Words on the Move

by John Mcwhorter

  But bit isn’t just an isolated thing. Down below, you get the same thing between bait and bat. That is, what about not ay, but eh? There’s bait but also bet—those are different sounds. And bet exists in the same relation to bait as bit to beat—a little farther down and a tad farther back. Like so:

  Now, in line with the tidy opposition between the front stack of beet, bait, and bat and the back one of boot, boat, and baht, you get the same alternate duo in the back. Between boot and boat sits book*—book is a little lower in your mouth, and also a bit more to the front than boot and boat. It fits in right across from bit:

  Then, what happens here is a perfect example of why spelling “can’t jump,” so to speak. Yes, there indeed is a vowel between boat and baht that is lower than boat and pulled farther to the front. But it isn’t spelled with a single letter. If we weren’t so snookered by spelling, it would be easy to glean what that vowel was, but reality requires me to just lay it out: bought. Or, in an alternate universe I would prefer that bought were spelled “bawt”—it’s the sound we most readily associate with aw. We put it here:

  With that, we have a nice parallel two “internal” stacks of vowels. Along with the a in cat, the ih, the eh, the oo in book (“eugh”?), and aw are, for better or for worse, what make English sound like English instead of Italian.

  Except we’re missing one last one: what you may have learned as “short” u, the vowel in bus and but. Where does that one go? (To see how misleading spelling is when it comes to vowels, note that this uh sound is the vowel in no fewer than three words I’ve just used—but, does, and one—despite the fact that only one of the words spells it with a u.) The uh sound is pronounced between the front stack and the back stack. Also, but is pronounced not as high up as beet and boot, but not as low down as bat and baht. Say bait, but, and then boat and you feel words pulling backward in an almost soothing straight line. This shows that uh goes smack in the middle, between front and back and between the top and the bottom:

  Here, then, is what and where our vowels are. Some readers may be waiting for the official phonetic symbols for the vowel sounds, intuiting that in a linguistics class we would be on our way to learning the International Phonetic Alphabet. However, for our purposes in this chapter, those symbols belong in the notes (where you will find them). Our interest is not in learning how to transcribe speech the way Henry Higgins in Pygmalion does, but in something simpler: internalizing a sense of how vowels sit in our mouths, in order to see the coherence in the different ways people pronounce words. For that, we just need something that takes us fleetly from the page to how we talk.

  No, Not “Bees in Your Mouth” Just Because Those Words in the Chart Begin with B

  The next thing to understand is that the diagrams we have seen are, like dictionaries, snapshots of something that never stays the same. In an illustration of an atom, we understand that the electrons shown surrounding the nucleus do not, in reality, sit frozen in those positions. The electrons whiz around the nucleus. In the same way, a vowel is something going on, not a thing situated. To be a vowel is to face a future, maybe sooner, maybe later, but sometime, of becoming another vowel.

  The diagram format obscures not only this, but the very nature of the environment the vowels have for this motility. The diagram implies “places” for the vowels, whereas in actuality vowels occupy an open field: your mouth. A vowel can pass through (note I did not write “sit”) anywhere on that field. Each language’s vowels consist of a mere fraction of the endless possibilities, and where a given vowel is pronounced in a given language at a given time is as much a matter of chance as what the temperature happens to be in Seattle right now, or exactly how many episodes of The Simpsons there will finally be. For example, there is no self-standing reason that the oh sound has to be exactly where Americans pronounce it, as if vowels fall into set places like marbles on a Chinese checkers board. Because there is no tin honeycomb of boxes in your mouth—if there is, do seek medical attention immediately—a vowel can be anywhere. There are many ways to be an oh.

  Even within American English, oh is hardly in one place. Take boat again: most Americans really say “boh-oot.” It’s Minnesotans who are famous for actually saying “boht,” with what really is a simple “oh” sound. Then, many Philadelphians and Baltimoreans say something more like “beh-oot,” where the sound is moving forward, toward the bet sound across from it. Many black Americans say something more like “baw-oot.” So what is the “real” oh sound? That’s just it—there’s no such thing. For example, the most common American pronunciation, “boh-oot,” is in no way basic, central, easier, or even earlier: for whatever it’s worth, Old English speakers would have found Minnesotans’ “boht” version the familiar one.

  The takeaway is that a vowel is a moving thing, not a thing that is. These critters are alive, like bees in a hive. Any oh we hear today is where speakers of that language—or actually, speakers of that dialect of that language—happen to have alit at that time. Vowels exist in a field with no boundaries, such that they are not blocked in any way from moving—and they do.

  More to the point, they couldn’t sit still. The reason they move is that new generations tend to reproduce what they hear with a bit of distortion. People are imperfect mimics. “Oh-oo” happens because oo is a sound that happens on the way to passing from oh to somewhere else, or even to silence—oh requires a kissy pose with the lips, and oo is that same pose with lips a little closer to closed. A generation might start riding on that little oo a bit; the next generation, more, and the next thing you know, “boht” has become “boh-oot.” Elsewhere, as speech flies out of your mouth you might not pull your oh back as far as Mom and Dad did—your boat starts to take on a redolence of the bet in front of it. After some generations, “boh-oot” has become “beh-oot.”

  There is a temptation to imagine that this vowel drift happens because of some kind of external influence, such as contact between people who speak in different ways. There is truth there. When different groups each lay their oh in different places, they may start affecting one another’s placement of the vowel, and then more distortion happens when a new generation of people hear all these different renditions and settle on some kind of compromise. Contact between people does make a language change more quickly than normally.

  However, external interventions like this are by no means a necessary condition for a language’s changing, such as its vowels drifting. The change is, in itself, a default condition just as it is for weather. Contact between people speaking different languages, or dialects, only makes the vowels move in different directions than they would have moved otherwise—that is, the vowels would drift even among those people stuck in a cave. People born in there, in the dark waving bats away, would still render their parents’ vowels somewhat inaccurately, adding bits here, clipping bits there, scooching things around a bit.

  However, vowels do not jump willy-nilly from one end of the field to the other. They move in orderly fashion, one step at a time, just like the meanings of words. Once you have a sense of where the vowels are on the field—as opposed to an arbitrary row of symbols like A E I O U—then what strike us as peculiar accents end up seeming as ordinary as the fact that the long hand on the clock is in a different place now than it was fifteen minutes ago. After all, why would it not have moved?

  Vowel Shifts Are What That’s All About

  Back to upstate New York. Why “keh-it” instead of cat? Or, let’s change it to “beh-it” for bat, to relate most directly to our chart. As you can see, bet is just a step up from bat. The bat vowel moved, as vowels will. Specifically, it floated up to where bet is—not way up to where beet is, or up over to where but is. In chess terms, vowels are kings, not rooks and certainly not queens. To pronounce bat as “beh-it” isn’t even as eccentric as it may seem. If you are American and don’t happen to be in the group with this “flat” a, then say cat and now say man. Do you really say man with the exact same vowel as you say cat with?
Some people do, but to most of us it sounds a little odd or at least British. Most likely, you say man in what you might think of as a “flatter” way—you “squeeze” that a a bit when you say man (or can, or ma’am, or spam, or damn). You’re saying something more like “meh-in.” If this describes you, you do the squeeze before m’s and n’s. Other people just do it before all consonants—hence “keh-it,” plus “keh-ip” for cap, “peh-ik” for pack, and beyond. You could even think of them as more consistent than the rest of us!

  The chess analogy continues: when bat moved up, it left an empty space, ripe for something from nearby to come fill it in. Something did: baht from right next door. This is why the same woman who says “beh-it” may very well also sound almost like she’s saying “black” when she says block. Or this explains why some Americans will talk about the importance of what sounds like “jabs” rather than jobs. In real life we hear “jabs” as a “nasal” way of saying jobs. But someone speaking with these vowel shifts isn’t talking any more through their nose than the rest of us. What’s different between them and us is simply that their vowels are in different places from ours. “Jabs” is just jobs said in the forward part of your mouth, as in moving baht to bat. Someone whose vowels have moved in this particular way sounds almost like they’re saying “the keh-it on the black” when they say “the cat on the block.”

  People’s vowels have moved in this particular way in a swatch of the northern part of America’s eastern half, which is why one hears it in upstate New York towns such as Buffalo and Rochester (it was Ithaca for me). It is also well established in Connecticut, and stretches westward to Cleveland, Chicago, Milwaukee, and most of Minnesota. It’s called the Northern Cities Shift, and the way people in those places say cat and flat, although what sticks out most to the passing observer is but a taste of something larger and, in its way, elegant.

  The vowels follow each other and get out of one another’s way in a pleasantly choo-choo train fashion. Once the baht slot is empty because a person is saying block as “black,” then up above that baht slot is bought, and indeed, people with this vowel shift also often pronounce bought more like baht, caught more like cot. This is all like a little train, you can see. So if bat, pulling along baht and bought behind it, has elbowed bet out of its area, then what happens to bet up front?

  Well, the clockwise creep continues: bet ticks back into the area of but. If you know someone from the Northern Cities Shift region, watch them speaking and you might notice that they say a word like bet, set, or met in a way that will seem in passing as if they are opening their mouths a little wider. But actually, they are pulling the sound backward—a foreigner with no idea of what English looks like on the page, having met no one but a person with these vowels, might think a person saying bet was saying but, or at least sounded like they were. To this foreigner, the sentence I’ll bet she caught that cat down the block would sound like something close to “I’ll but she cot that kyet down the black.” If you’re not from the Northern Cities Shift region and you can render that sentence without exaggeration, just taking it light and saying it with neutral affect like someone saying it sitting on some front steps one July afternoon, you may find a Milwaukee accent in your mouth that you never expected.

  This is not, however, an indication of anything odd about northern cities in the United States. It’s just the way vowels happen to have been moving there. Everywhere else, they are moving, too—just in different ways. Vowels are like bees. In a hive where the bees are not moving, a likely analysis is that they are not alive. Similarly, a language in which the vowels stayed the same would be a language spoken either by robots or by a people so demoniacally obsessed with keeping their vowels in place that they would have to devote all life’s energy to that task—which would be, essentially, death anyway.

  Is This Really How a Language Changes?

  I tell you that vowels are inherently, even ardently, motile. But then I exemplify it with the accent of people in Rochester and Milwaukee. You may wonder: is this business of how people sound here and there really a sign of how a whole language changes? Aren’t these little things just, well, mannerisms, tics, or, to put it in a cocktail hour way, just some random (ahem) crap?

  It seems like it—our lives aren’t long enough for us to see the whole picture. It can be hard to even perceive vowel shifts going on in the present. Vowel shifts are slow, incremental, and subtle. We likely catch a mere glimmer, such as “keh-it,” and hear it as an isolated anomaly. Even the people undergoing the vowel shift often hear themselves as talking just like everyone else. The Detroiter who says “jab” for job, if played a recording of another Detroiter saying “jab” for job, most often hears someone saying jab just as people from outside the Northern Cities Shift area do. They don’t think of themselves as people who say something more like “jab” than job. The difference is fine-grained, and writing exerts a powerful pull on how we hear language, even our own. Is calling a job a “jab” really the kind of thing that ushered English from Beowulf to Bellow?

  Yes, and the handiest way to see it is, as it happens, what most sources use as an introduction to the whole topic of vowel shifting. It’s hardly that other books on linguistics don’t mention that vowels move. However, the traditional way of getting the point across is to describe a vowel shift that happened in the fifteenth and sixteenth centuries, which is called, in fact, the Great Vowel Shift.

  I have refrained from opening with it, because when described alone, the Great Vowel Shift implies that vowel movements—I couldn’t resist just once—are something unusual, a museum exhibit safely removed in the past rather than life as it always is. Yet it was merely one phase in an ongoing process of which the Northern Cities Shift is a latter-day continuation. It is useful in showing that vowel shifting is normal, in that it stands out in a way that the Northern Cities Shift doesn’t—on paper.

  Namely, English spelling is hideous in large part because it represents what English was like before the Great Vowel Shift happened. That neatly demonstrates that vowels really are as liquid as I am presenting, and that what happens to them can neither be said to “ruin” the language nor be dismissed as bubblegum static, since no one wishes we could go back to talking the way Chaucer did.

  Here’s what I mean. Why would any sane person write mate and pronounce it “mayt”? We’re used to being taught that this is a “long” a and that the “silent” e is our clue to that. But clearly no one would design a system this way. If people in France and Spain and seemingly everywhere else on earth were writing the “ay” sound as e, what was the sense of instead bringing in an a and signaling that it, instead of e, is pronounced “ay,” and by putting a “silent” e at the end of the word? Why e? Or why use any sound at all as standing for absence instead of, duh, not writing anything there? While understanding that customs differ across the ages, we can be quite sure no writerly caste decided on such nonsense as a writing system. Sure, some eccentric medieval scribe could have arbitrarily decided on such a system, but why on earth would it have been accepted across England?

  When something makes that little sense, usually it was created amid conditions now past in which they did make sense. And indeed, time was that mate was actually pronounced the way one would expect: MAH-tay. The final -ay, unaccented, wore off over time just as the -ther in brother has worn off among men saluting each other such that guys of a certain demographic call each other “bruh” and one might call one’s sister “sis”—that’s easy. “Mahtay” became “maht.” But why don’t we just say “maht” today? Because likely the vowel moved, and this one did.

  “Maht” exited stage left: to mat, upon which it was poised to then do what bat did in the northern cities of something called America seven centuries later—it floated up. This time, it alit not on the bet slot but on the bait one. That, then, is why mate is spelled the way it is—mate represents how the word was said two steps behind—“MAH-teh.” Speech moved on; spelling stayed put.

  T
hen, once words like mate were in the bait slot, what was in that slot got pushed up, to the beet slot. Here is why, for example, meet is spelled the way it is. We’re so used to seeing ee as standing for the sound in beat, seize, etc. However, in any other language you’ve learned, e likely stands for ay or eh. Why is ee the vowel sound in beet just because there are two e’s? If e is ay, then shouldn’t ee be, if anything, “ayy”?

  Yes, and once, it was. Meet was once pronounced as it “should” be: “mate.” The two e’s instead of one just meant that it was more like “ayy” than “ay,” a difference which in earlier stages of English actually made the difference between one word and another. In Old English, if you said “may-tuh” it meant meat, while only by stretching it out into “mayy-tuh” could you mean meet.* But, edged out and above the ay slot, meet came to be pronounced the way it is now, but again, the spelling held on. So: mate and meet are both spelled crazily because of the Great Vowel Shift.

  The Great Vowel Shift was even beautiful, in the way that mathematicians and physicists often describe the workings of their solutions. Well, not quite that beautiful, but close. Over on the back row, wouldn’t you know, moose is pronounced the way it is for the same reason meet across from it is pronounced the way it is. Again, o is supposed to indicate oh, and so oo should by all rights stand for two ohs, “ohh.” That’s how an early medieval English speaker thought of things as well: moose was pronounced “mohhs.” One saw one of those oddly majestic creatures and said, “Look at that mohhs.”

  However, “mohhs” moved up to the boot slot in the same way that “mayyt” moved up to the meet slot over across, and thus “mohhs” became our moose:

  Then, finally, the Great Vowel Shift became a vowel spill, like out of a fountain. From both sides! If the pronunciation of “mayyt” moved it up to the beet slot, then where did what was already in the beet slot go? Mite was in it—pronounced, at first, “MEE-tay,” just as it would be in any normal language where i stands for ee. If you have studied Spanish, imagine how mite would be pronounced in it: “MEE-tay.” That’s normal. Why, then, is mite in English today pronounced like might?