I couldn't find an answer to my question because Google Search went downhill these years.

Why is g pronounced as y in a lot of Old English words? Is the reason native phonetics changes happen in a lot of languages or Old Norse influence?

I heard that g is pronounced similiar to y in Danish.

Is Old Norse the reason why this happened in Old English or just because it happened in two languages doesn't mean one is the reason since they're sister languages not to mention in dialectal German g is pronounced as y? thanks in advance.

  • 3
    It certainly did happen in Old English. It is an example of a very widespread phenomenon called palatalisation
    – Colin Fine
    Mar 30, 2022 at 15:24
  • I'm not a native speaker of English, are you saying Old Norse is not the reason why did g shift to y in Old English? Mar 30, 2022 at 15:37
  • 3
    G is not pronounced similarly to y in Danish in any of the many ways that statement can be understood. The letter <g> in modern Danish generally represents /g/, pronounced [g̊ ~ k] initialy. At the end of a phonological syllable, this phoneme is lenited and, depending on context, pronounced [ʊ, ɪ, j, ∅], the third of which is the same sound as represented by the letter <y> in English, as in ‘yes’. The letter <y> in Danish is [y(ː)], which doesn’t exist as a sound in English. You may be thinking of Swedish, where <g> before a front vowel represents /j/, the same sound as in English ‘yes’. Mar 30, 2022 at 16:46
  • 1
    It's a common-enough phenomenon in various Germanic dialects. Berlinisch was famous for the purported sentence Eine jute jebrannte Jans ist eine jute Jabe Jottes. Replace the J's with G's to see the pious Hochdeutsch.
    – jlawler
    Mar 30, 2022 at 20:04
  • 2
    @jlawler Berlin dialect (and Cologne dialect) are different from Frisian and Old English because the palatize their G's unconditionally (Schöne Jruß us Kölle), and don't palatize K's at all. Mar 30, 2022 at 20:49

2 Answers 2


Many of the surviving Scandinavian languages have palatalised g in similar environments, but loan words like give (which replaced the native word, which gave dialectal yive) show that at the point of borrowing, Old English must have already palatalised these g's, and Old Norse cannot have done so.

It's worth noting that standard textbook Old Norse is actually the Old Icelandic of the the Sagas, and dates to around the 13th-14th centuries; after the end of the Old English period, and several centuries the period of most intense contact in the Danelaw in the 9th-10th centuries. It is also this earlier period that standard textbook Old English dates to, mostly being based on texts produced in the West Saxon dialect during the reign of Alfred the Great and his immediate descendants.

Even the First Grammatical Treatise, our earliest description of Old Norse phonology, describes an archaic phonological system very different from textbook Old Norse (especially the presence of nasal vowels) and dates to the 12th century, well after the period of most intense contact, and does not appear to show palatalisation of g in these environments.

It should also be noted that, whilst most modern Scandinavian languages have palatalised g in some form or another, the reflexes of the initial g in Old Norse gefa vary between [ɡ̊] (Danish), [c] (Icelandic), [t͡ʃ] (Faroese), and [j] (Swedish and Norwegian), and with the exception of Swedish and Norwegian these are all still phonemically /g/. From this evidence we would probably reconstruct Proto-North-Germanic* to have had /g/ here.

So the early Old Danish spoken in the Danelaw almost certainly did not palatalise g in environments like this, and the two developments must have occurred in very different eras (before the 9th century in Old English, and various points after the 12th in Old Norse, and even then possibly not in all dialects).

So why did they both palatalise?

Because this an extremely common sound change cross-linguistically. It's the cause of "hard" and "soft" c & g in the Romance languages (apart from Sardinian), and numerous consonant alternations in the Slavic languages stemming from the Slavic First & Second Palatalisations.

  • 1
    It’s worth specifying that the reflexes of ON gefa phonetically start with [g̊, c, tʃ, j] in those respective languages, but apart from Swedish and Norwegian, those are all still phonologically /g/ to this day, just like in Old Norse. The palatalisation in Icelandic and Faroese is still allophonic and automatic. It’s only in Norwegian and (less prominently) Swedish where a distinction between /ɡ/ and /j/ before front vowels has become phonemic. Mar 30, 2022 at 16:52
  • For a parallel outside Germanic (in fact, outside Indo-European), Turkish 'ğ' (historically from /g/) is pronounced as an approximant or sometimes just as lengthening the vowel. When between front vowels, it is generally somethinig like /j/ (the English semivowel 'y').
    – Colin Fine
    Mar 30, 2022 at 17:09
  • @ColinFine that's more complicated because ğ is also lenited in all other positions, so it's unclear whether it was /g/ that palatalised, or an already lenited ɣ
    – Tristan
    Mar 31, 2022 at 8:43

No, it is not related to Old Norse, it is inherited from Anglo-Frisian. Old Norse supplied some words with "hard g" and "hard k" in environments where Anglo-Frisian developed palatals, e.g., all the native English words staring in /sk-/.

Norwegian and Swedish later also palatised some of their g's and K's, but this development is independent from the English one.

The sound shift itself is quite natural and similar sounds shift occur in other languages as well.

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.