I cannot help but notice some 'r'-s seem to have randomly disappeared in both German and English. What is going on there?

  • 6
    According to A Historical Phonology of English by Donka Minkova (p121): Sporadic loss of <r> occurs in some onset clusters: sp(r)ecan ‘to speak’, sp(r)æċ ‘speech’, p(r)æˉtiġ ‘clever, pretty’. Mar 11, 2021 at 17:10
  • @"Joyful Sadness", so, the vowel in "pretty" was originally long 'ae'? So, how did it came to be short 'i'? And why was the same phenomenon, the random disappearance of 'r', occuring in German? Mar 12, 2021 at 8:59
  • 1
    Since we have worship versus German Wirtschaft "economy, establishment" contrast so much, whereas Weihe is closer to the sliritual sense, it should be considerable if this was a case of substitution that might apply to Werald and Welt as well. Between PIE *wer-, *werh1-, *swer-, etc. a reasonable mount of overlapping semantics can be infered as much as for *wihros "man" > were and *wer-(?) > warden, warroir, etc. (cp. further warranty, promis, Ger. Versprechen, Vorsehen... Lat. spect-?)
    – vectory
    Mar 13, 2021 at 6:38
  • 1
    Re: why there is no r in PDE speak - we don't know why it happened. cf. Anatoly Liberman's remark that "Despite a sizable number of pairs like speak (in Old English, specan coexisted with sprecan: cf. German sprechen), the mystery of the fugitive r has never been solved." blog.oup.com/2011/06/pretty-2
    – Alex B.
    Mar 14, 2021 at 16:37

1 Answer 1


Different reasons. 'they speak/they spoke' in Protogermanic was something like 'sprekanþi/spurkun'. Note the different position of the 'r'. Our forebears made the word more regular with the Germans going for 'spr-' and '-k-' whilst the Angles going for 'sp-' but also '-k-'. 'world' and 'Welt' are both contactions of 'weraldiz' ('men-age' or 'generation'). Contractions are always fairly irregular. The Germans and the English contracted the word slightly differently.

  • 2
    Disagree. Simplification of the cluster after contraction from werald is imaginable at least if the /r/ was a Upper German alveolar trill/flap next to the lateral approximant /l/. But about speak; it is attested besides sprekan in OE, maybe a case of different though very similar etyma like answer ~ Ger. Antwort. See, sprek- compares well with spray, whereas speak might compare to spuck- "spit" (cp. spit it out, Ger. spuck's aus, i.e. "tell me"), or spüren "sense" versus Spuk or fühlen "feel" (Grimm's law applies), or PIE *werh1- vs. *bheh2-, etc.
    – vectory
    Mar 13, 2021 at 6:33
  • @vectory It isn't surprising that both specan and sprecan existed in Old English as it will have been at that time that the regularising was happening. However I'm in good company see Kroonen, Guus "Etymological Dictionary of Proto-Germanic" (Leiden: Brill, 2013) page 469. You are right that the simplification of /rl/ to /l/ is imaginable in German. Is it a regular sound law?
    – Ned
    Mar 14, 2021 at 9:52
  • It isn't surprising? There's a hefty amount of circular reasoning and false equation in that argument. Is /rl/ > /r/ regular (in German)? Beats me. Comparing Kalle ~ Karl, Polier << Fr. parler, Berlin [bɛɐ̯ˈliːn], [bɛʁˈliːn], (Berlinisch) [bɐ̠liːn], the elision is regular where the underlying backed resonant may be elided in syllable codas, similar to rhotic vowels of AmE rather than the alveolar trill /r/ (as is reconstructed for PGem). This extends into Middle Bavarian, which is an upper German dialect coming from Middle High German, so I don't know where the /r/ can be examined.
    – vectory
    Mar 15, 2021 at 22:02

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.