How is 'wl-' pronounced at the beginning of a word?

Of course, you just don't pronounce it at all, because there is no English word that begins that way and if there were, well, that's just not English there can't be any. Also, even though there are many possible arbitrary consonant sequences, some that just aren't in English words are naturally pronounceable (they just happen not to have any words with them). But 'wl-' just ain't one of them.

Except Oxford Dictionaries (makers of the Oxford Living Dictionaries, (not the OED)), just this morning posted a blog entry about the wonderful world of legitimate English words beginning with 'wl-'..

Some examples:

  • 'wlonk' - proud ('wlench' a verb for 'to make proud')
  • 'wlisp' - basically a variant of 'lisp'
  • 'wlat' - nauseous, loathsome

The difficulty is how to pronounce them in Modern English. Or for that matter in Old English. All the 'wl-' words, (as attested by the OED) are called obsolete, no one uses them nowadays, and looking at all the histories none were used past Middle English (after 1400s).

Is it analogous to 'wr-' where more recently the 'w' is dropped, but in older times there was an actual accepted natural way to utter these things?

Is the 'w' followed by a schwa? Or does it naturally glide liquidly to the 'l'?

The etymologies of all these words (in OED) is pretty sparse. They give connections to Old Dutch and Old Low German, but few hints. I was hoping for some cognates that are pronounceable in the other languages so that sound changes could predict the OE. Under the etymology for 'lisp', it gives "from late Old English awlyspian". But none of the other words has anything similar, so there is no corroborating evidence on a suggestion of how to pronounce it.

Frankly this feels like Russians saying 'Pskov', a cruel trick on foreigners to send them to the tongue doctor or be laughed at.

So, how is 'wl- intended to be pronounced in MSE, ME, OE, West Germanic, anything at all? Is there any definitive evidence (well beyond idle speculation)?

  • 1
    We cannot know for certain, but when saying wlonk you put your lips in position to say the /w/ but then proceed to the /l/. It's rather like saying "Vladimir" only with /w/ instead of /v/. Latin lana is glossed uul.
    – TimR
    Feb 8, 2019 at 16:15
  • @TRomano Sure, I can figure out how to do it, but is that the way it actually was done (and even if we can't know for certain, can we at least attempt a reconstruction that is consistent with Germanic phonological systems)?
    – Mitch
    Feb 8, 2019 at 16:27
  • 1
    @TRomano I didn't follow your statement: "Latin lana is glossed uul" and its relevance here (is the 'uul' supposed to be like 'wl'? 'lana' seems to only share the 'l' - what do you mean by 'gloss' here? An attempt at a phonetic representation?
    – Mitch
    Feb 8, 2019 at 16:30
  • 1
    Latin balbus was glossed uulisp, u-u-l-i-s-p.
    – TimR
    Feb 8, 2019 at 16:38
  • 1
    @TRomano Oh 'lana' is Latin for wool, right. So there is a lot of context missing with 'lana'. Do you have a Latin to ME dictionary that says that 'lana' is Latin for 'uul'? Which would be pronounced as something like /ʊl/ or /u:l/? That seems like a reasonable conjecture, that 'wlisp' is /ʊ 'lisp/, two syllables. But how would that account for one of the cognates in Old Low German 'gewlispan'?
    – Mitch
    Feb 8, 2019 at 16:39

2 Answers 2


The benefit of Old English and Middle English orthography is that it is usually phonetic and the consonants are pronounced. Lots of people know how the letter k before an n does mean that you pronounce the k (hence the funny Monty Python pronunciation of knight).

The same is true for wl-, tough as it is to produce. I learned to pronounce it as written. According to the text commentary book to From Old English to Standard English, "Old English pronunciation," all Old English consonants were pronounced. In the book's examples are a few of the /wl/ words with their pronunciation as conventionally understood (I substitute slashes for brackets to not throw off link formatting):

wlisp -> /wlisp/

wlanc -> /wlank/

wlitan -> /wli:tən/

wlitig -> /wlitij/

Similarly, the /wr/ preserved to today with silent /w/ in words like wrist would be pronounced. In both cases, imagine starting the /w/ sound at the front of the mouth and then moving the tongue up to produce the /l/ or /r/. Following how Old English is conventionally taught, someone pronouncing /wl/ should move from the labial to alveolar approximate.

How do we know this? The study of Old and Middle English phonology over the past couple of centuries has combed Old English orthography, compared it to Germanic and other similar languages, and worked from texts (most notably the twelfth-century Midlands text Ormulum) that are consistent and meticulous about matching spelling to orthography. The disjunction of spelling and orthography only grows in late Middle English. (See below for an example where wr and r alliterate.) From Old English texts and some basic assumptions that authors within dialect groups would share phonological and orthographic features, it's clear that consonants literally represent their sounds and the consonants rarely shift in what sound they represent, to the point that we can see in texts when the pronunciation does change. If a vowel was present, the Old English spelling would add a vowel.

From these sources we're fairly certain that /wl/ and /wr/ were the pronunciations of wl and wr. At most, there is the stray argument that maybe (as Jacek Fisiak claimed in 1967) /wl/ was a single phoneme. If that were so, then this line in Beowulf wouldn't be alliterative:

wliteseon wrætlic; weras on sawon.

It's the /w/ that alliterates, not the /wr/ or /wl/. See also:

wlitan on Wilaf. He gewergad sæt,

What happened to these sounds? A recent treatment of phonology in Donka Minkova's text A Historical Phonology of English (2014) describes the transition from /wl/ or /wr/ to /l/ and /r/ in Chapter 5, "Consonantal developments in the second millennium." She cites early Old English examples of splitting the cluster /wr/:

wrohte ~ worohte ('wrought')

She also notes Middle English moves to simplify the sound:

wrynkul ~ runkel ('wrinkle')

Or to suggest that wr only slowly shifts to alliterate with r, as it does in the late alliterative text Piers Plowman (14th c.):

riche: ryden: wrathe

In contrast, the words with wl mainly dropped out of the language rather than being simplified in form. She cites one exception:

Words with initial /wl-/ were very rare in the OE and ME lexicon, and the only survival is lisp < OE * wlispian.

The pronunciation of /wl/ and /wr/ is so ingrained that Minkova need not cover how they would be pronounced in Old English; meanwhile, she carefully documents how the sounds would have changed into Middle English.

  • Nice. So it's not a syllabic /w/ but really a consonant cluster with /l/. Can you comment on the possibility whether /wl-/ is close in articulation to /vl-/ (the 'Vladimir' idea)? /vl/, though not a native English cluster (in MSE words), is one that is easily pronounceable (but I find any attempt at a /wl-/ difficult). How close are they?
    – Mitch
    Feb 8, 2019 at 17:56
  • 3
    @Mitch: Consonant clusters at the beginning of words are generally short in modern English, but some at the end of words (belch, sixths) can take quite a bit of time to pronounce. I can pronounce /wl-/ fairly easily if I take more time than I'm used to with modern English word-starting consonant clusters. I don't really know what the situation was like in Old or Middle English. Feb 8, 2019 at 18:14
  • @Mitch: The /uu/ sound allows for a velar /l/. Think of a frog learning to talk, practicing on the word "gulp".
    – TimR
    Feb 8, 2019 at 18:58
  • Compare this alliterative line from the 14th c. Sir Gawain and The Green Knight with alliteration on /w/: Vmbeweued þat wyȝ vpon wlonk stuffe
    – TimR
    Feb 8, 2019 at 19:10
  • 1
    Expanding on my comment above, there are initial consonant clusters in Polish and Russian which take much longer to pronounce than any in English, for example wzgląd /vzɡlɔnt/. It's possible that Old English was also more tolerant of long initial consonant clusters than Modern English (probably not to the extent that Polish is). Note we've also simplified "kn" and "gn". Feb 9, 2019 at 13:41

As TaliesinMerlin mentioned, the pronunciation of "wl" in Old English is often simply reconstructed as /wl/. That may seem a bit tricky to pronounce, but when I try it, I think I can produce something approximately like [wl].

Polish is an example of an extant language that has "odd" clusters with [w]: there is a Polish word łgarza pronounced /wɡaʐa/, and there is even a word where /w/ comes between two consonants, płciami /pwt͡ɕami/. You could try to look into how /w/ is pronounced in Polish to get an idea of how it might be possible to keep it distinct from /u/ in these kinds of contexts.

  • It certainly isn't part of the current consonant cluster inventory of English. On the other hand, /vl/ isn't but is allowed.
    – Mitch
    Feb 10, 2019 at 23:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.