The letter swap responsible for worried wolves, lovely honey, and some monkeys.
One of the many skills that Yorkshire folk have over southerners like me is that they can rhyme something with wolf. In my accent, wolf is a lone word, with no partners in rhyme. In northern England, by contrast, the flavours of English speech are united against the widespread ‘foot–strut split’, and so maintain one vowel in bull and gull, put and putt, and wolf and gulf.
North and South nonetheless do agree that wolf doesn’t rhyme with golf – despite their shared spelling. The latter, a word of Scottish heritage but of ultimately unclear origin, seems sensibly spelled. Golf suits its letter O, just as you might read and write in dolphin. Wolf wears its spelling awkwardly; you can find the same vowel at the heart of bull and full, only there represented by a U.
Wolf is one way into a very different period in the history of English, albeit one responsible for an orthographic oddity that we’ve maintained to this day. It’s a story of circumstances, of the needs of readers and writers, and it deserves to be better known, if for no other reason than to reassure learners that English spelling has its reasons. Strange to say, wolf and its eccentric vowel are actually the product of an attempt at clarity.
A Wulf in Wolf’s Clothing
It bears repeating that writers in the European Middle Ages didn’t spell according to a rule book. As the small literate populations sought to communicate through ink as clearly and efficiently as possible, there was a great deal of teaching, influencing and copying going on, but also organic variation. This is what we see in texts from medieval Britain, especially after the shenanigans of 1066, which pushed written English to the margins for a few centuries.
It’s specifically in this linguistic period, which we call Middle English, that we find wolf first written with an O. Prior to the eleventh century, in the era of Old English, and sporadically afterwards, it’s spelled with a U – perhaps as uulf or as ƿulf, with the since-forsaken letter wynn.
We can trust this U, and that it truthfully represented a U-type vowel in speech, for the reason that it’s what we expect from the Germanic ancestry of the word. A wolf in English’s great-aunt Gothic, for example, is 𐍅𐌿𐌻𐍆𐍃 (wulfs). This U-vowel has continued until today in speech, but not in spelling.
The mistruthing begins in Middle English, and the generally accepted theory goes that new and fashioning styles of handwriting in this time are to blame for the O of wolf – or alternatively, to thank for it. In the quintessentially lettering of the High Middle Ages, Gothic, the greater weight given to upward lines had the potential to confuse readers. Forests of short vertical strokes, known as minims, could obscure their alphabetic ingredients. Readers might lose track of the minuscule letters m, n, u, w and undotted i among all the minims.
As Old English habits of writing fell from favour – wynn being one of them – ƿulf was due to become wulf. But this would begin with five minims, and readers might struggle to untangle its initial sequence of double U and single U. The solution of scribes, from the twelfth century onwards, was to swap the vocalic letter. O, with its clear rounded shape and its proximity in the mouth space, was the obvious option. Wulf became wolf.
To be clear: this was not a change in sound, or rather, not necessarily accompanied by one. What we have here is a phenomenon limited to the spelling practices of medieval Britain. Since that day, many products of multi-minim avoidance have been accepted into standard English spelling, and once you’ve been made aware of the tactic, you start to spot them everywhere.
Joining wolf are won and wonder, rhyming as they do with pun and plunder. Worry, likewise, does not rhyme with lorry, but rather with hurry. Some involve U and M, such as come, monk and some itself, which share no rhyme with either home or honk. Others were responses to the two minims of an N (honey, tongue) or of a consonantal U (love, then written loue).
When we mentally separate sound from habitual spelling, we hear how these words contain the short vowels that we also find in put and but. Such vowels would be better suited to a U.
A Wurd of Caution
When I first read about this tactic, my reaction was ‘woah’ – although a woah tinged with scepticism. This is my gut-response to any theories about medieval language change based on writing; the pool of writers was small and disconnected, but also generally committed to representing sound accurately. Nonetheless, the ‘O for U’ disambiguation is widely accepted, and seems the likeliest explanation in many cases.
The reality of the tactic draws strength from the fact that the swap was applied to a greater set of words. What we have today is just a subset of affected vocabulary. Our texts from medieval English exhibit not only huni alongside now-standard honey, but also hondred and hongry beside hundred and hungry. Just as Old English sum became some through this practice, so too might sommer have won out over summer. The medieval character of the nun, with her six minims, was an obvious beneficiary of the swap, and unsurprisingly the spellings non and nonne are documented too.
Considered altogether, this variation makes the swap seem like an active practice among the community of Middle English writers. The situation was in flux as standard spelling began to emerge in and from London in the fifteenth century. What variants happened to be most common there and then were those that made the cut. Honey and wonder were in, hunny and wunder were out.
But: the disambiguation of minims may not be the right or sole explanation for all English words that exhibit an apparent U-O mismatch. For example, I would include wood and wool among the affected, and have included done so in the past (such as in a book that’s sold tens of thousands of copies, so yay.) But the double O in wood and wool is suspicious. Wod and wol would be enough to avoid confusion. Having two Os is overkill.
Instead, there is occasional mention by scholars (e.g. Minkova 2014: 221) of a sound change that lengthened and lowered high vowels in certain words, wood being one. If this is correct, then the Os of wood and wool do have a basis in sound. We must make the further stipulation that their pronunciation with a short vowel today comes down to us from another medieval dialect. That is to say, we get their modern spelling from a dialect with the lengthening, and their modern sound from a dialect without it.
Relatedly, the O in among, rhyming with sung not song, might also have once been true to speech. This was before the vowel underwent a sound shift that estranged it from spelling. The same shift may be behind the “sound-symbol anomaly” that is the E in English.
A good gauge, I think, for whether the U-O scribal swap is the primary reason is whether the word in question was given a U in Old English. Many of the relevant words can be tracked to an Old English forefather consistently spelled with U: ābufan, cuman, huniġ, lufu, munuc, sum, ƿulf, ƿunnen for above, come, honey, love, monk, some, wolf, won. It’s only in later Middle English that we start to find O-variants in the record for this vocabulary. This can be taken as a sign of the swap in spelling, while their short U-vowels continued in speech.
Other words are more complicated. Worry, for example, goes back to Old English wyrġan, a physical and possibly fatal verb meaning ‘to strangle’. (It arrived at its modern meaning of ‘mental distress’ via metaphor.) The development of the high and forward Y-vowel in wyrġan was dialectally diverse in Middle English, but we don’t expect it to have emerged as an O. Perhaps there was a U-variant *wurġan in pre-Norman speech too, but it hasn’t reached us in text.
Elsewhere, we do have the variant attested. Why word rhymes not with lord but with curd, or why worth sounds like it’s written ‘wurth’ might also find an explanation in the U-O disambiguation. Complicating matters, dictionaries give their Old English ancestors as ƿord and ƿeorþ; these forms lack the U necessary to be affected by the change. However, variants with U are documented too. A century after 1066, awesome Orrm, the beloved witness to Early Middle English, supplies us with the spelling “wurrþi” for worthy. The same may be true of worm and worse, rhyming as they don’t with gorm and gorse.
There are also those words possibly affected that don’t have an Old English ancestor to consult, because they come into currency much later. Monkey is one, a word of very unclear etymology that first turns up in sixteenth-century English. If monkey is indeed derived from monk, then the swap is indirectly responsible for monkey too. But we’re really not sure.
Glove and other puzzle me, having always been written with an O. I suspect that they don’t belong to this group of respelled words. They might instead fall into the same category as blood, in that their O-spelling is actually faithful to their pronunciation prior to the Great Vowel Shift of the early modern period. This raised their vowel in the mouth to a long U. They would’ve ended up rhyming with groove, smoother and food today, had that vowel then not been shortened and eventually unrounded. This three-step journey of vowels (raising, shortening, lip-unrounding) would get us from Old English to my pronunciation of glove, other and also blood. It requires no letter-swapping scribes.
Lastly, it remains possible that both minim disambiguation collaborated with another factor to create our current spelling. The O of some does distinguish it from the mathematical sum, a borrowing of Old Norman soume or Old French somme. A French influence may also have partly contributed to the respelling of tun as ton, the two variants of which co-existed until they split into two words with two meanings (big barrel vs. big measurement) around the seventeenth century.
Having originated as a solution for ambiguity of reading, ambiguity of meaning was something else that the U-O swap could help with. Back in Old English, it was clear whether you were talking about male offspring (sunu) or the great ball of flame in the sky (sunne). As English’s grammatical endings eroded, O accidentally and gradually offered itself as a new way to separate a son from the sun.
As in every etymology, I advise caution in identifying the handiwork of minim-modifying medieval scribes, but the evidence does at least point to its being a factor in the history of many English words, and I can recommend enjoying it as a linguistic fact too.
The change offers me a good example of how every quirk of current English spelling has an origin and a cause – it just may not be a very good one, by our standards. Writers who think they’re doing something helpful for their generation can bequeath to the next a habit of spelling that makes much less sense. In the ongoing chapter of written English, what with its modern fonts, screens for virtual text, and ready availability of ink and paper, medieval disambiguation has become a source of confusion.
It would be remiss of me not to put the idea out there that we could change the spelling. We each could start/return to writing abuv, huney (or hunny), gluv, munk, sum, wulf, wunder and wurry; phonemically faithful writers in non-standard spelling may already do so. If even the briefest glimpse of such spelling disgusts your eyes, then, well, you’ve gained an insight into how oddities and irregularities endure.
END.
Texts
References
- Bosworth Toller’s Anglo-Saxon Dictionary online
- Minkova, D. (2013). Historical phonology of English. Edinburgh University Press.
- oed.com
- Upward, C., & Davidson, G. (2011). The History of English Spelling. John Wiley & Sons.
- Welna, J. (2014). <U> or <o>: A dilemma of the Middle English scribal practice. Contact, Variation, and Change in the History of English. 305–324. John Benjamins Publishing Company.
Images my own, from Wikimedia Commons or from credited sources. Cover picture: fourteenth-century Dutch illustration of a wolf. Only human intelligence (or lack thereof) was used in the creation of this article.
.jpg){kind=link}
Leave a Reply