Episode 160: Approximant-ly English

Welcome to the History of English Podcast, a podcast about the history of the English language.

This is episode 160, Approximately English.

In this episode, we're going to continue our look at the sound of Elizabethan English.

Specifically, we're going to look at a couple of sounds that modern linguists call approximates.

These are the L and R sounds, and they're some of the most challenging sounds in the language.

They're challenging because they exist somewhere in the gray area between vowels and consonants, and they have strange effects on the vowel sounds that surround them.

They also have a tendency to disappear over time.

So this time, we'll examine the history of the L and R sounds in English, and we'll see how they shaped the language we speak today.

But before we begin, let me remind you that the website for the podcast is historyofenglishpodcast.com, and you can sign up to support the podcast and get bonus episodes at patreon.com/slash historyofenglish.

Now last time we looked at the sound of Elizabethan English by exploring the writings of John Hart.

Hart was a writer and spelling reformer who argued that English words should be spelled phonetically.

He wrote three works about the topic in the mid-1500s, and those works are important because they describe in some detail how words were pronounced at the time.

Hart created a phonetic alphabet to illustrate those pronunciations, and he wrote long extended passages using that alphabet.

As a result, his writings provide us with a great deal of information about the pronunciation of English in the Elizabethan period.

In the last episode, we focused on a variety of consonant sounds, specifically those that are distinguished distinguished by their voicing.

Well, this time, we're going to look at a couple of other consonant sounds, the sounds represented by the letters L and R.

And these are actually some of the most difficult sounds in the language to analyze.

Hart described the sounds as liquids, a term that some linguists still use today.

And unlike most of the other sounds of English, Hart didn't really describe how those sounds were pronounced in any detail.

Now, if you've listened to the podcast from the beginning, you may have noticed that I haven't had much to say about the L and R sounds.

In fact, some of you have reached out to me over the years to ask if I was going to talk about the R sound at some point.

After all, the way that sound is pronounced and whether it's even pronounced at all is one of the key features that distinguishes the various accents of English.

But one of the reasons I haven't discussed that sound before now is because so little is known about the pronunciation of that sound before the modern English period.

And even where we have some evidence, it's often contradictory, or subject to differing interpretations.

The L and R sounds are tricky to analyze because they don't really work like normal consonants.

And as I noted, they exist in a gray area between consonants and vowels.

In fact, those two sounds, together with the W and Y sounds, are usually placed together in a special category of sounds called approximates.

And it's the vowel-like quality quality of those sounds that make them so challenging and fascinating.

Now my original plan was to discuss all four of those approximate sounds in this episode, but there was so much to cover that I decided to break them into two parts.

Phonetically speaking, the sounds associated with letters L and R are closely related to each other.

Again, they're sometimes called liquids.

And similarly, the sounds associated with the letters W and Y are also closely related to each other.

They're sometimes called semivowels.

So given the natural division between these sounds, I'm going to discuss the first pair in this episode and the second pair in the next episode.

But again, all of these sounds have similar qualities, and that's why they're sometimes grouped together under that general heading of approximates.

So why are they called approximates?

Well, it has to do with the way they're made or articulated in the mouth.

And we need to begin by recalling the basic difference between vowels and consonants.

As we've seen before, vowels are produced in the open cavity of the mouth, with no restriction or blockage.

The main thing that determines the sound of the vowel is the shape of the tongue.

As we raise and lower either the front or back of the tongue, we change the sound of the vowel.

But otherwise, the sound is open and fluid and unrestricted.

By contrast, consonants are produced with some type of constriction in the vocal tract.

So, for example, the sounds that we associate with the letters K and G are produced in the back of the mouth by briefly closing the vocal tract there and then releasing the air to make the sound.

K,

g.

The sounds of other letters are made in a similar way by blocking and releasing the airflow in the front part of the mouth, like bu,

p,

t, and d.

Again, those are all consonant sounds.

Then we have a group of sounds where the airflow isn't actually blocked, but it's tightly constricted, creating a hissing or whooshing sound, like the sound of letter S, s, and the sound represented by CH, ch,

and the sound represented by letter J, j.

Again, in all those cases, there's some type of blockage or constriction of the vocal tract to produce those sounds.

But then we have this separate group of sounds that are somewhere in the middle.

There's some type of constriction going on, but the constriction is so subtle or so weak that the tongue is actually doing much of the work to shape the sound.

So in that regard, these sounds are sort of like vowels since the constriction is subtle and slight.

And that's why linguists call them approximates.

Approximate means near or close by, and in this case, it refers to the various parts of the vocal tract that produce sounds, like the tongue and palate palate and teeth and lips.

When we make these approximate sounds, the various parts of the vocal tract tighten or narrow or restrict just enough to create the sound, but not enough to obstruct the sound or produce friction.

So let's think about the R sound for a second.

And when I say the R sound, I mean the sound that I'm using here, r.

As we'll see later in the episode, that's the most common R sound in modern English, but it isn't the only R sound.

Anyway, this particular R sound, R, is an approximate.

If you say the word ru, as in you'll ru the day, notice how little movement there is between the initial R sound and the U vowel sound that follows it.

R.

For the R sound, your tongue makes a specific shape, and there's a slight tension or constriction of the vocal tract.

And then, when you move to the vowel, the tension is released.

R.

But again, the change is very subtle.

As we'll see next time, something very similar happens with the W and Y sounds, which is why they're also called approximants.

But what about the sound we associate with the letter L?

L?

At first glance, there seems to be something a little different going on there.

When we make the L sound, the tip of the tongue actually touches the ridge behind the upper teeth.

Again, you can try it for yourself.

L.

We actually do something very similar when we make the sounds associated with the letters D and T.

We put the tongue in a similar spot.

D

t

l.

For all three sounds, the tongue touches that ridge, and it's that contact that actually produces the d and t sounds.

But with the l sound, that contact is not what actually produces the sound.

The l sound is made because the tongue narrows to allow the air to flow around it on each side.

So again, it's an open, flowing, continuous sound, and it's being shaped by the tongue, much like a vowel.

It just so happens that we make that tongue shape by lightly touching that ridge.

And sometimes we don't even do that.

If you say words like solve and tilt, you may notice that the tongue never actually touches that ridge behind the teeth.

Again, it varies a a little bit from person to person, but that's why the L sound is also considered to be an approximate.

It's because the various parts of the vocal tract are brought together in a way that they shape the sound without actually blocking the sound, or constricting it to the point where it creates friction or turbulence.

Now I mentioned a moment ago that the four approximate sounds are often divided into these two separate pairs, with the L and R sounds being closely related to each other.

Most linguists tend to think of them as sister sounds.

So why is that?

Why are those two sounds usually lumped together?

Well, think about the word roller.

It begins and ends with an R sound, and it has an L sound in the middle.

If you say the word roller, you may notice that the main thing that distinguishes the sounds in that word is the movement of the tongue.

The tip of the tongue starts in a position behind that ridge above the upper teeth, and then to make the L sound, the tongue tip reaches up briefly to touch that ridge.

That gives the tongue the shape it needs for the L sound.

And then at the end of the word, the tongue is released, so it can return to the R position.

And if you say the word very slowly and extend that L sound in the middle, you can feel how the tongue is shaping those two sounds.

And that's really the main difference between the two sounds.

Now, depending on your dialect, especially if you're from England or Australia or New Zealand, you may not actually make that R sound at the end of the word roller.

Now that's a whole different issue that we'll get to later in the episode, but either way you should be able to see how the L and R sounds are similar to each other.

They're both somewhat open sounds that flow continuously without any friction, and they're both shaped by the movement of the tongue to control the way air moves around it.

When John Hart considered these sounds in the mid fifteen hundreds, he called them liquids, and that term is also used by linguists today.

The use of the term liquid for those types of sounds can actually be traced back to descriptions used by the ancient Greeks, who described those sounds as being wet or moist.

Latin scholars used the Latin word liquidus to describe the sounds, which was then rendered in English as liquids by writers like John Hart and others in the 1500s.

Now, as I noted, the L and R sounds are fascinating because they don't really behave like other consonants.

Because of their vowel-like qualities, they tend to do weird things to the vowels around them, especially to the vowels that precede them.

They pull and tug on the vowels.

Sometimes they elongate the vowels and break them into diphthongs.

Sometimes they lower the vowels, producing a brand new vowel sound.

Sometimes they essentially merge with the vowel, creating a blended sound, like the er sound.

sound.

And sometimes they disappear altogether and let the vowel do all the work.

And the reason why this is so important to our story is because the L and R sounds essentially live by their own rules.

Even the systematic changes of the great vowel shift often broke down when a vowel was located next to an L or an R.

The rules are just a little different when dealing with these two sounds, so this time I'm going to take you through the history of these two sounds in English, at least through the Elizabethan period.

And if you stick with me, I think you'll discover a lot about the language that you didn't know and probably never even thought about.

So let's begin with the L sound, L.

As we look at the history of the L sound in English, I'm going to focus on situations where the L sound disappeared in a word, and sometimes reappeared.

I should note that the L sound also had a tendency to alter the vowel sound that preceded it, but I'm not going to explore that issue in this episode.

There was already enough to discuss without getting into those vowel changes, but I did want to mention that vowel sounds were sometimes altered through the process I'm going to describe.

Now it appears that the L sound was pretty stable in Old English, but we start to find some notable changes after the Norman Conquest when we enter the period of Middle English.

It was during that period in the 1300s that we have evidence that the L sound disappeared in an entire group of words.

In fact, these were very common words, like each, such, much,

and which,

W-H-I-C-H.

Now, today you would never know that those words once had an L sound, because none of those words are spelled with a letter L.

Of course, there's no L in those words because the L sound disappeared before the spellings became fixed.

But if you read Old English documents, you'll find that those words are almost always spelled with an L, either before or after the CH sound.

So which was pronounced more like hwilch,

each was Elch,

such was swilch,

and much was mutchel.

The L sound started to disappear in those words after the Norman Conquest.

For some reason, the close proximity of the L sound to the CH sound seems to have been the trigger for the change.

I should also note as an aside that the L sound in those words did hang on a bit longer in the north of England and in Scotland.

In those regions you can still hear the word mickle for much,

and mickle is just an older form of the word much where the L sound was retained.

Now after the L sound disappeared in those words, another group of words started to lose their L sounds as well.

And those are the words would and should.

Of course those words are still spelled with an L today, so the sound was lost a little later in those words after the spelling had become fixed.

I actually talked about that development back in episode 149, and if you have a good memory, you might recall that the loss of the L sound in those words also affected the word could.

Could didn't actually have an L sound in its original form, which was kuth,

and which later evolved into could.

But it acquired its L since it was closely associated with would and should.

Could, would, and should had similar grammatical functions, and since people thought of them as related words with related pronunciations, they also apparently thought that they should all be spelled the same way.

So could got its silent L to match would and should in the 1500s, shortly before the time of John Hart.

Interestingly, Hart spells all three words with an L in his phonetic spelling system.

Other writers who used a similar spelling system in the decades after Hart also show an L in all three words.

So that confirms that the pronunciation was consistent between them, even though it seems a little odd that could had picked up an L sound where one had never existed before.

Presumably, the influence of the older pronunciations wuld and schuld was so great that it also affected the pronunciation of could for a while in the Elizabethan period.

We also have to keep in mind that Hart's transcriptions transcriptions reflect the speech of the London upper class.

So whereas Hart and some of his contemporaries were still pronouncing the L sound in those words, it's very likely that many common people on the street had already dropped those L's.

Now around the time when John Hart lived and wrote, there was another much more widespread loss of the L sound underway.

But in this case, the loss was limited to certain specific situations.

Now the details here are not really important to our story, but just in case you're curious, the L sound disappeared in a word when it occurred after an A or O and before an F, V, M, or K.

See, I told you it was kind of specific.

Again, the details aren't really important, but it appears that the L sound disappeared in those words because of the specific phonetic environment created by those sounds.

That environment caused the L sound to be lost in words like talk, walk, chalk, folk, yoke, half, calf, behalf, save,

calm, and bomb.

Now you may be saying, wait a minute, the L sound is pronounced in some of those words.

Well, it's certainly possible to hear some of those words pronounced with a subtle L sound today, and again, that may come from the influence of spelling.

Also, some people people think that they hear an L sound, but they're actually hearing a vowel sound that has a texture that's similar to an L sound.

So for example, talk has an aw vowel sound, like in the word cot,

so there's not really an L in there.

We don't say talk,

we say talk with the aw vowel, or in some dialects you may say talk with the ah sound.

But again, there's no L there.

Now, this more widespread loss of the L sound appears to have gotten underway in the north of England in the 1400s, and it gradually spread southward.

John Hart only has a couple of examples of these words, specifically chalk and half, but in both cases he spells them phonetically with an L, implying that they were still pronounced with an L in educated London speech in the 1500s.

But other writers who came along immediately after Hart, like William Bullicar in the 1580s, indicated that the L's were silent in those words.

And another early linguist named Alexander Gill wrote in the early 1600s that the L's were pronounced in educated speech, but often dropped in the speech of most common people.

So again, Hart's transcriptions reflect an older, more traditional pronunciation where the L's were still being pronounced.

But they were definitely on the way out.

Now I should also mention one other important development in the 1400s and 1500s that complicates this discussion a bit.

As we saw in prior episodes, it was common in the Tudor period for writers and printers to change the spelling of English words that had originated in Latin or Greek.

They often reintroduced letters that had been lost over time because they were no longer being pronounced.

And by reintroducing those old letters, they brought the spellings more in line with the original Latin and Greek spellings.

But that also meant that those words now had silent letters that just stuck out like a sore thumb.

You might remember that that's how we got the B in doubt and debt, and the P in receipt.

Well, as it turns out, the L sound had also disappeared in a lot of words over the centuries.

So in the 1400s and 1500s, many of those loan words also got brand new L's to reflect their roots.

And as we saw in those earlier episodes, those new L's sometimes changed the pronunciation of those words.

People often pronounced words like they were spelled, especially if the words were newer loan words where the pronunciation was uncertain.

So let me give you a few examples to illustrate this development.

English had borrowed the word falcon from French in the twelve hundreds.

You might remember that falconry was a popular sport among the Norman nobility, so the word falcon came in immediately after the conquest.

Well, the word falcon was derived from the Latin word falconem with a distinct L sound.

But the word lost that sound in French, and when English borrowed the word as falcon it was spelled F A U C O N.

There was no L.

Then in the 1500s the L was added back into the word to reflect its Latin roots, and to day some people pronounce the word with the L sound as falcon, and some people pronounce it without the L sound as falcon.

The same same thing happened with words like fault, vault, and assault.

Those words were actually borrowed as falt, vaut, and assault, without an L in either the spelling or the pronunciation.

But all of those words had an L in their original Latin forms, and after the letter L was reintroduced in English, that L sound started to be pronounced again.

And today, most English speakers tend to pronounce those words with a slight L sound.

In some cases, like the words palm, psalm, and salmon, the words were borrowed into English without an L sound, and even after an L was reintroduced in the spelling, the L sound remained silent.

Now some people may say salmon today, thanks to the revised spelling, but the standard pronunciation omits the L sound, which was the way the word was borrowed from French.

Now again, when we look at the writings of the early Elizabethan linguists, we find a mixed bag when it comes to the pronunciation of these words.

John Hart actually spells the word fault both ways, one way with an L and one way without.

Again, as I noted a moment ago, Alexander Gill, writing in the early 1600s, says that the L in fault is pronounced in educated speech, but often dropped in colloquial speech.

So his comments suggest that educated speakers were trained to follow the spellings more closely than common speakers, and they had therefore picked up the L sound based on the revised spellings.

In the mid-1700s, Samuel Johnson published his Landmark Dictionary of British English, and he included a comment about that same word, fault.

He wrote that the L in the word was sometimes pronounced and sometimes silent.

So, almost two centuries after John Hart, the pronunciation of that word and many others like it was still unsettled.

So, in summary, the L sound was lost in many native words, like which and such and would and should, and talk and walk, and half and calf.

And the L sound was also lost in many words that were borrowed into English.

But over time, some of those loan words got their original L's back, and some of those native English words continued to be spelled with their original L's that had gone silent.

And that created a lot of silent L's in English.

And then, as inevitably happens, some of those L's started to be pronounced again because people tended to pronounce words like they were spelled.

So, as you can see, that L sound was very unstable in a lot of words in the 1400s and 1500s.

Now, let's turn our attention to that other closely related approximate sound, the R sound.

because it's also been very unstable over the centuries.

The R sound is actually one of the most difficult sounds in English to analyze historically, and there are several reasons why it's so challenging.

First, there isn't one specific sound represented by the letter R.

As we'll see, there are actually several different R sounds that are quite distinct from each other.

And those variations have existed in English for centuries.

And since they're all represented with the same letter R, we can't really rely on spellings to tell us which sound was being made at any particular time.

The second problem with trying to trace the development of this sound is that it can be difficult to describe exactly how these various R sounds are made in the mouth.

So even when writers in the past tried to explain the sound, they didn't really know how to describe it in a way that makes it clear to us today what the sound was.

The third problem is that not all dialects use the R sound the same way.

Some dialects usually drop the R sound altogether after a vowel.

That's one of the most distinctive features of the English spoken spoken in most of England, as well as in Wales, Australia, New Zealand, and South Africa.

So whereas an American might say birth or bear with a distinct R sound, someone in England might say birth or bear without that sound.

That R less pronunciation became widespread in England in the 1700s, after English had already been established in North America.

So the R sound doesn't always work the same way throughout the English-speaking world.

And that difference really complicates any discussion about the R sound because what might be true in one place isn't necessarily true in other places.

So hopefully you can start to see why I haven't tried to tackle the R sound before this point.

It's a challenging topic with a lot of historical uncertainty.

But as we move into the Elizabethan period, we start to come across more and more descriptions of the sound.

And even though there's still some uncertainty, we can finally get a sense of how and where the sound was pronounced, and the impact that it was having on the vowel sounds around it.

So let's begin our look at the R sound in English by considering the different versions of this sound in the language.

As I noted a moment ago, the most common R sound is the R sound heard in words like run and ride, and at the end of words like car and sir.

Now, mechanically, there are different ways of producing this type of sound.

In North America, it's common for people to produce the sound by bending the tip of the tongue upward and backwards a bit.

This is sometimes called a retroflex R.

There are other ways to shape the tongue to get a similar result, but ever how you do it, it's an approximant.

The tongue doesn't actually touch the palate or the teeth or any other part of the oral cavity.

So the tongue shapes the sound similar to the way it shapes a vowel sound, but there's a bit more constriction than is normally found with a vowel.

Now I'm going to call this type of r the standard modern English R, because it's the type of R found in most standard versions of English.

But it isn't the only R sound that's found in English, and it's probably not even the original R sound.

Another common R sound, which is probably the older and more traditional sound, is the trilled or rolled R, r.

This sound is also found in many languages.

A lot of people associate it with Spanish or Italian, but it's actually the most common type of R sound found around the world.

And as I noted, it was once very common in English as well.

In fact, it can still be found in English dialects in Scotland.

Here's a clip of a speaker from Glasgow that illustrates that trilled R in modern English.

By the way, this clip comes from the British Library Dialect Archive, which is available online.

There are words that aren't very, very Glaswegian, but you find that there are words associated with the river that change from Glasgow Green down into the harbour, down to the coast, different words, different places.

We take the word punt, P-U-N-T, a punt here is a small single rowing boat.

A punt down in Cambridge is a boat you stand on and shove away a pole.

A punt down in the harbour here, just down below the weir.

A punt is what we call a barge.

A barge down there is a motorised vessel.

So you actually find all over Britain there's river words that exist, but they have different meanings in different rivers.

Now, in addition to the trilled R that you heard in that clip, there's also a tapped or flapped R.

This sound is similar to the trilled R, but instead of an extended vibration, the tongue just briefly touches the ridge behind the upper teeth, one time.

I actually described essentially the same sound in the last episode when I talked about the way many Americans pronounce the T sound in the middle of words.

You hear it in pronunciations like butter and water.

Well, some speakers in parts of Britain do essentially the same thing when making an R sound.

They just briefly tap that ridge with their tongue, like merry for Mary.

Again, this can be heard in places like Scotland and Wales, and can sometimes even be heard in the standard received pronunciation of the South.

Here's a clip from a speaker from Aberdeen in Scotland who tends to use that type of tapped pronunciation.

Again, this comes from the dialect archives of the British Library.

Well, a tenement was at least three stories high, sometimes four stories high,

and

no one ever locked a door.

And people used to come and go all the time.

They'd walk into your house at any time, day or night, and you just said, Hi, I for you dee, and then you can, and that was it.

It was wonderful community living.

I wish we could go back to the old days.

We were all so close.

And we called all the wifies by their last name.

I had Howie, that was Mrs.

Howie, Strachan, Mrs.

Strachen, Brun, Mrs.

Brown,

and you just called them by the last name, and nobody objected to that.

It was a wonderful life.

Now, in addition to the R sounds that we've already explored, there are some languages that produce a different kind of R sound in the back of the mouth, in the throat region.

This is sometimes called a uvular R.

It's the type of R that you might hear in French.

Well, believe it or not, some English dialects also use or used a similar type of R.

In the northeastern part of England, it was once common to hear a type of back or uvular R sound.

It's largely disappeared over the past few years, but you still might encounter it in some pockets today.

Linguists referred to it as the Northumbrian burr, and it was often found at the beginning of words that began with an R sound.

So what did that Northumbrian burr sound like?

Well, as I noted, it's largely disappeared, so it's a little difficult to find audio clips to represent it.

But again, the British Library has archives of dialect samples that go back to the early 1900s, and those archives provide some examples.

Here's a clip of a speaker from the town of Wark that was recorded in the mid-1900s.

The speaker was actually born in 1889, so this is typical of an older form of speech.

And in this clip, you'll hear him speak about the robbery of a shopkeeper in a town near a hill called Raylis, but the speaker simply refers to it as the Ray.

And you'll hear him pronounce terms like Ray and Runoff and Robbed with that distinctive R sound at the beginning of those words.

About a hundred and fifty years ago,

there was a man come over

the Tain,

and went over into the woodside to a place called the Rare,

where there was an old woman kept a shop.

But they stopped half way on the hill

at at a stale, that's a round place on the hill where the sh where the sheep gone in for shelter.

And there was a haired laddie, he looked over her in the ware, and they saw him and the cuddy, two woman folk.

He he ran off, he ran away him, the laddie, and he told what he'd seen.

And uh these uh tinkles they went off right over Billsmoor and into the woodside and they broke into this poor old buddy shop and robbed them, knocked them senseless, and they couldn't say far away in the two women folk system again back on feddle.

So, as you can hear, the R sound in English can actually vary quite a bit.

And you might be tempted to think that some of those other pronunciations, like the trilled or tapped R, are just unique regional variations.

But that's not necessarily the case.

The bulk of the evidence suggests that English once contained a mixture of these sounds.

In fact, let me play another clip that you might find interesting.

This clip actually features an American, specifically a former American president.

William McKinley was president from 1897 until 1901.

His term actually ended when he was assassinated in office.

He was from Ohio, and he happened to be president at a time when audio recording devices were starting to record people's speech speech for the first time.

And of course, being president at the time, some of his speeches were recorded for posterity.

These are some of the earliest recordings we have of a sitting American president, so the quality isn't very good, but it's still fascinating.

At the time, there was more of a class difference in American speech than there is today, and McKinley spoke with a cultivated upper-class accent.

So his speech is a little bit different from those who were raised in a more working-class environment.

But when we listen to his speech, we find that he spoke with a mixture of trilled R's, tapped R's, and modern standard English R's.

So, check it out.

Our freed embraces an honest dollar,

an unparnished national credit,

adequate revenues for the uses of the government,

protection to labor and industry,

preservation of the whole market,

and reciprocity which will extend our foreign markets.

Upon this platform, we stand and submit its declaration to the sober and considerate judgment of the American people.

So, as you can hear, some of the various R sounds that I described earlier were once common in American English as well, at least in the speech of a certain cultivated type type of American English.

And that clip also points to a basic fact about the R sound.

As we go back in time, we find that the R sound was much more diverse and variable than it is today.

But here's the thing, no matter how it was pronounced, it was always spelled with a letter R.

And that poses a real problem for historical linguists.

As I've noted in prior episodes, English spelling was much more phonetic in the Old and Middle English periods.

Scribes tended to spell words like they sounded, and that actually helps linguists to determine how some words were pronounced in the past.

But phonetic spellings don't help if the scribes were using the same letter to represent all of those various sounds.

All we can really determine is that some type of R sound was being pronounced.

We just don't know which one.

Of course, it would have been helpful if writers had described that sound represented by the letter R, but we don't really have any English descriptions of that sound until the 1500s, and most of those accounts describe the sound with vague terms that can be interpreted many different ways.

So all of that makes it really difficult to pinpoint the nature of the R sound in early forms of English.

Now, having said all of that, most historical linguists think that the R sound was mostly trilled or tapped in Old and Middle English.

And you probably noticed in earlier episodes that I tended to use a trilled R when I read passages from those earlier periods.

And again, that's in keeping with the general view that the R was pronounced that way at the time, but it's really difficult to say anything definitive about the sound in those early centuries.

So why do most scholars tend to think the sound was trilled or tapped?

Well, it's partly because that's the most common way to pronounce the R sound across most of the world's languages.

It seems to be something of a default pronunciation, at least historically.

Also, the earliest descriptions of the R sound by English writers in the 1500s and 1600s also point to a trilled R, suggesting that it was still prevalent in the language in those later centuries.

We can also look to Latin for evidence.

It's generally agreed that Classical Latin also used a trilled R.

Of course, Latin is in English, but the Latin alphabet was applied to English.

For the most part, Latin and English had the same sounds, so the Latin alphabet was easily adapted to English.

But the Anglo-Saxons did make some minor changes to account for certain differences in pronunciation.

For example, we saw in earlier episodes that English scribes used the runic letter thorn for the th sound because Latin didn't have that sound and didn't have a specific letter for it.

So they did modify the Latin alphabet when they felt they needed to.

But they didn't modify the letter R.

They adopted it in a pretty straightforward way.

And that's further evidence that the English R was similar, if not identical, to the Latin R.

But again, none of that evidence is conclusive.

In fact, an alternate view is that English actually had a combination of R sounds using a variety of the sounds I described earlier.

And if that's true, the usage probably varied from region to region and even person to person.

But one thing that we can say with some certainty is that some kind of R sound was pronounced in words that were spelled with an R.

There's no evidence that speakers dropped the R sound after a vowel in Old English, like we hear in many dialects today.

So linguists would say that Old English was rhotic.

That simply means that the Rs were pronounced in words, no matter where the R sound appeared.

Accents that tend to drop the R sound after a vowel, especially at the end of a word or syllable, are called non-rhotic accents.

So general American English is rhotic, and the standard received pronunciation spoken in England is non-rhotic.

But historically, the English of England was rhotic.

The R's were pronounced.

The first evidence we have that some people were no longer pronouncing their Rs in certain situations comes from the 1400s.

Around that time, we start to find documents where some words were spelled without their typical R, especially letters and informal documents written by people in the southeastern part of England.

Remember that spellings tended to be phonetic during that period, so when an R was dropped in a word, it seems to suggest that the writer didn't pronounce the R sound in that word.

For example, you might remember that I talked about the letters of the Paston family in an earlier episode about the Wars of the Roses.

They lived in Norfolk in the same part of eastern England where the R's were often omitted.

You might remember that the Pastons were also involved in several property disputes with prominent nobles in the region.

In one letter from 1467, we find the word parcel, as in a parcel of land, rendered as passel, P-A-S-S-E-L,

and that reflects a type of pronunciation that's common throughout England today, but it was a rare exception in the 1400s.

Other documents from the 1400s reveal the word first rendered as as FUST, FUST,

MORSEL as MOSL, M-O-S-S-E-L-L-E,

Forrester as Foster, F-O-S-T-E-R,

Scarcely as Scosley, S-K-A-S-E-L-Y,

Porsche as POSHIN,

P-O-S-S-H-E-N-E,

and Dorset as Dosset, D A

In all of those cases, the R is omitted before an S or an SH sound.

Now, all of this evidence suggests that some speakers were starting to drop the R sound when it appeared after a vowel and before an S or similar type of sibilant sound.

So this wasn't a general loss of the R sound.

It was limited to certain very specific situations, and it appears that this pronunciation was also limited in scope.

It's mostly found in the region known as East Anglia in the southeastern part of England.

In fact, in the late 1500s, a scholar named William Bullacar referred to the loss of the R sound in that region as a, quote, widespread vulgarism, end quote.

But despite the limits of that pronunciation, its influence was strong enough that it actually produced a few R less words that have survived into modern English.

For example, the word bass for a type of fish is actually derived from the Old English word barse.

The r sound disappeared in the word in the late Middle English period.

And there were some other words that apparently evolved through that same process and gave us several word pairs where one version has its original r and the other doesn't.

So, for example, we have burst and bust,

curse and cuss, and arse and ass.

Now, modern linguists still debate whether this early loss of the arse sound in East Anglia is connected to the more widespread loss of the R sound throughout England in the 1700s.

But it's worth noting that the later, more widespread loss of the sound also appears to have its origins in the same general region in the eastern part of England.

From there, it gradually spread to London and then to most of the rest of the country.

One other interesting note is that many of the early settlers of the Plymouth colony in Massachusetts came from that same part of eastern England.

And today, the region around Massachusetts still has a very distinctive non-rhotic accent.

It's the Boston accent that so many people like to mimic when they say, pak, the ka, and havid yad.

Well, many scholars think that that tendency to drop the R sound after a vowel in parts of New England can be traced back to those early settlers from eastern England.

But we'll explore those developments in more detail in an upcoming episode.

In fact, since the widespread loss of the R sound after vowels occurred in England in the 1700s, I'll address that development in more detail in the future.

For now, it's only important to know that a more limited loss of the sound can be found in certain parts of the country in earlier centuries.

Now, I think the main thing to take from this discussion so far is that strange things happen when a vowel is followed by an R sound.

As we've just seen, sometimes the R sound disappears in a word.

But other times, the R sound actually changes the way the vowel is pronounced.

And that's the other phenomenon that has intrigued historical linguists.

Vowels just tend to do funny things when they appear before an R sound.

It may have something to do with the way the tongue shapes the sounds when they appear together.

But regardless, English has experienced some unique vowel changes when the vowel is followed by an R.

Some of the first evidence we have of vowel changes triggered by an R comes from the late 1300s and 1400s.

During that period, we find many documents where words typically spelled with ER received a brand new spelling with AR,

and that reflects a new pronunciation at the time.

So let me explain what was going on.

That was the Middle English period, and and just like today, the letter E could be used to represent the short E sound, e.

Remember that the short vowel sounds were not affected by the great vowel shift.

So today we use the letter E in the same way when we spell words like set and pet.

So the spelling ER represented the er sound at the time.

Well, beginning in Middle English and extending into the early modern period, words with that sound experienced two different vowel shifts.

First, in the Middle English period, the sound shifted from er

to r,

and that was the change reflected in those documents where the spelling shifted from er to ar.

That change is first noted in the late 1300s with the arrival of those new AR spellings.

Now, if you have a good memory, you might recall that I actually discussed that specific vowel change way back in episode 128, but this is a good time to revisit that discussion.

Thanks to that vowel change, a word like serve, pronounced serve at the time, was now sometimes rendered as serve, and a word like convert, pronounced convert at the time, was now sometimes pronounced as convart,

and learned, or learn at the time, became larn.

It appears that this vowel change originated in the East Midlands in the eastern part of the country, and it it soon spread into London.

But this particular vowel shift was never fully completed.

The older pronunciations did not entirely disappear.

So, for a time, both the air and r pronunciations existed side by side throughout much of England.

Some people said serve, and some people said sarve, and some people said larn, and some people said larn.

Gradually, over time, those differences were sorted out, but in many cases, it didn't happen until later in our overall story in the 1600s and 1700s.

Again, some of the words kept the new R sound, and some kept the older er sound.

Very generally speaking, the words that kept the newer R sound were actually the older words in the language, typically words from Old English.

That included words like heart, both the deer, H-A-R-T, and the organ in the body, H E A R T.

Both were pronounced as Hert before the vowel change.

It also included the word harvest, previously pronounced Hervest.

It included the word dark, previously pronounced derk.

It included star, previously pronounced stare.

It included yard, previously yard,

and farm, previously farm, and carve, previously carve.

You get the idea.

Again, those are very old words, going back to Old English, Old Norse, or in the case of farm, a very early loan word from French.

And notice that in all of those cases, the spellings have been modified over time to account for the modern pronunciations.

Those words are typically spelled with AR today.

Now, most of the other words that retained the original air sound, spelled er,

were loan words from Latin and French.

not always, but in most cases.

And those words experienced a second vowel shift from er to er in later centuries.

That included words like certain, servant, perfect, merchant, nerve, verb, and erb, or herb, depending on whether you pronounce the h.

Again, those words are still spelled with er to day, which reflects their original pronunciation with the er sound.

By the way, some English dialects still preserve that older vowel sound, so this discussion is mainly about the development of the standard varieties of English found in North America and England.

So, some very old words got a new R sound spelled with AR,

and some more recent loan words kept their original sound spelled with ER.

But of course, English is never quite that simple.

In some cases, a word kept the original ER spelling, but came to be pronounced with the newer R sound.

That's what happened with the word sergeant.

Of course, it's spelled S E R G E A N T, but it's pronounced sergeant, not sergeant.

Again, it was affected by the initial vowel shift, but the spelling was never revised to account for the change, probably because both pronunciations were common in the language until well after the spelling had become fixed.

And eventually sergeant died out, leaving us with sergeant and a mismatched spelling.

Of course, this type of thing happened quite a bit in England, where the spelling became fixed with the R while the pronunciation still varied, and eventually the pronunciation with the R sound won out.

But that didn't tend to happen in North America.

There the pronunciation tended to match the spelling.

That's why the word spelled C C-L-E-R-K is pronounced clerk in American English, but clark or clock in standard British English.

The same thing happens with the towns that Americans call Derby and Berkeley, but Brits call Derby and Berkeley, or Dobby and Bockley without the R sound.

Again, the spellings were mismatched in Britain because the spellings were fixed with ER before the pronunciations were finally settled.

Interestingly, since both pronunciations were common in England for many centuries, we have quite a few cases where both forms of the word survived.

I mentioned these in that earlier episode, but let me mention them again.

We have Vermin and Varmint, two different pronunciations of the same original word.

We also have university and varsity.

Varsity began as a shortened form of university, and it shows the switch to the R sound.

The words person and parson were also distinguished through that same process.

Of course, today parson has a specific religious meaning, but it was originally just the word person pronounced with that R sound.

Another example which I didn't mention in that earlier episode is the difference between the words dear and darling.

Of course, dear, D-E-A-R, is a term of endearment.

Endearment.

Dear is the root of endearment.

And you might refer to someone who you hold dear as darling.

Well, darling was originally a dearling, someone you hold dear.

But thanks to the vowel shift before the R sound, it became darling.

Now again, these competing pronunciations were still common around the current point in our overall story of English in the mid-1500s.

They weren't really resolved for another century or so, so it's common to find both forms in Elizabethan documents.

John Hart spoke with an educated London accent, and he generally avoided the newer R pronunciations, which he may have considered to be too common or colloquial for his tastes.

For the most part, he spelled most of those types of words with ER in his phonetic spelling system.

That included words like certain, person, perfect, verb, serve, and so on.

But in some words, the vowel change had become so ingrained by that point that heart does show the change in his spellings.

For example, he spells harvest with AR, which confirms that the modern pronunciation of that word was well established by that point.

He also used both spellings for the word heart, meaning the organ that pumps blood.

In one passage, he spelled it HERT, reflecting its older pronunciation as Hert,

and in another passage he spelled it HART, reflecting its modern pronunciation.

Again, that newer pronunciation was apparently well established by that point since Hart included it in his writings.

Maybe he pronounced it as Hart because his last name was Hart.

Who knows?

Now, based on John Hart's evidence, it's tempting to treat that newer R pronunciation as something only heard in the speech of the lower classes.

But that isn't the case.

It appears that even Queen Elizabeth used the newer R pronunciation sometimes.

In her personal letters, she spelled the word heard

as hard, H-A-R-D-E,

and she spelled person as parson.

And a few decades later in our story, Shakespeare showed the same development.

He rhymed deserve with starve, suggesting that he pronounced them as desarve and starve.

And he rhymed convert with art, suggesting that they were pronounced convart and art.

And he rhymed serve and carve, pointing to a pronunciation as sarve and carve.

So these alternate pronunciations were still common in the Elizabethan era.

Again, it would take about another century to sort out those pronunciations.

And speaking of the following century, that's when those words that retained that older er or er sound experienced that second vowel shift from er to the modern er.

Now that shift isn't really documented until the very end of the Elizabethan era in the early 1600s, so I'm going to reserve a discussion about that change for a future episode.

But there is one aspect of that later change that I want to mention here because it relates directly to the history of the r sound and the way that sound was pronounced in the Elizabethan era.

As I noted, words spelled with er and pronounced with the er vowel sound came to be pronounced with a new er sound in the 1600s, and that's the sound that most of those words still have today.

That er sound is really a combination of the r sound and that central neutral vowel called schwa.

Schwa is the

sound heard at the beginning of the word about

and at the end of the word china.

I talked about that sound before in the episodes about the great vowel shift.

It's formed by shaping the tongue in a somewhat neutral position, so it's not really a high vowel or a low vowel or a front vowel or a back vowel.

It just sort of sits there in the middle of the oral cavity.

And that tongue position is very similar to the tongue position used to make the standard modern r sound, r.

In fact, the schwa vowel and the r sound are so similar that when they appear next to each other, they basically merge together.

Modern linguists don't really think of the er sound as a vowel followed by an R sound.

They really think of it as a distinct vowel sound, a vowel with R-like qualities.

They actually call it an R-colored vowel.

And the reason why that's so important is because the standard modern R sound has a tendency to pull all of the short vowels into that position.

And that's actually what happened over the course of the 1600s.

Again, we'll explore the details in a future episode, but words spelled with er and pronounced er were pulled into that position and came to be pronounced as er.

And words spelled with ir and pronounced er were also pulled into that same position and also came to be pronounced as er.

And words spelled with or and pronounced or did the same thing, as did words spelled with ur and also pronounced or,

and words spelled with ear and pronounced er.

Thanks to the centralizing pull of that r sound, all of those vowel sounds merged together in the 1600s.

That's why words spelled with er,

like nerve and verb, have that sound, as do words spelled with IR, like bird and first,

and words spelled with or

like word and work,

and words spelled with you are, like hurt and nurse,

and words spelled with ear,

like earth and search.

In these cases, the R sound was like a collapsed star at the center of a black hole.

It sucked in everything around it.

And that's a major reason why English spellings are no longer phonetic in many words spelled with an R.

Before the 1600s, words like nerve, bird, word, hurt, and earth would have all had different vowel sounds, and the spelling would have been much more phonetic.

But today, that's no longer the case.

The R in those words has completely altered the vowel sounds over time.

Of course, given that the merger of those vowel sounds occurred a little bit later in our story in the 1600s, we would not expect to find any evidence of that merger in the writings of John Hart.

And that's actually the case.

In fact, part of the reason why scholars know that the merger occurred in the following century is because it isn't reflected in the writings of Hart and other linguists in the late 1500s.

Again, Hart's spellings preserve the original vowel sounds in words like bird, word, and nurse, suggesting that he used the older pronunciations, something like berd, ward, and nurs.

The other reason why I wanted to mention that upcoming vowel merger at this point is because it may provide a clue about the pronunciation of the r sound at the time.

Remember that there are several different R sounds, and scholars don't agree about the specific nature of that sound in earlier centuries.

As I noted earlier, the trilled R was apparently much more common in the past, even though other versions of the sound probably existed as well.

Of course, today we have that standard modern R sound, which is an approximate.

So when did that modern R start to replace the trilled R?

Well, some scholars think that the upcoming vowel merger in the late Elizabethan period was really dependent on the modern R already being in place.

They argue that the modern R sound was actually the trigger for that merger.

And if that was the case, the modern R sound was probably being used in the 1500s, at least at the end of words where this vowel merger took place.

So what's the theory there?

Well, the idea is that the modern R sound and the neutral schwa sound are pronounced with the tongue in similar positions.

And as we've seen, the two sounds can effectively merge together as a single single R-colored vowel, er.

So if the modern R sound was being used, then the schwa sound would have been a natural fit for it.

Even if you started with one of the other vowel sounds, you would tend to slide into the schwa sound on your way to the R sound.

So over time, there would have been a tendency to cheat a little bit and just begin the sound with the schwa, and that would have produced a somewhat generic er sound whenever a a short vowel appeared before the letter r.

Again, the modern r sound could have had a neutralizing effect in that position, and it might account for the merger of those various vowel sounds in the 1600s.

One of the leading advocates of this view is the Elizabethan English scholar Fausto Serciani.

This view finds some supporting evidence in the surviving descriptions of the R sound from the Elizabethan period and the period immediately after it.

Now, as I noted earlier, most of the descriptions of the R sound during the 1500s and 1600s are vague and subject to differing interpretations.

But one of the most notable descriptions comes from Ben Johnson, who was a poet and playwright and a contemporary of Shakespeare.

In a book he composed about English grammar in 1640, he provided a description of the R sound.

His description suggests that the R was trilled at the beginning of words, but was pronounced differently in the middle and at the end of words.

Here is his quote about the sound of letter R.

Quote, The dog's letter heareth in the sound, the tongue striking the inner palate, with a trembling about the teeth.

It is sounded firm in the beginning of words, and more liquid in the middle and ends.

Now again, the language is a little bit vague, and Johnson pulled much of that quote from an earlier French grammarian who wrote about the sound, but regardless, Johnson must have felt that the description was accurate.

It seems pretty clear that he was describing a trilled R initially, a sound that had been associated with a dog's growl going all the way back to the Roman period.

But then he says that the sound was more liquid in the middle and ends.

Well, that word liquid is subject to differing interpretations, but it could refer to something like the modern standard R.

And whatever that sound was, it was apparently different from the firm trilled sound used at the beginning of words.

That implies that a word like runner was pronounced something like runer.

And if that was the case, then we can see how that more modern R sound at the end would have provided an ideal setup for the er sound that was about to become so prominent as the various vowels merged together before that R sound.

We should also keep in mind that the modern standard R sound is the dominant R sound in England, North America, and Ireland.

And of course, it was during this same period in the 1600s when large-scale English settlements were established in North America and were expanded in Ireland.

So that provides some more evidence that the modern standard R sound was in place in England when that period of migration got underway.

Of course, that R sound was probably being used alongside the trilled R.

Writers were still describing a trilled R in the late 1600s and 1700s, and as we saw earlier, it was even being used in parts of North America in the late 1800s and early 1900s.

But the use of that trilled R declined over the course of the modern English period, and today its use is limited to a few places like Scotland.

So, in this episode, we've explored the development of the L sound and the R sound over the centuries.

As we've seen, those sounds have a tendency to disappear in certain situations, and they also have a tendency to alter the vowel sounds that come before them.

But there's one other interesting thing about those sounds that I need to mention before I wrap up this episode.

That other interesting note is that the L and R sounds are so similar that they are sometimes interchangeable.

If a word has two L sounds or two R sounds, that can be a little difficult for some people to pronounce, or maybe it just seems a little repetitive.

For whatever reason, there is a historical tendency to either substitute one of the sounds or to drop one of the sounds altogether to break up that pattern.

For example, in a word like surprise, there are two R sounds close together, S-U-R-P-R-I-S-E.

And you may have noticed that most people tend to drop that first R sound and just say surprise.

Of course, this is an example where non-rhotic accents like those in England and Australia and New Zealand would tend to drop that R anyway.

But even in regions like North America where the Rs are usually pronounced, the first R in surprise is often dropped because it appears so close to that following R.

But as I noted, speakers sometimes make a substitution to break up that pattern.

They'll replace one of the Rs with an L, or vice versa.

Believe it or not, words like purple, marble, turtle, and laurel all originally had an R sound at the end.

They derive from Old French poupora, marbre, tortre, and laurier.

And yes, Old French had a trilled r.

The modern French R pronounced in the back of the mouth didn't really emerge until the 1700s and 1800s.

But in those common words, we can hear how the problem of the repetitive R's was solved by substituting an L at the end.

The word riddle, as in to riddle something with holes or bullets, shows a similar development.

It comes from the Old English word ridere.

So again, just like those French words, it originally had two R sounds, but the one at the end became an L.

By the way, linguists have a fancy term for this phenomenon.

They call it dissimulation, but it's really just the substitution of one sound for another, in this case, the closely related L and R sounds.

So why did that happen so often?

Why did an R sound turn into an L sound and not some other sound?

Well, let me offer one possible explanation, and it has to do with the fact that the R sound was still trilled in these earlier periods when the sound switched to an L.

If we think about a trilled R, the tip of the tongue vibrates against the ridge behind the upper teeth.

Well, that's the exact same place that you put your tongue to make an L sound.

So, in a word that had two trilled R's close together, there was a lot of vibrating going on.

So, it would have been natural to simply hold the tongue against the ridge for a brief moment rather than letting it vibrate both times.

It would have been a convenient shortcut.

But holding that tongue there effectively converted the sound into an L sound, from

to

so in a word like,

there would have been a trilled R at both ends.

So to minimize that tongue vibration, it might have been tempting to just hold the tongue against the ridge on the second R.

So from Redder to Reddl, and that would have produced the sound change.

Anyway, regardless of the specific process, this shift from the R sound to the L sound was very common, especially when there were multiple Rs in a word.

We've actually encountered quite a few examples of this sound change over the course of the podcast series.

One of the first words I discussed in the early episodes also shows this development.

It's the word pilgrim.

The word was formed within Latin from two elements.

The first element was per, meaning beyond or through, as in perpetual, permanent, and preserve.

The second element was agri, meaning land or country, as in agriculture and agrarian.

When you put those two elements together, you got peregrinus, someone who travels beyond the country into other lands.

But peregrinus had those two R sounds, one from per and one from agri.

So in late Latin, the first R was replaced with an L, and the pronunciation shifted from Peregrinus to Pellegrinus,

and from there Pellegrinus became pelgrim and then pilgrim.

Another example of this process is the unusual word pair grammar and glamour.

Again, I discussed this development in an earlier episode, but glamour actually evolved out of the word grammar, and it evolved out of a distinct pronunciation where some people replaced the first R in grammar with an L, thus glamour.

The connection has to do with the specialized knowledge that grammarians had in the Middle Ages.

Their ability to speak and read foreign languages seemed almost magical to some people, and that gave the word grammar a sense of magic and wonder that was then extended into the word glamour.

Another example of this process can be found in the words incarcerate and cancel.

Believe it or not, they come from the same Latin root word.

The root word was carcer, meaning prison.

It survives intact with both R's in the word incarcerate.

But in Latin, an alternate form emerged where the second R was replaced with an L.

And then the first R was also dropped over time, so the resulting word was consulis, meaning lattice or something with a crisscross shape.

Of course, we see how that word is connected to a prison, which has bars or planks with a similar crisscross shape.

Well, it became common in the Middle Ages to mark through deleted passages in a document by drawing crisscross marks over the deleted passage.

And of course, we sometimes do the same thing today,

and that's how the word consulus, meaning lattice, evolved into our word cancel.

But ultimately, cancel and incarcerate are cognate, and they're distinguished in part thanks to the substitution of an L for an R.

We can find a similar process at work in the nickname we associate with the name Margaret.

Of course, Margaret also has two R's, but notice that a common nickname for Margaret is Molly.

Again, we see an L being used for an R in that nickname.

By extension, the same thing happens with other common names.

Dorothy becomes Dolly, Sarah becomes Sally, and Harold becomes Hal.

Again, this is a common pattern in the language.

Now, all of the examples I've given so far involve an R being replaced with an L,

but sometimes it worked the other way.

An L got replaced with an R.

That's what happened with the words tulip and turban, which I discussed a few episodes back.

You might remember that those words are also derived from the same root word.

The original word was a Turkish word for the turbans that were commonly worn in that region.

The word was tulbent.

That gave us the word tulip because tulips originated in that same region, and the Europeans who brought them back to Europe thought they resembled turbans when they blossomed.

so they used the same word tulip for the flower.

But then the original word passed into into Italian where the L sound switched to an R sound, and that gave us the word turban.

We can also see how an L switched to an R when we compare the words lineal and linear.

They have very similar meanings and they're really just two different pronunciations of the same root word.

That word was the Latin word linealis.

Linearis developed later, probably because linealis had two L sounds, so the second was replaced with an R sound.

And that gave us lineal and linear.

And I want to conclude this episode with one last example of this same process.

And I saved this example for the end because I get asked about this word a lot.

And now I can finally tell you why this word has such a strange pronunciation.

It's the word colonel, as in an officer in the military.

Of course we pronounce it colonel, but it's spelled with an L, C-O-L-O-N-E-L.

Well, if that spelling has always confused you, it's because of that close relationship between the L and R sounds.

The word colonel is actually related to the word column, as in a column of soldiers.

That's why colonel is spelled with an L.

It was originally colonello in Italian.

Again, it was derived directly from the word columna or column.

But colonello had those two L sounds.

So when the word was borrowed into French, the first L sound was replaced with an R sound, and that gave French the word coronal.

Of course, English borrowed a lot of words from French, but it also borrowed a lot of military terms from Italian in the 1500s.

So English ended up with both versions of the word, both the Italian version with the L sound and the French version with the R sound.

The Italian spelling was adopted early on as spellings were becoming fixed in the 1500s, but the French pronunciation eventually won out.

So English ended up with the word spelled C-O-L-O-N-E-L, but pronounced Coronel.

Remember, French also had a trilled R at the time.

But then, remember what happened in the 1600s, after the Elizabethan period.

By that point English probably had the modern r sound in that position, so the word was probably pronounced more like coronal.

And then we had that vowel merger, where the short vowel sounds before r came to be pronounced er.

That included words spelled with o r, like word and work

and world,

and here the or in coronal did the same thing, and that gave us the modern word colonel, still spelled the Italian way with its original L as C-O-L-O-N-E-L.

So that's the reason why Colonel has such an odd pronunciation in English.

By the way, we saw earlier that the word sergeant also has a mismatched pronunciation thanks to an R sound.

Remember that the presence of that R changed the preceding vowel from air to R,

and that gave us two competing pronunciations, sergeant and sergeant, and the spelling was fixed before the pronunciation was resolved.

So we ended up with sergeant, spelled with er and pronounced AR.

So between those two military ranks, colonel and sergeant, we can see how the L and R sounds have wreaked havoc over the centuries.

They've disappeared in words, reappeared in words, been substituted for each other, altered the vowel sounds around them, and contributed to wholesale changes in the way words are pronounced in English.

Now next time we're going to continue our look at Elizabethan English by examining the two other sounds that are sometimes called approximants.

Those are the W and Y sounds.

They also have a long history together.

In fact, those two letters were once the same letter in ancient Greek.

And the sounds associated with those letters are so difficult to categorize that most linguists call them semivowels because they also have vowel-like qualities.

So next time, we'll look at the history of those letters and sounds, and we'll see how they shaped the language we speak and write today.

Until then, thanks for listening to the History of English podcast.