Silent "w" in words starting with "wr-"
My eldest is a beginning reader. Yesterday we read one of my favorite books, The Wreck of the Zephyr. He pointed at wreck and asked me why that one looked like it said "wuh-reck." I explained that spelling is funny like that sometimes.
This didn't satisfy my curiosity though. Silent w is not uncommon—we see it in the question word 'who' for instance—but it often appears in the combination wr-, and this is what I am curious about.
Was the w ever pronounced in these words (and if so, how)? Do these kind of words all share a common lineage that has some unique sound represented by this combination? (I am thinking of, for instance, that someone told me once that most words with ph come from Greek.)
Not My Field, so subject to correction:
In Old English the “voiced labiovelar approximant” /w/ was in fact pronounced in the initial clusters /wr/ and /wl/. Lass, Cambridge History of the English Language describes the loss of this pronunciation in the context of “Onset-cluster reduction” (III, page 122):
Witch/which, not/knot, Nash/gnash, rite/write are homophones in most varieties of English (see below on the first pair); conservative spelling preserves an earlier state. During our period [1476-1776] English underwent the most extensive simplification of onset clusters in any Germanic language. Old /wr,wl/ and /xn,xr,xl/ were lost in many other dialects, but /kn/ was generally retained (E *knee /ni:/ v. German, Swedish, Dutch /kni:/).
By late Middle English /wl/ had reduced to /l/ (wlispian > lisp), and /xr,xl,xn/ to /r,l,n/ (hracu > rake, hlūd > loud, hnacod > naked). The only (from a modern perspective) ‘exotic’ clusters remaining were /xw/ (hwilc ‘which’), /wr/ (wrītan ‘write’), and /kn,gn/ (cnāwan ‘know’, gnagan ‘gnaw’). All except /xw/ (> /hw/:3.5.1) simplified in the sixteenth and seventeenth centuries; /hw/ remained for some southern speakers until well into this century, and is still stable in Scotland, Ireland and parts of North America.
The first post-Middle English simplification is of /wr/: while most sixteenth-century sources are uninformative, Coote (1596) gives wrest/rest, wrung/rung as homophones. There is sporadic retention in Hodges (1644), and Jones (1701) seems to be the last mention of possible /wr/. In general, /wr/ > /r/ during the seventeenth century.
German developed similarly, but not contemporaneously. Joseph Wright, Historical German Grammar (1907), I,119:
§229. Germanic w = Engl. w in wet (generally written uu, uv, vu, vv in OHG. manuscripts) remained initially before vowels in OHG. and MHG. as OHG wahsan, to grow, wëg, way, wësan, to be. It became the labio-dental spirant v (written w) = Engl. v in vat, in late MHG., and this has remained in NHG. [...] Initial w had disappeared before l,r in prehistoric OHG., as OHG. ant•luzzi, Goth. wlitz, face, countenance; OHG rëhhan, Goth. wrikan, to persecute.
StoneyB, Lass, and Wright have outlined the recent history of initial WR simplification. The ancient history of how they got that way is interesting, too.
I checked the American Heritage Dictionary of Indo-European Roots and found, to my surprise, that all the words beginning with wr- in the American Heritage Dictionary (with etymologies traced to Proto-Indo-European, which includes all the words under discussion here) come from just two PIE roots:
*wer³- Conventional base of various Indo-European roots; 'to turn, bend'. Derivatives include stalwart, weird, vertebra, wrath, wrong, wrestle, briar, rhapsody, and worm.
*werg- 'to do'. Derivatives include work, urge, energy, allergy, wrought, irk, wright, bulwark, and boulevard.
Note that both of these roots have a vowel between the W and the R. That's not always true in the words they form, however. That's because of the Schwundstufe.
Indo-European, as far as we've been able to figure out, used vowel switching patterns regularly (the phenomenon is called Ablaut, a German term), so that for a given root, it is common to find both "E-Grade" and "O-Grade" words in daughter languages, like Latin pedis vs Greek podos, both meaning 'foot'.
There's a third Grade, however, called "Zero-Grade"; that's where the vowel is neither E nor O but rather absent. In the original German it's Schwundstufe, the 'disappearing grade'; sounds both mystical and official.
And that's what wright and wrought and wrath and wrestle and so on come from. They're Schwundstufen. They come from processes or previous alternants of words with that root where the vowel became superfluous and was dispensed with. These always conform to the pronunciation norms (the phonology) of the people speaking them at the time; when these norms change, as Lass explains in StoneyB's answer, things happen.
These are generally words of Germanic extraction, where the German w is pronounced like our English v.
The German equivalent of English wreck is Wrack, pronounced roughly “vrahk”. German also has wringen, pronounced “vringen”, which is equivalent to English wring.
What most people find problematic about the wr- blend (and what they are usually most curious about) is how it could have ever possibly been pronounced to begin with, since w to r requires a complete repositioning of the tongue and lips, which would naturally insert an intrusive vowel between them to sound like "werist", for example. Although we can't know for certain, as the w died out early in Modern English, it was mostly likely not pronounced as a blend at all, but instead, the w colored the r so that the tongue would be positioned for the r, but the lips would be rounded to form w, resulting in a phoneme that no longer exists in Modern English.
Some linguists speculate that the wr- blend is a remnant of a time when r was a flip instead of a liquid, and in that case the w would be pronounced independently. This may have been true early on, but the r was probably a liquid by Late Middle English and the w was still pronounced, so that the repositioning of the mouth as described above would have been still been likely.