump, even though it usually contains internal clues as to how it should sound (Chinese characters often split into a left half and a right half, and the right half often offers a clue to the pronunciation by referencing a simpler character that most people know how to pronounce).   There are a few other languages that tolerate spelling irregularities – Scots Gaelic, classical written Mongolian – but nothing I know of comes close to English. 

 

Many people have tried to reform English spelling, and few have met with any success.  The reason is not just that English speakers are used to the current spellings.  English spelling does actually have its own bizarre logic that makes at least some of it hold together.  If you reformed spelling so that the word ‘sign’ was written differently from the first part of the related word ‘signature’ – for example if you used ‘sine’ and ‘signachur’ (or ‘signatyure’) to spell the words phonetically – then the visible logical and historical links between the words would be lost.  You could drop the ‘g’ sound and change ‘signachur’ to ‘sinachur’ to bring the words back together, but if such things were done consistently, then thousands of words would need to change their pronunciations.  ‘Sine’ as a replacement for ‘sign’ would also be spelled the same as the geometrical function ‘sine,’ as in ‘sine wave,’ creating a new source of possible confusion.   

 

The conservatism of our spelling is even more deeply rooted than these practical thoughts would suggest.  I’ll return to this topic below.  Just look for the key word ‘political.’

 

What needs to be admitted right away is that our spelling drives many people mad, including plenty of native speakers.  The playwright George Bernard Shaw famously pointed out (though he may have encountered the idea in conversation) that if a person mixed up bits of English spelling, a new word ‘ghoti’ could be composed that would be pronounced ‘fish:’ ‘gh’ with an ‘f’ sound as in ‘cough,’ ‘o’ with an ‘i’ sound as in ‘women,’ and ‘ti’ with a ‘sh’ sound as in ‘ignition.’   Later skeptics have noted that ‘gh’ pronounced as ‘f’ never occurs at beginnings of words in English, and that ‘ti’ pronounced as ‘sh’ only occurs if the ‘ti’ is followed by another vowel, preferably ‘o.’  Such sober criticisms miss the point, though, since ‘ghoti’ is as much a satirical joke as it is a criticism. 

 

The ultimate use of humour against the idiocies of English spelling is a long poem called The Chaos, or a Dutchman’s Difficulty with the English Language, written by Dutch author and schoolteacher Gerard Nolst Trenité.  The poem was an appendix to a Dutch-language textbook about improving your English pronunciation, and it was expanded in successive editions, reaching its final 274 lines of hilarious spelling criticism in 1944.  The entire poem is easily found online, though my first encounter with it was in the bestselling book I always get my Sin, het bizarre Engels van Nederlanders (‘the bizarre English of Netherlanders’) by Maarten H. Rijkens, BZZTôH Press, the Hague, 2005.  The poem begins: 

 

Dearest creature in creation

Studying English pronunciation,

I will teach you in my verse

Sounds like corpse, corps, horse and worse.

I will keep you, Susy, busy,

Make your head with heat grow dizzy;

Tear in eye, your dress you'll tear;

Queer, fair seer, hear my prayer.

Pray, console your loving poet,

Make my coat look new, dear, sew it!

Just compare heart, hear and heard,

Dies and diet, lord and word.

Sword and sward, retain and Britain

(Mind the latter how it's written).

Made has not the sound of bade,

Say - said, pay - paid, laid but plaid.

Now I surely will not plague you

With such words as vague and ague,

But be careful how you speak,

Say: gush, bush, steak, streak, break, bleak…

 

and ends: 

 

Pronunciation - think of Psyche! -

Is a paling, stout and spiky.

Won't it make you lose your wits

Writing groats and saying ‘grits’?

It's a dark abyss or tunnel

Strewn with stones like rowlock, gunwale,

Islington, and Isle of Wight,

Housewife, verdict and indict.

Don't you think so, reader, rather,

Saying lather, bather, father?

Finally, which rhymes with enough,

Though, through, bough, cough, hough, sough, tough??

Hiccough has the sound of sup...

My advice is: GIVE IT UP!

 

As you can see, Trenité went out of his way to dig up some obscure words and pronunciations to mix in with the mainstream of English chaos.   The items you may or may not recognize in the quoted segments include ‘hough,’ pronounced ‘hock’ and meaning the ankle joint of an animal (or, as a verb, to cut the hamstring), and ‘sough,’ pronounced ‘sauw,’ meaning ‘to sigh,’ but usually used to refer to the sound of wind in the trees.  ‘Housewife,’ beside being pronounced like ‘house’ + ‘wife,’ can be pronounced ‘hezef’ in some parts of the English speaking world, and ‘rowlock,’ besides being pronounced like ‘row’ + ‘lock,’ can be pronounced ‘rolleck.’  ‘Gunwale,’ the upper edge of the side of a boat, is always pronounced ‘gunnel.’  A ‘paling,’ pronounced ‘pay-ling,’ is a fence made of upright sharpened boards or sticks.   The geographic name ‘Islington,’ despite resembling ‘island,’ is pronounced ‘izlington.’   ‘Isle of Wight’ is pronounced ‘ile of wite,’ that is, almost like ‘aisle of white.’  ‘Hiccough’ is indeed pronounced, and now sometimes spelled, ‘hiccup.’  ‘Ague,’ a fever with chills and sweating, is pronounced ‘ay-gyue.’  ‘Bade,’ a now seldom used past-tense form of ‘to bid,’ is usually pronounced ‘bad.’  (Most people now use ‘bid’ as the past tense of ‘to bid,’ e.g., from the web, “Then I bid on another camera and on Aug 26, I won it for a whopping $51.00.”).   For the rest of the words, if you don’t know them yet, I must leave you on your own to the dictionaries. 

 

Humour aside, there are an amazing number of ways to spell most basic English sounds.  These are all laid out in handy tables these days in the Wikipedia article on ‘English orthography.’  For example, the sound ‘sh’ as in ‘shock’ can be seen in different forms in:  shin, nation, special, mission, expansion, tissue, machine, sugar, conscience, ocean, schmooze, and crescendo.  In this list, you can see many of the factors that account for the variation or chaos in English spelling.  First you have the basic sound ‘sh’ in a Saxon-derived word in ‘shin.’  Then you have a slurred English pronunciation of some words imported from Anglo-Norman French or Latin: ‘nation,’ ‘special,’ ‘mission,’ ‘expansion,’ ‘tissue,’ ‘conscience,’ and ‘ocean.’  Many of these slurred words were already slurred in French, though their Latin equivalents were crisp.  ‘Ti,’ ‘si’ or ‘ce’ letter combinations in French, when followed by another vowel – for example, in ‘direction’ or ‘expansion’ – all tend to yield the same ‘si’ sound.  This sound has a natural tendency to slur towards ‘sh’ in everyday use.  Linguists call this phenomenon ‘yod-coalescence.’ ‘Yod’ is the Hebrew letter (written  ‘י’ ) yielding a short ‘y’ or ‘i’ sound – the type of sound heard in the ‘-tion’ of ‘direction.’  The idea in the academic phrase is that the little ‘i’ sound fuses into the letter before it and changes it, as happens when the ‘ti’ in the Canadian English ‘direction’ becomes ‘sh’ (‘de-rek-shen’).  In French, the little ‘i’ sound tends still to be present, but the letter before it may be slurred anyways.  The French ‘direction’ can sound like ‘dee-hrek-shiohn’ though very proper pronunciation is more like ‘dee-hrek-ssiohn’ (I am crudely showing the throaty Parisian ‘r’ here as ‘hr’ and the nasal ‘o’ vowel indicated by ‘-on’ as ‘ohn.’   For those not familiar with French, it has many sounds not found in English, and in my opinion is one of the most difficult of all languages for English speakers to pronounce.  This makes learning it rather fun.) 

 

The same slurring occurs in French words like ‘tissu’ (‘tissue’) where an ‘s’ sound precedes the French ‘u’ sound.  This ‘u’ sound is not found in English at all.  It is represented as ‘ü’ in German and as ‘y’ in the International Phonetic Alphabet (known to its friends as IPA), and it has a quality similar to the French ‘i’ or the similar English ‘long e’ as in ‘ee.’  One French teacher I had advised forming the mouth to say ‘ee’ and then saying ‘oo’ instead, which is not a bad starting point for learning this vowel.  Anyways, since this ‘ü’ vowel comes from the same part of the mouth as the French ‘i,’ it causes the same slurring of ‘s’ and ‘t’ sounds coming before it.  The slurring, in English, became an all-out ‘sh’ sound in ‘tissue’ and many other words, or a ‘ch’ sound in words like ‘future’ where a ‘t’ was being slurred.  At the same time, English generally imported the ‘ü’ vowel from French as a ‘yu’ sound combination, which is why formal British pronunciation of ‘tissue’ is ‘tis-shyue’ and ‘future’ is ‘fyoochyuh’ – different from the common northern American ‘tishoo’ and ‘fyoochur.’  Another word on the list of ‘sh’ sounds, ‘sugar,’ is also an English slurring of a French ‘sü’ word, ‘sucre,’ which in everyday French may sound like ‘sshyük-hreh’ (where the ‘eh’ at the end is a barely pronounced neutral vowel or shwa.)   In this case, we may have inherited our slur from the Norman version of the word, ‘chucre’ (pronounced ‘shük-reh’) even though our spelling was based on the French word.  Since ‘sugar’ is traditionally considered by protesters to be one of the English spellings that makes no sense at all, you can see the trouble our language got into even when it was fairly consistent about how it imported words from other languages.   I will come back later to ‘sugar,’ since it is a very sweet example of language chaos in action. 

 

The list of ‘sh’ sounds then goes on to include an importation of the ‘official,’ proper French ‘sh’ sound in ‘machine,’ still spelled in the French way as a ‘ch.’  This is followed by a direct import from Yiddish, ‘schmooze,’ ‘to chat with friends or business connections,’ rendered in German phonetics with ‘sch-’ at the beginning.  This German form would have been a transcription for the convenience of non-Jewish Europeans:  the original Yiddish word began with the Hebrew letter ‘shin,’ ש, in the verb ‘shmuesn,’ (שמועסן) to chat (yes, the beginning letter is at the right hand side of the word in Yiddish, just as in Hebrew).  This was actually a ‘real’ Hebrew ‘sh,’ since ‘shmuesn’ was derived from the Hebrew word for rumours or gossip, ‘shemu'ot’ (שמועות, pronounced with a throaty sound in the middle as ‘shemoo/strangled-‘a’/ot’ – do you remember the letter ‘‘ayn’ or ‘strangled-a’ from the last chapter?). 

 

Then there is a direct import from Italian, ‘crescendo,’ still wearing its original Italian spelling.  The English pronunciation is very close to the Italian, with the ‘scen’ in the middle pronounced ‘shen.’  Here you see English as a sticky language in action, picking up not only interesting words but also different spellings for the same sound.

 

Now, you might ask yourself, “when Anglo-Norman words like ‘direction’ and ‘special’ came into English, why didn’t the speakers automatically correct the spelling to ‘direcshon’ and ‘speshal?’  After all, English speakers were not afraid to call the city of Firenze (‘Florentia’ in Latin) ‘Florence,’ Roma ‘Rome,’ Paris (‘Pahree’) ‘Paris’ (‘Pairiss’), Moskva ‘Moscow’ and Praha (Praga in Latin) ‘Prague.’  Anglicizing foreign words was, to some extent, routine.  Conventional wisdom on this question states that English tended to preserve historical spelling.  Scholars sometimes even tried to restore verbal history, as was done after the Middle English ‘dette’ was given a new silent ‘b’ to become ‘debt’ in modern English.  This was to make it more like its Latin ancestor ‘debitam,’ ‘thing owed.’  What is left out in this ‘historical links’ viewpoint is the question of why English became so stuck on historical spelling.  When you look at what may have happened as Anglo-Saxon and Anglo-Norman societies delicately began to fuse together, a question arises:  who was going to assume the authority to change Anglo-Norman spellings to Anglo-Saxon or vice-versa?   Such actions would involve not only changing spellings, but also saying, in effect, that one social group could have its familiar verbal forms erased within the English language while the other took over.  The Anglo-Normans had their own superior language to shepherd, and had no strong reason to dig into Middle English, regularizing spelling by changing ‘shin’ to ‘chinne,’ ‘shone’ to ‘chône,’ ‘should’ to ‘choede’ and so on.   Meanwhile, the Anglo-Saxons had every incentive to appear sophisticated and able to deal with the French of their rulers and the Latin of the educated people.  If they went to school, they studied French, not English, regardless of their background.  Before long, they began adding their grammatical tags to French words, as seen in the last chapter with the participle ‘ycointised.’  On the other hand, making large numbers of spellings conform to an ethnically biased rule may have seemed arrogant.  Changing French words to Saxon forms may even have seemed rebellious. 

 

The upshot of all this is a ringing conclusion.  I’m going to make it number one in a series of bold statements about English spelling. 

 

Bold Statement (BS) number 1.  English is the only language in the world where the spelling mainly arose out of ‘political correctness.’ 

 

Middle and early modern English spellings represent a ‘politically correct’ diplomatic compromise between two nations with incompatible spelling conventions.  Compromise overruled consistency.  Once this compromise was made, it made no difference if a few words with spelling conventions from other languages were added in, as long as the words were pronounceable by Anglo-Saxons. 

 

I’d suggest, in fact, that there were two levels to the correctness.  There was the national level, where the preservation of two types of spelling was, in effect, a negotiation about the relative equality of the Saxon and Norman people.  True, the French were placed on a slightly higher plane, but the English held their ground and kept their own words.  And then there was the schoolteacher level of correctness.  Changes in pronunciation, you see, probably arose quite unofficially in both the Norman and the Saxon parts of English.  No educated people ever officially allowed Anglo-Norman ‘division’ to mutate from ‘dee-vee-syohn’ to ‘di-vi-zhun.’  Nor was there any moment when Anglo-Saxon ‘laugh’ formally became ‘laff’ instead of ‘laa-uekh,’ as it was in early Middle English.  These mispronunciations were, in effect, bad behaviour, at least at first.  It didn’t matter so much when the lowly Saxons mispronounced their own words, but mispronunciations of the French words were never officially accepted.  The British English accent today that attempts to be the most high class in sound is called ‘conservative received pronunciation (CRP);’ this accent features ‘yod retention,’ the retaining of the little ‘i’ sound in the ‘-ion’ of words like ‘faction.’  Also in classic CRP, ‘suit’ is pronounced ‘syute,’ ‘duke’ is ‘dyuke,’ ‘visual’ is ‘vizyooal’ and ‘tune’ is ‘tyune.’  Other than ‘tune,’ which is a mysterious variant of the word ‘tone,’ these words are all directly extracted from Anglo-Norman words. 

 

CRP is only spoken by circa 2% of the British population, making it a very rare prestige accent.  Nonetheless, it carries ancient traditions forward.  People who de-French their pronunciation, who give in to ‘divizhun,’ ‘fakshun,’ ‘sute,’ ‘vizhual,’ and especially ‘dook’ and ‘toon’ (to rhyme with ‘spook’ and ‘moon’), show themselves to be relatively low class.   Yes, even WITHIN individual English words, the more Anglo-Norman-French the pronunciation, the higher class the word.  Now, how can spelling reform come along and turn ‘suit’ into ‘sute’ rather than ‘syute?’  The pronunciation ‘sute’ is wearing a T-shirt, not a suit.  To reform English this way would be to convert us into a linguistic low-rent district.  Or, at least, so we are told by the guardians of political correctness in English spelling.

 

When you go to the trouble of learning the hundreds of oddly spelled English words, you are thus participating in the national success story of England, the two people who became one.  You are buying into the democratic system that uses the Language of Commons, Anglo-Saxon modern English, as its popular parliament (a word that literally means ‘a speaking,’ from the French ‘parler,’ ‘to speak’) and the Language of Lords, Anglo-Norman modern English, as its upper house of speaking. 

 

It doesn’t help you learn the language.   But it isn’t every day that conquering and conquered peoples avoid fighting wars with each other forever afterward.  So when you enter into the diplomatic peace represented by English spelling, you can be very proud of the human race.   There is hope for us all (Anglo-Saxon).  Potential for cosmopolitan amity exists (Anglo-Norman). 

 

Incidentally, when one thinks of the history of Britain and its peoples, eventually one’s thoughts drift northward and encounter the Scots.  I have said very little about them so far, even though I am partly of Scottish ancestry and spent some of my pre-teen and teen years playing the bagpipes in a kilted marching band.  The reason they have had such a low profile in this book so far is that their language, Scots, is virtually another topic.  I am one of many people who grew up with the impression that the original language of the Scots was Erse Gaelic, and that the distinctive Scots accent and vocabulary in English had arisen when Gaelic-speaking people took up speaking English.  It turns out my impression was completely wrong.  The Scots language, in the form that is more or less understandable by English speaking people everywhere, is actually a separate language related to English.  It’s a direct descendant of the Anglian English of the ancient Northumbrian region, and has very little Gaelic influence.  Gaelic speakers mostly lived in northern and western Scotland, and were mostly ethnically cleansed out of Scotland in the 18^th and 19^th centuries.   I suspect my ancestors were in that group.  Scotland now has fewer than 60,000 Gaelic speakers in a population of 5.1 million.   The Scots language, by contrast, has 1.5 million speakers in Scotland and 30,000 in Northern Ireland.  It originated in the Scottish lowlands and nearby areas of England, and it has persisted through the centuries despite often being treated as a quaint and uncivilized dialect.  It retains the verb ‘ken,’ to know, and many other features of its ancestral tongue that were lost in forms of English that evolved further south.  As with Galician (the Portuguese-like language of the northwestern province of Spain) in contrast to standard Portuguese, one could argue today about whether Scots is a dialect or a language, especially since it varies from place to place and blurs with standard English near its boundary areas.  In this book, I am going to treat it as a language in its own right.  It is not, however, a language I speak myself, though I have heard and appreciated a lot of it.  The fine poetry of Robert Burns is written in it, and I recommend that to you if you have not encountered it yet. 

    

I’ll give you some examples of contemporary Scots from the official website of the Scottish Parliament.  You will see immediately why, when I am talking about English spelling, I am not talking about Scots!   

 

Ye are walcome tae visit the Pairlament tae hae a keek roon or find oot aboot whit wey the Pairlament warks.  (You are welcome to visit the Parliament to have a look around or find out about how the Parliament works.) 

 

We want tae mak siccar that awbody is able tae visit the Scottish Pairlament biggin.  (We want to make sure that everyone is able to visit the Scottish Parliament building.) 

 

Tae mak Scotland wark for ye, the Scottish Pairlament needs tae ken yer views on thir issues.  (To make Scotland work for you, the Scottish Parliament needs to know your views on their issues.) 

 

Scrieve til yer MSPs or a committee.  (Write to your Members of the Scottish Parliament or a committee). 

 

I chose these particular examples because they have Anglian words in them that are long lost from English English but are nearly identical to words found in Dutch.  The first sentence has ‘keek,’ similar to the Dutch verb ‘kijken,’ to look.  The second has ‘siccar,’ similar to Dutch ‘zeker,’ meaning ‘sure.’  The third has ‘ken,’ similar to Dutch ‘kennen,’ ‘to know in the sense of to be acquainted with’ – though here the Scots word is used in a way that would be translated as ‘weten,’ the other ‘to know’ verb, in Dutch.  Finally, the last sentence has ‘scrieve,’ to write, similar to Dutch ‘schrijven’ with the same meaning.  I hope this convinces you, at least on an informal level, that Scots is really another language entirely. 

 

One more interesting bit that I will refer to again later: 

 

The committee micht:

• • bid ye tae come in an speak til the committee
  • speir anither committee or the Scottish Executive tae luik intil yer petition

 

(The committee might invite you to come and speak to the committee, or ask another committee or the Scottish Executive to look into your petition.)

 

This passage contains the word ‘micht,’ meaning ‘might.’  This is the part I’ll return to later.   Also, though, notice the verb ‘speir,’ meaning ‘ask,’ and notice ‘til’ and ‘intil’ for ‘to’ and ‘into.’  ‘Speir’ and ‘til’ are both words derived from Norse; compare the Icelandic ‘til,’ meaning ‘to,’ and the verb ‘spyrja,’ (pronounced ‘spear-ya’) to ask, as seen in ‘spyrja spurningar,’ to ask a question.   Yes, there are separate bits of Viking talk in Scots that English English knows nothing of.  Another is the word ‘biggin’ for ‘building’ in the second passage quoted above.  Actually, English does have the same Norse word in it that gave rise to ‘speir:’ we have it as the verb ‘to spur,’ meaning ‘to stimulate action,’ as well as the noun ‘spur’ for the objects horse riders attach to their boots.  Clearly, though, the word went in two completely different directions in southern English and Scots. 

 

So, OK, let’s go back to English English, from now on known as English. 

 

You may have noticed by now that I occasionally give the pronunciation of a word, and there seems to be almost no system to how I do it.  There is a good reason for this.  There are several sounds in English that can’t be reliably written in a way that makes sure readers can understand what’s expected.  For example, there is no way to distinguish ‘th’ as in ‘them’ from ‘th’ as in ‘thin,’ and, as I mentioned before, there is no way to represent the ‘s’ sound in ‘confusion’ without using the non-English combination ‘zh.’  The worst problem, though, is that the vowel heard in ‘took’ and ‘put’ cannot be written in any English-language way that can’t also be pronounced as another sound.  Let’s say you had a new word, ‘tut,’ that rhymed with ‘put’ and ‘soot.’  If you wrote the pronunciation clue as ‘tut,’ you’d have a spelling that could rhyme with ‘but;’ if you gave it as ‘toot,’ it would become the word for a car horn sound, rhyming with ‘boot.’  If you gave it as ‘teut’ or ‘teuht’ or ‘tuht,’ your intention could still easily be misunderstood (‘tyoot,’ ‘tyuht,’ ‘tut’).  Clarity is impossible unless you actually state what word ‘tut’ rhymes with.  I promised earlier to come back to the word ‘sugar.’  Like ‘tut,’ this word can’t be spelled in a way that shows how it sounds unless you invent new symbols for English writing.  The closest you can come is ‘shuhgur,’ which might be taken to rhyme with ‘bugger.’ 

 

Long ‘u’ as in ‘flute’ can also be a problem, especially when it is trapped between consonants.  If I was to give you the pronunciation of the Portuguese surname Da Cunha, I could write ‘da koonya,’ but since ‘oo’ can also be the sound in ‘took,’ that is not completely clear.  ‘Da kune-ya’ is also possible, but people from many linguistic backgrounds might be tempted to pronounce the ‘e.’ People speaking CRP might pronounce this spelling ‘da kyoon ya.’

 

There is actually a single, simple factor that is a profound influence on the tortured spelling of our language.  And here it is. 

 

Bold Statement (BS) number 2.   English is written in the wrong alphabet. 

 

Old English wasn’t written in the letters we use today, at least, not in its early period.  Our current alphabet stems from the decision of religious scholars to adopt the Roman alphabet for use in writing English.  The Roman alphabet, not surprisingly, was designed to convey the sounds of just one language:  Latin.  Latin had no ‘ch,’ no ‘sh,’ no ‘th’s’ of any kind, etc., etc.  I haven’t checked this in great detail, but I doubt that the Roman alphabet as used by Romans can convey the sounds of any language other than Latin.  As soon as you get into any other language, some improvising needs to be done to invent ways to represent the sounds.  Take the sound ‘sh,’ which is ‘sh’ in English and Albanian, ‘ch’ in French, ‘sz’ in Polish, ‘s’ in Hungarian (normal ‘s’ in Hungarian is written ‘sz’), ‘sch’ in German and Romansh, ‘sj’ in Norwegian, ‘ş’ in Romanian, Turkish and Gagauz (a language related to Turkish, spoken in southern Moldova), ‘š’ in Czech, Slovak, Slovene, Croatian, Latvian and Lithuanian, ‘sj’ or ‘sch’ or ‘ch’ in foreign words imported into Dutch (‘sch’ in Dutch is normally pronounced ‘skh’), ‘sc’ in Italian, and ‘x’ in Portuguese, Basque, Maltese, Mayan, Zapotec and Mixtec.  There’s no consistency whatsoever about the ‘sh’ sound among the languages that use the Roman script, except that ‘s’ tends to be involved.  It all reminds me of a colourful quotation from the Christian bible: “No one puts new wine into old skins, otherwise the new wine will burst the skins, and it will be poured out, and the skins will be destroyed.”  I suppose it’s very convenient that most of Europe writes using the same script, but on the other hand, almost no one ever learns how the letters are pronounced in all those languages.  People who can sound out all the European languages written in our familiar alphabet are one in a million.  The letters are not used in anything remotely resembling a standard or uniform way.  I recently spent some time in Istanbul and was delighted by all the French words that have been incorporated into the names of Turkish shops and tourist sites.   Most of these words are in English as well, with similar pronunciation.  How many can you recognize based on the phonetic Turkish script?   Şoför, kuaför, şale, büfe, otobüs, şarküteri, and, with some Turkish word endings, avukatlık bürosu (Answer:  chauffeur = driver, coiffure = hairstyling, chalet = small palace, buffet = ready-to-eat food place, autobus = bus, charcuterie = delicatessen or pork-butcher [actually often a bottle-shop in Turkey since alcohol and pork are both un-islamic], bureau des avocats = lawyers’ office or, translating French literally, ‘advocates’ bureau’).  Your final Turkish-French-English test: şövalye (answer:  chevalier = knight) and, most difficult, bone (bonnet = swimming cap).  The script is familiar, but sometimes that’s not much help in recognition!  

 

The religious scholars should have known they were doing a very rash thing by pouring all sorts of new languages into the old skin of Roman script.  English is the worst-case scenario:  the new language violently burst the skins it had been poured into and sprayed spelling chaos all over its vocabulary.  That was partly because it was actually a blend of two languages, making it twice as bubbly as any of the other languages. 

 

The script Old English was first written in is called Futhorc.  Essentially the same script was also used by speakers of the Old Frisian language in what is now the Netherlands.  The script is related to the so-called ‘runic’ scripts used by Vikings and some Germanic groups.  Like the Viking runes, it consisted mostly of letters with a stick-like appearance, easier to carve into wood than curvy letters.  There were from 24 - 33 letters, depending on where and when it was being used.  The script was based on an earlier Old Norse runic script called ‘Elder Futhorc’ but added four to five extra letters to represent Anglo-Saxon vowel sounds.  The Anglo-Saxon version was not a phonetically ‘perfect’ script, since it still failed to distinguish some sounds, but in general, it was well fitted to Old English.  The number of texts known to be written in it is rather small, but it did loan a particularly vital letter to the later writing of Old English in Roman script: þ, called ‘thorn.’ In Old English written in Roman script, this letter was partnered with a second letter, ð, called ‘eth;’ these two letters were used interchangeably.  They indicated the unvoiced ‘th’ sound, as in ‘thin,’ and the voiced ‘th,’ as in ‘them.’ The sound was usually unvoiced (‘th’ as in ‘thin’) except when it fell between vowels, which made it become voiced, as in the modern ‘rather.’ 

 

In later versions of English, these ‘th’ letters were squeezed out of the alphabet.  They are still used in modern Icelandic, where þ is always the unvoiced ‘th’ and ‘ð’ is always the voiced one.  It would be very handy for learners, at least, if we still had these letters in English.  If you look closely at the names ‘Futhorc’ and ‘eth,’ you will soon realize that you don’t know how to pronounce them.   ‘Futhorc’ could have either a voiced or unvoiced ‘th’ in it, and it also could be pronounced ‘futt-horc.’  ‘Eth’ could be as in ‘ethnic,’ or as in ‘tether.’  If it were a word imported into English from Hindi or Sanskrit, it could be pronounced ‘et.’ On the other hand, when you know that ‘futhorc’ is fuþorc’ and ‘eth’ is ‘eð,’ using the Icelandic system, then you can see without doubt which way the pronunciation should go.  There’s no fuss, no boðer, when you use boþ letters. 

 

These days, almost all languages can be written in the IPA, the international phonetic alphabet, if you want to see exact pronunciation.  For conventional literary purposes, though, the IPA is far too sensitive, since it reflects every different sound made in every different accent and dialect.  (See the handy Wikipedia ‘IPA chart for English dialects’).  English is very diverse, so if IPA were used to represent it, there would be dozens of spellings for every word.  Even the way one person said a word would take on different spellings, for example, if the person adjusted his or her accent when speaking to a foreign-born relative.  Another factor making IPA too specific is that, curiously, we have strong differences among some sounds in English that we pay no attention to at all in our writing.  For example, the ‘p’ in ‘park’ and the ‘p’ in ‘spark’ are very different.  The first one includes a strong release of air and is called an ‘aspirated p,’ (meaning ‘breathy p’) while the second releases minimal air and is an ‘unaspirated p’ (‘non-breathy p’).  Insert the ‘p’ from ‘park’ into ‘spark’ and you would get a word sounding like ‘sp’hark.’  Take the ‘s’ directly off ‘spark’ and you would get a word that would sound like a slightly mispronounced ‘bark’ to most listeners.   And, indeed, many languages distinguish these letters, particularly the East Asian languages.  Mandarin Chinese pinyin script represents ‘aspirated p’ as ‘p’ and ‘unaspirated p’ as ‘b.’  Unlike a real ‘b,’ ‘unaspirated p’ involves no use of the vocal cords.  If we represent the unaspirated ‘p’ for a moment as (p), the real pronunciation of the Chinese capital Beijing (北京; pinyin with tone marks Běijīng) can be clarified.  It is approximately ‘(p)ěidtzhīng’ and is much crisper sounding than the anglicized version. 

 

Numerous languages, such as Dutch and Inuktitut, only use the unaspirated ‘p.’  With Dutch, this yields some interesting moments when English words are adopted into local speech.  There is a very popular music festival each year in the Netherlands called Pinkpop; the first part of the name comes from ‘Pinksteren,’ the Dutch name for the Pentecost holiday, and the second from the English ‘pop.’  It is quite interesting to hear the festival’s name, which looks so predictable to English readers, referred to consistently in Dutch English as something sounding much like ‘de binkbobp faystival.’  And remember, Dutch people tend to speak English very well. 

 

Besides breathy and non-breathy ‘p,’ English also contains a breathy and non-breathy ‘t’ (as seen with ‘ton,’ where ‘t’ is breathy, and ‘stun,’ where it’s non-breathy) and ‘c’ or ‘k’ (‘can,’ breathy, vs. ‘scan,’ non-breathy).  Again, we don’t distinguish the members of these pairs of sounds, but each sound has its own Pinyin letter in Mandarin Chinese.  In the order I have mentioned them, they are ‘t,’ ‘d,’ ‘k’ and ‘g.’  Our regular, voiced ‘d’ and ‘g’ don’t occur in Mandarin at all.  (When these sounds need to be shown in the Pinyin script the Chinese government has sponsored for the Burmese-like Yi or Nuosu languages of Sichuan and Yunnan provinces, they are written ‘dd’ and ‘gg.’)  So now you know how Guangdong (廣東, 广东, Guǎngdōng) province of China came to be called ‘Canton’ in English:  early listeners, who happened to be Portuguese commercial adventurers, quite reasonably heard the unaspirated ‘k’ at the beginning of ‘Guang’ (‘broad’) as an oddly spoken ‘c’ or ‘k’ sound, and also took the unaspirated ‘t’ at the beginning of ‘dong’ (‘east’) to be a ‘t.’  Their ‘Portuguizing’ of the word omitted a few inconvenient sounds, such as the ‘ng’ sounds, and thus a typical Western-style name was formed. 

 

Now let’s look at another unnoticed difference in English.  I have to begin with a short explanation.  Our vowel sounds are conventionally divided into ‘long’ sounds, like the ‘a’ in ‘plane,’ and ‘short’ sounds, like the ‘a’ in ‘plan.’  Some linguists nowadays refer to this distinction as ‘tense’ for the long ‘a’ and ‘lax’ for the short ‘a.’  The reason is that English differs from many languages, like Dutch and Cantonese, that really do distinguish long and short versions of the same sound as long and short vowels.  Good examples are sàan (山, ‘mountain’) vs. sàn (新, ‘new’) in Cantonese – sounding rather like ‘sahn’ and ‘sun’ – and maan (moon) vs. man (man or person) in Dutch – sounding almost like ‘mahn’ and ‘mun’.   English long and short vowels are quite distinct in sound quality:  the spoken word ‘plane’ is not just a dragged-out version of ‘plan.’  Interestingly, though, even though we note the distinction between the ‘a’ in ‘mate’ and the ‘a’ in ‘mat,’ we don’t notice the distinction between the ‘a’ in ‘mate’ and the ‘a’ in ‘mare.’  We would typically call each of these last two sounds ‘long a.’  Yet, in northern North American pronunciation, if we were to substitute the ‘a’ in ‘mate’ into ‘mare,’ we’d get a word sounding like ‘mayyr.’  IPA distinguishes the sounds as ‘eɪ’ for the sound in northern North American or British CRP ‘mate’ and ‘ɛ’ for the sound in ‘mare.’  Ordinary speakers notice no distinction.

 

Here’s another interesting item related to vowels.  One of the baffling experiences involved in being a Canadian is that every once in awhile, curious Americans ask you to say the words ‘out’ and ‘about.’  Then they say, with a gleam of discovery in their eye, “when you people say the words ‘out’ and ‘about’, you always say ‘oot’ and ‘aboot’, don’t you?”  

 

“No,” we reply, “we don’t say ‘oot’ and ‘aboot’ at all, we say ‘out’ and ‘about.’  “You just said it!” cry the Americans excitedly, “oot and aboot!”  “Did not,” say the Canadians, “only Scots say that.”  But we can never convince them.  The first person to confront me with this was a newspaper editor, and after conversing with me, he could still have written a strong editorial about how Canadians all say “oot and aboot,” even if they then, for strange Canadian reasons, deny it.

 

This conversation makes more sense if you can see the key words written in IPA symbols. “When you people say the words ‘out’ and ‘about’ (IPA ‘aʊt’ and ‘ə’baʊt,’ rhyming with trout, but the American pronunciation sounds to us like ‘auwt’ and ‘abauwt’), you always say ‘oot’ and ‘aboot’ (IPA ‘ut’ and ‘əbut,’ rhyming with ‘boot’), don’t you?”

 

“No,” we reply, “we don’t say ‘ut’ and ‘əbut’ at all, we say ‘ʌʊt’ and ‘əbʌʊt.’”  Now, this is a little hard to explain in print, but here goes: ‘ʌʊt’ sounds similar to the American ‘aʊt’ but it is shorter, and starts with a sound, written in IPA as ‘ʌ,’ that is like ‘u’ in ‘cut.’ The American aʊt is longer and starts with something more like the ‘a’ in ‘father.’ Either the ‘ʌ’ or the ‘a’ combines with a ‘ʊ’ (the vowel in ‘foot’) to make the diphthong (two-part vowel sound) that we conventionally write as ‘ou.’ 

 

Amazingly, unless our American questioners have trained as linguists or have a very sensitive ear, they can’t distinguish our hockey-playing, Nanaimo-bar-eating ‘ʌʊt’ and ‘ə’bʌʊt’ from the kilted, bagpipe playing ‘ut’ and ‘ə’but.’   They come away thinking we haven’t been telling them the trout. (I’ll sign your book copy if you ‘got’ that nasty pun instantly – ‘truth’ is pronounced ‘troot’ in some dialects like New York City English.) 

 

Now, just in case you’ve forgotten the real ‘oot’ and ‘aboot’ from the Scots sentences above, here’s another fun piece from the Scottish Parliament’s language policy, written in Scots:  “This policy is wrocht tae caw doun the leid barriers that micht stap folk that bide in Scotland but that arena fluent in English frae takin pairt in the darg o the Pairlament or findin oot mair aboot the Pairlament’s ongauns.”  (“This policy is wrought [made] to knock down the language barriers that might stop people who live in Scotland but who aren’t fluent in English from taking part in the work of Parliament or finding out more about Parliament’s proceedings.”)  Note that the word ‘arena’ there is not our usual Latin ‘arena’ (pron. ‘a-ree-na’) meaning a place where a game such as hockey is played, but rather ‘are-na’ (pron. ‘ar na’) meaning ‘are not.’

 

One last example of ignored pronunciation changes.  What English speakers do to ‘t’ sounds in the middle of words is almost comical; it seems many people don’t like the idea of a middle ‘t’ at all.  ‘Butter’ is pronounced very close to ‘budder’ by most of the people I know.  My home city Toronto is casually called ‘Toronno,’ ‘Taranna’ or ‘Tranna’ by most of its inhabitants.  My sister in law is one of several people I’ve met who calls Italy ‘I’ly,’ substituting a heavy glottal stop for the ‘ta.’   This throat-stopper sound is normal in Cockney London English, though the east Londoners might say ‘I’oly’ rather than ‘I’ly’ – but my respected relative is from Cape Breton Island in eastern Canada.  Her brother, my partner, grew up in the same house but is more in the ‘Iddly’ school of ‘Italy’ pronunciation. 

 

 Interestingly, then, even if we English speaking people were going to reform our spelling, we would probably completely ignore some of the most significant sound differences in our language.   We have very little awareness of how our language actually sounds.  Our spelling habits are probably at least partly the cause of this disconnection between our sounds and our ideas about our sounds. 

 

Bold Statement (BS) number 3:  writing English according to the way it sounds would baffle all users and damage the international status of the language. 

 

The more English spreads around the world, the more types of differing pronunciation it accumulates.  Attempts to make phonetic versions of English for literary purposes, that is, for writing things to people, would probably be an error.   Take, as an extreme example, the word ‘khaki,’ a dusty greenish colour used in military clothing.  It’s pronounced ‘cacky’ in the U.S., ‘caw-kee’ in England, and usually, ‘carkee’ in Canada.  Despite these differences, when the word appears in written English, it remains in its original transcription as ‘khaki’ from the Hindi and Urdu languages, and is the same for everyone.   Originally, it was pronounced khaa-kee with the initial ‘kh’ like the ‘ch’ in Loch Ness – but no English speaker outside of the Indian subcontinent would dream of saying it that way. 

 

As I noted earlier, the status of English word spellings is quite similar to that of Chinese characters.   There are many Chinese languages in China: they are dogmatically called dialects by politically-correct Chinese people but are actually distinct, related languages.  Well known examples, besides Mandarin and Cantonese, are the Wu language of the Shanghai area and the Hakka language of many overseas Chinese.   As long as the users of all these languages use the same Chinese characters in the same sentence structure, then they can all write to each other perfectly happily, even if they cannot understand each other’s speech.  (Mao Zedong complicated this by introducing a new set of simplified characters in the mid 20^th century, but we’ll ignore that for now.) 

 

English is not as diverse as Chinese, and we can usually quickly learn to understand most people who speak with different regional accents.  There is still much convenience, though, in the way we generally communicate using the same spellings.  Yes, charming novels and short stories can be written with their dialogue or even their narration in regional English, written out according to its sound.  Publishing something like an avionics textbook in that format, though, would be pretty unthinkable.  Ah’m thinkin,’ fo’ ‘sample, that y’all don’ need me to go redoin’ this heah whole book in anuthah kahnna English.  (‘I’m thinking, for example, that all you [readers] don’t need me to redo this whole book in another kind of English.’  I am roughly imitating a tour guide who did the announcing on a ‘swamp tour’ I went on near New Orleans one day.)

 

“Still,” people might say, “at least the extreme idiocies could be removed from English spelling.  What about that ‘gh’ from ‘cough’ that was so silly that it became the first part of Shaw’s ‘fish’ spelling, ‘ghoti?’”

 

Well, there’s a whole story behind that ‘gh.’  Maybe the reformers are right; maybe ‘gh’ spellings should be changed.  I will tell you the story and you can judge for yourself.  It is summed up in the following pronouncement. 

 

Bold Statement (BS) number 4:  English spelling looks so odd partly because the English people completely forgot how to pronounce two of the most common letters in their alphabet. 

 

Think about it.  The Hebrew alphabet as we know it now has been around since the first century C.E. and it closely mimics the earlier Aramaic alphabet.  The Jewish people have been through every conceivable hardship and book-burning, and yet, at least among the educated, they have never forgotten how to pronounce the letters in their alphabet.  Ordinary Israelis from northeast European stock (Ashkenazim) normally mispronounce a few letters in daily speech, but those from Mediterranean stock (Sephardim) can still pronounce the whole alphabet.  The English people have distinguished themselves by universally forgetting two consonant pronunciations completely.  One sound was forgotten during the Old English era and the other one in the Middle English era.  I am quite familiar with which words included these forgotten consonant sounds, because they are all pronounced in Dutch more or less as they would have been in early forms of Old English. 

 

The first sound to get lost was the original ‘g’ sound, which was pronounced ‘gh’ much like the Dutch ‘g’ (this sound is like the ‘ch’ in ‘loch’ but pronounced with some voice in it; the IPA symbol is ɣ̞).  It turned into a normal hard ‘g,’ or a ‘y,’ a ‘j’ sound or, in a few cases, a ‘w.’  For example, have a look at yellow, which is ‘geel’ (pronounced ‘gheil’) in Dutch, ‘geolu’ in Old English (originally probably said ‘gheyolu,’ later ‘yeyolu’) and ‘yelwe’ in Middle English.  Then compare ‘day,’ which is ‘dag’ (‘dahgh’) in Dutch, ‘dæg’ (‘dagh,’ later becoming ‘day’) in Old English and ‘dai’ in Middle English.   It’s no surprise, then, that yesterday began as ‘geostran dæg’ (‘gheyostran dagh’ --> later ‘yeostranday’), where the first word, ‘geostran,’ was related to the current Dutch ‘gisteren’ (‘ghee-steren’), ‘yesterday.’ 

 

Sometimes ‘g,’ instead of becoming ‘y,’ disappeared completely, such as the second ‘g’ in the word ‘youth:’ ‘geoguð’ (pronunciation ‘gheyoghuth’ --> ‘yeyowuth’) in Old English, ‘jeugd’ (‘yeuhght’) in Dutch, ‘yougth,’ ‘yowith,’ ‘ghowith’ and various other forms in early Middle English, and ‘youth’ in later Middle English. 

 

In a large number of words, especially where the ‘g’ was followed by something other than ‘e,’ the early Old English ‘gh’ sound became a normal hard ‘g.’  Here are some examples, with Dutch word first (where the ‘g’ is alway pronounced ‘gh’) then the Old English (abbreviated OE; where the early pronunciation of the first letter was ‘gh’ and the later pronunciation ‘g’) followed by modern English.    Dutch ‘gaan,’ OE ‘gan,’ English ‘go;’ Dutch ‘gat’ (meaning ‘an opening’), OE ‘gæt,’ English ‘gate;’ Dutch ‘geit,’ OE ‘gat,’ English ‘goat;’ Dutch ‘grond,’ OE ‘grund,’ English ‘ground.’

 

As you can see, this growing allergy to the ‘gh’ sound didn’t do much harm to spelling, because the spelling tended to change as the pronunciation did in these older words.  A few famously strangely spelled English words, however, did arise this way, to give a taste of things to come.  For example, the Dutch word ‘geest’ (pronounced ‘ghayst’), meaning ‘spirit,’ is related to Old English ‘gāst’ (‘ghaast’ --> ‘gaast’), which gave rise to the strangely spelled modern English ‘ghost.’  It seems that, in this case, the original ‘gh’ sound was pointed to by the use of ‘gh’ as a letter combination.  ‘Gh’ was rare in English but was well known to anyone who could read medieval Dutch.  Where modern Dutch uses the letter ‘g’ for the ‘gh’ sound, medieval and renaissance Dutch often used the two-letter symbol ‘gh.’  I suspect that the English name ‘Ghent’ (pronounced hard-g ‘gent’) for the Belgian city of Gent (pronounced ‘ghent’) was spelled as it was for similar reasons.  If it had been just a matter of ‘protecting’ the ‘g’ from the following ‘e’ to make sure no one pronounced the name as ‘jent,’ parallel to ‘gentleman,’ this could have been done systematically in the Anglo-Norman French way by calling the city ‘Guent.’ The insertion of an unpronounced, protective ‘u’ would make ‘Guent’ parallel with words like ‘guest’ (Dutch gast, OE ‘gæst’ or ‘giest,’ Anglian dialect ‘gest’) and ‘guess’ (Dutch ‘gissen,’ Middle English ‘gessen’).  We almost never use ‘gh’ to protect a hard ‘g’ in Germanic words.  I can think of just one more word where we spell the Dutch ‘gh’ sound but pronounce our own hard ‘g:’ that is ‘gherkin,’ the small cucumber pickle.  It derives from the regional Dutch plural ‘gurken,’ pronounced, of course, ‘ghurken’.  Usually, an ‘h’ is only used to ‘protect’ a hard ‘g’ from turning into a ‘j’ sound in words stolen directly from Italian, like ghetto and spaghetti.  Adding an ‘h’ is the normal linguistic condom used for this purpose in Italian, whereas French, like its stepchild Middle English, typically uses ‘u,’ as seen in ‘guerre’ (‘war’), Guillaume (William) and many other words.   

 

The words related to ‘ghost,’ like ‘ghoul’ and ‘ghastly,’ all have the same mysterious ‘h’ in them.  It is tempting to say that it’s the ghost of the original pronunciation. 

 

Another classic weird spelling that came from the ‘gh amnesia’ period was the spelling of ‘weigh.’  In this case, the Dutch word is ‘wegen’ (pronounced ‘way-ghen’ – the ‘w’ at the front is halfway between ‘w’ and ‘v’ in sound), the Old English was wegan (way-ghan), and Geoffrey Chaucer’s Middle English is weyen (‘And in balance weyen ech montayne’ [and in a balance weigh each mountain], Monk’s Tale, line 698).  Yet the ‘gh’ letter combination, representing the Old English ‘g,’ found its way back into modern English spelling.  I don’t know why: perhaps it was to help distinguish ‘weigh’ from ‘way,’ which appears in both Dutch and Old English as ‘weg.’  We also have the milk product ‘whey’ (Old English ‘hwæg,’ Dutch ‘wei’), so our language gets complicated when it comes to words sounding like ‘wey.’  Newcomers to English may think that ‘whey’ doesn’t add much to the problem, since it’s a very uncommon word.  In fact, though, most English speaking children are raised with the nursery rhyme:

 

Little Miss Muffet sat on her tuffet

Eating her curds and whey.

Along came a spider and sat down beside her

and frightened Miss Muffet away!

 

This rhyme remains wildly popular because it has infinite possibilities for running tickling little fingers over the child and starting a fine round of giggling, or as we say it in Dutch, ‘giechelen’ (‘ghee-khelen’).  So when you weigh it all together, ‘whey’ is not such a wayward, out-of-the-way word.   

 

‘Weigh,’ then, is sort of early prototype ‘ghoti’ word – a word where a ‘gh’ appears in a seemingly nonsensical place because people were avoiding pronouncing a ‘gh’ sound, but where they still, for mysterious reasons, wanted to display the historical pronunciation.  

 

Another twist is added to the story by words like ‘saw,’ the wood-cutting tool.  It is ‘zaag’ (‘zaahgh’) in Dutch, ‘sagu’ (sahghu --> saa-u) in Old English (related to ‘saex,’ knife, the word that gave the Saxons their name) and ‘saw’ in English.  The change from ‘sagu’ to ‘saw’ could be seen as the dropping of the ‘g’ sound, or it could be seen as the substitution of another consonant, ‘w,’ for the ‘g.’  You know that the ‘avoid gh-sounds at any cost’ idea is becoming pretty strong when unrelated consonants get thrown in where the letter ‘g’ used to be.  Parallel to ‘saw’ is the past tense of ‘to see:’ in Dutch, you say ‘ik zag je,’ or more formally ‘ik zag u,’ where the English speaker says ‘I saw you.’  It seems that somewhere back in history, the ‘gh’ had to be sawed off all these words.  

 

The situation became much worse when the English later forgot how to pronounce yet another, related letter.   This was the Old English letter ‘h,’ (letter ‘eolh’ in Fuþorc runes) in all the places where it indicated a ‘kh’ sound as in Scottish ‘loch.’   Of course, ‘kh’ and ‘gh’ are very similar: ‘kh’ is more or less the unvoiced version of ‘gh.’  Modern Dutch, in fact, does not distinguish the sounds.  Historically, in Dutch, ‘g’ is a ‘gh’ sound and ‘ch’ is a ‘kh,’ but now people tend to pronounce both of these as ‘gh’ if they come from the north, and as ‘kh’ if they come from the south.  I learned to pronounce ‘kh’ from Iranians, so when I speak Dutch, I sound like I’m trying to be a northerner. 

 

With the avoidance of ‘kh’ sounds, we reach the ‘heart of darkness’ at the middle of the English spelling conundrum, the singularity that sucks all common sense into oblivion within our spelling black hole.  Somewhere in here is the answer to the Gerard Trenité’s dramatic question:

 

Finally, which rhymes with enough,

Though, through, bough, cough, hough, sough, tough??

 

The notorious ‘-ough’ words are only the beginning, though:  there are also ‘-ought’ words like ‘thought’ and ‘drought,’ ‘-aught’ words like ‘caught,’ ‘daughter’ and ‘laughter,’ ‘-eight’ words like ‘eight,’ ‘freight,’ and ‘height,’ ‘-eigh’ words like ‘sleigh,’ ‘-ight’ words like ‘bright’ and ‘-igh’ words like ‘high.’ These are mostly words that have Dutch and Old English equivalents where the ‘kh’ sound represented by ‘ch’ is preserved.  Occasionally a ‘g’ appears in the Dutch instead of the ‘ch.’ Either way, quite a few of these words are verbs in the past tense or past participle form.  ‘Thought,’ for example, is related to the Dutch past participle ‘gedacht’ (pronounced ‘ghedakht’), meaning ‘thought;’ the present tense is the very different sounding ‘denken’ (pronounced exactly as it looks, except that the final ‘n’ sound can be dropped) clearly related to the equivalent English present tense, ‘think.’  ‘Drought’ is related to ‘gedroogd’ (pronounced ‘ghedroe -ghd’), a past participle meaning ‘dried;’ the present tense word ‘dry’ is one of the many cases where a Saxon ‘g’ turned to a ‘y’ in English.  The original Dutch word for dry is ‘droog’ (‘droe-gh’).  Other words are simply direct translations with changed sound:  the Dutch partner of ‘eight’ is ‘acht;’ ‘freight’ is ‘vracht,’ ‘daughter’ is ‘dochter,’ ‘to laugh’ is ‘lachen,’ ‘light’ is ‘licht.’   You can pronounce these Dutch words well enough (Dutch ‘genug,’ pron. ‘ghenuegh’) if you just remember that the ‘ch’ represents that ‘loch’ sound, ‘kh.’ 

 

‘Sleigh’ and ‘caught’ from the list above are later imitations of the silent ‘gh’ form; neither word is descended from a word with a ‘kh’ or ‘gh’ sound in it. 

 

I can more or less reproduce Gerard’s rhyme in Dutch.   The exact meanings of the words may deviate from the English.

 

Finally, what rhymes with genug,

Doch, door, boog, kuch, hak, zwoeg, taai? (pron. dokh, dor, boegh, kuekh, huk, zwoogh, tie)

 

Unfortunately, the modern loss of ‘kh’ in ‘taai,’ meaning ‘tough,’ ruins the rhyme.  It would work better if switching was done to make

 

Finally, what rhymes with genug,

Doch, door, boog, hak, zwoeg, taai, kuch??

 

The passage comes out much better with Old English words, all of which end with a ‘kh’ or ‘gh’ sound, with or without the verbal ending ‘-en’ or ‘-an’ tacked on.  ‘Coughen’ is from Middle English; an Old English equivalent along the lines of ‘cogan’ is presumed to have existed but is not known from existing texts.  

 

Finally what rhymes with genog,

þeah, þurh, bog, ‘coughen,’ hoh, swogan, toh??

 

Some of these words kept their Old English ‘kh/gh’ pronunciation in Scots.  ‘Sough,’ for example, is found in Robert Burns’ 1785 poem ‘Cotter's Saturday Night’:   “November chill blaws loud wi’ angry sugh.”  (“November chill blows loud with angry sough,” i.e., with angry whoosh of wind through the trees).   In fact, one thing you’ll notice in the Scots Parliament quotations given above – and I promised I would return to this point – is that the word ‘micht,’ meaning ‘might,’ appears regularly: “The committee micht bid ye tae come in an speak til the committee.”  “This policy is wrocht tae caw doun the leid barriers that micht stap folk….”  Clearly ‘micht’ is much more faithful to the original pronunciation of ‘might’ than our current ‘mite’ pronunciation is.  The closest Dutch word, incidentally, is ‘mocht,’ which is similar in meaning to ‘might’ but used somewhat differently.  And look, in that last Scots sentence, there’s ‘wrocht’ as well, which in English is ‘wrought,’ pronounced ‘rot.’  In Scots, I don’t know if the ‘w’ has gone silent or not, but the ‘ought’ is as it ought to be, ‘okht.’  Clearly our twin language in the northeast retained its ability to pronounce a ‘kh/ch’ sound. 

 

Our own avoidance of the ‘kh/ch’ sound has led to farfetched extremes, such as substitution with an ‘f’ sound in ‘laugh,’ ‘tough’ and ‘cough.’  The rare substitution of ‘k’ and ‘p’ sounds, as in ‘hough’ and ‘hiccough,’ show that the trend almost had no limit in how strange it could get.  It’s been pointed out that ‘f’ was only used in cases where the vowels before ‘gh’ originally were ‘a,’ ‘o’ or ‘u’ sounds.  These are vowels from relatively far back in the throat, the so-called ‘back vowels,’ as opposed to ‘front vowels’ like ‘i’ and ‘e.’  For front vowels, the ‘gh’ predictably disappears – that is, it isn’t pronounced – as in bright, light, might, sleight, high and sleigh.  In thinking about this, I realized that it would somehow feel wrong, even to me, if ‘kh-dropping’ meant that ‘laugh’ would become ‘la,’ laughter ‘latter,’ tough ‘tuh,’ cough ‘coh,’ rough ‘ruh,’ and so on.   Almost all of the words where ‘gh’ became ‘f’ really do seem to be covered by a rule, which is, roughly,  “words that have back-vowel sounds in front of a ‘gh’ that can’t be pronounced any more still need to have a definitely pronounced consonant as an ending.”  In words where ‘gh’ follows vowels or diphthongs that are pronounced in or near the front section of the mouth (or, technically, the upper part of the back section), whether we’re looking at long ‘i’ in ‘high,’ the long ‘u’ sound in ‘through’ or the ‘ou’ sound in ‘bough,’ the ‘gh’ can disappear from the soundscape.   But before short ‘a,’ ‘u’ or ‘o’ sounds, however they are spelled (‘u’ as in ‘rough,’ ‘o’ as in ‘cough’), an ‘f’ sound or something even more extreme (such as ‘p’ in ‘hiccough’) has to be inserted to end any word that would otherwise end with a vowel.   This rule applies especially well in British English, where the ‘a’ sound in ‘laugh’ is much throatier than the North American ‘a’ sound in that word.   To continue the rule:  “for back-vowel ‘gh’ words that already have another consonant to end them in their basic form, such as ‘thought’ and ‘caught’ with their ‘t’ endings, no ‘f’ sound is necessary.” 

 

No language authority ever made this amazing rule; it was the spontaneous, democratic decision of the English people.  I hope you don’t find it la’able (laughable).  And that’s enuh about that! 

 

With the singular exception of ‘Loch Ness,’ then, a name made famous by a prehistoric monster, English speakers have been avoiding ‘kh’ sounds consistently ever since Middle English times.  To English people, ‘kh’ IS a prehistoric monster.  As welcoming as English is to other tongues, the thought of letting ‘kh’ sounds come in to our language is impossibly scary. 

 

The medieval decision against this sound is still a major influence.  In a relatively recent example of ‘kh’-escape, the original name of the Canadian region I grew up in, Secwép (pronounced ‘Sekhwep’), was imported into English as ‘Shuswap’ as soon as Europeans encountered it in the late 19^th century.  This name is now officially pronounced ‘Shue-swap’ in English, but local people often say ‘Shue-shwap.’   Apparently they know that ‘Shuswap’ is too cleaned-up and not quite right, but since they can’t say ‘kh,’ they resort to ‘sh’ instead.  Nearby is Skaha (pronounced ‘Skaw-haw’) Lake, which if rendered in the original Secwépemctsin (the so-called ‘Shuswap language’) should be Sqéxe (pronounced ‘Sqekhe’) Lake.  The meaning is ‘Dog Lake.’  English speaking people bravely adopted this name even though both the Secwépemc ‘q’ (similar to the one in Arabic, Persian, Inuktitut, and Quechua) and the ‘kh’ (which is actually one stage deeper in the throat than a typical ‘kh’) were beyond their pronunciation skills.  To be fair, none of our ancestors ever had a ‘q’ in their speech, but the same cannot be said for ‘kh.’  If we spoke Scots rather than English, we could at least have managed ‘Skacha’ (‘Skakha’) rather than Skaha. 

 

By the way, the reason the two ‘e’ sounds in ‘sqéxe’ became ‘a’ sounds in English is that when your voice is on the way back into your throat to pronounce a consonant like ‘q’ or the Secwépemc ‘x’ (deep ‘kh’), where the sound comes from the back end of the throat, vowels like ‘e’ become ‘darkened’ and merge into back-of-the-throat vowels like ‘a’ as in ‘father.’  I have a hunch, if I can go back to the word ‘sugar’ for the third time, that the odd sound of the ‘u’ vowel there is an effect of vowel-darkening brought about by the ‘g,’ which is one of the throatiest of our pronounceable consonants.  At least in my mouth, ‘g’ goes considerably further back than ‘k,’ its supposedly unvoiced equivalent.  Many European languages have front vowels in this word, like the ü in French sucre (described above) and the ‘ui’ diphthong in Dutch ‘zuiker’ (pronounced approximately ‘zeüiker’), or they have middle vowels like the ‘u’ in Spanish azucar and Italian zucchero (‘zue-kero’).  We, on the other hand, have our strange unwriteable near-the-back vowel of the ‘u’ in ‘put’ and the ‘oo’ in ‘soot.’  This throatiness factor may be the final part of the story of how ‘sugar’ became one of our oddest conflicts between spelling and pronunciation.

 

To get back to ‘kh,’ though, two reasonable questions could be asked.  One is, ‘why did we give up saying it?’ and the other is, ‘why did we keep on printing the letters referring to it?’  

 

My attempts to answer the first question will take a few minutes, so let’s deal with the second one first.   It was already partly dealt with earlier under the general theme of ‘why did we retain traditional spellings of all kinds?’  The aspect that has not been mentioned so far is that when printing presses were invented and written matter became more available than it had been, efforts started to be made to stabilize English spelling.  English was a late-bloomer in terms of getting dictionaries going, but in 1604, the schoolteacher Robert Cawdrey produced ‘A Table Alphabeticall,’ which listed a modest number of potentially troublesome words.  The first truly thorough English dictionary was Samuel Johnson's ‘A Dictionary of the English Language,’ printed in 1755.   Johnson was a firm believer in traditional spellings, and was a major influence on how English looks today.   A sample entry in his dictionary for one of our ‘ghoti’ words is: “Cough: A convulsion of the lungs, vellicated by some sharp serosity. It is pronounced coff.”   Clearly, Johnson was not about to tell the English people how to re-spell their words.  His American successor, Noah Webster, in his 1828 ‘An American Dictionary of the English Language,’chose to do a tiny bit of spelling reform, which yielded the standard American (but not Canadian) spellings used today.  For reasons unknown to me, though, he mainly attacked French elements of English rather than Saxon oddities like ‘cough.’  He turned the French-styled ‘centre’ and ‘theatre’ around and made them ‘center’ and ‘theater.’  He also removed the Gallic two-vowel ending from words like ‘colour’ and ‘humour’ (which are ‘couleur’ and ‘humeur’ in French), simplifying them as ‘color’ and ‘humor.’  But only underground comic strip authors in the 20^th century dared to write “coff, coff!” and “that’s enuf!”   The Saxon words in English are the common peoples’ words, and anyone who needs to win wide public approval leaves them alone. 

 

Now let’s get back to the topic of how we became allergic to ‘kh’ and ‘gh.’  We could possibly blame the French, at least for the loss of ‘kh.’  Middle English was a fusion between a language that was losing ‘gh,’ Old English, and one that lacked both ‘gh’ and ‘kh,’ Norman French.  When I first thought about the word ‘wait’ in terms of the trends I’ve been talking about, I thought, “there’s another typical example of English kh-avoidance, since the same word is ‘wachten’ (pron. ‘wakhten’) in Dutch.”  It turns out, though, that English got ‘wait’ not from Old English but rather from the Norman ‘waitier.’   The Normans had already long ago stripped the ‘kh’ from the Frankish Germanic root word ‘wahton’ (‘wakhton’) to make this word.    Perhaps, then, the stripping of ‘kh’ sounds from Middle English was a gesture of welcome to French speakers who could not pronounce this sound.  On the other hand, first of all, the French learned to pronounce the impossible ‘th’ sounds in English, or at least their descendants did.  Secondly, ‘kh’-loss occurred around the same time as the much larger trend, which I discussed in the last chapter, called the ‘Great Vowel Shift.’ 

 

As you remember, in the course of that ‘Shift,’ a large number of vowels that were often similar to Dutch or Latin vowels moved forward or upward in the mouth.  Several became new sounds not commonly heard outside English, like long ‘a’ in plate (though French ‘é’ comes close) and long ‘i’ in ‘bright.’  Even the sounds that remained stable moved to different letters:  for example, the ‘æ’ of Old English ‘dæg,’ pronounced like our modern ‘a’ in ‘flat,’ became the normal short ‘a’ sound, which had previously been like the ‘a’ in ‘father.’   The modern short ‘a’ sound, especially in its North American version, is quite unusual: Dutch people cannot say it at all (they pronounce ‘flat’ as ‘flet’) and the nearest European languages with similar sounds are the Nordic languages and Finnish.  As if to recognize this sound as exotic, Finnish decorates it with dots as ‘ä,’ while an undotted ‘a’ is pronounced like the ‘a’ in ‘father.’   Finnish is not related to other European languages, except Estonian, Hungarian, Sami and a few minority languages of Russia like Mordvin.   English, then, with the ‘Great Vowel Shift,’ was going very far out on a linguistic limb with its sounds.

 

Incidentally, when I say that vowels moved ‘forward’ in the mouth, I do literally mean that the part of the mouth they are pronounced from moved closer to the teeth.  When I say they moved ‘upward,’ that means that they moved from a position where the tongue is low in the mouth to a position where the tongue is held up close to the roof of the mouth.   Note the word ‘close’ there.  Linguists call the vowels spoken with the tongue up high ‘close’ vowels (rhymes with ‘dose,’ not ‘those’).  Our English ‘short o’ as in ‘bought,’ ‘long o’ as in ‘bone’ and ‘long u’ as in ‘flute’ are all pronounced at the back of the mouth, but the first is relatively low (called an ‘open-mid’ vowel by linguists), the second is higher (called a ‘mid’ vowel) and the third is quite high –  that is, the gap between the tongue and the roof of the mouth is quite narrow (called a ‘close’ vowel).  Vowels affected by our English ‘Shift’ mostly went either forward, or upward (more ‘close,’ less ‘open’) or both.  Vowels that went upward are said to have been ‘raised,’ and the process in which this happened is called ‘tongue raising.’ 

 

Some of the vowels that went forward or upward attracted a second vowel as part of their pronunciation and became ‘diphthongs.’  These are double vowel sounds as heard in words like ‘boil’ and the above-mentioned ‘out’ and ‘about’ (IPA sound aʊ or ʌʊ; note the two-symbol combinations).  The current English long ‘a,’ for example, as in ‘bake,’ is technically considered a diphthong, represented ‘eɪ’ in IPA (at least as pronounced by northern Americans, Canadians and upper crust English people).  Both parts of the diphthong are pronounced higher in the mouth than the word’s original ‘a’ sound, which was like the ‘a’ in the North American ‘father.’  The ‘e’ part of the diphthong is also distinctly further forward in the mouth than the original middle English ‘a.’  Looking at the whole history of this ‘a,’ then, from middle English to modern English, it moved upward and distinctly forward into the first, ‘e’ half of the new diphthong, and moved upward and perhaps slightly forward into the second, ‘ɪ’ half.  Just to explain the symbols in the diphthong, an IPA ‘e’ by itself is an uncommon sound in modern English, but linguists use it for the Scottish English pronunciation of ‘e’ as in ‘bed’ (sounds like a clipped version of ‘baid’ to a Canadian.)  To me, the sound is much like the ‘e’ in the Canadian question word ‘eh,’ if that word is said casually and quickly: ‘so I went t’ the hockey game, eh? ‘n’ it was goin’ along ok ‘n’ our guys were doin’ real good, eh? but then…”  Yes, that’s how we talk in our secret meetings.  The ‘ɪ’ part of the diphthong is like short ‘i’ in ‘pit.’  I hope you don’t sprain your tongue experimenting with all this stuff, but I think if you try a few vowels, you’ll see what I and the linguists are getting at.  A ‘linguist’ literally means ‘a tongue person,’ so they should understand your wet oral flapper pretty well.   

 

The ‘Great Vowel Shift’ was inconsistent, and it also left inconsistent spellings behind.  You can see this very clearly in Trenité’s series “though, through, bough, cough, hough, sough, tough,” where the same ‘ou’ indicates upper level vowels as in ‘through,’ middle sounds as in ‘bough’ and low, ‘open’ sounds as in ‘tough.’  Linguists have pointed out that, like floating logs trapped at the side of a rapidly moving river, a number of words got stuck in their older pronunciations while all the others with similar spelling moved past them into the new way of speaking.  Many words with ‘ea’ in them didn’t move up front in the mouth the way ‘each,’ ‘teach,’ ‘sear’ and ‘meat’ did; some stayed behind, such as ‘great,’ ‘break,’ ‘swear’ and ‘bear.’  Many of the latter group stayed close to Dutch pronunciations of the same words:  the Dutch ‘beer,’ meaning a bear, is pronounced much like ‘bear;’ compare also ‘ik breek het,’ pronounced ‘ik brake et,’ meaning ‘I break it,’ and, in parallel format, ‘ik zweer het,’ pronounced ‘ik zware et,’ ‘I swear it.’  ‘Great’ went in another direction in Dutch, becoming ‘groot’ (pron. ‘ghrote’).    

 

‘Tear,’ the ripping action, stayed behind, while ‘tear,’ the drop from the crying eye, moved forward.   To make things still more complicated, ‘ea’ in some words, especially words starting with consonants like ‘d’ and ‘th,’ placed right at the front of the mouth, became short ‘e’ sounds, as seen in ‘dead’ and ‘threat.’   ‘Head’ managed to butt itself into that group as well, and ‘sweat’ presumably made it in at the last minute.   Similar things happened with ‘oo’ words, where ‘oo’ in ‘cool,’ ‘soon,’ ‘boot’ and ‘root’ moved upward (though some dialects still have ‘root’ rhyming with ‘put’) while ‘oo’ as in ‘cook,’ ‘book,’ ‘soot,’ ‘foot,’ ‘hood,’ and ‘good’ didn’t move so far up (‘oo’ as in ‘food’ is a ‘close vowel,’ while ‘oo’ as in ‘foot’ is only ‘near-close’) and also became shortened in length.  A few words got diverted closer to the lower back end of the mouth, as in ‘flood’ and ‘blood.’  Other famous left-behind words in the Great Vowel Shift are ‘broad,’ which should have advanced to rhyme with ‘road,’ and ‘father,’ which should have kept pace with ‘lather.’  ‘Rather’ is a curious case, more-or-less rhyming with ‘father’ in conservative upper-crust British, but with ‘lather’ in North American (as well as in the very different sounding Australian, where to the Canadian ear it sounds like ‘retheh’).  Apparently, it had contrary feelings, deciding it would ‘rather’ go forward in the mouth in some cases, and in other cases would ‘rahther’ not.  A headstrong word. 

 

Actually, ‘rather’ raises a topic that I am not planning to deal with in detail here, which is the differences among English regional accents in how words with short ‘a’s are pronounced.  If you’re interested in why ‘cat’ and ‘trap’ have a different ‘a’ sound from ‘bath,’ ‘chance,’ ‘laugh’ and ‘grass’ in some parts of England, you can have a look at the Wikipedia page on the “phonological history of the English short ‘a.’” The latter group of words is pronounced, in North American phonetics, ‘bawth,’ ‘chawnce,’ ‘lawf’ and ‘grawss’ in parts of England that have the so-called ‘Broad A.’   The group containing ‘cat,’ ‘trap’ and ‘man’ has the typical English short ‘a’ in most of England and the Americas.   

 

Ultimately, the question of ‘how did we lose kh?’ has to be considered in a broader context, namely “why were many of the sounds in Middle English trying to get away from the throat and move ever closer to the teeth and the top of the mouth?”  You’ll notice, by the way, the ‘f’ in ‘enough’ and ‘cough’ is precisely pronounced with the teeth.   ‘Kh’ and ‘gh,’ by contrast, are quite far back in the throat. 

 

The French don’t immediately seem to be to blame for this tooth-wardness in Middle English pronunciation.   The results of the ‘Great Vowel Shift’ don’t resemble French.  Also, French has several famous guttural sounds.  Parisian in particular requires a deep ‘r’ sound that is as throaty as any sound you can learn.  The nasal ‘un’ serving as the number ‘one’ in French is also pretty throaty, even though half of it is aimed for the sinuses.  No, even taking into consideration that Norman differed from Parisian, the vowel shift is something peculiarly English.  I think.  (Other languages have had different vowel shifts, usually on a smaller scale than the English one.)  

 

I am going to go very far out on a limb and make a suggestion about where the pronunciation changes in Middle English came from.  First, though, I have to explain a little about the general kind of phenomenon that I think kh-loss, and perhaps even the ‘Great Vowel Shift,’ may fall into.  I use a special expression for this type of phenomenon.  Even though I have tried to write this book in a very easy-to-understand way, with few literary words, I think it would be best if I introduced this expression.  It is ‘epistemic adducts,’ which means, ‘cultural ideas that clamp onto people’s minds and permanently distort the way they see things.’  (‘Epistemic’ refers to ‘how we know what is true,’ and ‘adduct’ means ‘something that has attached itself.’ In my field, we use the expression ‘DNA adducts’ to refer to chemicals that clamp onto genetic material and cause mutation damage that gives rise to cancer.)   If you want a simpler term, you could think ‘mind clamps’ or ‘brain Velcro®.’  Or, though it sounds much too harsh, ‘thought cancers’ – mutant thought patterns that eliminate reality by displacing it.

 

My own last name actually provides an excellent illustration of epistemic adducts at work.  I was very interested, when I worked in the Netherlands, to discover that nearly everyone there who heard me say the name ‘Summerbell’ could spell it correctly.  What could be simpler, ‘summer’ and ‘bell,’ two common English words?  The interesting thing is that people who grew up speaking English as a first language find this name almost impossible to spell.  I will spell it out for them very carefully and still see them, with wrinkled brow, doggedly writing ‘Summerv...’ or ‘Summerh....’ or ‘Somerv...’, just to give the three most popular variants.  And even if, with some self-correction, they can get through ‘Summerbell,’ I have to watch them like a hawk, because they will then start to add an extra ‘e’ at the end to make the name Summerbelle.  ‘Oh sorry,’ they say, ‘oh sorry.’   They have no idea why I would present them with such an impossible name, when their brain was all ready with Summerville or Summerhill or Somerville.  They mess up ‘Summerbell’ two or three times before getting it right and become quite embarrassed.  No exaggeration.   ‘Summerbell’ literally boggles people’s minds.  But not in the Netherlands.  There, it is easy.   

 

When did the English-speaking peoples of the world become complete mental slaves to the name ‘Summerville?’  In fact, though, there is a lot of that sort of thing going on.  My partner and I got a dog a few years ago from Canine Vision Canada, the people who train seeing-eye dogs for the blind.  Their policy is to let people take young pups home as ‘foster puppies’ for one year so that the pups can get used to ordinary life before they go back for specialized training as guide dogs.  After the training, the dogs are assigned to a blind person and the ‘foster’ owners seldom if ever see them again.  Our plan was to raise our foster puppy for a year, then get a dog of our own.  The foster pups were not allowed to have human names like ‘Max’ or ‘Fred’ and we finally hit on the idea of naming ours ‘Practice,’ since he was to be our practice dog before we got a full-time dog, so to speak.  In fact, our foster puppy had genetic leg problems and we ended up getting to keep him after all.  He was beautiful looking and attracted a lot of attention in downtown Toronto.  This meant we had to explain the name ‘Practice’ to hundreds and hundreds of people.   An eerie pattern soon emerged.  

 

On first introduction to his name, 90% or more of people would say, “Practice makes perfect!”   Sometimes with insight, sometimes with conspiratorial glee – the emotions were many, but the phrase was always the same.  We certainly heard it at least 1200 times.  Even more strangely, when they met Practice again, a day or more later, they would again all say the same follow-up phrase: “how’s Patience?”

 

If we corrected the name, we’d see the same thing as I was used to with Summerbell.  A terrible mental struggle would be visible, like someone with Alzheimer’s syndrome desperately trying to remember something, followed by something like “oh yes, Practice.”   But the next time the person encountered him, he might well be ‘Patience’ again.  At later meetings, there was a lot of, “Patience, no, what was that name?” and red-faced confusion. 

 

It seems people file unusual names in their mind by first letter or first familiar word, and, when called into action, pull out the commonest name in their archive that begins with the correct letter or word.  Summerbell thus becomes ‘Summerville,’ and Practice ‘Patience.’  This doesn’t explain the whole phenomenon, though.  A dog-walking friend named Ingrid reports that people strongly tend to forget her name and then come up with ‘Brigid’ as a substitute.  Very mysterious.  Perhaps the parallel ‘i…g…i’ hooks their memory, or perhaps, in a sort of ‘Practice makes perfect’ effect, they file her under B for ‘Bergman.’

 

Practice flew to the Netherlands to stay with me there.   Mental archives have different stuff in them in the Netherlands, and people who met the dog for the first time, after asking his name, would say ‘Oh, Prectice.’  No one ever said, “Prectice makes perfect.”  The next time they saw the dog, they would say, “Ah, Prectice.”  They treated the name like learning another bit of English, which they are skilled at.  It did not change. 

 

Epistemic adducts also work on a larger scale.  Within my profession, mycology, some fascinating books were written by an amateur researcher named R. Gordon Wasson, who travelled the world looking for interesting patterns in how people got along with mushrooms.  He wrote classic books on hallucinogenic magic mushrooms, which have made him a cult hero in some circles.   Also, with his Russian wife Valentina Pavlovna Wasson, he co-wrote Mushrooms, Russia and History (1957), which looked into peoples’ differing viewpoints on edible wild mushrooms.   In the last topic, the Wassons found some cultural prejudices that were as deep and strong as any that have ever been described.  It turns out that some nationalities love wild mushrooms, and others hate them.  ‘Mycophilic’ peoples regard mushroom hunting as great fun, and enthusiastically learn how to recognize the edible and poisonous species.   They have many experts who can help them; in some places in Europe, paid civil servants are on duty to help people distinguish edible from poisonous mushrooms.  Dozens of mushroom species have individual common names in their languages, and these names are familiar to many people.  The names appear in novels and have literary significance.  Wild mushrooms are widely sold on the market.  Cultures that are like this include the Scandinavians, Finns, Slavs of all nations, Dutch, Belgians, French, Italians, Turks, Chinese (though in China certain minority nations, such as Tibetans in northwest Yunnan province, are especially considered to be woodland mushroom experts), Japanese, and many Africans (who eat the huge Termitomyces mushrooms that grow out of termite mounds.)   Arabs are also included, though in some of their scorchingly dry home nations, they mostly collect underground desert truffles (Terfezia) rather than above-ground mushrooms. 

 

‘Mycophobic’ cultures, who detest, fear or simply ignore wild mushrooms, include the English, the Celts, some of the nationalities of the Indian subcontinent, and most native American nations.  English literature contains many descriptions of slimy, venomous woodland toadstools, and English speaking people know few or no mushrooms by name.  Most English common names for wild mushrooms have been made up by scientists since 1980; traditional culture lacks such terms.   These days, if asked what mushrooms they know, native English-speaking people say something like “the ones I buy at the store” and giggle.  

 

Here’s a classic piece of English mushroom detestation from John Gerard’s Great Herbal, 1597: 

 

Galen affirmes, that they are all very cold and moist, and therefore to approach unto a venomous and murdering facultie, and ingender a clammy, pituitous and cold nutriment if they be eaten.  To conclude, few of them are good to be eaten, and most of them do suffocate and strangle the eater. 

 

‘Pituitous,’ a word not used in English today, means ‘snotty’ or ‘mucous.’  Galen (129-199 A.D.) was a Roman physician and prolific writer on medicine and nature.

 

Most native North American nations connect mushrooms with ghosts; for example, the Coast Tsimshian language of northern coastal British Columbia divides mushrooms into ‘gaaydi baa’lkh,’ ‘ghost hats,’ and ‘gaaydi tsu’u’uts,’ ‘penis hats’.  Needless to say, not many people want to go out collecting ghost or penis hats for the dinner table. 

 

Up until a few years ago, as with “Practice makes perfect,” the same words would pop out of many English-speaking people when they were asked about wild mushrooms – generally, ‘slimy,’ ‘cold’ and ‘poisonous’ were on top of the list.  Recently, there has been a lot of press about porcini and other tasty wild species, and cultivated Asian woodland mushrooms like shiitake have shown up at the grocery store.  This has caused some changing of attitudes, and yet our original cultural attitude is still widely prevalent.  I belong to a society of mushroom hobbyists in Toronto, and most of the membership grew up at least partly in continental Europe.  A younger generation has recently emerged in the society that has a strong Chinese component.

 

When it comes to ‘kh,’ there is, again, a set phrase that one hears over and over.  That phrase is “clearing your throat.”  If you speak a bit of Dutch to English speaking people, you will soon hear that phrase.  English speaking people actually find the Dutch ‘g’ disgusting.  It suggests phlegm, disease, and uncouth social behaviour to them.   To the Dutch, of course, it is just another letter of the alphabet, but they have great fun teasing English speaking tourists with it.  I remember one Amsterdam canal boat tour guide amusing tourists with “in English, it’s so romantic, you can say ‘I love you,’ but in Dutch we have to say ‘ik mag je graag’ (which he exaggerately pronounced ‘ik maaaghghgh ye ghghraaaaawwghghgh’).”  This wasn’t actually true: ‘ik mag je graag’ actually means ‘I like you,’ not ‘I love you.’ If you want to say ‘I love you’ in Dutch, you need to say ‘ik houd van je’ (‘I hold of you’).  But the comic horror on the faces of the tourists showed that the guide had maximized the entertainment with his tonsil-scraping, “ik mag je graag.” 

 

Word for word, by the way, ‘ik mag je graag’ means ‘I may you willingly,’ which not only reveals another disappeared English ‘gh’ sound (mag --> may), but also the fact that Dutch can have very non-English phrasing at times.   As closely related to English as it is, it can be pretty disorienting to speak.  One of its many mysteries is it is the only language I know that has no word corresponding to ‘to like.’  This absence requires dozens of verbal dodges such as ‘ik vind het leuk’ (‘I find it nice’) when ‘it’ is an inedible item, and ‘ik vind het lekker’ (‘I find it tasty’) when ‘it’ is food.  That’s why a simple ‘I like you’ becomes the sleep-apnea expulsion ‘ik mag je graag.’  (You see, even I can play the ‘gh is disgusting’ game!) 

 

Now, where did we English speakers get this idea that one of the original letters of our alphabet, and one of the most popular ones, was actually a disease symptom? 

 

When you look at the time all this language change was occurring, something else was also going on.   As I mentioned earlier, half the population or more of many areas was dying in plague epidemics.   In England, at least two million of a population of four or more million people died in 1349-1350.  Plague was often seen as the bubonic form, mainly distinguished by swollen lymph nodes producing large bumps on the body surface, followed by a red rash that turned dark.   There was also, however, a pneumonic form of plague, which was a like a quick-killing, terrible cold, or a rapidly deadly pneumonia.  Other epidemics also came along, including smallpox, but, as any student of English literature or Italian opera knows, one of the most dramatic epidemic horrors that came along in later years was tuberculosis.  People would start to cough and keep coughing for months or years until one day they coughed up buckets of blood and died. 

 

Before this all began, Old English was already finding ‘gh’ uncool or uncouth – for whatever reason, it was turning gh’s into g’s and y’s.  Then it merged with a language that had no ‘gh’ or ‘kh’ sounds, Norman French.  At the same time, as people wheezed to their death all over the known world, any suggestion of clearing the throat must have been horrifying.  It was not a good time to say, “dauchter, micht ye go licht some candles – a knicht comes to visit and he micht plicht his troth to ye!”  (‘Plight his troth’ is an old expression meaning to propose his engagement for marriage.)  I think the back of the throat became a bad place for English people.   Nowadays, in literature, if we say someone speaks in ‘guttural’ tones, that is, in throaty tones, it is the next best thing to swearing about them.  It means they are a bad person.  They are speaking from the gutter, from the sewer of their mouth.  A new sense of throat hygiene may have swept the language, causing everything that was once guttural to become clean and bright – vowels nearly hopping over each other, in fact, in their effort to move forward and upward in the mouth and become ever brighter and toothier.  A Scot or a Dutchman coming in and spraying the back of his throat with ‘gh’ sounds may have seemed like a ghoulish reminder of death in the family.  Certainly, the idea that ‘gh’ is ‘clearing the throat’ is completely hard-wired in our culture now.  

 

Historical conjecture depends on literary support; scholars look through all the written documents from an era and see what notions were documented.  I don’t know what sort of track record ‘clearing the throat’ may have in connection with ‘gh’ and ‘kh’ sounds.   But there are some things, too, that people don’t talk about.   You will find no literary evidence at all of people confessing that they have a forceful, almost insane compulsion to write ‘Summerville’ when someone tries to spell the name ‘Summerbell’ for them.  Yet the Summerville-o-maniacs completely rule the English-speaking earth at this time.  Likewise, there is no predicting that nearly everyone meeting a dog named Practice would be culturally hammerlocked into saying “Practice makes perfect.”  Yet no political party could possibly gain the support level that this impulse had.  English sounds may have fled the throat and clustered in a refugee mob around the front of the mouth, and no one may have written a word about it.  We shall see. 

 

Contemporary use of the ‘horrible throat’ adduct isn’t hard to find.  I recently read an entertaining book called The Rude Story of English, by Tom Howell, and found this meme in the text very early on.  In mentioning an alternate candidate for the first English word ever recorded in Britain, Howell discussed some runic script found written on an animal knuckle-bone from about the mid 400s C.E. in East Anglia, eastern England.  The runes seemed to spell ‘raihan,’ which might be pronounced ‘raaeekhahn.’ Howell comments, “Charles Wrenn…noted that ancient Germanic people often mixed their letter h’s with an extra w sound, as in ‘why’ and ‘what,’ … pronounced ‘hwy’ and ‘hwat.’  If ‘raihan’ similarly contained a w sound in its middle, then an ‘attractive conjecture,’ as Wrenn called it, was to see the words spelling out the word “rahwhan.” … scholars know it as a magic word in Ancient Norwegian, a bit like ‘abracadabra.’ It was spoken as ‘rah-hhuh-wah-han’ with a throat full of phlegm [in other words, ‘rakhawakhan’ in the notation used in this book – RCS], and elsewhere…formed the first element of ‘Ragnarök,’ which was the name Norwegian storytellers gave to a coming apocalypse, the day when the gods would explode… (but) no one needs to gargle the vowels and consonants of ‘raihan’ and ‘rahwhan’ for long periods to produce the Celtic-British-Welsh translation of … Rowena…Rhonwen (etc.)  … [the daughter of Hengest].” 

 

The horror is all there, even in the midst of humour.  And the vowels are included. 

 

You may ask, of course, why didn’t this horror of throaty sounds take hold in other languages?  Well, my tentative answer is that cultures really ARE different, and they were even more so before the age of mass communication.   Even today, in the Netherlands, absolutely everything is done somewhat differently than in Canada.   There is not a single thing in life that can be approached in exactly the same way.  When it’s your birthday in the Netherlands, for example, YOU treat everyone you work or go to school with to some goodies or snacks – later, you may get some presents yourself at your birthday party (which you are more or less obliged to have) but you are always a giver on your birthday.  Then, at the party, all the arriving guests go around and wish birthday congratulations not just to you, but to everyone there.  It’s an extraordinary appealing culture, but it takes a long time to learn your way around it.  I think that the Dutch may have managed to express their anxiety about epidemics by turning disease names into swear words.  As I mentioned earlier, you are getting into the very worst language when you start saying ‘pest’ (plague), ‘kanker’ (cancer), ‘tering’ (tuberculosis) or ‘tyfus’ (typhus), especially in poignant combinations like (from the internet, cursing a television network’s failure to broadcast certain sports matches) ‘kankerhoerige tyfuszooi’ (‘cancer-whorish typhus-mess’).   (Profound apologies to all Dutch speaking readers who dislike bad language.)   I expect the notion of avoiding their own throats never occurred to the Dutch.   This was a peculiarly English form of cleanliness, one of the earliest of many Anglo-American attempts to take the animal nature out of the body and turn us into spotless, sexless, squeaky-clean religious-hygienic mannequins.  Dutch people can also talk about sex much more easily than English-world people can.  Saying ‘gh’ is not on their radar as an act of filth.

 

Many people who have speculated about the Great Vowel Shift have assigned some sort of role to plague epidemics.  It’s been proposed, for example, that many survivors whose ways of life had been destroyed by plague moved south towards London and picked up an emerging London middle-class accent.  Another idea was that the mingling of social groups caused by the need to find marriage partners after plague outbreaks led to a mixing of accents from the upper and lower classes of society.  

 

There are also explanations that don’t cite the role of plague.  The French element of English society was becoming assimilated and starting to speak English as a home language around the time of the plague outbreaks, and suggestions have also been made that the Vowel Shift was an attempt to mimic a prestigious French sound. 

 

These ideas may not all be correct, but they are all compatible with one another and with the idea I’ve put forward above.  The one and only special strength of my proposal, so far, is the very high number of primarily English-speaking people who will say “sounds like you’re clearing your throat,” or very similar words, whenever you tell them how the Dutch ‘g’ and ‘ch’ are pronounced.   Try saying the slightly atypical but understandable Dutch sentence ‘graag wil ik de gegrilde gerechten krijgen’ (pron. ‘ghraagh wil ik de ghe-ghrilda ghereikhta kreiygha’), ‘I will gladly take the [available] grilled foods.’  This should give you a good response from people carrying the epistemic adduct. 

 

I admit that I can’t prove this widespread cultural impulse to say ‘clearing the throat’ isn’t something that arose in our culture a very short time ago.  Since it is very widespread, though, even though it has not been popularized in any conspicuous way, I suspect it is a very old and broadly established idea.  I might think differently if Humphrey Bogart had famously said, “Clear your throat again, Gerard” (pronounced ‘Gheirart’) to a Dutchman in Casablanca, but nothing like that has ever happened, to my knowledge.   This impulse is much more basic than cool.  Rejecting ‘kh’ lies somewhere in our deepest cultural roots. 

 

Curiously, just when the English were moving away from ‘kh,’ the Dutch started making a move towards it.  Consider the case of the Dutch word ‘achter’ (pron. ‘akhter’) meaning ‘after.’  Back in the year 900, the Old English for ‘after’ was ‘æfter’ (pronounced ‘after’) and the Old Dutch was almost identical as ‘aftar.’  Some time around 1200, ‘after’ and numerous other words in Dutch began developing a ‘ch’ where ‘f’ had always been, even in older languages like Old Saxon.  The prototype word to Dutch scholars of this f --> kh shift is the word for an urban canal, ‘gracht’ (pronounced ‘ghrakht’). It began as ‘Graft,’ a place name having to do with digging.  The name was related to the Dutch verb ‘graven,’ to dig, and more distantly to the English words ‘grave’ and ‘engrave.’  The ‘f’ in this word migrated in the direction of the Dutch ‘ch’ in the 13^th century along with the other words showing this pattern.   Another example is ‘zacht’ (pron. ‘zahkht’), the Dutch word for ‘soft,’ which was ‘senifte’ in Old Dutch, ‘softe’ in Old English, and ‘saechte’ in Middle Dutch.   A couple more?   Number one: English ‘shaft,’ Dutch ‘schacht’ (pron. ‘skhakht’), Old English ‘sceaft,’ and Old Dutch ‘scaft.’ Number two: English ‘rafter,’ Dutch ‘rachter,’ Old English ‘ræftras’ (pron. ‘raftras’), and Old Dutch ‘rafter’ or ‘rachter.’   The Frisian language in the northern Netherlands was immune to this trend and, for example, has kept ‘efter’ for ‘after’ and ‘sêft’ for ‘soft’ to the present day.  There seem to be two things to be learned from this:  1) just when a mysterious socio-linguistic program was making English people increasingly avoid ‘kh’ in the Middle Ages, the reverse program was running in Dutch and building up the number of ‘kh’ words; and 2) the hidden connection between ‘kh’ or ‘gh’ and ‘f’, seen in words like ‘laugh’ and ‘cough,’ isn’t peculiar to English.   ‘Gh’ words may develop ‘f’ sounds, as in ‘laugh’, or ‘f’ words may develop ‘kh’ sounds, as in ‘achter.’  Words with ‘back vowels’ like ‘a,’ ‘o’ and ‘u’ always seem to be involved.  Linguists might say, “well, ‘kh’ and ‘f’ are both fricatives,” meaning sounds made by forcing air through narrow passages produced when we nearly close parts of the mouth.   That’s true, but how important is it?  One sound, ‘f,’ is pronounced near the front of the mouth and the other, ‘kh,’ near the back; there are several other fricatives, like ‘th’ and ‘sh,’ lying in between.  ‘Kh’ and ‘f’ don’t seem much like the same letter, and are pronounced quite separately one after the other in words like the Greek ‘thank you,’ ‘Ευχαριστώ’ (‘efkharisto,’ roughly ‘truly graced,’ found in English as the Christian term ‘eucharist’) and the Arabic term for the Coast Guard, ‘خفرالسواحل’ (‘Khfar as-Suwahhal’).  In any case, it shows that if English is illogical in producing ghoti-fish words where ‘gh’ has turned into ‘f,’ the strain of illogic goes back a long way in the West Germanic languages and isn’t unique to English itself. 

 

Spelling reformers are often gung-ho to change the ghoti words, as well as the words like ‘bear’ and ‘broad’ that got sidelined in the Great Vowel Shift.  They soon run into problems: in particular, it seems the words involved are especially likely to have sound-alikes with other spellings.  Correct ‘bear’ to ‘bare’ and you make a naked beast as a noun (for example, ‘I wuz out hiking and I saw a black bare’) or an act of stripping as an abstract form of carrying (for example, ‘my oeld frend, I bare you no ill will’).  If you read that some large trees had many bows (rhymes with ‘cows’), instead of boughs, this might lead you to imagine these trees were constantly bending over in Japanese displays of respect.  You could also misread and end up imagining the trees had many bows (rhymes with ‘toes’), that is, many ribbons decoratively tied to them.   Slangy North Americans saying ‘I’ve been coughin’ all day’ would end up saying, “I’ve been coffin’ all day,’ which would make their head colds seem deadly.  All right, I admit, none of this is more absurd than the existing way that English spells things, but it seems pointless to exchange old absurdities for new.   Especially when the old absurdities have a powerful genie in them, which is the history of the Germanic language family. 

 

Of course, there are other possible ways of dealing with our ghoti words.  For example, instead of introducing spelling reform, we could introduce pronunciation reform:  insist that English speaking people begin to pronounce ‘gh’ again as the Scots do.   Yes, ha ha ha.  You micht well lauch to see such an idea broucht up.  Still, it’s no less practical than spelling reform, which also requires popular consent. 

 

Or, thinking broadly, the United Nations could get all the languages that use the Roman alphabet to rationalize the correspondence of letters to sounds, making it possible for us all to use the same keyboard.  That way, we could correct the global spelling irregularity and make dozens of languages easier for students.  Here’s a fantasy phonetic script where the English ‘long’ vowel sounds are written with an acute accent (play = plá, see = sé, high = hí, low = ló, sue = sú) and reasonably similar sounds in other languages are written with the same symbols. Consonants are standardized in the English form. The French ‘u’ as sucre is ‘ü,’ the neutral schwa vowel is ě (sucre = sükrě). Let’s put together a sentence that switches from language to language. 

 

Mársé bókú, méyá améché é améká, ik vint dázě édá leük, ama üzgünüm, Í dón’t think Í’ll implěment it túdá. 

 

This sentence breaks down as: “Merci beaucoup (‘thank you,’ in French), miei amici e amiche (‘my friends [male and female],’ in Italian), ik vind deze idée leuk (‘I like this idea,’ in Dutch), ama üzgünüm (‘but I am sorry,’ in Turkish), I don’t think I’ll implement it today.”  Such a system would do most violence to French, which has a very different and original use of Roman letters. If that caused protest bonfires to be lit on the Champs-Elysées (Shanhz-Elézá), we could simply agree to write everything in French phonics (with fudges where necessary, for example to represent ‘th’ sounds): 

 

Merci beaucoup, mié amitchi é amiqué, hic vinde déze idée leuc, ama uzegunume, Aï dône’t thhinque Aï’lle eimplemainte hitt toudé.  

 

Becoming practical for a moment, I think it’s easiest just to recognize that English spelling is relatively stable.  The best way to learn it, other than brute memorization, is to make it interesting.   I hope that by revealing the stories it tells, I am making it easier for any of you who are trying to learn to spell some of our words.  Now, test yourself.  Can you recognize, in that last sentence beginning with “I hope,” three words that probably got stuck and stayed behind while the other words were moving in the Great Vowel Shift? 

 

Look at ‘any,’ which could have shifted to ‘aynie’ to rhyme with ‘zany;’ the Dutch form is ‘enig,’ pronounced ‘einigh.’  The ‘e’ sound in ‘enig’ lies part way between the ‘a’ sound in ‘any’ and the one heard in ‘zany.’  The ‘a’ in ‘any,’ therefore, is about as close as you can come to the Dutch sound within the range of typical English sounds.  ‘Learn,’ which started off as the Old English ‘leornian’ (pronounced ‘leh-ornian’) could have shifted to ‘leern’ but instead is even throatier than its Dutch equivalent ‘leren,’ pronounced ‘leiren.’  The other ‘-earn’ words, ‘earn’ and ‘yearn,’ also stayed down in the throat, as did ‘earth’ and ‘early,’ but not ‘nearly,’ ‘ear,’ ‘fear’ and many others.  ‘Word,’ based on Old English ‘word,’ could have moved to rhyme with ‘ford’ (from Old English ‘ford’), ‘sword’ (from Old English ‘sweord’) and ‘lord’ (from Middle English ‘loverd,’ Old English ‘hlaford’) but instead, it is much closer in sound to its Dutch equivalent ‘woord,’ pronounced approximately ‘woert.’

 

Dutch immigrants are famous for assimilating very well into English speaking societies. In our language, though, as shown by the words above and by some of the examples mentioned earlier, like ‘bear,’ ‘swear’ and ‘father’ (Dutch ‘vader’, pron. ‘vah-der’), there are many ancient proto-Dutch words that are still stubbornly wearing their wooden shoes and eating whole herrings in one gulp.  Yes, a language full of windmill-words with their North Sea Germanic pronunciations intact or nearly so.  Now you can recognize some of them. 

 

Words starting with ‘w’ are a particularly busy nest of old-fashioned pronunciations.  The example ‘word’ that I gave above is only the beginning of the story.  Among words starting with ‘w’ or with ‘sw,’ there are some normal long-‘a’ words like ‘wary’ and ‘wake,’ but there are only two commonly used words with a normal short ‘a,’ namely, ‘wag’ and ‘wax.’  There are also a few slangy words, ‘wangle,’ `wacky` and ‘wanker.’  Nearly all the others are deviant, either saying ‘a’ as in ‘father’ or short ‘o’ as in ‘or:’  ‘wad,’ ‘waddle,’ ‘waffle,’ ‘waft,’ ‘walk,’ ‘wall,’ ‘wallaby,’ ‘wallet,’ ‘walrus,’ ‘wallop,’ ‘wallow,’ ‘waltz,’ ‘wampum,’ ‘wan,’ ‘wand,’ ‘wander,’ ‘want,’ ‘war,’ ‘warble,’ ‘ward,’ ‘warm,’ ‘warp,’ ‘warrant,’ ‘warren,’ ‘wart,’ ‘wash,’ ‘wasp,’ ‘wassail,’ ‘watch,’ ‘watt,’ and ‘wattle.’  There is also ‘was’ with, in northern North American English, a short ‘u’ sound.  A few extra words like ‘swarm,’ ‘swallow,’ ‘swan,’ ‘swap,’ ‘swat,’ ‘twaddle’ and the nasty ‘twat’ (insulting slang for ‘vagina’) can be added that start with other letters.  Many of the words have similar sounding Dutch counterparts like ‘wafel’ (‘waffle’), ‘wand’ (‘wall’), ‘wandelen’ (‘wander’), ‘warm’ (pronounced ‘w[v]ahrum,’ meaning ‘hot’ rather than ‘warm’), ‘was,’ (pron. ‘w[v]ahss,’ ‘was’), ‘wassen’ (‘wash’), ‘zwan’ and ‘zwaluw.’ 

 

Also in the ‘wo-’ group, the normal English short ‘o’ is rare in commonly used words, and is mostly known from ‘wobble.’ Otherwise, it is found in Australian import words like ‘wobbegong’ (a flattened Pacific Ocean shark with whisker-like growths on its face and attractive spots on its back) and ‘wombat.’  Long ‘o’ occurs here and there, as in ‘woke.’ With the rest of the ‘wo-’ words, short and long ‘u’ sounds rule, along with the unnamed ‘put’ vowel, as in ‘wolf,’ ‘woman,’ ‘womb,’ ‘won,’ ‘wonder,’ ‘word,’ ‘work,’ ‘world,’ ‘worm,’ ‘worry,’ ‘worse,’ ‘worship,’ ‘worsted,’ and ‘worth.’  Again, the Dutch often roughly corresponds, as with ‘wolf,’ ‘gewonnen’ (‘won’), ‘wonder’ (which means ‘a miracle’), ‘woord,’ ‘werk,’ ‘wereld,’ (pronounced ‘w[v]eirelt,’ ‘world’), and ‘waard’ (pron. ‘w[v]aahrt,’ ‘worth’).   The most curious aspect of this is that the pronunciation ‘woh’ (‘w’ + short ‘o,’ to rhyme with ‘straw’) that is so rare in ‘wo-’ words is the normal North American pronunciation of ‘wa-’ words like ‘wander.’  So it’s not as if ‘w’ – which sometimes alters pronunciation of things that follow it – could have blocked the ‘woh’ sound as a pronunciation for words starting with ‘wo-.’   Similarly, words like ‘war,’ ‘warm,’ ‘ward’ and ‘wart’ could be phonetically spelled ‘wor,’ ‘worm,’ ‘word’ and ‘wort’ in English, but instead, this ‘wor’ phonetic spelling is rare and other pronunciations occupy most of the spellings listed (there is no ‘wor’ in English, but ‘wort’ can be either a plant, such as a liverwort, or the liquid that is fermented to make beer.  In either case, it can be pronounced ‘wert’ or as the phonetic ‘woart.’) 

 

Somehow, then, ‘w’ seems to have served as a German preservation license for Middle English words.  Perhaps this was because the sound was unknown in French.  ‘Wh-’ and ‘qu-’ (pron. ‘kw’) words also got caught up in this trend, though, even though many of the latter derived from French.  We have, for example, ‘wharf,’ ‘what,’ ‘who,’ ‘whore,’ ‘squab,’ ‘squabble,’ ‘squad,’ ‘squat,’ ‘squawk,’ ‘quadrangle,’ ‘quaff,’ ‘qualify,’ ‘quality,’ ‘qualm,’ ‘quantity,’ ‘quarrel,’ ‘quash,’ ‘quasi-’ and ‘quaternary.’   You can pick out a few good Latin words in there that mostly came to us via French, such ‘quality’ and ‘quantity.’  Again, the trend was broken up by a few rare exceptions that became normally English in sound, but most of them are sound-imitating words like ‘quack,’ ‘wham,’ ‘whack,’ ‘whap,’ ‘whomp’ and ‘whop.’  Ironically, ‘quagmire’ is among the few words that managed to escape this quagmire (swamp) of Germanic pronunciations.  ‘Swamp’ itself, as you’ll notice, did not escape.  Too bad, because this scary linguistic swamp also includes the monstrous sasquatch, otherwise known as ‘bigfoot,’ ‘yeti’ or ‘abominable snowman.’ 

 

The letter ‘w’ does get credit for helping to make ‘women’ a particularly strange word.  The word began as ‘wifmen’ (‘wif’ being related to the word ‘wife’), the plural of ‘wifman,’ and condensed into ‘wimmen’ in Old English.  Gradually, however, according to Douglas Harper, the author of the wonderful Online Etymology Dictionary, “the rounding influence of ‘-w-’” changed the pronunciation of the singular form to ‘woman,’ with the ‘o’ standing in for the unspecifiable ‘put’ vowel.  A spelling change followed.  The original pronunciation of the plural, ‘wimmen,’ remained even after the spelling had changed.  Some feminists in the 1970s began to spell the word this way again, or as ‘wimmin’ (to remove the unwanted comparison to males), but it’s not yet clear if this spelling will live on.   No matter how bizarre the spelling of an English word is, in any case, there is always some sort of an explanation.   It’s a bit like a teenager who can excuse all bad behaviour. 

 

To the extent that there is regular, sensible spelling in English, it tends to be restricted to words that obey our English version of the old Germanic rules for when vowels should be long or short.  Before I talk about the English spellings that distinguish long from short vowels, I’ll just mention, for comparison, what is done in Dutch.  There is an important overlap with the English system despite differences that have developed over the centuries. 

 

Dutch is a nearly phonetic and relatively grammatically regular language, so its system is quite simple.  Syllables (parts of words) are classified as ‘open’ if they end in a vowel, or ‘closed’ if they end in a consonant.  In a word like ‘poten’ (‘animal feet or furniture legs’ pronounced ‘poe-ten’, probably related to ‘paws’ in English) the first syllable, ‘po’ is open, and the second, ‘ten,’ is closed.  If you have two consonants at the junction point of the two syllables, then the first syllable automatically becomes closed.  The example here is the word ‘potten’ (meaning ‘pots,’ pronounced very close to what it looks like, ‘potten,’ if you remember to pronounce the ‘t’s).  In real life, the ‘n’ at the end may not be distinctly pronounced.  In any case, in principle, the first syllable ‘pot’ is closed, as is the ‘ten’ that follows.  Now, the rule for long and short vowels is that a vowel in an open, stressed (spoken as the loudest syllable in the word) syllable is always long, while a vowel in a closed, stressed syllable is normally short.   The long ‘o’ sound in ‘poten,’ ‘poe-ten’ (which would rhyme with the English word ‘voting’ if that were said in North American casual slang as ‘votin’’) arises because ‘po-’ is an open, stressed syllable.

 

‘Poten’ is a plural, and it so happens that that the singular word for ‘animal foot’ is a one-syllable word rhyming with the English word ‘vote.’   In Dutch, it can’t be written ‘pote,’ since that would be a two-syllable word pronounced ‘poe-tuh.’  Instead, a double vowel is used to show that the vowel is long: ‘poot’ (still pronounced ‘pote’ to rhyme with ‘vote,’ not like the dreaded English quasi-word ‘poot’ to rhyme with ‘boot’ – this ‘word’ was used relentlessly by hippie-era humor and comic strip writers to indicate the sound of a fart, and became famous in music through scatological rocker Frank Zappa’s description of young teen-boy activities – “whizzing and pasting and pooting through the day.”) 

 

Similarly, Dutch has one ‘boot’ (pronunciation and meaning both ‘boat,’ as you may remember from previous discussion) and two ‘boten,’ boats.  The rule about open, stressed syllables being long applies to all sorts of words, not just plural nouns.  A good pair of examples are the verbs haken (ha - ken, pronounced haahken, literally ‘to hook,’ meaning ‘to crochet’) and hakken (hak - ken, pronounced almost as ‘hucken,’ meaning ‘to chop’).   The noun corresponding to ‘haken’ is, predictably, ‘haak,’ meaning a ‘hook.’

 

The main remnant we have of this system in English is the ‘silent e’ system for making long, or so-called ‘tense’ vowels.  As humour songwriter Tom Lehrer wrote in a children’s song: 

 

Who can turn a cap into a cape?

Who can turn a tap into a tape?

A little glob becomes a globe instantly,

If you just add Silent E.

 

This system arose in Middle English, when the silent ‘e’ was still pronounced as a short neutral vowel, a schwa.  In a passage like the one below from Geoffrey Chaucer, all the ‘e’s at the ends of words were probably pronounced (also, ‘cleped’ was probably pronounced ‘clep-ped’). 

 

This yonge man was cleped Piramus,

And Tisbe hight the maid, Naso seith thus;

And thus by report was hir name y-shove

That, as they wexe in age, wex hir love.

 

The Legend of Thisbe of Babylon, lines 19-22. 

 

(Translation:  This young man was named Piramus

And Thisbe was the girl’s name, [the poet] Ovid says thus;

And in this way, in gossip, were their names together shoved:

that as they grew in age, so grew their love.)

 

It doesn’t really read well in the original unless you read it pronouncing the e’s, in the Middle English way:

 

This yongeh man was clep-ped Peeramus

And Tisbee heekht the maid, Naso saeeth thus;

And thus bee report was hir naameh ee-show-veh

that as they wexeh in aahjeh, wex hir low-veh.

 

Now, if you’re more interested in spelling than in love stories, you will have noticed that the words that end in ‘e’ in the passage mostly consist of open syllables.   ‘Name’ breaks down as ‘nah-meh,’ ‘shove’ breaks down as ‘sho-veh’ and ‘love’ breaks down as ‘lo-veh.’  Each word has the Middle English version of a long vowel in the first syllable.  There is also ‘wexe,’ but because ‘x’ actually represents a fusion of two letters, ‘k’ and ‘s,’ it breaks down as ‘wek-se,’ with short vowels.   In the Great Vowel Shift, ‘name’ was moved forward to become ‘naym,’ while ‘shove’ and ‘love’ are among the exceptional words that went backwards to their current pronunciation.   Several words with similar spelling have a long ‘o’ today, as in ‘stove,’ ‘grove’ and ‘rove.’  The English ‘silent e’ convention, then, represents a relic of a system that produced open syllables, and thus produced the long vowel sounds that went with those open syllables.  We say ‘bake’ as ‘bayk,’ with a long ‘a,’ because Middle English had ‘bake,’ pronounced ‘bah-keh,’ as a Dutch-style word with two open syllables.  Here’s Chaucer again, from Summoner’s Tale lines 1730 and 31, with a religious person talking about about the torments of hell, where, according to medieval thought, your flesh may be ripped with hooks, or burned or baked.  

 

“Ful hard it is with flessh hook or with oules

To been yclawed, or to brenne or bake.” 

 

Very hard (on you) it is, with meat-hooks or with awls,

to have been clawed, or to (be) burned or baked.

 

An ‘awl’ is a pointed device used to make holes in leather. 

 

English also took up many French and Latin words into this spelling system:  for example, the Latin ‘globus’ became the English ‘globe,’ with a suitable silent ‘e’ to indicated the long ‘o’ sound in the middle of the word.  In the Tom Lehrer verse quoted above, ‘cape’ is also from Latin, while ‘tape’ is from Old English.  The silent ‘e’ system was part of the blending of Germanic and Romance elements within English. 

 

I’ll come back to Latin in a moment, but first, I want to note that the Dutch-style format of using double letters for long vowels, as seen above in ‘poot’ and ‘haak,’ is still very much a part of Anglo-Saxon-derived English.  Strangely, though, we only hung on to this format in its normal form for the letter ‘e.’   Thus we have many words like ‘beet,’ ‘cheek,’ ‘deep,’ ‘reed,’ and ‘steel.’   This format didn’t take over all the one-syllable long-e words, though.  Many words have overlapping sounds in Modern English, but came from different sounding words in Old English:  examples are ‘beat,’ ‘read’ and ‘steal.’   ‘Read,’ for example, comes from West Saxon ‘rædan’ (pronounced ‘raddan’) or Anglian ‘redan’ (pronounced ‘ray-dan,’ related to Dutch ‘raden,’ pron. ‘raw-den,’ ‘to advise’), while ‘reed’ comes from Old English hreod (pron. ‘hrayode’ or ‘khrayode,’ related to Dutch ‘riet,’ pronounced ‘reet.’).  Only in Modern English do these two words meet up as having the same sound.  As you can see, English, in its mildly crazy form of logic, has maintained a difference between the two words only in the way they are spelled.  This does help us to distinguish them in text, I suppose, though it’s hard to imagine when this would be useful.  “As I snuck through the flooded library at dawn after the hurricane, a large duck flew out of the reads.”  If English expected to win a logic prize with this difference, though, it then immediately ruined its chances by having the past tense of ‘read’ (pron. ‘reed’) spelled as ‘read’ (pron. ‘red’).  Yes, a verb with a Great Vowel-Shifted present tense, and a stuck-in-the-starting-gate past tense with the same spelling.  Now that’s just weird.  In sentences like ‘the women read the book’ there is no immediate way to know if the reading is ongoing now, or if it happened in the past – though this is mainly a textbook problem, since in any real situation, the ongoing present would be expressed as ‘the women are reading the book.’  Still, I wouldn’t submit English for a ‘Language of the Year’ prize with this sort of messy stuff to catch the judges’ eye.  But you can picture the dilemma of some poor 12^th century monk sitting there with writing quill in hand, thinking, “hmm, shall I write ‘yesterday I red the book’?  No, that won’t work, looks like a red ink spill – I give up, leave it as ‘read.’”

 

The use of ‘ee’ became well enough established that it moved out of closed syllables into open ones, like ‘see’ and ‘free’ (‘seon’ and ‘freo’ in Old English) and even became extended to some Norman words, like ‘discreet’ (Old French discret, pronounced ‘deescreh’).    

 

As further trouble for our logic prize, we have the double-o words like soon, moon, took, and book.  None of them has an ‘oak,’ ‘own’ or other long-‘o’ sound as would be predicted from the Dutch system, and many of them actually bear the sound of long ‘u,’ so that ‘soon’ usually rhymes with ‘June.’  One common feature the long-‘u’ ‘oo’ words all have, though, it that none of them have a ‘yoo’ sound as many other long-u words have in conservative English dialects.  ‘Cartoon,’ will never be pronounced ‘cartyune,’ whereas a word spelled ‘cartune’ might well be pronounced in that way, making it sound like a song on the car radio.  A raccoon, unlike a ‘raccune,’ has no danger of becoming a ‘rakyune’ (almost like ‘rancune,’ the nasty French word for a grudge) and most importantly, a coot will never be a ‘cute.’  (Actually, the coot, a black coloured water-bird that lives in the Amsterdam canals and builds its nest there out of plastic bags fished from the water, is rather cute, but when we call an elderly gentleman an ‘old coot,’ that certainly shows that we don’t think he’s an ‘old cute.’)  The Australian town of Coober Pedy, the Opal Capital of the World, is protected against becoming Cuber Pedy, a name that would force it to give up gems and take up obsessively solving Rubik’s Cube.  Again, then, there is a little method in the madness of English spelling. 

 

One of the mildly hilarious features of English is that it has rather thoroughly applied the Germanic ‘stressed open syllable is long’ rule to Latin words and to words derived from Latin.  Words from languages related to Latin, such as Portuguese and even, in many cases, French, also are pronounced using this system.  I have already mentioned the case of the botanical tree name Pinus (pron. ‘pye-nuss’ in English, ‘pee-noos’ in Latin) for the pine tree, vs. penis (pronounced ‘pee-niss’ in English and ‘peh-neess’ in Latin).   Even English peculiarities like our substitution of ‘yoo’ for long ‘u’ get involved in this – as you can see in the Latin-derived words ‘peculiarities’ (‘pe-KYOO-lee-AIR-i-teez’) and substitution (‘sub-sti-TYOO-shyun’ in many dialects, including mine).  The ‘y’ sounds did not exist in the Latin ‘peculiaris’ (‘having to do with personal property,’ from ‘pecu,’ meaning ‘cattle herd’) or ‘substitutus’ (‘put in place of another, from ‘sub-,’ meaning ‘under,’ and ‘statuere,’ meaning ‘set up’).

 

In fact, our rules for pronouncing Latin and Latin-like words from the Romance languages (let’s say ‘Latinesque’ words to cover them all) in English are complicated, and are often disobeyed.  You are getting close to the genius level in linguistics when you can perceive all the patterns involved, and all the irregularities that break the patterns.  The rules, as in the Anglo-Saxon words we’ve been dealing with so far, relate to which syllables within a word are stressed.  If you are curious and not afraid of detail, the main rule for Latinesque words in English is that ‘the second-to-last syllable is stressed unless it is an open syllable ending in a short vowel.’  Here we’re talking about a short vowel in the corresponding bit of Latin itself, not in the English pronunciation, though they often correspond.  For real Latin words, a Latin dictionary will show which vowels are short and which are long.  Diphthongs like ‘ae’ count as long.  The rule continues:  ‘whenever an open syllable ending in a short vowel is encountered as second-to-last syllable, move the stress to the third-to-last syllable.’  For typical (sometimes Graeco-) Latin-derived words with stressed second-to-last syllables, try ‘axis,’ ‘asylum,’ ‘aorta,’ ‘panacea’ and ‘rhododendron.’  Then, compare ‘genius,’ ‘camera,’ ‘medium’ ‘chrysanthemum’ and ‘platypus’ for words where the stress jumped back to the third-to-last syllable because the second-last was open, with a short vowel. 

 

Another bit of rule is ‘a final, open syllable is pronounced long, like the ‘go’ in ‘ego’ and the long ‘i’ that ends ‘octopi.’ 

 

Ready for more?  ‘A vowel followed by two consonants is usually short, since the first of the two consonants automatically closes the syllable it is in – unless the two consonants are “p,” “t,” “c” or “k” followed by “l” or “r.”’  (Too much, stop!  I’ll come back to that one).  Lastly, ‘a vowel becomes long when it is in an open syllable that is followed by another syllable consisting of just a vowel.’  This is seen in ‘Leo,’ ‘pleonasm’ (meaning ‘a verbal redundancy,’ like ‘Sahara desert,’ which means ‘Desert desert’), ‘theory,’ ‘myopia’ (short sightedness), ‘cloaca’ (pron. ‘cloe-AY-ca;’ the part of a bird that allows urine and feces to mix together and be aimed collectively at your head) and ‘creation.’

 

With this set of rules, you can interpret difficult words like ‘ignominious’ (meaning ‘dishonourable’).  The stress in this word is on the third-to-last syllable ‘min,’ not the short-vowelled, open, second-to-last syllable ‘i.’ That gives us ‘igno-MIN-ious’ instead of ‘ignomin-EYE-us.’  Since the stressed syllable ‘min’ is ‘closed’ (ending in ‘n’, a consonant), the vowel ‘i’ at its centre is in the short form, so that ‘min’ rhymes with ‘sin,’ not ‘sign.’  The front part of the word, ‘igno,’ is nearly treated like a separate word.  It consists of the semi-stressed, closed syllable ‘ig’ and the unstressed open syllable ‘no.’  The ‘no’ that ends the part-word is an open syllable, so the ‘o’ becomes long.  This ‘no’ is usually reduced to a neutral vowel or schwa in normal pronunciation of ‘ignominious,’ but if you were shouting the pronunciation at someone, you would say ‘ig – noe – minious,’ making the long vowel in ‘no’ clear to anyone who was listening.  Long vowels in the unstressed second syllable are even more clearly heard in similarly constructed words like ‘prototypical’ ‘photochemical,’ and ‘topographical,’ as well as in the differently stressed ‘monosyllabic.’   Nonetheless, all the vowels in these second syllables are shown as schwas in the Merriam Webster dictionary, suggesting that Merriam and Webster listen to very hasty speakers.    

 

This pronunciation of ‘ignominious’ may seem very odd to anyone who understands the Latin roots of the word. It might seem that ‘no’ should have the ‘m’ from the next syllable attached to it, making it ‘nom.’  ‘Ignominious’ literally means ‘loss of a good name’ and the ‘nom’ you see there is the part of the word that means ‘name.’  But breaking the word down as ‘ig-nom-in-i-ous’ is simply not how English handles syllables (I don’t think we had better get into the topic of how to break a word down into syllables in English; just check a dictionary).  The formation of ‘min’ leaves us with ‘igno’ as the front part of the word, whether or not this makes etymological (word origin) sense.  Pronunciation goes according to English rules, not according to Latin logic. 

 

There are many puzzling aspects to the English pronunciation of Latinesque words.  It took the genius-level reflections of Noam Chomsky and Morris Halle, in their 1968 book The Sound Pattern of English, to express the rule they called ‘trisyllabic laxing.’  The word ‘genius’ here serves as a warning to you that the next bit of writing is going to be a challenge for you to get through, just as it was for me.  Chomsky and Halle noticed that one- or two- syllable words with a long vowel in the final syllable would almost always be related to some three-syllable words with a stressed short (they preferred to say ‘lax’) vowel.  In the longer word, the syllable with the vowel of interest had to be followed by one unstressed syllable and a third syllable, which could be stressed or unstressed.  This combination of three syllables is, of course, what the word ‘trisyllabic’ in Chomsky and Halle’s phrase refers to.  For example, when the verb ‘provoke’ (‘pro-VOKE’) gave rise to the three-syllable adjective ‘provocative,’ (‘pro-VOC- a -tive’) the long ‘o’ of ‘provoke’ disappeared and the ‘voc’ syllable with its stressed short vowel was followed by two syllables, the unstressed ‘a’ and the unstressed ‘tive.’   Chomsky and Halle produced an impressive string of parallel examples, such as serene vs. serenity, impede vs. impediment, profane vs. profanity, grateful vs. gratitude, provide vs. provision, and the slightly deviating ‘school’ vs. ‘scholastic.’   The rule even applied to diphthongs, such as ‘profound’ vs. ‘profundity’ and ‘pronounce’ vs. ‘pronunciation.’  These ‘trisyllabic laxing’ words do mostly appear to follow the general rules given two paragraphs above, since their stress is mainly on a closed third-to-last syllable preceding an unstressed open syllable with a short vowel, as seen in ‘pro-VOC-a- (short, open) -tive.’

 

In fact, in many cases, you can turn the short vowel in that second-last syllable into a long vowel just by changing the final syllable to a two-syllable suffix, as seen in ‘provocation,’ ‘prov-o-CA-tion.’  By the way, my breakdown of syllables here generally follows the Merriam-Webster dictionary.  Oddly, however, although ‘-tion’ is generally considered one syllable in English, pronounced ‘shyun’ or ‘shun,’ it seems to function as an open, short ‘ti’ plus a closed ‘on’ in its effect on pronunciation of Latinesque words.  It makes stress jump to the third-to-last syllable and, if that syllable is open, turns the vowel long.  See also derive, derivative and derivation; divine, divinity and divination; profane, profanity, and profanation.  Not to mention ‘pronunciation,’ which we have already seen.  The Latin origins of the ‘tion’ and ‘sion’ suffixes certainly lie in two-syllable stems, as seen in Latin words like ‘pronuntiatio,’ meaning ‘proclamation.’  The final ‘ti-o’ consists of two distinct syllables.  In English, if ‘tion’ were treated as one syllable instead of two, ‘derivative’ and ‘derivation’ would be completely parallel, and the latter would be ‘de-RIV-a-sh(y)un’ due to ‘trisyllabic laxing.’   

 

Isn’t this fun?   (‘No!!’ you may be saying.  Well, it’ll be over soon.)

 

No one memorizes these rules unless they are studying linguistics in university, and yet, all native English speakers develop a ‘feel’ for them as they become more educated.  In my profession, we regularly come up with new Latin or Graeco-Latin words as names for newly described sorts of fungi, and the native English speakers among us rapidly come up with a standard pronunciation.   Our pronunciation is usually typically English and sounds barbaric to, for example, our German colleagues.  We have no doubt that ‘Rhizomucor,’ a type of mould, should be pronounced ‘RYE-zoe-myoo-cor’ (or, more orthodoxly, Rye-zoe-MYOO-cor), not the classical ‘REEZZ-oh-MOO-cor,’ the German ‘RHIT-zo-moo-cor,’ the French ‘Hree-zoh-mü-cohr’ or the Japanese ‘rye-zuh-mü-cohrü.’  (The Japanese tend adopt the English long vowels in such words – they treat Latinesque technical terms as imported English.)  We do, however, maintain random bits of classical pronunciation: we are not entirely consistent, and have little field-specific traditions that no one explicitly notices.  Take, for example, our specialized technical name for the most common mould spore in outdoor air worldwide, something that you always have at least five of in your lungs at any given moment.  The name is ‘Cladosporium,’ pronounced Clad-o-SPOR-i-um (meaning ‘little branching spore’), where the short ‘a’ in ‘Clado’ (Greek ‘klados,’ ‘branch’), rhyming with ‘bad –oh,’ attempts to preserve the classical Graeco-Latin use of short ‘a’ as in ‘father.’  No classical language actually contains the English short ‘a’ sound that North Americans pronounce there, but we don’t care about that.  In fact, we would never dream of using the actual classical pronunciation, which would sound like ‘cloddo’ to our ear, suggesting a lump of soil.   Meanwhile, the much more broadly known scientific word ‘cladogenesis’ (meaning ‘generation of branches on the conceptual diagram of the “evolutionary tree of life,”’ with humans branching out of the ape family line, and so on) has been anglicized according to the normal rules, making it ‘clay-do-GEN-e-sis.’  Cladogenesis is shown on diagrams called ‘cladograms’ (‘CLAY-do-grams’) and is analysed by people who may call themselves ‘cladists,’ (‘CLAY-dists’), performing the type of analysis called ‘cladistics’ (cla-DIS-tics, where the second-last syllable is closed and thus kidnaps the stress away from the ‘a.’).  Only fungus names are ‘claddos’ rather than ‘claydos.’   This fungal rule only pertains to the front half of a word, though:   Cladobotryum and Cladophialophora are usually ‘claddos,’ but Ulocladium is a ‘clay’ word, ‘Yoo-low-CLAY-dee-um.’  It seems there is no limit to the English capacity for producing inconsistency in language. 

 

On looking through the list of fungi causing human disease, I realized we even have some regional accents in our botanical Latin:  currently most of us have a pseudo-classical pronunciation of the beautiful name Exserohilum (meaning ‘little peg sticking out’): we say ‘Ex-seir-o-HEEL-um.’  People from the southern United States, however, are very likely to use the more conventional English pronunciation ‘Ek-ser-o-HYE-lum,’ which with a full southern accent becomes ‘Ek-ser-uh-HAH-lum.’  The fungus mainly causes nasal sinus infections and miscellaneous medical problems in people living in the southern states and Texas, so it may be unconsciously promoting this pronunciation of its name by giving southern doctors and scientists something to say at conferences.   So far, though, its campaign to have its name become typically Anglo-Latin has not had wide success.  

 

There are many mini-rules involved in the pronunciation of Latinesque words, but I will just show you a couple of examples.  It would be mind-numbing to try to document them all.   As I briefly noted above, the letters ‘p,’ ‘t,’ ‘c’ and ‘k’ tend not to split away from an ‘r’ or ‘l’ that follows them when words are being split up into syllables.   This head-scratcher rule explains why ‘microscope,’ which by the normal rules ought to be ‘MICK-ro-scope,’ is actually ‘MY-cro-scope,’ with an open first syllable that has a long vowel.  Likewise, ‘cyclone’ is ‘SIGH-clone’ not ‘SICK-lone,’ thanks to this rule.  English speakers instinctively apply this rule even to very non-Latin words like ‘Popocatepetl,’ a Mexican volcano named ‘smoking mountain’ in the Nahuatl language.  We tend to say ‘poe-poe-CAT-e-pet-ul,’ whereas our Anglo-Latin rules would agree with the Nahuatl that the stress should be on the ‘te’ syllable, ‘poe-poe-ca-TE-pe-tul’ (there’s an unstressed, open second-to-last syllable ‘pe,’ so the stress jumps to the third-to-last, ‘te;’ we would have a long ‘e’ there, whereas Nahuatl has a short ‘e.’).  The name is officially written in Mexican Spanish as ‘Popocatépetl’ to show this accent.  The trick is, though, that, we perceive the ‘tl’ as a consonant-cluster that can’t be split, like the ‘cr’ in ‘microscope,’ and this makes ‘petl’ one syllable for us in a word that splits up as ‘poe-poe-CAT-e-petl’.  Thus the stress jumps back to the new third-to-last syllable, ‘cat.’  And indeed, in Nahuatl itself, ‘petl’ is one syllable, since the ‘tl’ at the end is not really ‘tl’ at all, but rather a single sound, a ‘lateral t’ (a ‘t’ pronounced from the side of the tongue when the mouth is in the position we use to make an ‘l’ sound.)   So if Popocatapetl actually were a Latin word, our handling of it as an import into English would be correct.  The ‘tl’ would function as a unit like the Greek ‘ch’ and ‘ph’ (see below).   

 

Another rule is that two-letter combinations that have a single sound in Greek count as one letter.  This rule covers ‘rh,’ ‘th,’ ‘ph’ and ‘ch.’  The last of these is not our English ‘ch’ but rather χ, the letter ‘chi,’ pronounced ‘kh’ in Greek and ‘k’ in Latinized Greek.    This is why ‘pathos’ in our theatrical dramas is ‘PAY-thos’ rather than ‘path-os’ (some people, however, do say ‘path-os’) and why our pronunciation of ‘psychology’ starts with ‘sigh’ rather than ‘sick.’  An ‘aphid’ must be an ‘AY-fid,’ not an ‘AF-fid.’  On the other hand, the single Greek letters ‘xi’ (ξ) with an ‘x’ sound and psi (ψ) with a ‘ps’ sound are treated as two letters, as are their Latin equivalents.  Thus ‘ochraceous’ (coloured yellow brown like ochre stone, Fe₂O₃) starts with long ‘o’ (‘oh-KRAY-sh[y]us’) while oxalic acid (a chemical that we eat every day in vegetables or nuts but that sometimes turns against us by forming kidney stones) has a short ‘o’ (‘ok-SAL-ic’). 

 

Speaking of Greek, even though most of the Greek words in English have come to us modified by contact with scholarly Latin, they are somewhat distinct in spelling from pure Latin import words.  They can often be recognized by the letter combinations ‘rh’ and ‘ph,’ which don’t occur in native Latin.  The ancient Greek ‘rh’ is a typical rolled ‘r’ as heard in Spanish or Dutch, while ‘ph’ is an air-filled, aspirated ‘p’ as heard in English ‘pan.’  Neither of these noises was native to Latin speech.   Latin also lacks a ‘ch,’ so any word generally looking like it might be Latin but containing this letter group is probably Latinized Greek instead.  A few examples are ‘Christmas,’ ‘chromatogram,’ ‘choreography,’ ‘schizophrenic,’ ‘oligarchy,’ and ‘monarch.’  When the ancient Romans encountered Greek culture, they recognized that they had encountered a highly developed civilization, and they imported all sorts of cultural ideas.  To support these ideas, large numbers of Greek words were imported into Latin, and some whole chunks of Latin grammar, such as the Greek subsections of the first and second declensions (Latin systems of modifying words for different genders, numbers and grammatical cases), became dedicated to dealing with Greek words.  Medieval scholars continued the tradition of treating Greek words as imports into Latin.  They routinely changed spellings and word endings to fit the Greek words into the Latin mold.  Even today, when scientists are coining a new name for a plant or fungus, our international code of naming rules strongly recommends that the names be ‘sufficiently Latinized.’ 

 

When scholarly words filtered into the languages of ordinary people, any Greek words that were included all came in Latinized form.  That is why, apart from a handful of modern Greek words like ‘baklava’ and ‘ouzo,’ all Greek that is imported into English is treated as Latin in terms of pronunciation and syllable stress.  Greek words in their original form do have characteristic stressed syllables, but the Greek pattern of stress is never imported into English.  This makes life easier for us.  In fact, we tend to take liberties and also treat the neighbouring Slavic languages as if they were Latinized Greek.  This is why we always say ‘BOR-is’ for the Russian name Борис (Boris), which is really pronounced ‘Bah-REESS,’ and, most barbarically, say ‘EYE-van’ for Иван (Ivan), which is really ‘Ee-VAHN.’   Well, now you know the explanation:  in Ivan, for example, the open, second-to-last syllable ‘I,’ which happens to be the first syllable, tends to get the stress and a long vowel sound according to the Anglo-Latin rules I’ve given above.  Just as in ‘idol’ and ‘item.’  If Russians think differently, we can’t deal with it. 

 

In general, our language seems to have the underlying assumption that if words aren’t Anglo-Saxon, they must somehow be derived from Latin.  In the last chapter, I mentioned the heavy Latinization of our classical Arabic words like ‘algebra’ and ‘alchemy.’  If one looks at place names derived from languages that English speakers find impossible to pronounce, most of them have a distinctly Latin look to them.  For example, in British Columbia, my home province of Canada, we have a city named Coquitlam.  This name has a distinctly Latin ‘c’ and ‘qu.’ The original term in the Sto:lo (pron. Stoe-loe) language is now spelled Kwikwetlem, and means ‘red salmon going upriver.’   This is an easy one.  In a more extreme case, the fishing port and tourist destination Ucluelet, near Pacific Rim National Park, is a fancifully Latinized representation of Yuułu’ił’athh (IPA ju:ɬuʔiɬʔatħ) from the Nuučaan̓uł (Nootka) language, where the ‘ł’ – just like the Welsh ‘ll’ for those who know it –  is a lateral ‘s’ (prepare your mouth to say ‘l’ as in ‘lemon,’ then hiss air out the sides of your mouth) and the apostrophe (’) marks are glottal stops (as in Cockney ‘bo’ol’ [bottle] or household English ‘’uh’uh!’ meaning ‘no no no’).  The name, so difficult for the English speaker, renders reasonably well into Welsh as ‘Iŵllŵ’ill’atch,’ as long as the Welsh speaker doesn’t mind inserting foreign glottal stops.   Coquihalla, our handy highway pass through the Cascade Mountains, represents the Sto:lo place name ‘Kw'ikw'iya:la’ (to make a kw’ sound, close the back of your throat so that you can’t breathe, round your lips up to where they’d be if you were making an English ‘w,’ and then make a popping ‘k’ sound inside your mouth.  The ‘a:’ is a conventional long ‘a,’ ‘aah’).  Now look at the stresses on those native names in English:  Co-QUIT-lam, U-CLUE-let, Co-qui-HAL-la – 100% according to our Latin rules, with stressed second-last syllables that are either closed, or open with a long vowel.   There you have it, the secret Latin of aboriginal British Columbia. 

 

Dutch is also a challenge to pronounce, which gives us quasi-Latinisms like ‘The Hague’ (‘the HAYG’) for Den Haag (‘Den Haagh’) and ‘YOO-trekt’ for Utrecht (‘Oo-trekht’).  (‘Hague’ is an interesting invention, though:  note the Anglo-Saxon-derived ‘silent e’ at the end, making the ‘a’ long, and the Anglo-Norman style ‘u as hard-g protector’ that precedes it.  Quite the hybrid.  Similar to English ‘vague,’ from Latin ‘vagus.’)  Still-unlatinized Dutch city names like Nijmegen (‘Neiiy-may-ghen’) only baffle us, though I solved the problem in my own mind by thinking of it, in a Cockney music hall accent, as ‘Nidgy Megan.’  We could always use its original name, Noviomagus (Latinized ancient Celtic for ‘Newfield;’ the full name was Ulpia Noviomagus Batavorum, meaning the Batavian [ancient Dutch] Newfield of the Roman Emperor Trajan’s Etruscan family clan, the Ulpii).  One way or the other, we have to Latinize in order to grasp these names.  

 

Only Chinese words, which can’t be Latinized without producing bizarre flights of fancy like ‘Confucius’ (孔夫子; Kǒng Fūzǐ, pronounced Kǒng Fū-dzěuh), tend to appear as pseudo-Anglo Saxon, as seen in ‘chow mein’ (炒面, ‘cháau mihn’ in Cantonese, chǎo miàn in Mandarin, meaning stir-fried wheat noodles).  Early missionary schemes for writing Chinese in Roman letters tended to look somewhat Graeco-Latin and somewhat Germanic, giving us approachable names like Hong Kong and Peking.  Later American academic scripts tried to be more phonetic without completely departing from Latin, giving us, for example, the Cantonese equivalents Hèung Góng and Bàkgìng.   The Communist-era Pinyin scheme for writing Mandarin Chinese, however, is a departure.  It is based, like the Maltese version of the Roman alphabet, on the principle of scavenging otherwise un-needed Roman letters for new purposes – hence ‘q,’ not needed for any ‘qu’ sounds, is assigned to a ‘ts’ sound coming off the roof of the mouth (hard palate), ‘c’ to a ‘ts’ sound coming off the alveolar ridge of the mouth (the ridge that holds the tooth sockets), ‘x’ to an ‘hs’ sound from the roof of the mouth, ‘r’ to what is often a ‘zh’ sound, and so on.   (This last one may seem bizarre, but it has a European precedent in the Czech ř, pronounced approximately ‘rzh,’ as seen in the name of the composer Antonín Dvořák). The now routine use of Pinyin spellings like Beijing (though not Xiāng Gǎng, the Pinyin version of Hong Kong) in English text is just one of the many ways in which we are finally getting away from Latin, after all these years.   I’ve already mentioned our other main de-Latinizing strategy, which is to express as much as possible in acronyms.   Even there, though, the most successful and natural-looking acronyms usually look like Latin.  What could be more Latin than ‘quantas,’ the normal pronunciation of Qantas, the acronym for the Queensland and Northern Territories Aerial Services?  Latin is, of course, the source of the words ‘quantum’ and ‘quantity.’  If ancient Romans were with us today, they would surely all be happy to fly on Qantas. 

 

Latinized though many of our imported words may be, we still tend to pronounce ‘ch’ as in ‘church’ except in our words officially imported from classical Greek, where ‘ch’ becomes ‘k’ as in ‘charisma.’  We also make a few exceptions for imports from modern French, where ‘ch’ becomes ‘sh’ as in ‘chardonnay’ and ‘charades,’ and Italian, where it is a ‘k,’ as in ‘orchestra’ and ‘chiaroscuro.’  Outside those restricted zones of different ‘ch’ sounds, the ‘church’ sound rules.  We have no doubt, for example, that another import from aboriginal American, ‘Cochabamba,’ (a Bolivian city whose original name, Qhochapampa, means ‘lake plain’ in Quechua) is ‘koe-cha-BAM-ba,’ not ‘koe-ka-BAM-ba.’  Perhaps this is one of the reasons why, in the academic world, it is considered very bad form to produce a new word that combines primary word roots from two different languages, such as Greek and Latin.  Scholars try hard to avoid making new hybrid words like ‘television’ (mentioned earlier) and ‘homosexual’ (first root Greek, second root Latin – yes, the word itself is an established, non-traditional couple).  We try to retain our Latin words and our Greek words as separate linguistic environments, each with its own distinct properties.  

 

Our words derived from Greek and Latin are largely based on using a small group of well known roots over and over.  This makes it easier for us to recognize automatically when we are dealing with a word that needs to be given a Greek pronunciation.  Of course, this type of recognition takes practice, but I would guess that most native English college students, when they first run into an unfamiliar word like ‘stoichiometry’ (pronounced ‘stoy-kee-OM-etry’ and meaning, ‘calculating the numbers of various sorts of atoms involved in chemical reactions’), recognize automatically that the ‘ch’ needs to be pronounced as a ‘k.’  The ‘meter’ root at the end of the word is connected to a whole nest of related Greek-derived words like ‘chronometer’ (‘time measure,’ note the ‘ch’ with a ‘k’ sound),  ‘diameter’ (‘across measure’), ‘barometer’ (‘weight measure’), ‘perimeter’ (‘around measure’), ‘optometry’ (‘vision measurement’), ‘thermometer’ (‘warmth measure’), ‘trigonometry’ (‘three angle measure’), ‘parameter’ (‘alongside measure’),  ‘metronome’ (‘measure ruler,’ with the same ‘nom-’ root as we saw earlier in our discussion of ‘economy,’ ‘household rule’), ‘isometric’ (‘equal in measure’), ‘symmetry’ (‘together measurement’), ‘pentameter’ (‘five measure,’ referring to poetry where lines have five beats), ‘odometer’ (‘road measure’) and ‘metric,’not to mention the name of the popular dinosaur with the giant upright fin on its back, ‘Dimetrodon’ (‘two measure tooth,’ i.e., ‘two tooth sizes’).  It’s not that each student thinks, “These words are Greek;” rather, he or she just reacts to a language environment where familiar things happen, and those familiar patterns happen to be based in Greek.  Even naughty hybrid words like ‘speedometer’ (half Anglo-Saxon, half Greek) haven’t erased the long-established patterns.

 

Incidentally, a speedometer should be a ‘tachymeter’ in all-Greek or a ‘speedmat’ (or maybe ‘speedmathe’) in all-Saxon.  No matter: it’s clearly far too late to stop the hybrid ‘meter’ word phenomenon.  A look at today’s Google top hits for ‘o-meter’ reveals a mortgage lender’s ‘implode-o-meter,’ a television channel’s ‘gay-o-meter’ (‘to find out how much of you is dying to come out of the closet’), a ‘stress-o-meter,’ and a potential alcoholic’s ‘drink-o-meter.’  If there are any truly conservative language scholars left, they must surely feel the needle jump on the ‘headache-o-meter’ when they see these innovations.  

 

One bit of Greek that causes some spelling trouble is the diphthongs.  These are groups of two vowels that are common in Greek and that, for many years, were dutifully represented as double letters in Greek words adopted into English.  In the original Greek, the problem diphthongs consist of two letters printed in the ordinary way as αι (ai) and οι (oi).  In Latin, these foreign sounds were transcribed as special symbols consisting of two letters shoved especially close together (‘ligatures’), æ and œ.  These same ligatures were introduced into Old English when it was first written in Roman letters, but their original purpose had nothing to do with Greek.  They were used to represent the Fuþorc runes called ‘æsc’ (pronounced ‘ash’) and œðel (pronounced ayðel, you know, ‘ay,’ ‘th’-as-in-‘the,’ ‘el’).  The latter was little used but the former was a common letter.  Its sound was later taken over by the ordinary short ‘a’ as in ‘ash,’ and it fell from use.  In the meantime, as English began importing Latinized Greek words, the two ligatures were given their Latin function of representing the Greek diphthongs.  In later centuries, many English speakers decided they had no time to deal with these Greek diphthongs in English words, especially when the symbols that were needed were left off the standard typewriter keyboard.  In American spelling in particular, the æ and œ diphthongs fell completely out of use and were replaced by the letter ‘e.’  Just to give a starting example, ‘diæresis’ (pron. dye-ER-e-sis) is often written simply as ‘dieresis.’  That’s the word for the two little dots above the second vowel in words like naïve, where two vowels that usually form a diphthong are pronounced separately. 

 

These problem diphthongs, however, have remained dear to the hearts of classical scholars, since they sometimes distinguish similar words.  For example, our little clump of common Greek root words includes ‘pedon’ (πέδον), the soil, as well as ‘paed-’ or more formally pæd-, the usual combining form (based on the genitive case) of ‘pais’ (παίς), the boy.  Neither of these should be confused with our Latin root ‘pes,’ the foot, which combines with other words as ‘ped-’ – see pedal, impede, quadruped and centipede.  The Greek ‘boy’ root is often generalized to include girls as well, giving it the meaning of ‘children.’  It has given rise to a series of words referring to education, including ‘pædagogy’ (‘teaching,’ literally ‘child leading;’ pron. ‘PED-a-go-jee’) and ‘encyclopædia’ (which with finesse says ‘all-round child teaching tool’ but at the literal level of root-words says, ‘in-a-circle boy-process.’).  Anyone who mostly reads American writings will encounter these words as ‘pedagogy’ and ‘encyclopedia.’  The same ‘boy’ root has also given us an English version of the ancient Greek word ‘pæderast’ (‘boy lover-eroticizer’), an adult sexual lover of teenaged boys, and the relatively modern scientific term ‘pædophilia’ (‘child loving attraction’).   This last word, coined by Austrian psychiatrist Richard von Krafft-Ebing in 1886, has shifted in meaning since about 1980 from the original ‘sexual attraction to preadolescent children’ to the more general ‘sexual attraction to preadolescents and adolescents’ and finally to the current newspaper emotion-zapper, ‘uncontrolled sexual aggression against children.’  Linguists have often stated that properly speaking, the ‘soil’ root ‘pedon’ should combine into ‘pedo-’ words, as it does already in ‘pedology,’ the study of soil properties, ‘pedogenesis,’ soil formation, and ‘pedosphere,’ the ecological soil zone.  Whenever the ‘boy’ root is used, they say, it should combine into ‘pædo-’ words.  Thus, ‘pedophile,’ the typical American spelling for ‘pædophile,’ should literally mean a lover or a friend of the soil.  Headline writers, however, prefer shorter words to longer ones.  Most of these terse people are not pitching in to help save the diphthong in ‘pædophile’ and may even reduce the word down to ‘pedo.’  

 

The Greeks had four quite distinct words that are all translated by ‘love’ in English.  ‘Philia’ (φιλία), the root of words ranging from ‘pædophilia’ to ‘Philadelphia’ (‘[place of] the love of brothers’), wasn’t strongly connected to the sexual.   The philosopher Aristotle publicized the word as meaning dispassionate, virtuous love, such as the love involved in loyalty to one’s community.   Scientists typically ‘love’ what they work on in this sense, and I would proudly call myself a mycophile.  This well-established word, also mentioned earlier in this chapter, translates as ‘fungus lover.’  Coming back to the diphthong problem, though, I wouldn’t recommend that a soil scientists’ society print joke T-shirts saying ‘We are the true pedophiles.’  The wording might excite people other than diphthong advocates.

 

In a similar vein of language mischief, I’ll suggest that we men shouldn’t let our shirt collars become encyclopedic (stained inside with a complete ring of soil).   

 

The list of fairly common words where the diphthong problem comes into play includes (just to give the ligatured versions) æsthetic, anæsthesia, hæmoglobin, cœlacanth (pron. ‘SEAL-a-canth’) and diarrhœa.  The blood-filled body cavity of insects, the hæmocœl (pron. ‘HEEM-o-seal’), wins a prize for having both.  Fortunately, all English speakers have forgotten that ‘federal’ and related words, such as ‘federation,’ originally came into our language as fœderal, fœderation, etc.  Canadians are very multicultural these days, but you won’t soon see us rewriting ‘confederation,’ the process that brought our country together, as ‘confœderation.’ 

 

For practical purposes, the words with Greek diphthongs in them can mostly be written in their American forms (esthetic, anesthesia, hemoglobin, diarrhea) unless a formal scientific name is involved, as in coelacanth or the fungal group coelomycetes.  If you want to use British English, though, you are best advised to employ the diphthongs, except where, as in fœderal, they are completely extinct in the written language.  You can just type the normal letters in:  you need not use a ligature character unless your editor insists. 

 

Interestingly, American scientists have begun spelling ‘hæm,’ the oxygen-carrying molecule of blood, as ‘heme,’ but can’t yet bring themselves to write ‘hæmocœl’ as ‘hemocele.’  Instead, you will see ‘hemocoel,’ as well as other related words such as ‘blastocoel’ (the liquid centre of a very young human embryo).  Those pesky diphthongs are very hard to round up and herd completely out of our language, even in laboratory cowboy country. 

 

It’s been pointed out that Americans rigorously stick to the diphthongs in ‘paean,’ ‘amoeba,’ ‘oedipal,’ and ‘Caesar.’  To that list, I can add ‘phoenix,’ ‘Aesop,’ ‘aegis’ and the all-important ‘subpoena.’  It’s very pleasant to know that Americans share the general English language talent for spelling irregularity. 

 

So far in this chapter, I have dealt with some broad trends that have contributed to the weirdness of English spelling and pronunciation:  the political development of spelling, the use of a foreign script, the diversity of English and the impracticality of going phonetic, the development of ‘kh-avoidance,’ the Great Vowel Shift, our Great-Vowel-Shifted, West Germanic pronunciation of words derived from Latin, and the tendency to treat words from many different languages as that same type of mutant Latin.  Now, I’d just like to spend a little time talking about some smaller effects, some oddball mini-trends that also helped make English spelling the delightful mess it is today.

 

Many of these mini-trends resulted in silent letters.   One of the features that sometimes make English seem like a language junkyard is that there are leftover, unpronounced letters all over the place.  Many of these old bones are in our Anglo-Saxon words.  A few are in our Greek, and a very few came from elsewhere.  Perhaps the most fun is the sometimes-silent ‘g’ in ‘gnu,’ the name of an African antelope more commonly called the wildebeest.  This ‘g’ began its existence as a click made with the tongue on the roof of the mouth in a Bushman word now spelled ‘!nu’ – the ! indicates the click in question.  OK, if you must know, you put the tip of your tongue on the roof of your mouth up at the top, just back of the last of the series of little ridges that start behind the teeth.  The back of your tongue then naturally curls up to touch the back of the roof of your mouth, making an air enclosure above your tongue.  Then just briefly tighten the air seal with the tip of your tongue and pull it down sharply with a loud ‘click!’ (sounds like a ‘tok!’).   Whew, that was hard.  The word was then made a little easier in related Khoikhoi or ‘Hottentot’ languages of far southern Africa, where it appeared as i-ngu (pron. ‘i’/ ‘ng’ as in ‘sing’/ ‘u’) or t’gnu.  Thanks to explorer Georg Forster in 1777, it entered the Dutch language as a typical raspy ‘gh’ in the word ‘gnoe’ (pron. ‘ghnoo;’ the ‘oe’ in Dutch is always as in our English word ‘shoe’).   On its entry into English, it ran into a pre-existing idea that ‘g’ before ‘n’ at the beginning of a word is always silent, as in ‘gnash,’ ‘gnat’ and ‘gnome.’ It was therefore eliminated from pronunciation.   The resulting word, however, was uncomfortably similar to ‘new,’ regardless of whether the latter was pronounced ‘noo’ or ‘nyoo.’  Some people now solve this problem by humorously adding an exaggerated ‘g’ at the beginning when they say ‘gnu,’ just to emphasize that they are pronouncing a silent letter.  It is the world’s most forcefully pronounced ‘silent’ letter.  Maybe we should just go back to the click. 

 

The silent ‘g’ pattern that ‘gnu’ blundered into had already been established among Anglo-Saxon words.  There are only a few silent ‘g’ words, but there are fair number of silent ‘k’ words, which are similar in origin.  When you see these words, you can be sure that nearly all of them will have a Dutch equivalent where the ‘k’ is pronounced.  For example, ‘knight’ in English corresponds to ‘knecht,’ an old Dutch word for a male servant, very properly pronounced ‘k’nekht.’   ‘Kn,’ if you pronounce it, causes the tongue to suddenly change direction, and many of the ‘kn’ words are for things or concepts that also do that, in particular, joints:  see ‘knee’ and ‘knuckle’ (‘knie’ and ‘knokkel’ in Dutch).  A ‘knot’ (‘knoop’ in Dutch, pronounced ‘k’nope’) also makes a twist, as does something that is ‘gnarled.’  The word ‘gnarled’ is interesting: it may have been invented by William Shakespeare, and its twisted, knotted nature makes an interesting contrast to a mean-spirited public official in this little essay from his play, Measure for Measure.

 

(Isabella says):  “Could great men thunder

As Jove himself does, Jove would ne'er be quiet,

For every pelting, petty officer

Would use his heaven for thunder;

Nothing but thunder! Merciful Heaven,

Thou rather with thy sharp and sulphurous bolt

Split'st the unwedgeable and gnarled oak

Than the soft myrtle: but man, proud man,

Drest in a little brief authority,

Most ignorant of what he's most assured,

His glassy essence, like an angry ape,

Plays such fantastic tricks before high heaven

As make the angels weep; who, with our spleens,

Would all themselves laugh mortal.”

 

(That is, if officials and rulers had the powers of a weather god like the Roman Jove, they would be making thunder and lightning all the time just so they could look intimidating.  Real lightning and thunder don’t selectively attack the weak and vulnerable, but instead often attack the ‘gnarled’ and useless items that resistantly clutter the landscape, like hilltop oaks too twisted to be used for timber.  People, given authority, pretend to know a lot, and are most fierce where they have least knowledge.  Their actions then are silly and cruel and would make angels cry; but if the angels had human dispositions, they would laugh until they lost their immortality.)

 

“Gnarled” probably relates to a pre-existing Middle English word ‘knar,’ meaning a knot in wood.  Shakespeare seems to have adapted it to make it more ‘gnarly.’  Yet another gnarly ‘kn’ word is ‘knead,’ the word for folding dough over with your hands in baking, and similar processes where you cause soft materials to buckle.    

 

‘Kn’ also sounds percussive, which may explain how it got into ‘knock’ and ‘knell’ (the sound of a bell).  Dutch has ‘knokken’ and ‘knallen;’ the latter means to making a popping or kabooming sound.  Dutch also has other percussive ‘kn’ words, like ‘knarsen,’ to creak or to grind the teeth, the equivalent of our English ‘gnash.’     Arguably, ‘knagen’ (pron. ‘k’nawghen’), to ‘gnaw,’ is also a percussion word based on the sound of rodent teeth scraping away at something.  The Dutch van Dale Etymologische Woordenboek (van Dale’s Etymological Dictionary) calls most of these words ‘klanknabootsende’ (pron. ‘klunk-na-boat-send’), meaning ‘sound-imitating.’ ‘Knorren’ is Dutch for the oinking of pigs, and to my ear, it sounds quite realistic.  Very klanknabootsende. 

 

‘Knetteren,’ the verb meaning to make sudden loud sounds, combines into the richly textured word ‘knettergek’ (pron. ‘k’netter-ghek’), meaning ‘stark raving bonkers,’ or noisily ‘barking mad.’  The similar English word ‘to natter’ is only known from 1829 at the earliest, but it probably derived from a northern English dialect word ‘to gnatter,’ to chatter or complain.  (It seems the silent ‘g’ can disappear from spelling if it is only found in dialect.)  Dutch happens to have a sound-imitating ‘kn’ word meaning to nag or constantly complain, ‘kneuteren,’ so if you listen at a distance to your family members nagging in English or Dutch, you may hear some sounds that sound like gnattering or kneutering.  Perhaps in modern English where ‘kn’ sounds are gone, they only natter or, in the case of disagreeable spouses, neuter. 

 

All these dramatic twisting, knocking and grinding effects are lost if the ‘k’ or ‘g’ goes silent.  Sometime around the time of the Great Vowel Shift, though, the English people clearly tired of chewy, earthy language and went for more heavenly sounds.  This meant not only raising vowels and moving consonants away from the throat, but also removing startling consonant clusters from the beginnings of words.   Therefore, we lost the fun of being able to gnaw on our language with a hard ‘g,’ and to make it knock with a hard ‘k.’ We are smooth, smooth, smooth.  Only our spelling is left as a bumpy place, a chewed material strewn with historical rodents’ nests.

 

A very few ‘kn’ and ‘gn’ words are not Anglo-Saxon in origin, but rather Greek and Latin.  Greek has given us ‘gnosis,’ transcendent knowledge.  Perhaps from this, the 16^th -century German occult physician Phillip von Hohenheim, under his nom de plume (writer’s name) Paracelsus, spun out the fictional word ‘gnome’ for the spooky, wise earth spirits that, these days, live as little statues in the garden and wear red hats.  The Greek island of Crete gives us the place name Knossos for the site of the magnificent temples of the lost Minoan civilization.   

 

Classical Greek is a major source of our silent letters. This is partly because it had many words starting with the trident or pitchfork letter Ψ, ‘psi,’ which said ‘ps’ as it does in modern Greek today.   These ‘ps’ words included the root words of psychology, pseudonym, psoriasis and psalms.  When you get into specialized terminology, the homeland of our Greek, then you can put together an amazing collection of rare silent letters.  ‘Chthonic,’ pronounced ‘thonic,’ is quite a find; it has a silent Greek χ (‘chi’) at the beginning, rendered in English as ‘ch.’  In Greek, the ‘ch’ is pronounced, of course, as the dreaded ‘kh.’  The word means ‘related to the spiritual underworld.’  Then there’s ‘phthisis,’ a word for any disease that causes you to gradually lose weight and ‘waste away’ to nothing.  In English, it’s pronounced ‘þizis’ (‘thizis’ if you’re confused by my use of the fuþorc letter ‘thorn’ there) but in Greek, it’s ‘f þissis.’  The punchy Greek words ‘pneuma’ for the breath and ‘pneumon’ for the lungs have given us nice, smooth words with silent ‘p’s for ‘pneumonia,’ ‘pneumatic,’ the AIDS-defining fungal infection ‘pneumocystosis,’ and so on.   And let’s not forget ‘pnigophobia,’ the phobic fear of being smothered.  If using in a hiphop lyric, pronounce the ‘p.’

 

One of my long time favourite silent letter words is ‘mnemonic,’ with silent ‘m.’  It refers to something used as an aid to memory, and is therefore, like the word ‘curt,’ self-referring – since what could be more distinctive and memorable than a silent ‘m’ word?  I tend to pronounce the ‘m’ anyways, since without it, in my North American pronunciation, the word really does sound just like ‘pneumonic,’ as in the deadlier form of plague.  Also, it’s more fun that way.  And, still collecting gems of Greek silence, you have ‘pterodactyl’ (‘wing fingers’), a flying dinosaur, and many other technical words based on Greek ‘pteron,’ meaning a wing.  The ‘pt’ is pronounced in mid-word, as in ‘helicopter’ (‘helical wing’).  Poking around dictionaries has revealed ‘pterylology,’ the study of how birds’ feathers are arranged.  There’s a specialization for you.  A psychrophile (pron. ‘sigh-crow-file’) is someone or something that loves the cold.  A Canadian, for example.  ‘Ptomaine’ food poisoning is well known, but who knows ‘ptochocracy,’ government by the poor?  Maybe it’s never happened.  It sounds like ‘toke-ocracy’ when said out loud, which suggests the equally unlikely scenario of government by marijuana puffers.  ‘Ptosis,’ the condition of having a drooping eyelid, is a nice one.  The strangest of all is ‘ptarmigan,’ an arctic game bird related to grouse.  The name derives from the Scots Gaelic ‘tàrmachan’ and the silent ‘p’ was added by someone who either mistook the word for Greek, or simply loved silent ‘p’s.  Pnice work.  Finally, saving the best til last, there are the jellyfish words.  I remember as a boy seeing creatures washing up on the seashore near Ucluelet, British Columbia, that looked like gelatinous, see-through grapes with an antenna inside.  As an avid little biologist, I soon found out they were the ‘ctenophores’ or ‘comb jellies.’  Pronounced, of course, as ‘10-0-4s’ or ‘10:04s.’  They have two long swimming appendages with comb-toothed structure: the Greek root for ‘comb’ is ‘κτεις,’ ‘kteis,’ combining as ‘kteno-.’  Then there are the regular, normal jellyfish, technically known as the cnidarians (pron. ‘nid-AIR-ians’).  There’s some poetry in this:  deep in the silent seas, the jellyfish all have silent ‘c’s. 

 

Incidentally, don’t be fooled by the Italian-English word ‘gnocchi,’ lumpy pasta that is often made from potato flour.  There is no silent ‘g’ here.  The Italian ‘gn’ is a distinct sound: it sounds like ‘ny’ if you pronounce it carelessly, or a little more like ‘ngy’ (‘ng’ as in ‘sing’), if you’re more careful.  The ‘ng’ is reduced to a sticky little trace of the full sound as heard in ‘sing,’ and the whole thing comes together as ‘n(g)yoak-key.’  The same sticky pasta sound heard at the beginning is also found near the end of ‘lasagna.’  Technically, this sound (ɲ in IPA) is like an ‘ng’ pronounced further forward in the mouth, with the tongue against the hard part of the back of the roof of the mouth (the palate) rather than further back against the soft part (the velum), which is where the English ‘ng’ hangs out.   Welcome to the English language, ‘palatal nasal,’ as linguists call the pasta-‘ng’ sound.  The regular ‘ng’ is ‘velar nasal.’ 

 

Luckily, no English word begins with the other odd Italian ‘g’ sound, ‘gl’ as in the Italo-English words ‘intaglio’ (carving so that the image of interest is sunken into the stone) and ‘seraglio’ (an Ottoman-style harem).  The letter ‘l’ in English is normally pronounced lightly off the sides of the middle of the tongue, with the tip of the tongue coming up to touch the roof of the front of the mouth.  Merging ‘g’ with ‘l’ in the Italian way forces the air constriction back to the back of the tongue.  This change makes it very heavy sounding: it nearly stops the breath as it is being pronounced.  Many English speakers ignore the heavy ‘l’ and just say ‘intalio’ and ‘seralio.’  There is, though, a long-standing trend for Italian in English to be both spelled and pronounced as in Italian.  It’s a prestige thing, especially with musical terms like ‘andante.’  Now you know why I got so upset as a 10-year-old hearing some of my classmates talking about eating ‘ps-ghetti.’  Italian in English is a test of your status as a citizen of the world and as a bearer of intelligence, and those children were evolving back towards the apes.  So yes, unless you’re the sort of person who says, ‘them damn yurra-peeans,’ try to stick some proper sticky pasta gluten in your pronunciation of ‘gnocchi,’ and learn to apply the air brakes to the back of your tongue when you’re reading words off the menu like ‘tagliatelle’ (strip-cut pasta from Emilia-Romagna province in northeastern Italy, pron. ta-[g]lee-a-tell-ay) and ‘zabaglione’ (egg custard with a shot of sweet wine, pron. ‘zaba-[g]lee-oh-nay).  Incidentally, ‘zabaglione’ (also seen on menus under its French name ‘sabayon’) may be the only contribution to our language of the long-extinct Illyrian language that was spoken in Roman times in the area around modern Albania.  The original word was ‘sabaium,’ meaning a type of beer.  That gave rise to the ‘zabagli’ part of ‘zabaglione’, and then the Italians added the ‘-one’ as a suffix meaning, more or less, ‘the great one.’  

 

Coming back to the comb jellies, they have interesting silent letters no matter where you look, because ‘comb’ is just as odd a word as ‘ctenophore’ is.  It turns out that in Middle English during the Great Vowel Shift, a few consonant clusters were often interpreted as making vowels placed before them take on the long form.  ‘Comb,’ derived from Old English cámb, is one of the few ‘mb’ examples; ‘dumb’ was held back in the throat while ‘climb,’ raised itself to a higher sound.  Vowels in similar words that came from French didn’t go into the long form: ‘tomb’ went off in its own unique direction, while ‘bomb’ and ‘aplomb’ kept the short vowel sound.  ‘Tomb’ had actually been spelled ‘tumbe’ in Anglo-Norman and ‘tumba’ in Latin, consistent with its modern sound, but was spelled ‘tombe’ in Old French; it seems someone ‘corrected’ its Middle English spelling.  Like ‘tomb,’ most ‘mb’ words originally derived from two-syllable words, e.g., ‘climb’ from ‘climban’ and ‘comb’ from a two-syllable West Germanic root that also yielded the verb ‘cemban’, ‘to comb.’  ‘Bomb’ was from Latin ‘bombus,’ a buzzing or booming sound, which explains why the scientific name of the bumblebee is Bombus.  You needn’t worry that bumblebees will explode.  The expression ‘to do something with aplomb,’ meaning to ‘do it with confidence so it’s lookin’ good!’ in the most modern terms, comes from the idea of building construction being done exactly right with the ‘plumb line’ (piece of shaped lead metal on a string) – the French for this was ‘à plomb,’ from the Latin ‘plumbum,’ lead.   But just when the ‘mb’ story seems to tie together, the exceptions come along.  The words ‘thumb’ and ‘limb’ had been ‘þuma’ (‘thuma’) and ‘lim’ in Old English and picked up a ‘b’ for no good reason.  It’s interesting, though, that ‘limb’ acquired its ‘b’ just about at the same time English imported another ‘limb’ word, meaning the edge of a disc-shaped piece of equipment like a navigator’s quadrant, from Latin.  The Latin word ‘limbus’ for a border or edge may thus, through confusion, have supplied a ‘b’ transplant for the Old English ‘lim,’ making it also become ‘limb.’ 

 

While I was looking at jellyfish articles in researching this chapter, I happened across the wonderful site, jellieszone.com (I hope it lasts a while) and found the following description of comb-jelly feeding: “Small individuals (of the species Lampea pancerina) glomb onto the bodies of salps and consume pieces, much like a parasite.”   Note the ‘glomb’ there:  that’s the slangy verb ‘to glom,’ a piece of criminal underworld slang that recently graduated into casual English, meaning to snatch or grab onto something.  The ancestor of this word was the Scots Gaelic ‘glam,’ to snatch or devour.  ‘Glom’ seems to have acquired a ‘b’ in the author’s text for the same reason ‘thumb’ and ‘limb’ did centuries ago:  it just feels like the sort of word that ought to have a silent ‘b’ at the end.  Aren’t we English speakers strange?  Incidentally, just to make the quoted piece of text understandable, a ‘salp’ is a small floating marine animal.  If the sentence seemed slightly odd when you read it, that’s because it has a serious construction error:  it should read (even with the ‘glomb’), “Small individuals glomb onto the bodies of salps and consume pieces, much as parasites would (do),” or even better, “Small individuals, acting much like parasites, glomb onto the salps and start to consume their bodies piece by piece.”  We’ll come back to that sort of thing later, in chapter 4.     

 

‘Mb’ was rather half-hearted as a vowel-raising cluster.  A more consistently successful one was ‘ld,’ as in ‘child,’ ‘wild,’ ‘mild,’ ‘gold,’ ‘fold,’ ‘told’ and so on.  It didn’t work for ‘a’ words, as seen in ‘bald’ and ‘scald,’ though the ‘a’ there is the old ‘long a’ sound from Old English.  ‘E’ was also unchanged, as in ‘meld’ and ‘weld,’ though these words both had unusual sources (meld from the Old French ‘melder,’ to mix, not from Anglo-Saxon, and ‘weld’ from ‘well’ as in ‘make well’).  In any case, ‘baled’ and ‘wield’ already exist, so it’s good that ‘bald’ and ‘weld’ didn’t change their sound.  Also, ‘meld’ pronounced as ‘mealed’ would sound related to ‘meal.’  Inconsistency seems to have worked out well for clarity with these words.   Meanwhile, miraculously, the Old English ‘feld’ actually managed to change its spelling when it became raised to ‘field.’ 

 

The last such consonant group of note is ‘nd,’ which raised Old English ‘blind’ (rhymes with ‘tinned’) to the modern ‘blind’ (rhymes with ‘mined’).  There is a pile of words with similar histories, such as mind, kind, behind, and rind.  There is also a group of verbs where the present tense and past tense were raised together.  ‘Find,’ ‘grind,’ ‘wind’ (the verb) and ‘bind’ were ‘findan,’ ‘grindan,’ ‘windan’ and ‘bindan’ in Old English, rhyming with ‘tinned man,’ and their past tenses were ‘fand,’ ‘grand,’ ‘wand’ and ‘band,’ sometimes appearing as ‘fond,’ ‘grond,’ ‘wond’ and ‘bond.’  These vowels in these past tenses were all raised to the diphthong ‘ou’ in ‘found,’ ‘ground,’ ‘wound’ and ‘bound.’  The same effect was also seen in ‘pound’ and ‘hound,’ which had been ‘pund’ and ‘hund’ in Old English.  Not all ‘-ound’ words fit this pattern:  some, like ‘round’ and ‘sound’ came from very different Anglo-Norman origins (in these cases as ‘rounde’ and ‘soun,’ related to Old French ‘roont’ and ‘son’).  The ‘nd’ ending did nothing to words that ended up as ‘a’ words, like ‘sand,’ ‘land,’ ‘band’ and ‘hand.’  ‘Wand’ was the odd man out.  Like ‘bald’ up in the ‘ld’ words, it took on the old long form of the vowel.  Its unusual origin, from Old Norse ‘vondr,’ a rod, may excuse its different evolution.   Or, as mentioned earlier, it may just have conformed to the general Germanizing effect found in words starting with ‘wa-.’

 

Words related to the ‘mb,’ ‘ld’ and ‘nd’ words with raised vowels weren’t necessarily consistent with them.  ‘Child,’ for example, had the unraised ‘children’ (we all know how difficult children are to raise) as its plural.  This was because its pair of closed syllables, ‘chil/dren,’ blocked the change to a long vowel.  This blocking of vowel-raising by a double syllable was a general phenomenon.  ‘Comb,’ for example, is nearly unrecognizable as a close relative of ‘unkempt,’ which now just means messy but originally meant uncombed.  On the other hand, the long vowels survive the adding-on of interchangeable suffixes, as in ‘binding,’ ‘bindery,’ ‘mindfulness,’ ‘childlessness,’ ‘kindliness’ and so on.  

 

One very sad story associated with these words is the story of ‘wind,’ the noun meaning ‘air movement.’  It began in Old English rhyming with ‘tinned’ but successfully became raised to rhyme with ‘dined.’ In the 18^th century, though, it dropped back to the short-vowel form we use today, perhaps to harmonize with ‘windy.’  As the author of the Online Etymology Dictionary points out, this was “a sad loss for poets, who now must rhyme (wind) only with sinned and a handful of weak words.”  John Donne had been able to say, in a poem published in 1633:

 

Go and catch a falling star,
Get with child a mandrake root,
Tell me where all past years are,
Or who cleft the Devil's foot;
Teach me to hear mermaids singing,
Or to keep off envy's stinging,
And find
What wind
Serves to advance an honest mind.

 

Or in Donne’s original spelling:

 

Goe, and catche a falling starre,  

  Get with child a mandrake roote,  

Tell me, where all past yeares are,  

  Or who cleft the Divels foot,  

Teach me to heare Mermaides singing,         

  Or to keep off envies stinging,  

          And finde  

          What winde  

Serves to advance an honest minde.

 

(Go and do all sorts of magic, including catching meteors and getting the mandrake plant, Mandragora officinarum, pregnant – though according to legend, it, being shaped like a man, was supposed to help women become pregnant.  Donne portrayed the mandrake as male in his poem, The progress of the soul, so we know he wasn’t confused about its magical male nature in European herbology.  And, in any case, if you can do these magical things, do the equally powerful act of keeping Donne’s character from feeling envy, since, as the later verses of this poem reveal, women in his life have been unfaithful to him with other men.  The ultimate act of power would be to find what power, what wind, would serve to turn people’s minds to honesty.) 

 

Especially in the version with Donne’s spelling, you can see that find and mind are supposed to rhyme with the wind: words with long vowels tend to have an ‘e’ at the end even in cases where that ‘e’ is lost today.  On the other hand, he seems to be fudging the rhyme of ‘roote’ and ‘foot’ just as he would be today – though as I mentioned earlier, ‘root’ is pronounced to rhyme with ‘foot’ in some dialects.  This poem is a song lyric that seems to satirize the ultra-romantic poetry of the same period, and Donne may have felt entitled to have a little fun, even with his rhymes.  ‘Winde,’ however, seems to be a straightforward rhyme with ‘finde.’ 

 

The strange love affair of the Middle English with ‘mb,’ ‘nd’ and ‘ld’ went on to another phenomenon, the insertion of these consonant clusters into new words in order to make those words more satisfying to pronounce.  For example, with ‘mb,’ Old English ‘bremel’ became ‘bramble,’ ‘nǽmel’ (pron. ‘nammel;’ the ‘ǽ’ represents ‘æ’ in the long, dragged-out form) became ‘nimble,’ and ‘slumere’ became ‘slumber.’  As with ‘thumb’ and ‘limb,’ which you may remember picked up an extra ‘b’ for no good reason, these words just impressed someone as verbal flowers that wouldn’t be complete if they didn’t have a ‘b’ in them.  A decision purely based on the beauty of ‘b.’  Sounds rather Sesame Street, the children’s television show, but there it is. 

 

These insertions of extra consonants didn’t cause new spelling problems in most cases.  When ‘þunor’ became ‘thunder,’ ‘spinel’ became ‘spindle,’ ‘ealre,’ the shrub, became ‘alder’ and ‘ganra,’ the male goose, became ‘gander’ – to give some ‘ld’ and ‘nd’ examples – spelling was none the worse.  It was a different matter, though, when less magical consonant clusters got involved.  The cluster ‘st,’ for example, got involved in one of these obsessions, perhaps becoming an early example of the English difficulty with pronouncing a ‘t’ in the middle of a word.  The Old English ‘hlysnan’ had an extra ‘t’ inserted to become ‘listen’ in Modern English, but as you’ll notice, you can listen to the word as closely as you want and you will never hear that ‘t.’  A ‘t’ was also inserted into Old English ‘glisnian’ to form ‘glisten,’ the verb that is used when a surface reflects light brilliantly, but the ‘t’ sound was polished back out of it by the general trend to brighter English.  In a related trend, many Old English words that had good working ‘t’s in them saw these ‘t’s fall silent in the glistening new English that was invented after 1066.   Consider ‘þistel,’ now spelled similarly as ‘thistle,’ but lacking the thorny ‘t’ sound near the end.   Then there’s ‘þrostle,’ now ‘throstle.’ This is a European bird, also called the song thrush, whose name is a synonym for sweet sounds; the silent ‘t’ seems appropriately melodious.  Then we can add the Old English ‘gristle,’ now pronounced in a smoother, less chewy way, ‘hwistlian,’ now ‘to whistle,’ ‘byrst,’ now ‘bristle,’ ‘wræstlian,’ now ‘to wrestle,’ ‘nestlian,’ ‘to nestle,’ and the christmassy smooch plant ‘mistiltan,’ now perversely given a trace of foot odour as ‘mistletoe.’   Whenever these words have Dutch equivalents, the ‘t’s are still clearly pronounced:  have a look at ‘distel’ (‘thistle’), ‘fluister’ (‘whistle,’ pron. ‘fleouster’ [try to mix the ‘e’ sound of ‘jet’ into ‘ou’ as in ‘out,’ in a single sound – not easy!]), ‘luister’ (‘listen’), ‘borstel’ (‘brush,’ related to ‘bristle’), and worstel (‘wrestle’).  The Dutch for ‘mistletoe’ is ‘maretak,’ but German and Swedish both have ‘mistel’ (with a capital ‘M’ in the German version).

 

The mania for silent ‘t’s in ‘st’ words caught imported French words like ‘castel’ (‘castle’) as well as imports from Greek, via Latin, like ‘epistele’ and ‘apostolos,’ which became the New Testament words ‘epistle’ and ‘apostle.’  It also swept through a whole cluster of words that were made up to bring out the cute and funny aspects of common actions.   These words, which linguists call ‘frequentative,’ usually ended in ‘dle,’ ‘ble,’ ‘tle’ or ‘gle.’  For example, ‘wade’ gave rise to the cuter and funnier ‘waddle,’ ‘gab’ gave rise to the zoophonic (brand new Anglo-Greek word I just made up on the spot, meaning ‘animal sounding’ – you can play this game too) ‘gabble,’ ‘prate’ (complain irritatingly) gave rise to ‘prattle’ (complain idiotically), and  ‘toot’ to the very British ‘tootle’ (excellently used, according to the author of Gems of Japanized English, in a classic police bulletin seen in Japan in the 1930s, advising motorists “When a passenger of the foot hove in view, tootle the horn; trumpet at him melodiously at first, but if he still obstacles your passage, tootle him with vigour and express by word of mouth warning, ‘Hi, Hi.’ ”).  I also much appreciate the change of ‘snug’ to the cosy ‘snuggle,’ ‘dab’ to the almost philosophical ‘dabble’ (meaning to dab or dip into topics and not learn them thoroughly) and many others.  Where this smiley-faced trend ran into ‘tle,’ it tended to produce yet more words with silent letters.  A complicated example is ‘bustle,’ which, as the Online Etymology Dictionary explains, is “frequentative of the Middle English ‘bresten,’ ‘to rush, break,’ from the Old English ‘bersten’ (or ‘berstan,’ ‘to burst’), influenced by Old Norse ‘buask’ ‘to make oneself ready.’”  ‘Jousting,’ knocking knights off their horses, became mere ‘jostling.’  The Dutch word ‘hutselen,’ to knock or throw small objects around, got a toehold in English via a trickster game called ‘hustle-cap,’ where coins were shaken around inside a cap.  This carnival word eventually gave rise to busy, shady slang words like ‘hustler,’ a teenaged gay male prostitute, and ‘to hustle someone,’ meaning either to trick them out of their money or to pressure them for sex.  It popped out into clean living as a word for working with great energy and finally, in the 1970’s, as a chirpy, cheery dance craze. 

 

Curiously, one word entered the English language in 1272 and hung on to what should have been a silent ‘t’ for the next 744 years, and counting.  That was ‘pestle,’ the word for the short rod of stone used to grind seeds or spices in bowl called a ‘mortar.’  It was derived from the Latin word ‘pistillum’ for the same object.  In 1572, we picked up a similar word, ‘pistol,’ apparently from central European sources (German ‘Pistole’ from Czech ‘píšt’ala,’ firearm, pipe,’ pron. ‘peeshtshyala’), and, as you can see, it forcibly refused to adopt the normal English ‘tle’ ending and have its cracking ‘t’ sound silenced.   In 1718, scientists imported, via French, the Latin word ‘pistillum’ yet again, this time as ‘pistil,’ referring to the mortar-shaped female organ of typical plant flowers.   Again, the word adhered rigidly to its fully pronounced form.  I think these words may be having their ‘t’s and their unusual spellings protected, in English, by their sonic similarity to our bad word for urine, ‘piss,’ and the especially nasty word ‘pizzle,’ referring to the penis of an animal.   This last word is not considered swearing and many people don’t know it.  Those who do, though, will agree that a pizzle is better not confused with something you would use in food preparation.  In fact, one dictionary meaning of ‘pizzle’ – and I am not making this up – is a medieval torture instrument, ‘the pizzle,’ a bull’s penis used as an especially brutal tool for whipping someone.  Ugh.  (Note traditional English disgust expressed with throat sounds.)

 

Clearly, ‘pestle’ must be defended against becoming the gross-sounding ‘pessle’ by its sturdy dental broadsword, ‘t.’  To be shot by a silent-‘t’ ‘pistle,’ (pronounced ‘pissle’) instead of a ‘pistol’ would be humiliating beyond belief.  Whole genres of American movies, including Western, cop, spy and angsty-drama, would become sources of laughter.  At the moment, the U.S. movie industry routinely uses large, shock-and-awe inspiring pistols to indicate that dramatic stories are rooted in reality.   If actors could say ‘don’t point your pissle at me,’ it would all fall apart.  Fortunately, the word ‘handgun’ has been introduced to make sure these beloved power symbols can keep their dignity. 

 

Even though the seedy ‘hustle’ didn’t enter English until 1684, there also seems to be a ban against words becoming ‘hostle.’  We’ve brought in both ‘hostel’ and ‘hostile’ from French, and neither word has gone this route, even though some people pronounce them identically as ‘hostel.’  There is an old word ‘hostler,’ or ‘ostler,’ for someone who provides hospitality for horses at an inn; perhaps it got the linguistic copyright on this sound.  It does have a silent ‘t.’ It is also rarely used to refer to the innkeeper (hotel owner) himself, but the inn is never called a ‘hostle.’  And that’s quite amazing, since inns have been called many things related to the word ‘hostler’ over the course of English history.  ‘Hostel’ was just the beginning. 

 

In fact, ‘hostel’ is a first-class example of how greedy English is for new imported words.  We have actually imported variations on the Latin words ‘hospitale’ and ‘hospitum,’ meaning a large house, guest-house or inn, four separate times.  These words are all derived from ‘hospitalis,’ the Latin way of saying ‘of a guest or host.’  ‘Hostel’ was a 1232 importation via an Old French word also spelled ‘hostel.’  It remains with us today thanks to the poet Walter Scott, who revived the word from the dead in 1808 in his poem The Hostel, or Inn, part of a larger work called Marmion.

 

Soon, by the chimney’s merry blaze,

Through the rude hostel might you gaze;

Might see, where, in dark nook aloof,

The rafters of the sooty roof

   Bore wealth of winter cheer;

Of sea-fowl dried, and solands store,

And gammons of tusky boar,

   And savoury haunch of deer. 

 

(In the primitive country hotel, with the fireplace as its only light, you can see the winter food supply hanging from the wooden beams of the ceiling, including preserved northern seabirds, wild boar hams and cured deer legs.)

 

As you can see, Walter wanted to supply his description of a country inn, forgotten by time, with some old-fashioned words as décor.  He not only woke up the long-dormant ‘hostel’ but also pulled out ‘solands,’ an old name for a seabird now called the gannet.  The gannet belongs to a group of seabirds called the ‘boobies,’ a name that would fit the poem’s rhythm – but poetically, I think Walt was better with ‘solands.’  Then, as his Anglo-Norman proof of high status, he went on to ‘gammons,’ an obscure English relative of the French ‘jambon’ for ham.  I’ve already mentioned, in chapter 1, the related Norman word ‘gambe,’ meaning leg.     

 

The 1242 re-importation of ‘hospitale’ from the Old French variant ‘hospital’ didn’t need help from a poet to stay in circulation.  It gave rise to our current word for houses of medical treatment; at first, though, hospitals were shelters for the poor.  By 1644, meanwhile, the French language had smoothed ‘hostel’ out into ‘hôtel’ (approximately pron. ‘oh-tel’) and English sucked up the word yet again, this time for government residences and commercial guest-houses.   The second meaning soon took over the word, and the roof-like French circumflex mark over the ‘o’ was eventually dropped, even though it had many English fans for awhile (much like the arched eyebrow in ‘drôle,’ meaning ‘drily funny;’ the French spelling still competes in literature with the English ‘droll.’).  Finally, in 1818 we grabbed onto the French ‘hospice,’ the latest variant on the ‘hospitalis’ theme, referring to rest homes for travellers.  Eventually we turned it into yet another medical word, a home for the chronically or terminally ill.  ‘Hospital’ in French is now ‘hôpital,’ and it’s been over 150 years since our last importation, so the word is set up to be imported yet again in a new form.  I recommend ‘hip-hôpital’ (pron. ‘ip-opital’), a rest house for victims of inner city passions and injuries.  This could perhaps be introduced as the name of Québec TV series and eventually be anglicized as ‘hiphoppital.’  Now, it sounds like I’m criticizing hiphop and actually I’m a big fan.  The ‘gangsta’-culture it mixes with, though, isn’t my favourite part.  Too many pissles involved. 

 

Nearly all individual letters and combinations of letters in ‘hospitalis’ went silent somewhere during the construction of ‘hostel,’ ‘hospital,’ ‘hotel’ and ‘hospice,’ so it is a wonder that no ‘hostle’ ever came along.  Mind you, as a misspelling of ‘hostel,’ ‘hostile’ and ‘hustle,’ ‘hostle’ gets over 11,000 web hits today, so perhaps it will slowly work its way into the language. 

 

I’ll come back to silent letters again in a few paragraphs, but first, I want to mention a couple more magical consonant clusters that, when placed at the ends of words, tended to turn short vowels into long.  These clusters are ‘-gn,’ which involves a silent letter whenever it’s used, and ‘-ll.’ 

 

The ‘-gn’ cluster is an odd one.  The magic consonant clusters like ‘-ld’ that we encountered above mostly affected Anglo-Saxon words.  This one affects words from Old French that are based in Latin.  We’ve already seen one good example, ‘sign,’ from the Old French ‘signe,’ from Latin ‘signum,’ meaning a symbol or other meaningful visible mark.   This word is typical of the ‘-gn’ group in that it links with related words where the ‘g’ is pronounced and the vowel ‘i’ is therefore short.  Good examples include ‘signature’ and ‘signatory.’  At the same time, related words that end in ‘-ign’ also take the long ‘i:’ consider ‘design’ and ‘resign.’  Here’s another example:  the Latin word ‘pugnus,’ meaning a fist, and ‘pugnare’ (pron. ‘pueg-nah-ray’), ‘to fight,’ which gave rise to the Old French ‘impugner,’ which in turn gave rise to a good English ‘-gn’ word with a long vowel, ‘impugn’ (pron. ‘imPYOON’).  Nowadays this word usually means to attack someone’s reputation.  Some related words like ‘pugnacious,’ meaning ‘likely to fight,’ have the combination of short ‘u’ and clearly pronounced ‘g.’  ‘Repugnant,’ meaning ‘disgusting,’ is a related word; its original sense is ‘(something that makes your mind) fight back.’  In ‘pugilist,’ a fistfighter, the ‘g’ softens up as it usually does when an ‘i’ follows it.

 

English took in a lovely pair of Latin logical twins in the early 1300’s, ‘malignus’ and ‘benignus,’ meaning ‘badly born’ and ‘well born.’  These words came through Old French (‘malign,’ ‘benigne’) and entered English as ‘malign’ and ‘benign.’  ‘Benign’ came to mean ‘having good and kind intentions’ while ‘malign’ diversified.  At first it was an adjective meaning ‘with evil intentions’ but then also became a verb, ‘to say nasty things about someone.’  Physicians picked up the related ‘malignant’ directly from Latin to describe cancerous growths, giving English a word where the ‘g’ in ‘malign’ was pronounced. 

 

Perhaps you can see a trend developing in these ‘-gn’ words, especially if you look at the French involved. 

 

‘Align’ is a good rhyme for ‘benign’ and ‘malign,’ with even the silent parts rhyming.  Its French root is ‘aligner.’  English also has some ‘-gn’ words where the ‘g’ is silent and the vowels are in one of our long forms, but there was no vowel-raising involved:  the vowels were imported from Old French more or less as they are.   These words include ‘reign,’ from Old French ‘reigne,’ ‘feign’ from ‘feindre,’ ‘arraign’ from ‘araisnier,’ and ‘campaign’ from ‘campagne.’  The last word, ‘campagne,’ means ‘open countryside,’ and we imported it twice, once as the military word ‘campaign’ and once as the geographic name ‘champagne,’ applied to the famous sparkling wine of the Champagne region.  I prefer the second one, myself. 

 

Spotting the trend in these words might be made easier if I tell you that the verb ‘feindre’ often has a ‘gn’ in it as it goes through all its tenses and forms.  ‘We feign,’ meaning ‘we pretend,’ is ‘nous feignons;’ ‘I was feigning’ is ‘je feignais.’  Also, the Old French ‘araisnier,’ a much mutated form of the Latin ‘adrationare,’ ‘to bring to account,’ was also sometimes written as ‘aragnier.’  If this all starts to smell a little like Italian cooking to you, then you’re on the right track.  Yes, it’s the ‘gnocchi’ sound again, the ‘lasagna’ noise, the ‘palatal nasal.’  The French words that gave English its ‘-gn’ words with silent ‘g’s all included this sound.  I think that when you are raising your tongue up to the middle of the roof of your mouth to make this sticky sound, it automatically starts to make a ‘y’ sound that tends to influence the vowel that comes before it.  Try it and I hope you’ll see what I mean.  I don’t think this effect explains all the vowel raises seen in the ‘-gn’ words, since there’s nothing about it that automatically, for example, raises ‘maligner’ (pron. ‘malee[g]nyei’) to ‘malign’ (pron. ‘ma-line’).  That shift must have been just general trend-following within the Great Vowel Shift.  But, like the French ‘u’ (‘ü’) that translated into English as a ‘yoo’ sound – as heard in ‘cute’ (‘kyoot’) –  ‘gn’ has a hint of a ‘y’ sound that can’t be avoided.  That’s true even in words like ‘lasagna’ (pron. ‘laza[g]nya’) where the written vowels (all ‘a’ in this case) are relatively low.  This built-in ‘yod,’ if you remember that linguist’s word for a little trace of ‘i’ or ‘y’, is frankly acknowledged in the Spanish version of the sound.  The letter ‘ñ,’ as in mañana (pron. ‘mahnyahnah’) is always explained as similar to ‘ny.’  Suddenly, it seems perfectly reasonable that ‘impugn’ should be pronounced ‘impyoon’ instead of ‘impuggen.’  The ‘u’ is trapped between two ‘y’ sounds, one tacked in front of it as its heritage from ‘ü,’ and one tacked behind it as part of the French ‘gn.’  ‘Impugn’s’ Old French ancestor ‘impugner’ was pronounced ‘amhpyü(g)nyeih,’ so ‘impugn,’ is just a shortened version that’s easier for the English mouth to say.  The ‘g’ happens to be silent because, as English people, we can’t pronounce a ‘palatal nasal’ ‘gn’ sound unless we’re deliberately putting on a foreign accent, which may seem like showing off.  We’re modest, so we stick to mispronunciation.  ‘Champagne’ has to be ‘sham-pain,’ not ‘sham-pa(g)nyeh.’  Meanwhile, words like ‘pugnacious’ and ‘malignant’ skip over the French stage and go right back to our favourite source language, Latin.  They take on typical English-Latin spellings and pronunciations. 

 

We come close to saying the French palatal nasal in words like ‘cognac,’ where we are forced to.   Luckily, it’s not as heavily pronounced as the Italian ‘gn’ and could pass for an easily pronounceable ‘ny’ in some French accents.  By the way, notice that most North Americans and some other English speakers will raise the ‘o’ of ‘cognac’ to its long form, yielding ‘cone-yak’ rather than the more francophone ‘cahnyack.’  This is basically just giving the stressed open syllable a long vowel, as we usually do with our latinizings, but it may be influenced by the raising effect of the ‘gn’ as well.

 

I mentioned the -ll type of vowel-raiser above when I introduced the ‘-gn’ group.  This is another strange case, since the vowel-raising effect only appears to work on one vowel, ‘o.’  This gives us a list of words including ‘roll,’ ‘stroll,’ ‘droll,’ ‘poll,’ ‘troll,’ ‘toll,’ ‘boll’ and ‘knoll.’  This is a linguistic mixed bag, containing Old English words like ‘knoll,’ Old Norse words like ‘troll,’ and words of French origin like ‘droll.’  Most of the words seem to derive their long ‘o’ from a similar sound in the ancestral word, as seen in the corresponding Old English ‘cnoll,’ Old Norse ‘troll’ and French ‘drôle.’  ‘Roll’ comes from Old French as a derivative of the Latin ‘rotula,’ ‘little wheel.’  You can see the relation to ‘rotate,’ which also has a long ‘o.’ ‘Stroll,’ like ‘hustle,’ is a so-called ‘cant’ word from a secret language used by criminals or ‘travellers.’ The latter group is made up of people who professionally move around from place to place, often working on the boundary between criminal and regular life.  The source word is ‘strollen,’ from a German dialect, related to the more cosmopolitan German ‘strolchen,’ ‘to stroll or hang about.’  ‘Strolchen’ in turn relates to ‘Strolch,’ at one time meaning a hobo or fortuneteller, but now meaning a hoodlum.  The ultimate source, related to fortunetelling, may be the Italian ‘astrologo,’ ‘astrologer.’ 

 

It’s remarkable how regular the ‘-oll’ vowel-raise is.  Of the few and odd exceptions, none is a typical Anglo-Saxon, Anglo-Norman, or Anglo-Viking word.  Two are from a particularly unlikely source.  Around the 1500’s, English people developed a habit of turning personal names with an ‘r’ in the middle into slangy pet names ending in ‘l.’   ‘Harold,’ for example, yielded ‘Hal,’ and ‘Sarah’ yielded ‘Sally,’ ‘Sal’ for short.  Similarly, Mary generated ‘Moll’ and ‘Molly,’ not to mention “Polly,” while ‘Dorothy’ spun out ‘Doll.’  ‘Moll’ changed to mean ‘prostitute’ in the 1600’s, and slid even further down the social ladder to ‘criminal’s girlfriend’ a couple of centuries later.  ‘Doll’ equally quickly became ‘easily obtained pretty girl’ and later, in 1700, ‘child’s pretty toy girl.’  The proud Dorothys of the world soon reacted by becoming ‘Dotty’ instead.  In any case, mutant nicknames were the source of half the known short-o ‘-oll’ words. 

 

The other two words fitting this pattern are exotic imports.  One is ‘loll,’ from Dutch ‘lollen,’ referring to what a dangling tongue does, especially when licking a ‘loll’-ipop. The other is ‘quoll,’ pronounced ‘kwahl’ to rhyme with ‘ball.’  This is an aboriginal Australian name for a small, night-hunting marsupial predator, sometimes called a ‘marsupial cat.’  As seen in photos, it’s a very pretty little animal with a spotted coat and an opossum-like, pointed nose.  As seen in Australian wildlife reserve display pens, it is invariably curled into a spotted ball and sound asleep with the nose hidden well out of sight.  The word comes from the Guugu Yimidhirr language of the Hopedale area of tropical northern Queensland.  In 1770, the explorer James Cook had had to make some repairs to his ship, the Endeavour, after it struck the Great Barrier Reef.  While the ship was being worked on, his staff botanist Joseph Banks interviewed the local Guugu Yimidhirr people and collected some bits of their language.  His most famous collection was ‘ganguru’ (pronounced gang-ooroo and not gang-guru; it’s often written as ‘gaŋuru,’ with the IPA symbol for the ‘ng’ sound, to make this clear).  That word, suitably mispronounced, became the English ‘kangaroo.’  As you can see, since you’re getting to be quite expert about these things, the ‘g’ at the front of ‘gaŋuru’ was a voiceless (no larynx action), unaspirated (no puff of air) one, and thus turned into a ‘k’ in English.  This is just like what happened when the unaspirated ‘g’ in the Cantonese Hèung Góng became the normal English heavily aspirated ‘k’ in Hong Kong.  Joseph Banks also collected the word ‘dhigul,’ which he wrote down as ‘je-quoll.’  With the ‘qu,’ you can see that Guugu Yimidhirr ‘g’ making trouble for the English mouth again. 

 

The English ‘-ell’ words are regular – that is, they have short vowels, as seen in ‘well’ and ‘bell’ – and so are the ‘-ill’ words, like ‘sill’ and ‘kill.’  The ‘-ull’ words are a mixture: ‘cull,’ ‘lull’ and ‘dull’ are not like ‘bull,’ ‘pull’ and ‘full.’  The other real trend in the ‘-ll’ words, then, is what happens to the ‘-all’ words.  Almost all of them take the ‘ah’ sound that is similar to the Old English long ‘a:’ in no particular order, ‘hall,’ ‘call,’ ‘mall,’ ‘fall,’ ‘gall,’ ‘squall,’ ‘stall,’ ‘small,’ ‘spall,’ ‘ball,’ ‘pall,’ and ‘thrall,’ plus some compounds like ‘recall’ and ‘enthrall.’ These words mutated up through Middle English from various Old English pronunciations, such as ‘þræl’ (rhymes with ‘pal’) for ‘thrall,’ ‘getæl’ for ‘tall,’ ‘steall’ (pron. rather like stay-all) for ‘stall,’ ‘heall’ for ‘hall’ and ‘galla,’ Anglian dialect for ‘gall.’  A few other stray sounds came into the mix as well:  the Norwegian ‘skval,’ a burst of water, for ‘squall’ and the French ‘apalir,’ ‘to become pale or to make someone pale,’ for ‘appall.’  So there are three distinct vowel configurations there, ‘a’, ‘æ’ and ‘ea,’ all converging on the modern ‘-all.’  Then, since this is English, there is the one exception: ‘shall.’  It is derived from an ‘ea’ word, ‘sceal’ (pron. ‘shayal’) and has no strong excuse for not rhyming with the other ‘-all’ words.  The potential sound-alike ‘shawl’ is a Persian import from 1662, so that is not the factor that made ‘shall’ go deviant.   ‘Shall’ is so odd that it rhymes only with a few exotic words like ‘pal’ (a word from the Romany ‘gypsy’ language, which is related to Hindi), ‘canal’ (Latin ‘canalis,’ a pipe) and ‘corral’ (Spanish ‘corro’ and Portuguese ‘curral’). 

 

The ‘-all’ word with the best origin story is ‘mall.’  Long before suburban American teenagers found the perfect place to hang around, there was a broad, pleasant walkway called ‘The Mall’ in central London’s St. James Park.  Now, it is merely a place to stroll, but earlier in history, people used it to play a game resembling croquet.  This game came from Italy named ‘pallamaglio’ – ‘palla’ was a ball, and ‘maglio’ (which you now know how to pronounce) was the mallet used to hit the ball along the ground.  The name came through French into English as ‘pall-mall.’  Then, over time, the ‘mall’ from the ‘mallet’ part of the word became attached to the playing ground itself (circa 1737), and then to the walkway, and then to any broad, shady place where many people might stroll.  After being seduced and promoted by marketers, the word became the #1 label for a suburban shopper’s paradise.  And that’s the miracle of linguistics:  a croquet bat can become a shopping centre within 350 years. 

 

Getting back to spellings and sounds, as we saw above with ‘bald,’ ‘scald’ and perhaps ‘wand,’ that vowel-raising letter combinations often seem to turn ‘a’s into ‘ah’ sounds.  In parallel with the ‘-ll’ words, in fact, there is a series of ‘-lt’ words that also feature regular short ‘e’s and ‘i’s, raised long ‘o’s, and ‘medical patient’ ‘a’s that say ‘ah.’  Some examples of the last two categories are dolt, bolt, colt, jolt, volt, revolt, and smolt, along with halt, malt, salt, and exalt.  It seems, in general, that a very impressive fraction of the irregularly pronounced or spelled English words derive from one of these vowel-raising groups.  In all these groups mentioned so far, the consonant immediately after the vowel is sonorous, that is, you can hum it: m, n or l.  ‘L’ seems particularly powerful as a vowel raiser: even a single one can raise ‘o’s, as seen in the French-derived words patrol, extol and control.   More strangely still, even a silent ‘l’ can raise ‘o’s:  this occurs in folk, yolk, and the family name Polk.

 

The speakers of Middle English and early Modern English got so busy inventing and extending vowel-raising clusters that they made one – ‘nt’ – that was almost exclusively used for contractions.  First, in the 1400’s, came the adjective ‘wont,’ meaning ‘accustomed’ (‘he was wont to have tea at 4 p.m.’). It was a contraction of the Old English past-participle ‘wunod,’ meaning ‘dwelt’ or ‘was used to,’ related to the Dutch word ‘wonen,’ ‘to live (somewhere).’ Even though we North Americans, on the rare occasions we joke around with the word ‘wont,’ say it just like ‘want,’ conservative British pronunciations place it in the long form, which to our ear may sound somewhat like our ‘won’t’ or like ‘wehwnt’ (IPA ‘wəʊnt’). This word was a pronunciation oddball for a couple of centuries, but in the 1630’s to 1660’s, just after the time of Shakespeare, the English invented two similar sounding contractions to give it moral support: ‘don’t’ and ‘won’t,’ (‘dehwnt’ and ‘wehwnt’ in the British CRP) the short forms of ‘do not’ and ‘will not.’  At around the same time, ‘shan’t’ (British pronunciation ‘shahnt’) arrived to support the long-isolated spelling ‘want,’ which until then had only had rhymes in ‘–aunt’ words like ‘vaunt,’ ‘haunt,’ ‘daunt,’ ‘gaunt’ and ‘flaunt.’  This addition of ‘shan’t’ to the English lexicon clearly showed that ‘nt’ was being put to new work as one of the clusters that could produce the archaic ‘long a’ sound – the ‘ah’ – as seen with other clusters in ‘salt,’ ‘chalk,’ ‘ball,’ and so on.  ‘Shan’t’ was followed a few decades later by ‘can’t’ (conservative British pronunciation ‘cahnt’), when the efficient English began to find that saying ‘cannot’ was too much work.  Sorry, mum, I just cahn’t say ‘cannot’ one more time.  Meanwhile, over in a little linguistic corner of its own was the word ‘pint,’ a fully raised ‘i’ word that has no exact rhymes in English to this day.  It does, of course, have half-brothers and –sisters in the raised ‘nd’ words like ‘grind.’ We really should, though, have invented ‘I min’t’ as a contraction for ‘I might not’ or ‘I don’t mind’ (‘I mind not’) to give it a full sibling. 

 

There is a sense of regularity about all these irregular vowel changes that makes an interesting puzzle.  What linguistic intuition drove it forward?  You could almost imagine that the sonorous, hummable letter at the beginning of the raising cluster serves as an extra vowel, giving rise to a double-vowel combination that explains the change in pronunciation. Just as ‘rod’ becomes ‘road’ and ‘bet’ becomes ‘beet’ and ‘beat’ when you add in a suitable second vowel, so ‘bot’ becomes ‘bolt,’ ‘mid’ becomes ‘mind’ or ‘mild,’ ‘bid’ becomes ‘bind,’ ‘pal’ becomes ‘pall,’ and so on.  Maybe the English mind, at some time in history, intuitively broke letters not into vowels and consonants, but into singables and interruptions.  Whenever two compatible singables came together, they could alter the pronunciation of the first one to the ‘long’ form – or not, depending on whimsical custom. 

 

Unfortunately for this bold attempt to find regularity in irregular English, there are some vowel-raising clusters that have no hummable consonants in them.  In particular, an ‘s’ can sometimes signal at least a weak vowel-raising cluster, like the ‘st’ seen in post, most, ghost and similar words, but not in lost, cost, frost and their allies.  Combination with ‘o,’ the most raisable vowel, seems to be necessary.  There is even one double-s word that is raised: gross.  The others, however are a total ‘loss’ and can be given the ‘toss.’  They’re all short-‘o’ words. 

 

(‘Gross’ has a very strange parallel among the ‘a’ words.  ‘Bass’ rhymes with ‘class’ when it’s a fishing word and with ‘case’ when it’s a musical word. That’s even though the musical word derives from Italian ‘basso,’ which rhymes with ‘mass-oh’ when it’s used as an English opera term – and by no means rhymes with lasso, which rhymes with shoe and canoe and glue – oops, that brot my inglish speling hedayk bak.)

 

The letter ‘o’ is so volatile in short words, so likely to raise up and go long, that most short words with the short-‘o’ sound avoid using the letter.   You can see why when the Persian and Urdu word ‘شال’ (‘shal’ with ‘a’ as in ‘ball’) for a broad scarf was imported into English in the 1660’s, it was imported as ‘shawl’ rather than ‘shol’ or ‘sholl.’  (It could have been given the spelling ‘shall,’ if that spelling hadn’t been blocked by a very unusual word, as discussed above.)  Granted, British English tends to differ from North American by distinguishing the sounds ‘mall’ and ‘moll.’ The Anglo-Indian origin of ‘shawl’ tended keep it away from short-‘o’ spellings.  You can see, though, that if an American had found ‘شال’ first and spelled it ‘shol’ or ‘sholl,’ it would have been in great danger of drifting towards the sound of ‘shoal.’  Likewise, clearly, a ‘stalk’ couldn’t safely be a ‘stolk’ or a ‘hall’ a ‘holl.’   (In fact, English speaking people do well with the Dutch family name Stolk, except that they don’t know it’s a two-syllable word, ‘Stoe-lek.’)   If the short ‘o’ is pronounced in a short word, it is often bolstered with a sonically doubled letter after it, as in ‘odd,’ ‘off,’ ‘stock’ and ‘toss’.  It’s fine for ‘from’ to have a deviant pronunciation, but if one spelled it ‘fromm,’ it could only rhyme with ‘Tom.’  English doesn’t even possess any ‘-omm’ words, but we can all feel this clearly.   ‘Mom’ can take the short ‘o’ sound when North American kids are worried or complaining, as in the three-syllable ‘Mò-òh-óhm!’ (better written ‘mà-àh-áhm’) that I used to use as a boyish protest, but otherwise it settles in to sounding like the British ‘mum.’  ‘Mawm’ or ‘malm’ or ‘mahm’ would not have the same freedom.  ‘O’ is just a letter with a mind of its ‘one,’ oops, I mean, ‘own.’  

 

The second-most-common English word pronounced ‘mawm’ is the sensibly spelled surname ‘Maugham.’   

 

In the vowel-raiser examples, you can see the enormous amount of work that went into making English a massively irregular language.  There are all sorts of islands of logic and consistency amidst all the irregularities.  Of course, as soon as there is a relatively regular trend in the language, English speakers are forced to go to the Dutch or to the Guugu Yimidhirr people of northern Queensland to obtain some deviant words to break it up.  Perhaps our love of democracy requires this anarchy.  Impossibly irregular spelling and pronunciation may represent freedom for us.  

 

It seemed sensible for me to talk about some of the vowel-raising clusters that lacked silent letters along with the ones that were producers of silent letters.  Now I’m going to go back to pure silent letters.  Here they all are:  (two minutes of silence follows).  Actually, there are not so many left now. 

 

The silent ‘w’s, though, are a reasonably big category.  They come before ‘r’ in words like ‘write,’ ‘wrath,’ ‘wrack,’ ‘wrong,’ and ‘wretched.’  Linguists are not sure what the ‘wr’ combination means, historically speaking, since no one is sure how Old English speakers pronounced their ‘r’ sounds.  This can be quite a problem.  There are probably more different sounds running under the banner of ‘r’ than there are under any other single letter of the Roman alphabet.  Every time you learn a language, you have to learn its ‘r’ or ‘r’s.  Dutch, for example, is heard beautifully spoken in radio broadcasts with a rolled ‘r’ that, as I like to joke, has exactly four ‘hits’ in it, moments where the tongue hammers the roof of the mouth.  Many English speakers find this sort of tongue gymnastics impossible.  Worse still, most Dutch people use the German ‘r,’ similar to the Parisian ‘r’ from the back of the throat, for at least a few words, such as the common boy’s name ‘Roel.’ (pron. ‘hRool,’ rhymes with ‘spool;’ the ‘h’ I’ve written there is not pronounced but merely guides you to the part of the mouth that you should say the ‘r’ from).   Among the extreme ‘r’s I have known is the Korean ‘r,’ which is a flap of the tongue-tip against the roof of the mouth, almost like the flap Canadians use instead of a ‘t’ when they say the word ‘butter’ (‘alveolar flap’ in linguistics).  The Korean ‘r’ at the beginning of a word can sound almost like ‘n’ or ‘l.’ Some books on Korean call it an ‘l.’ Then there’s the Secwepemc ‘r,’ which sounds like an English ‘w,’ except that most of the work that the lower lip does in making a ‘w’ is done by the tongue instead, giving the sound an ‘r’-like quality.  This letter comes into personal pronouns (‘me,’ ‘you,’ etc.) and the Secwepemc equivalent of ‘the,’ so it is everywhere in speech.  A typical sentence like ‘ri7 ren sqeqxe,’ ‘It’s my dog’ (pronounced ‘w(r)ee/full stop/ w(r)en sshqeqkhe’), is very hard for an English speaker to ‘catch,’ that is, to identify what the words are and how they’re being said.  Our circuits are not programmed to hear that ‘r’ well.  (By the way, Secwepemc is often written with a ‘7’ as a glottal stop, since 7 is the closest keyboard imitation of the IPA glottal-stop symbol, which looks like a question mark with no dot.  You saw two of the latter above, in ‘ju:ɬuʔiɬʔatħ,’ the parent word of the town name Ucluelet.) 

 

English is a challenge because it was born of a fusion of dialects, and it is a native language in several countries; thus, there are several common English ‘r’s.   For example, there’s the flat North American ‘r’ with the sides of the tongue up near the sides of the upper teeth, and the tip of the tongue more-or-less curled up.  Then there’s the more relaxed ‘r’ that people from England typically say when they pronounce it (often where it’s not supposed to be, as in ‘Oi ed en oideyr,’ ‘I had an idea[r]’) , and the disappearing ‘r’ that darkens final vowels when they say things like ‘bar’ (more-or-less pron. ‘bah’) and ‘war’ (‘woah’ or ‘wah’).  The ‘r’ system of the neighbourhood African-American accent in the USA is similar to the one used in much of southern England: it was probably picked up from English people rather than being a leftover bit of African pronunciation.  ‘Star’ is ‘stah’ in that accent just as it is in London.  Scottish people speaking English (as opposed to the Scots language) have a rolled ‘r,’ or a ‘trill’ as linguists like to put it.  It’s similar to the ‘r’ of radio Dutch or standard Spanish.  Linguists consider the ‘r’s North Americans pronounce in words like ‘search,’ ‘work,’ ‘standard’ and ‘dinner’ to be ‘r-coloured vowels,’ essentially, a vowel sound that has some ‘r’ buttered over the top of it. 

 

Getting back to the ‘wr’ now, one theory about it is that Old English had two ‘r’s, one where the lips were rounded – written ‘wr’ – and one where the lips were unrounded – written ‘r.’ The rounded one would have been similar to the modern American and English ‘r,’ while the unrounded one was something else, maybe a trill as in Scottish or Dutch.  Over time, the rounded sound took over all the ‘r’ sounds in the language, making ‘wr’ and ‘r’ sound the same. 

 

This may be true, but many of the ‘wr’ words have sister-words in languages like Frisian, Dutch or German where the ‘w’ is pronounced.  In German, ‘w’ sounds like a ‘v,’ and in Dutch, as I have mentioned, it is halfway between an English ‘v’ and ‘w.’ The ambiguous Dutch letter can still be heard in words like ‘wrakgoed’ (‘wreckage,’ pron. ‘w[v]rak-ghoot,’ to nearly rhyme with ‘jackboot’), ‘wreken’ (‘retaliate,’ pron. ‘w[v]rayken,’ related to English ‘to wreak [revenge]’) and ‘wriemelen’ (‘to wriggle,’ pron. ‘w[v]reemelen’).  I think this may just be one of those disruptive, scraping letters that was cleaned out of English, like ‘kh,’ ‘gh,’ and the initial ‘k’ in words like ‘knock.’  The ‘Great Consonant Clean-up,’ you might call it.  That’s not to say that bending the lips to make a ‘w’ might not have influenced how our ‘r’ came out, but there’s no way to be sure. 

 

English seemed to have an informal policy that highly inconvenient letters in classical Greek or Latin words could be dropped from pronunciation.   Given what we know about the smoothness of Middle and early Modern English, is it any surprise that that there are silent ‘g’s in the Greek imports ‘paradigm,’ ‘diaphragm,’ ‘phlegm’ and ‘apothegm’?   Even ‘paradigmatic’ (pronounced ‘pair a dime attic’) and ‘phlegmatic’ (‘flem attic’) keep the same silence going.  ‘Phlegm,’ which could be simplified to ‘flem’ in spelling reform, looks marvelously like a word that really needs to clear its throat – there are far too many large, loopy, extra letters clogging up its passages.  The ‘phragm’ in ‘diaphragm’ gets to be pronounced in our fungal word ‘phragmospore’ (pronounced as it looks, ‘frag mo spore’). ‘Diaphragm,’ though most commonly used to mean the sheet of muscle dividing our chest from our gut, literally means ‘cross-barrier’ or ‘cross wall;’ meanwhile, phragmospores are spores with many cross walls dividing them into multiple cells.  There you are, now you’re ready to talk to the mould analyst when your house springs a roof leak and starts to grow fuzzy stuff – giving you, just to tie it all together, lots of phlegm.   Switching over to Latin, imagine saying ‘autumn’ and pronouncing all of its letters.  It could be spoken to rhyme with ‘summon,’ but we English speakers tend not to pronounce extra syllables when we see consonant groups.   Our rule is that if consonants are together, we pronounce them together.   If they’re a couple, we treat them as married.  Alas, the ‘n’ is an absentee lover in ‘autumn,’ and consummates its relationship with the ‘m’ only in ‘autumnal.’  

 

Dutch, by contrast, does tend to stretch out some consonant clusters, namely those beginning with an ‘l’ or an ‘r.’  A palm tree is noted as a ‘palm’ in both Dutch and English writing, but in speech, it’s pronounced ‘pallem’ in Dutch and ‘pahm’ in English.  Salmon in Dutch is ‘zalm,’ pronounced ‘zallem,’ sounding very different from our ‘sammen.’  You can immediately see why the ‘l’s in these words became silent in English but not in Dutch.  The story is a little more complicated than that, though, as we’ll see in the next few paragraphs.  ‘Half’ is a similar situation, a two-syllable word in Dutch (pron. ‘hallef’) and one syllable with a silent ‘l’ in English (‘haf’).  We just can’t split those consonant pairs up, even when the letter-couples, in terms of pronunciation, don’t get along at all.  Various French words with Latin roots also got into the same pattern in English; a good example is ‘calm.’  The ‘l’ is pronounced in the two-syllable French ‘calme,’ which sounds like ‘callma.’  ‘Balm,’ descended from Old French ‘basme,’ Latin ‘balsamum’ and, ultimately, Hebrew ‘basam,’ was a word that had a silent ‘l’ stuck into it by a devoted student of classical languages.  I’ll say more about that trend below.  The situation seen with ‘l’ is also seen with ‘r,’ but only in accents like typical English, Australian and New Zealander.  In most of England, ‘harm’ is ‘haahm,’ ‘warm’ is ‘wohme,’ and ‘heart’ is ‘haht,’ more or less.  The British English ‘r,’ though, is unpronounced at the end of words (‘stir’ = ‘steuh’), while ‘l’ is almost always pronounced there (‘still’).  In some British accents from east London and the Thames Estuary, though, ‘l’ follows the trend set by ‘r’ and becomes vocalized, sounding more or less like a ‘w:’ ‘Aw stiww think e’s aww roait’ = ‘I still think he’s all right.’ 

 

In the Netherlands, ‘r’ combinations often become split into separate syllables.   The scientific institute I worked in when I first moved there was in a town called Baarn, and our visitors had difficulty buying a train ticket to reach us.  Though the ticket sellers spoke English perfectly well, they couldn’t understand the town’s name when non-Dutch people pronounced it ‘Barn.’  Only Dutch speakers knew it had to be a two-syllable ‘Baah-ren’ to be understood.  Preferably with a strongly rolled ‘r.’  Saying a rolled-r ‘Baarn’ without making it two syllables is as hard as pronouncing ‘autumn’ in full.   Scots speakers, though, say something similar to a one-syllable ‘Baarn’ with their Viking-derived word for a child, which is ‘bairn.’  As the Scots language Wikipedia says, “A bairn, wean, child or littlin is a youthie body, a lad or lass.”  ‘Bairn’ must be pretty pronounceable if it became the number one word over ‘wean,’ ‘child’ and ‘littlin.’  Mind you, Scots poetry doesn’t often need a rhyme for ‘chitlin’ (the southern US pronunciation of ‘chitterling,’ a portion of cooked pig intestine), so that may have counted against ‘littlin.’

 

Medieval and early modern-age scholars introduced an amazing number of extra letters into English words.  This was especially true for words introduced via French, whether Norman French or otherwise.  The scholars seem to have had a campaign to make English words look less like their French equivalents and more like Latin or Greco-Latin text.  I’ve already mentioned the ‘b’ that was put into the word ‘debt’ to make it like the Latin ‘debitam’ rather than the Old French ‘dete’.  The silent ‘b’ in ‘doubt,’ derived from French ‘douter,’ arose the same way, as did the ‘p’ in ‘receipt.’  These insertions did tend to link these words with related words, so that ‘doubt’ with a ‘b’ ‘indubitably’ got a better ‘reception’ from classical scholars than the ‘b’-less form.  Efforts were made to get people to pronounce these revised words properly.  This unlikely campaign is ruthlessly mocked by William Shakespeare in the comedy Love’s Labour’s Lost.  He has a character called Holofernes criticize a rival in laughably old-fashioned scholarly language (inverted commas are mine):

 

He draweth out the thread of his verbosity finer than the staple of his argument.  I abhor such fanatical phantasimes, such insociable and point-devise companions, such rackers of orthography, as to speak ‘dout,’ sine ‘b,’ when he should say, ‘doubt,’ ‘det,’ when he should pronounce, ‘debt,’ ‘-d, e, b, t-,’ not ‘d, e, t:’  he clepeth a calf ‘cauf;’ half, ‘hauf;’ ‘neighbor’ vocatur ‘nebor,’ ‘neigh’ abbreviated ‘ne.’  This is abhominable, which he would call ‘abominable.’

 

(Translation: “He makes his argumentative wording more complicated than the logic that supports it.  [The underlying metaphor is from weaving:  he makes his verbal thread thinner and more refined than the coarse mental wool fibre from which it is spun.]  I hate to associate with such demented people living in a fantasy world, such incompatible and demanding people who torture our language by saying ‘dout’ instead of ‘doubbt’ and ‘det’ instead of ‘debbt.’  He calls a ‘callf’ a ‘cahf,’ ‘hallf’ ‘hahf.’  ‘Neighbour’ he speaks out as ‘nebor,’ ‘neigh’ abbreviated ‘ne.’ This is ab-hominable, which he would call abominable.”) 

 

The last sentence is witty satire of a scholarly fool, since Holofernes clearly believes that the word ‘abominable,’ meaning ‘unbearably bad,’ is based on the Latin ‘ab,’ ‘away from,’ and ‘hominis,’ meaning ‘of humans.’  Many medieval scholars shared this unsophisticated misunderstanding.  In fact, ‘abominable’ is based on ‘ab’ and ‘omen;’ the latter is a warning of bad fortune.  It’s related to the Latin ‘abominari,’ to avoid something as a bad omen.  So Holofernes is trying to academically ‘correct’ English in a way that makes a word, ‘abhominable,’ that has no legitimate reason to exist.  Mind you, it’s not entirely clear if Shakespeare knew that this problem existed with ‘abhominable,’ a word some people actually used in his time.  His humour works well even if he didn’t know, but it’s at its best if he did.  Was he ingenious enough to write a joke that would suddenly become funnier later in history, after linguists had done some some studies?  Maybe so! 

 

Holofernes, as you can see, also wants to impose pronunciation of academically inserted letters on everyone, even if this would yield difficult pronunciations like ‘debbt.’ 

 

Just before we look further into such efforts to reform pronunciation, here are some explanations of other words in Shakespeare’s passage, for those who are curious. ‘Verbosity’ is skill with multiple words, ‘staple’ is fibre, usually wool; ‘phantasimes’ are madmen (a particular madman called Monarcho is described with this word earlier in the play); ‘point-devise’ means ‘precise, exacting’; the rack is a medieval torture device, ‘orthography’ is the proper form of words; ‘sine’ is Latin for ‘without;’ ‘clepen’ was an old-fashioned verb meaning ‘to call’ (you may remember “This yonge man was cleped Piramus” from earlier in this chapter); vocatur is Latin for ‘is called.’  It’s not entirely clear what Holofernes is getting at with ‘neighbour:’ is he actually trying to re-introduce the spoken ‘gh’ in this word, or is he just offended that his rival has used the short vowel ‘e’ rather than the diphthong ‘ei’?  According to the bible of Shakespeare interpretation, the Arden Shakespeare series of annotated plays, other literature from the same historical period suggests that Holofernes is only criticizing the vowel sound and is happy to leave the ‘gh’ silent. 

 

Even the name Holofernes is a joke: it derives from an unfortunate, overconfident character in the bible, an Assyrian general who decided to celebrate his initial successes in battle against a Judaean city by having sex with Judith, a young woman from the city.  She then cut his head off with a sword after he fell asleep.  It’s a great name for a pompous, luckless fool. 

 

Clearly, though, characters like Shakespeare’s Holofernes had a strong influence on English at a certain point in history.  A Spanish linguist named Isabella de la Cruz Cabanillas has posted a paper on the web showing all the ‘etymological respellings’ in English she could track down, that is, existing English words that had letters inserted in them to make them more like their Latin or Greek root words.  Some of these inserted letters failed to change pronunciation of the words involved; examples include the ‘l’s in ‘could’ and ‘salmon’ (French ‘saumon,’ Latin ‘salmo’), the ‘s’s in ‘island,’ ‘aisle’ and ‘scissors,’ the ‘b’ in ‘subtle,’ the ‘h’s in ‘anchor,’ ‘heir,’ ‘honour’ and ‘rhyme,’ the ‘g’s in ‘sovereign’ and ‘foreign,’ the ‘c’s in ‘indict’ and ‘scythe,’ and the ‘w’s in ‘whore’ and ‘whole.’  The ‘s’ in ‘island’ is an insertion Holofernes might like, since it’s based on a mistaken attempt to link the word to the Latin ‘insula’ rather than its true source word, the Old Norse ‘igland.’  The ‘g’ in ‘sovereign’ (from Vulgar Latin ‘superanus,’ overlord) is also a mistake, based on the word’s apparent similarity to ‘reign’ (from Latin ‘regnum,’ with genuine ‘g’ as you can see).  

 

Even though Holofernes was backing a losing cause trying to get people to say ‘debt’ with a ‘b,’ his real-life equivalents did succeed in changing many pronunciations with their insertions of classical letters.  For example, we say ‘amethyst,’ ‘anthem,’ and ‘apothecary’ with a ‘th’ thanks to scholarly corrections.  Just to single out one of these words, ‘amethyst’ came into English from the Old French ‘ametiste.’  The earlier Latin form was ‘amethystus,’ adapted from the Greek ‘amethystos.’  The purple amethyst stone was supposed to prevent people from getting drunk, and the roots of its Greek name are ‘a-’, meaning ‘not,’ and ‘methyskein,’ meaning ‘get drunk.’  The name really should apply to coffee rather than a rock, but that’s just my modern perspective.  Cup of amethyst, please, bartender.

 

The plastic surgery to put back the classical letter was not quite so successful with ‘arctic’ and ‘antarctic.’  The original Greek ‘k’ in ‘arktikos’ came from ‘arkos,’ a bear.  The word had nothing to do with polar bears and meant ‘in the direction of the Great Bear,’ a northern star constellation.  The bearish ‘k’ clawed its way into Latin and curled up as a ‘c’ (‘arcticus’) but then went into a deep hibernation sleep in medieval Latin (‘articus’).  It stayed asleep when French took up the word as ‘artique’ and when Middle English adopted it as ‘artik.’   Indeed, for many English speakers, it is still asleep, and they routinely pronounce the word ‘artic.’  The ‘c,’ however, was prodded out of its sleep by scholars in 1556, 165 years after ‘artik’ was first recorded, and we’ve all had over 450 years to get used to having it wandering around again.  My partner, who grew up on Cape Breton Island in Nova Scotia, eastern Canada, assures me that most people there habitually say ‘artic.’   It seems the extra ursine consonant hasn’t yet managed to swim across the Strait of Canso separating Cape Breton from the mainland.  Or perhaps it has, but the Cape Bretoners find that bookish pronunciation, as you might say, unbearable.

 

Some silliness was unavoidable in these changes.  For example, attempts were made to bring back the ‘d’ in the Latin prefix ‘ad’ (meaning ‘to,’ ‘at’) in words that had lost the letter while they were in their Old French phase.  Thus ‘auenture’ was restored as ‘adventure,’ from ‘ad’ (to) and ‘venire’ (come), Latin ‘res adventura,’ ‘the thing about to come along.’  Many other ‘ad’ words such as ‘administer,’ ‘admonish,’ ‘advance,’ ‘advantage’ and ‘advice’ were also restored to their classical glory.   Thrown in among these words was ‘admiral,’ which as you recall from chapter 1 actually came from the Arabic ‘amir al-bihar,’ ‘marine commander.’  It seems the scholar who added the nonsensical ‘d’ thought the word was related to ‘admire.’  Thank you, Holofernes. 

 

You may wonder if the ‘h’ in words like ‘whistle’ should be treated as a silent letter, since many English-speaking people strictly pronounce ‘wh’ as ‘w.’  The ‘wh’ in most cases is a survival of the Old English consonant cluster ‘hw,’ as seen in ‘hwistlian,’ the forerunner of ‘whistle.’  In fact, there are patches throughout the English-speaking world where ‘wh’ is still pronounced as a distinct consonant.  Scotland and parts of the U.S. south are the main strongholds of the sound.  It forms a rare case in which an Anglo-Saxon feature has prestige, since pronunciations with ‘wh’ sound slightly higher class or more educated than pronunciations where the same words are pronounced with ‘w’ alone, especially if this ‘w’ is emphasized.  The interesting thing about ‘wh’ is its rarity in the languages of the world; only English has it.  Thus, it seems to be something of a linguistic treasure, a rare species of sound.   Various ‘wh’ words are related to words that begin with ‘hv’ in Nordic languages and that were also ‘hv’ in the ancient Gothic language.  ‘White’ is a good example:  it was ‘hwit’ in Old English and Old Frisian, ‘hvitr’ in Old Norse and ‘hveits’ in Gothic.  In Danish today, it is ‘hvit.’  “Whet,’ ‘to sharpen on a stone,’ was ‘hwettan’ in Old English and ‘hvetja’ in Old Norse; the related verb ‘ga-hvatjan’ in Gothic meant to sharpen or incite.  ‘Whale’ is linked to ‘hwæl’ in Old English, the Old Norse ‘hvalr’ and ‘hvalfiskr’ (‘whale-fish’) and the modern-day Danish ‘hval.’  In several words, ‘wh’ seems to be in imitation of a sound, such as ‘whisper,’ ‘whine,’ ‘whistle,’ ‘whizz’ (the sound), and ‘whiff.’  Some of these words were unique to English, and some had relatives in Nordic languages: for example, the Old Norse verb ‘hvina’ meant ‘to make a whizzing sound.’  Finally, ‘wh’ ended up in quite a collection of important grammatical words, like ‘who,’ ‘where,’ ‘why,’ ‘what,’ ‘while’ and ‘whether.’  Its sound has definitely become ‘h’ in several words where an ‘o’ follows it as a vowel, e.g., who, whole, and whore, but it remains available for its original pronunciation in ‘whomp’ ‘whoosh’ and ‘whopper.’  All in all, then, ‘wh’ is a pretty special sound, and if we can manage to save it from linguistic extinction, we’ll all be able to say a huge collective ‘whew!’  Which would sound ridiculous if it was said ‘wew!’

 

English inherited some truly silent letters from French silent letters.  The French language, in fact, is systematically addicted to silent letters.  For example, you can almost never say ‘they do something’ or ‘they did something’ (they eat, they slept, they wept) without using at least three of them (in the same order, ils mangent, pron. ‘eel mahnzhuh,’ ils dormaient, pron. ‘eel dohrmeh’ and ‘ils pleuraient,’ pron. ‘eel pluhreh’ – note silent ‘s’ on ‘ils,’ the silent ‘nt’ on ‘ils mangent’ and the silent ‘ent’ on the other words).  The Old French and Norman words that came into Middle English deposited relatively few silent letters in our modern language.   There are some silent ‘h’s, which originally became silent in French before the words were picked up by English, in a few words like ‘honour,’ ‘honest,’ ‘hour’ and ‘heir.’  Some French silent ‘h’s also show up in the middle of words, such as ‘exhibition,’ ‘vehicle’ and ‘abhor.’  All these ‘h’s have been kept around in spelling because of the Latin words they connect to:  for example, ‘hour’ connects to ‘horoscope’ while ‘exhibition’ connects to ‘inhibition’ and ‘prohibition.’  Like so many English words, besides serving their function in communication, they are also a lesson in classical literature.  The Latin textbook is seldom seen in schools any more, but it is still built into our language. 

 

There is much more silent-letter action in words that have been adopted recently from modern French.  A few words from older French also retain French silent letters other than ‘h.’   I will just mention one common feature of French silence, as seen in words routinely used by English speakers.  If the last letter of a French word is ‘s,’ ‘t’ or ‘x,’ it is usually silent when a vowel comes before it.  From this trend we get words like the help-yourself feast ‘buffet’ (pron. buffay in English, büffeh in French), the traditional weight system ‘avoirdupois’ (pron. ‘av wahr doo pwah’ in English, roughly ‘avwahrdüpwa’ in French, keeping in the mind that the ‘r’ is deep in the throat) and the love letter ‘billet-doux’ (pron. ‘beeyei dooh,’ in English, ‘beeyyeh du’ in French [yy indicates a hard ‘y’ where the back of the tongue is raised to nearly block the breath, and ‘du’ sounds like ‘dooh’ but is much shorter]; literally, the expression means ‘sweet note’).   The ‘exact right thing to say,’ something we are in search of here, is called the ‘bon mot,’ the ‘good word.’   The French pronunciation of ‘bon’ includes a nasal vowel that can’t be represented by any combination of English letters, though I have roughly represented it in the text so far as ‘-ohn.’  The ‘n’ is neither silent nor normally sounded, but instead gives the ‘o’ a nasal quality.   This is accomplished by producing a short ‘o’ sound as in ‘gone’ and at the same time raising the back of your tongue to direct the sound to the point where the top of the back of your mouth connects to your nasal passages.  When English speaking people say ‘bon mot,’ they often try to produce this sound, especially if they are Canadian.  After all, we Canadians have French as one of our two official languages, even though over 80% of us don’t speak it competently.  Our best chance to fake remembering our school French is to say ‘bon mot’ reasonably Frenchly, that is, à la française.  Those who use the French pronunciation say ‘bohn moh’ or, as one might represent it, ‘boⁿ moh.’  Those who aren’t used to nasals may say ‘bonn moe’ or ‘boe moe’ to rhyme with ‘toe toe.’  We always reserve our English right to mispronounce other languages when necessary.  In the case of the word ‘chiffon,’ which ends with a nasal ‘-on,’ English speakers have given up completely and just say ‘shiffonn.’  But then again, with that word, we have taken a French term for an old rag and raised it up to mean ‘thin, silky high quality cloth.’  Like a beggar girl who has become a movie star, ‘chiffon’ has naturally re-styled her pronunciation.  It’s ‘Chiffonn,’ dahling. 

 

The French hard ‘y’ seen in ‘billet doux’ is fairly common in words that have come into English.  Some people may know the sound I am referring to if I say that it is similar to the Spanish ‘ll’ in ‘llama’ (Spanish pronunciation ‘yyahma’, at least in dialects where ‘ll’ hasn’t turned into a ‘zhy’ sound).  Upscale restaurants happily serve us bouillabaisse (‘booeeyyabess’ if meticulously pronounced, otherwise ‘booya-base’) and the pastry dessert ‘mille-feuille’ (pron. roughly ‘meel-fuiyyeh’ by the careful and ‘mill-fay’ or ‘mil-whatever’ by the intimidated; the ‘u’ in ‘fuiyy’ is as pronounced in ‘put’ rather than as in ‘fuss’ or ‘futon’).  ‘Mille-feuille’ actually means ‘a thousand leaves,’ a great name for this confection made of many layers of vanilla cream fondant and thin phyllo pastry (‘Phyllo,’ obviously similar to ‘feuille’ when you sound it out, is the Greek for ‘leaf’).  A third common hard-‘y’ word, ‘ratatouille,’ was made very popular by an excellent animated movie.  The rodents infesting the movie firmly established their slack pronunciation ‘ratta-twee’ as predominant over the careful one, ‘rahta-tooeeyyeh,’ in our culture. 

 

Other French words in English may not give us trouble with silent letters or unusual sounds, but just surprise us with their deviant spelling.   For example, the line-up word ‘queue’ – which in French means ‘tail’ and is pronounced ‘kuh’ (with ‘-uh’ sounding somewhat like the ‘oo’ in ‘foot’) –  is something we happily pronounce ‘kyew,’ just as we pronounce the word ‘cue’ and the name of the letter ‘q.’  The spelling, however, seems overblown for such a simple word.  In French, ‘qu’ is always ‘k,’ not ‘kw,’ and then you have ‘eu’ as the basic ‘uh’ vowel sound and an extra ‘e’ to show that the word is feminine in grammatical gender.  It all makes sense somewhere, but not in English.  ‘Quay,’ a ship dock, is a constant source of amazement:  it looks like it should be pronounced ‘kway’ but instead it is ‘kee’ like the word ‘key.’  The French version is ‘quai.’  Now you know that this is a French ‘qu’ here, ergo it sounds as a ‘k,’ and French ‘ai’ sounds close to ‘e’ as in ‘jet.’  The ‘ee’ in the English version is our mispronunciation of the ‘ai.’ It almost, sort of, kind of makes sense, a little.  The word is odd to start off with:  it’s a Gaulish Celtic word from ancient France, ‘caium,’ related to the Welsh ‘cae’ meaning a fence.  ‘Quay’ was commonly spelled ‘key’ in Middle English and seems to have had scholarly surgery done to restore its amputated French limbs.  The regular word ‘key’ for a lock-opener is from guttural Old English ‘cæg’ (‘kagh’) and is unrelated. 

 

French, as a mother-language of English, has high privileges inside our language and so is entitled to give us both exotic pronunciations and exotic letters.  In terms of pronunciation, we have well-integrated English words like ‘foyer’ (pron. ‘fwa-yay,’ whereas, if it were pronounced as in regular English, it would be ‘foi-er’), ‘sabotage’ (‘sabo-tahzh,’ not ‘sabo-tayge’ or ‘saybotahzh’), ‘vague’ (‘vayg,’ not ‘vag-yoo’) and ‘routine’ (‘rue-teen’ not ‘rue-tine’ to rhyme with ‘wine’).  These pronunciations are all based on French pronunciations, though some, like ‘vague,’ have become mispronunciations over time.  In terms of letters, we commonly see the French acute ‘e’ in English, as in ‘risqué,’ ‘résumé,’ appliqué, blasé, cliché and many other words.   Other French accent marks are occasionally seen, as in ‘derrière,’ ‘crêpe,’ ‘déjà vu,’ ‘ménage à trois,’ ‘tête-à-tête,’ ‘papier mâché,’ ‘soupçon,’ ‘naïve,’ ‘vis-à-vis’ and, as la crème de la crème, based on having the most different accent marks, ‘crème brûlée.’  Words from more distantly related languages like Czech have their accent marks shaved away like a bad haircut before they are allowed to come into English, but French can get away with just about anything.   As long as we can say it wrong when we need to, we’re happy. 

 

And now, right here at the end of our ghost story series about silence and eerie spelling, we encounter a group of people who are even harder on spelling and pronunciation than scholars:  the military.  The people who gave us the ac age (if you remember that from chapter 1) by coining ‘flak’ and ‘radar’ have also made noble contributions to puzzlement and head-scratching among students of English.  One of their most puzzling words, though, may have been a scholarly effort.  ‘Colonel’ came into English as ‘coronel’ from medieval French; its pronunciation soon became ‘kernel’ and stayed that way.  In 1583, the spelling was changed to make the word more closely resemble the equivalent Italian word, ‘colonnella.’ The source of the Italian name for the rank was the expression ‘compagna colonella,’ ‘small column company.’  This, of course, referred to a column of soldiers forming the unit size called a ‘company.’   When the spelling changed, people who had to pronounce the word on a regular basis paid no attention at all.  There was no military advantage in starting to speak Anglo-Italian all of a sudden.  The army simply walked around the letter ‘l’ and carried on its way. 

 

The military, indeed, was linguistically French in its terminology, presumably because William the Conqueror took the island in 1066.  The word ‘army’ itself is French.  Typical silent French letters are seen in ‘corps,’ pronounced ‘core.’  ‘Sergeant’ is French, though it should be pronounced ‘sair-zhyahn,’ rather than ‘sar-jent.’  Just teasing, ha ha – ‘sarjent’ is here to stay, sir!  It derives, though, from the Old French pronunciation of the Latin ‘servientum’ for a servant.   It is, after all, a relatively low rank.  The spelling and pronunciation changes do a great job of masking this embarrassing fact.  Somehow, I think military discipline would be disrupted if we relabelled every sergeant ‘Servant.’  Yes, sir, Servant, sir!  ‘Soldier,’ meanwhile, has become ‘sole-jer’ through ordinary yod-coalescence, if you remember that term:  the fusion of a little ‘i’ sound with the consonant before it, softening the consonant up.  This is the same phenomenon that turns what would be a ‘z’ sound in ‘confusion,’ though it’s written with an ‘s,’ into a ‘zh’ sound.  So if you’re confused about yod-coalescence, the very word for your condition –‘confusion’ – offers you a clue.  And since we’ve mentioned a ‘condition,’ so does the softening of the ‘t’ in that word to an ‘sh.’  You’ve been ambushed by understanding coming at you from all directions, right here in the military paragraph.

 

One of the oddest military words is ‘lieutenant.’  This is a pure piece of French meaning ‘place’ (‘lieu’) ‘holding’ (‘tenant,’ present participle of ‘tenir,’ ‘to hold’).  The strict CRP pronunciation in English is ‘lyoo-ten-ent;’ more relaxed dialects may say ‘loo-ten-ent.’ Americans still respect this nearly French pronunciation, perhaps because the French general Gilbert du Motier, the Marquis de Lafayette, came to the aid of their country’s independence.   They would never go so far as to say the nasal vowel at the end of ‘tenant’ (pron. ‘tenahn’ in French) or to pronounce ‘lieu’ in the French way as ‘liö’ (almost like lee-uh, with ‘uh’ like ‘u’ in ‘put’), but other than that, they use the French version.  English Canadians and many Britons are another matter; we say ‘lef-tenant.’  Why?  No one really knows, but earlier in this book you’ve seen a word where an ‘eu’ was pronounced as an ‘ef.’  That was ‘Ευχαριστώ’ (‘efkharisto,’ spelled ‘eukharisto’), the Greek word for ‘thank you.’  Modern Greek, in fact, turns all ‘eu’ words into ‘ef’ or ‘ev’ words.  No sooner was the euro issued as a currency than it became an ‘efro’ in Greek, sometimes transcribed as ‘evro.’  In a magnificent gesture of new friendship, the Turkish language then picked up ‘evro’ as its word for the euro, too.  The same sort of thing happens with ‘au’ in Greek as well:  my favourite sweet Greek dessert wine is μαυροδάφνη, ‘maurodafnē,’ pronounced ‘mafrodafnee’ and usually transcribed in English as ‘mavrodaphne’ or ‘mavrodaphni.’  British soldiers have been lurking around the eastern Mediterranean off and on since the Crusades, so perhaps somewhere along the line, local Greek speakers began to call the lieutenants ‘leftenant.’ Or, perhaps there is always some tendency in any language for ‘eu’ to become ‘ev;’ Classical Latin script did not distinguish between ‘u’ and ‘v,’ and whether the letter ‘و’ in Persian is pronounced ‘u’ or ‘v’ depends on the context.  This letter, called ‘waw’ in its original Arabic designation, doubles in Arabic as both the ‘u’ in words like ‘Mahmud’ and the ‘w’ in words like ‘wadi.’  (The distinction between ‘u’ and ‘w’ is ignored in various other languages as well.  For example, in French transcription, both sounds become ‘ou,’ as in Ouagadougou, the capital city of Upper Volta, which if it had first been written in the Roman alphabet by an English person would have been ‘Wagadugu.’).  Hebrew has a parallel situation with the letter ‘vav’(ו) , which can be ‘u’ or ‘v,’ though the two sounds can be distinguished by a little extra dot that is used to indicate the vowel sound in formal text.  Wikipedia says, “The rare Old French variant spelling ‘luef’ for Modern French ‘lieu’ (‘place’) supports the suggestion that a final ‘w’ of the Old French word was in certain environments perceived as an ‘f’.”  So the change of the ‘u’ in ‘lieutenant’ to an ‘f’ sound is not so much a surprise as a ‘whodunnit:’ the mystery is, who was the person or group of people who first introduced this ‘u to v’ switch in an English word, and what motivated them?   That’s a true military secret. 

 

Since naval forces and associated mariners were the first-ever speakers of English, as seen in the ‘cyulis’ stories in chapter 1, they can have the last word(s) in this chapter, too.  Their story is very simple: they just like to shorten words down.  ‘Captain,’ as often as not, is pronounced ‘cappen’ (usually written ‘cap’n’).  The merchant marine title ‘boatswain’ has been pronounced ‘bosun’ for so long that some dictionaries are starting to spell it that way.  ‘Ensign’ has also shortened up to ‘ensen’ rather than ‘en-sine.’  ‘Gunwales’ were spoken of as ‘gunnels,’ the forecastle of a ship as a ‘fo’c’sle’ (pron. ‘foke-sil’), the mainsail as a ‘mains’l,’ the nuclear torpedo as a ‘nudo,’ and so on.  Actually, I lied about that last one.  But it may happen any day.  

 

So there you have it.  In most languages, words are used to tell stories.  In English, they ARE stories, and their use in text is just an extra layer of storytelling.  When some English words became phonetic in the middle ages and lost their stories about relating to Latin, Greek or Germanic, scholars quickly changed the spelling to keep the stories intact.  If that meant introducing silent letters, no problem.   The silent letters actually told the historical story of the word.  To anyone interested in language, they spoke louder than the letters that were voiced. And why shouldn’t words comment on themselves?  In 1979, the mathematician Douglas Hofstader wrote a difficult but irresistible best-selling book called Gödel, Escher, Bach, an eternal golden Braid.  It was about logical self-reference and its consequences, as illustrated in the paradoxical statement, “this statement is false.”  The way English words comment on themselves – for example, the way ‘debt’ states, via its silent ‘b,’ that ‘this word’s spelling owes a debt to the Latin debitam’ is the sort of thing that Hofstader found delightfully clever.  It provides an extra dimension, increasing the truth value of the word.  As a left-leaning academic author might say, ‘instead of arrogantly pretending to be a universal Idea, each word admits to having an earthy, historical origin.’  Thank you, then, student of English, for learning these thousands and thousands of linguistic tales as you learn your English spelling.   You have no idea – or at least you didn’t, until now – how much culture you are keeping alive. 

 

Chapter 3 --> https://justpaste.it/43xln