“The history of language,” German instructor Matthew Bauman tells Urbo, “is one of things being weird. And that’s good.”
One example of things being weird: the word “knight.” Its spelling has confused new readers and speakers of English for centuries. Why are those letters there anyway?
Because we used to say them all, to put it simply.
If you listened to a telling of Beowulf back in the hoary days of Old English or read a chivalrous romance in Chaucer’s Middle English, the word knight would have sounded something more like cneehht. It started on a hard C, continued with the long E of see, followed by a guttural blast of the Ch still heard in the Scottish loch, and ending with a breathy T. You could have even heard some English speakers pronounce it that way if you walked around 200 years later in Shakespeare’s London.
For the Modern English word knight comes from the Old English cniht, which originally meant “boy” or “servant” before its rank was dramatically elevated. And every single letter of it was pronounced, something we still see today in its close kin, the German Knecht. (The German, as it happens, also preserves its root “attendant” meaning.)
Old English, German, guttural blasts, and servant boys? That’s already a lot to unpack—but even through this one messy word, we can get a brief history of the English language.
“Think about what it says about the human brain that you can do all that stuff without thinking about it,” Bauman says. “The fact of the irregularity of language is not a sign of stupidity. It’s a testament to the human brain being able to handle these complex rules and still get your point across. It’s amazing.”
You say “pater,” I say “father.”
Let’s take it from the beginning. Old English, the forebear of Modern English spoken roughly between 450 and 1150 AD, comes from dialects spoken by Germanic tribes—Jutes, Angles, and Saxons—from Denmark and the Low Countries when they invaded the British Isles in the mid-400s.
Previously, Celtic tribes were speaking various Celtic languages throughout the modern-day UK, and they left a mark, though small, on English, surviving largely in place names (Kent) and geographical features (crag).
And what, exactly, makes English Germanic? Linguist Kyle Gorman breaks it down for us:
“There was a community of people who spoke a language in pre-history…and they resided somewhere in northwest to north-central Europe. All of the languages which descend from the language they spoke are termed Germanic. We get the name Germanic simply because that’s what the Romans called them, these sets of tribes. And there are certain things that all these languages share because they share a common ancestor.”
The Germanic languages—including the likes of Dutch, Danish, and Icelandic—share an even older ancestor, Proto-Indo-European. This is why we find all sorts of common vocabulary and grammatical structures across tongues as diverse as English, Welsh, Russian, Latin, Greek, Persian, and Hindi. (Yes, even Persian and Hindi are related to English. But let’s not get ahead of ourselves.)
One place we can see English’s Germanic-ness is in its word-initial consonants.
In the early 19th century, German philologist Jacob Grimm—whose name also lives on through the fairytale collection he compiled with his brother—noticed a certain pattern across Indo-European words. Germanic languages, Grimm saw, had initial F/V sounds (English father, German Vater, and Dutch vader —hence Darth Vader) where other Indo-Europeans language have Ps (Latin pater, Greek pater, Sanskrit pitr). The English foot, German Fuß, or Dutch voet? Latin has pedis, Greek has podos, and Sanskrit has pad.
On the surface, P and F/V seem like very different sounds, but pay close attention if you say them out loud, and you’ll notice they are produced in the same area of the mouth.
Due to whatever idiosyncrasies or accidents in pronunciation some time in the first millennium BC, Germanic tribes made this subtle, but consequential, shift from saying Ps in words they inherited from Proto-Indo-Europeans as F/Vs.
Grimm also observed Germanic words feature Th/D sounds where other Indo-European languages have a simple T, for example, English three and German drei compared to the Latin tres.
Grimm formulated these three sound shifts into what we now call Grimm’s law. The technical linguistics of this law gets a little hairy for the layperson: Proto-Indo-European voiceless stops (P, T, K) became fricatives (F, Þ, H) in Germanic languages.
But the pattern is remarkably consistent, and scholars have since extended it to other consonant shifts in Indo-European languages. The pattern is also the key characteristic that sets Germanic languages apart from their Indo-European cousins.
Same goes for H and K sounds: English hound, German Hund, and Latin canis. That’s right, canine and hound, if you go back far enough in time, are the same word.
Easier for the non-linguist to spot, though, are some of the core words—those humble, little words we use every day—in Germanic languages.
“The two languages don’t always look that similar,” explains Bauman. “But if you go to a pet store or a farm, for example, you can see the Kuh, that’s the cow. People out at the park, they have the Hund. Going to the zoo? The tiger is just a big Katze. You sit on a Stuhl,” cousin to English’s stool.
“To study German is not to stray too far afield from one’s native English,” Bauman adds.
Talk, talked. Sing…sang?
Another place we can easily see English’s Teutonic (that is, Germanic) roots are in its verbs—those seemingly random ones that always trip us up and can make English hard to learn.
“Sing, sang, sung. That’s inherited pretty straightforwardly from Proto-Indo-European,” Gorman says. He’s speaking of a linguistic process called ablaut. We think of it as unusual, or irregular—as we call these verbal oddballs that change their vowels to mark the past tense or past participle (e.g., I sing the music, I sang the music, or the music was sung by me). But Gorman notes that this sort of vowel-hopping is actually pretty common in language.
As far as Indo-European languages are concerned, historically, there were a number of different patterns this vowel-hopping—this ablaut—could take. Drive/drove/driven and ride/rode/ridden follow one pattern. Throw/threw/thrown and grow/grew/grown follow another. Joining sing/sang/sung are the likes of swim/swam/swum and stink/stank/stunk, among others. “It just happens that a large number of these are well preserved in English and in German as well,” Gorman says.
It’s the exception—albeit one that took millennia to unfold—that became the rule. Germanic languages “developed this other alternative,” Gorman continues. “It uses a dental suffix, a T or D, to form the past of various forms. Talk, talked, anything of that form. That’s a Germanic innovation.” And that’s what became the normal way to form past tenses in English.
“These things do decay with time,” according to Gorman. The past tense of help used to be holp. Climb? Clamb. Broadly speaking, economy and convenience tend to tidy up inconsistencies and whittle away at complexities, regularizing help with helped and climb to climbed.
“However, for the most frequent ones, the decay is unbelievably slow,” Gorman explains. Which is why, in part, irregularities like sing/sang/sung hang on. And, very rarely, we muddle a perfectly ordinary verb like sneak, whose regular past is sneaked. We kept things interesting, though, by making it snuck.
But there were some outside forces at work that really sped up the decay: The invaders got invaded.
The Incredible Case of “They”
From late 700s to the early 1000s, Vikings invaded England in several waves, hitting northern parts of the island particularly hard. By living and working together and by intermarrying, Norse and Anglo-Saxon peoples became pretty close.
So close, in fact, that Scandinavian words replaced some very bread-and-butter English words. Sky, egg, and window—some of the earliest words we learn—displaced their native Old English equivalents. So did the verbs get, give, and take.
English even took on they from Old Norse. Now, it’s one thing for a language to borrow words for a new food, object, or idea. Tomato, for instance, comes from an Aztec language, bible comes from Greek, and algebra from Arabic. But it’s another thing entirely to borrow a function word—a word embedded into the deep grammar of language—like a pronoun.
The Old English third-person plural pronoun was hie, kicked out by the Scandinavian þei, now they. Old English’s he was hē, she was hēo, and it was hit, so the Norse they probably provided a useful point of contrast for English speakers.
Still, they demonstrates just how intertwined the two peoples—and their languages—had become.
What’s up with the M in whom?
Babies are made to learn language. That’s why children have no problem picking up the many tones in Cantonese, the clicks in Bantu, or the twists and turns of Arabic triconsonantal roots. But it’s a lot harder for adults for acquire a second language.
Imagine over a millennium ago an Anglo-Saxon woman buying some fish from a Norse fisherman. In her Old English, fish was fisc. In his Old Norse, fish was fiskr. Old Norse—parent of the modern Scandinavian languages—and Old English were both Germanic languages, remember, so a great many words closely overlap. (And while we’re at it, here’s a fun chance to apply Grimm’s law. If we see Germanic words beginning with an F, what should we expect, say, in Latin? That’s right. A P, and indeed, the Latin for fish is piscis.)
Anyway, depending on how our Anglo-Saxon and Viking pair were using them, fisc and fiskr would take on special case endings. Case endings are sounds tacked on to a word to convey what grammatical role it’s playing, such as subject or object, in an utterance. (Remember the nominative puella and accusative puellam from Latin 101? No? Well, stay with us anyway.)
Old English had four such cases, which quickly grow into a rash of these case endings when you account for plural forms, gender, and different patterns for different nouns.
For their fish transaction, the Anglo-Saxon and Viking would get pretty confused by all these endings. Nor would each have likely learned each other’s system fluently, as that takes a lot of time and adults just aren’t as skilled at doing it. (Skill is another word we get from Old Norse, by the way.) So, maybe they’d drop them off for the raw noun. With just plain fisc and fiskr, no strings attached, they’d achieve some mutual intelligibility. Add in a good deal of gesturing and they’d communicate.
Repeat that with hundreds of thousands of people over a long period of time, and English’s case endings wither away, allowing different peoples to talk to each other.
But as Gorman taught us, some forms can be very stubborn, especially in very high-frequency words.
“There used to be a who for every case,” Bauman observes. “[In] Old English, who was hwa. There was a hwam for whom. There was a hwæs for whose.”
That’s why there’s that pesky M in whom: “It stuck around,” he sums it up. And despite the most curmudgeonly efforts of schoolmarms and grammar scolds, that sticky, vestigial M in whom is largely falling off today.
Why, then, don’t we have a problem with whose? It’s because we still use the old possessive case-marker of yore: the apostrophe. The girl’s book, the dogs’ bones. Old English didn’t use an apostrophe, but that S comes from the same sort of business as the M in whom.
For Simplicity’s Sake
Did you catch that strange-looking Hw in the ancestor of who, hwa? Hhhwaah: That’s what who would have sounded like a thousand years ago.
And that Hw is another strange-sounding feature, just like the Kn in knight, that gives us clues to the history of the English language.
“Old English had a bunch of crazy onsets,” Gorman says. “Hn, Hw, Hl, Hr, Cn.”
Hn became N, as seen in the evolution of hnut to nut.
Hw became W and with its spelling flipped around, though you can still hear some speakers pronounce words like white whale with an extra burst of air. Cool whip, anyone?
Hl was reduced to L. For instance, the word lord, to the best of our knowledge, began as hlafweard—keeper of the hlaf, or loaf—and lady starts out as hlæfdige, or “loaf-kneader.”
Hr turned into R, as seen in hring becoming ring.
“There are often pressures to simplify these things and it happened slowly over time,” says Gorman.
We like efficiency when we’re speaking. We like to smush sounds together or leave them out entirely. (Whadjyeet? That’s how we very well may ask our co-worker “What did you eat?”) And so the K sound in knight gradually became subtler and softer, as did its guttural Gh, eventually disappearing altogether. The same forces were at work in hlafweard’s remarkable transformation to lord.
Words shift in meaning, too, as we see in the case of lord—or knight. We saw knight begins in Old English as “boy” or “servant.” Servants are like “attendants,” and so knight came to be used for military “followers” of a king in the 1100s.
Throughout the Middle Ages, these followers grew in importance, prominence, and, eventually, stature, gaining what we now think of as the noble rank of knight.
We raise cow, but we eat beef.
That Anglo-Saxons were sacked by Vikings likely accelerated some of the simplifications of English. So did the Norman French.
As Albert Baugh and Thomas Cable put it their classic 1935 text A History of the English Language: “Toward the end of the Old English period an event occurred that had a greater effect on the English than other in the course of its history. This event was the Norman Conquest in 1066,” when William, the Duke of Normandy, conquered the English and ushered in French rule for the next few centuries.
The most conspicuous place you can see this effect is in English vocabulary. The Normans spoke a form of French—itself essentially a modern version of Latin—that became the language of government, administration, nobility, commerce, literature, study, science, and art. Case in point? English borrowed government, administration, nobility, class, commerce, literature, study, science, and art all from the French.
In fact, English borrowed so many words that over half of the lexicon has Latin roots. That’s right, half of the vocabulary of our Germanic language is ultimately Latin, an Italic language.
The language of power and prestige (two words also borrowed from French) tends to diffuse throughout the rest of the population, either by imposition or aspiration.
Consider the dinner table. Our words for livestock are earthy and Germanic: cow, pig, sheep, chicken. But our words for the food we prepare from them are more refined, as it were. They’re French: beef, pork, mutton, poultry.
Going Through Changes
Over the 500 years following the Norman Conquest—the period known as the Middle Ages—English was in great flux as an enormous French vocabulary was grafted onto a Germanic structure that was gradually shedding sounds and case endings. Then the Great Vowel Shift threw English into even more tumult.
Starting in the late 1300s and into the early 1700s, certain English vowels underwent—due to whatever accident of history and language—a massive change. Meat was originally pronounce with the long A in mate. Bite was originally pronounced with the E in beet—hence the change in the vowel of knight.
“This created a gap in the system,” according to Gorman, dragging other vowels up, too. English is crowded with different vowel pronunciations, and “so if you have a very small apartment and you want to move something, you have to move everything else, too.”
Name used to be said with the a sound in father. Boot used to be pronounced like boat. Down, like dun. “This was finishing up as spelling standardization was really coming together,” Gorman adds.
We changed how we were pronouncing our words after our spelling of those words had already largely settled into place. As Gorman says, “English preserves a spelling that would have been appropriate 400 or more years back.”
And that’s why knight has all those extra letters we don’t say.