Post History

A peculiar feature of Icelandic is that it distinguishes vowel length, not just for pure vowels, but also for diphthongs.  (Vowel length does not distinguish meaning, or at least not directly; it is determined from the context, such as syllable stress and coda composition, rather mechanically.)

Any diphthongs in unstressed syllables are always short.  Being short, they have a tendency to be reduced to barely recognizable diphthongs, or often even to pure vowels.  This holds for both loanwords and for domestic vocabulary (of Old Norse origin).

It may well be that unstressed diphthongs get reduced more often in loanwords then in domestic words, but I have no proof nor reference nor explanation for that.


It is even less clear to me what determines how a given loanword enters Icelandic (i.e., "kóróna" vs. *"korona"), although some light is thrown on that topic by the thesis/[article](https://core.ac.uk/download/pdf/19204203.pdf) "I prestiti di origine latina in islandese moderno: un glossario" by Matteo Tarsi.  For "kóróna" specifically, Tarsi attests the word in Icelandic by the 13. century at the latest, when this spelling still indicated a long pure vowel  /ó/, whereas the same spelling would probably shift to be pronounced /au/, same as in current Icelandic, by the 14. century or so.  The word can be assumed to be a direct borrowing from Latin (and ultimately from Greek).  The second "ó" is both long and stressed in Latin, so it is unsurprising that it was kept long in Icelandic.  The first "ó" is short and unstressed in Latin, and stressed in 13. century Icelandic; for a reason unknown to me, it received the long vowel in Icelandic, too.  However, the point is that the diphthongization of the long vowel (demonstrable in stressed syllables, at least) came about only later.