Two Linguistic Oddities.

1) This post by Anatoly Vorobey (in Russian) describes an interesting detail of Russian morphophonemics: feminine words whose stems end in consonants + НЯ [nʲa] lose the palatalization on n in the genitive plural. Thus песня [ˈpʲesʲnʲə] ‘song’ has genitive singular песни, dative singular песне, and so on, with palatalized [nʲ] throughout, but the genitive plural is песен with unpalatalized final [n]. But! There are four exceptions:

барышня ‘girl of gentry family, miss’ – барышеНЬ
боярышня ‘boyar’s daughter’ – боярышеНЬ
кухня ‘kitchen’ – кухоНЬ
деревня ‘village’ – деревеНЬ

And the pull of these exceptions, with the “expected” palatalization, can attract other words, for example башня ‘tower,’ for which Anatoly uses the “incorrect” genitive plural башень in place of the traditional/”correct” башен in speech, which led him to write it that way in a recent post. And he’s not alone: “поиск в корпусе русского языка находит небольшое, но реальное количество старых книг и авторов, которые предпочитали писать именно “башень”, очевидно потому, что так говорили” [a search in the Russian language corpus finds a small but real number of old books and authors who preferred to write “башень,” obviously because they said it that way]. I love this stuff.

2) The delightful NY Times story “A Monkey Is on the Run in the Scottish Highlands” (archived) explains that the Japanese macaque in question “escaped from an enclosure in the Highland Wildlife Park in Kingussie, Scotland, and fled into the Scottish highlands,” later adding:

Amused residents, who have given the animal the nickname “Kingussie Kong,” have found themselves invested in its fate, and journalists have followed animal keepers as they have swept the hills.

I assumed Kingussie was pronounced kin-GUS-si and thought “Kingussie Kong” was slightly off, but then I looked it up and discovered it’s actually /kɪŋˈjuːsi/ (king-YOO-see), representing Scottish Gaelic Ceann a’ Ghiùthsaich. Now “Kingussie Kong” makes perfect sense, and I thought I’d pass along that unusually unexpected spelling/pronunciation matchup.


  1. /kɪŋˈjuːsi/ is also unusual because English doesn’t really do /ŋ/ immediately before a stressed vowel. I remember first noticing this during a discussion a ways back on the Log about whether one should say “Shandongese” or “Shandongnese”.

  2. But it’s not immediately before a stressed vowel, it’s before a consonant (y). It’s exactly like “King Jussi” (if Jussi Björling had been a king).

  3. But /juː/ is one vowel phoneme so many (not all!) such phonotactic rules still apply to it? Anyway it feels awkward to pronounce to me.

    Also clearly this rule doesn’t hold at word boundaries, so “King Arthur” or “checking in” is fine. On the other hand, “Long Guyland” is a thing, perhaps because it’s one word to the people who say it…

  4. But /juː/ is one vowel phoneme

    No, it’s a consonant followed by a vowel. That’s why we say “a Jussi Björling recital” or “a Yukon Gold potato,” not “an.”

  5. Wikipedia agrees with you. And yes, I was thinking of the article when I said “not all phonotactic rules”. But then English speakers have no problem saying “view” but insist on “vee-yenna”. It seems like /juː/ is allowed in places where j+another vowel isn’t.

  6. The Highland Wildlife Park was in the news recently for its program to breed and release the endangered Scottish wildcat into the Highlands. That’s the fierce little animal that has led some old Scottish families to adopt the motto “Touch not the cat bot a glove” (‘Na bean don chat gun lamhainn’). The hardest part of the reintroduction was they had to convince everyone in the target area not to keep domestic cats outdoors.

    I went to the Highland Wildlife Park a number of years ago. Worth a visit if you like animals. Kingussie also has the Highland Folk Museum.

  7. PlasticPaddy says

    re Kingussie
    This type of place name = head + [Tree] is found in Ireland, e.g. Kinnaderry, Kinnewry, Kinsale. I would have expected syllabification as Kin + gussie, but the ng would appear in casual speech.

  8. Jen in Edinburgh says

    Interesting – I would have said that Scottish ‘kin’ names tended towards coastal features (Kintyre, Kennacraig, Kentallen), but I don’t know if that would stand up to investigation.

    But Kingussie doesn’t really split like that, because ‘ghiuthsaich’ starts with a y- sound – it seems to be a combination of that and the more nasal(?) gaelic ‘n’ that suggests -ng in English.

    (I think you’d have to be being quite fastidious to get in the -ng AND the y-, though – I say KINg- OOSSie, like the people here, and if I try to get the ‘y’ in it comes out more like ki-NYOOSSie. I need to go down to Waverley Bridge later anyway, I’ll try to time it to find out what Scotrail say!)

  9. I wonder if “ceann” in Kin- names is really the oblique form “cinn”?

  10. Ladfoged agrees with F that [ju] is a phoneme. It simplifies accounts of /Cju/ but complicates /#ju/ — not just ‘a’ vs ‘an’ but also linking /r/.

  11. The monkey has been recaptured in Insh.

  12. Jen in Edinburgh: the more nasal(?) gaelic ‘n’

    I think the word you’re looking for is “velarized”, going by Wikipedia on Scottish Gaelic phonology. But others here may know better.

    FWIW, this Online Scots Dictionary gives a Scots spelling of Kineussie, and pronunciation [kɪnˈjusi], with n instead of ŋ.

  13. F: English doesn’t really do /ŋ/ immediately before a stressed vowel.

    That might be a reason for difficulty with another Scottish place name: Longannet, pronounced /lɒŋˈænɪt/. John Wells blogged about Longannet when he heard it mispronounced by politicians as “Long-gannet” with a g, and he pointed out the analogy with Kingussie.

    Scottish Church Heritage Research thinks it could be from Lann Na H-Annaide, meaning ‘former church enclosure’.

  14. As noted above, Kingussie may be proximately from Cinn a’ Ghiùthsaich, as here. (With frozen ‘locative’ cinn in place names, some say. I am very interested by this formation. but don’t have time to look into it now. Maybe someone else can. Perhaps originally dative cinn in independent locatival use, delabialized Old Irish variant of dative ciunn? Cf. Thurneysen, A Grammar of Old Irish §251.3, p. 161.)

    If from Cinn a’ Ghiùthsaich, then perhaps we can invoke Gaelic slender -inn > Scots -ing. From Paul Johnston ‘Older Scots phonology and its regional variation’, p. 106, in Charles Jones, ed. (1997) The Edinburgh History of the Scots Language:

    (a) The Palatal Nasal /ɲ/

    There seems to be no question that /ɲ/, taken in from French or Gaelic, survived at least before original or prosthetic /a/ > /i/ all across Scotland throughout the period, as spellings implying /n/ only start appearing after the Scottish standard variety was in process of being replaced by the English one, and and spellings (though the last could be spelling pronunciations) survive in cunyie, linyie, minyie into the nineteenth, if not the twentieth century (Robinson 1985: 128). In final position after high vowels, the native development was to /ŋ/, possibly through a palatalised velar the same way as Gaelic /ŋj/ (written -inn) developed, with benign, condign, malign, reign, sign all being spelled with -ing as early as the fifteenth century. These forms were later replaced by /n/ forms from English (Murray 1873:124; Zai 1942), but were probably General Scots in earlier times. Those items that do not take /n/ or /nz/ spelling pronunciations today now have /ŋ/, such as ingan, Menzies (as if Mingis); spellings implying this development also date from the fifteenth century.

    (Scots ingan [ˈɪŋən], ‘onion’; surname Menzies.)

  15. PlasticPaddy says

    For your example, you have to postulate delabialisation. However, at least in Ireland there are examples where that is not necessary, e.g., Ring > an Rinn and Ringaskiddy > Rinn an Scídigh. Maybe Ringawaddy in Co. Down is more relevant.

  16. I don’t think the problems with Kingussie and Longannet are phonotactic, I think they are mis-segmentations of the written form, which is also how I read Wells.

    The issue with /ju/ is probably that it has two or three origins, /iw ~ ew/ and /y/.

  17. I don’t think the problems with Kingussie and Longannet are phonotactic, I think they are mis-segmentations of the written form

    I agree.

Speak Your Mind