Nuckelavee

The first question is not whether the monster is real. It is how old the word is. The January 1891 article is the oldest printed appearance of "Nuckelavee" that I have been able to date, and the searches that failed are worth stating with their controls, because a nil result proves nothing on its own. Thomas Edmondston's An Etymological Glossary of the Shetland & Orkney Dialect (1866) — exactly the book in which the word should be, if it were current — returns zero for nuckelavee and any OCR-mangled form of it; the control in the same file is positive: "Niggle, a sort of water-kelpie", with Edmondston's guess at a derivation from "gneg, a horse, and el, water". John R. Tudor's The Orkneys and Shetland (1883) also returns zero, against 34 hits for Stronsay, 14 for trow and a glossary entry "Nuggle or Niogle, O. and S., a kind of water kelpie". Jamieson's Etymological Dictionary of the Scottish Language and Samuel Hibbert's Description of the Shetland Islands return zero on the same test, with the same family of water-horse words present. Every one of these searches was run twice. The Dictionary of the Scots Language has no headword nuckelavee at all: the word appears in it exactly once, as a quotation dated 1891 from the Scottish Antiquary, filed under NEUGLE — where the earliest dated quotation of the family is a "neagle" from the Statistical Account of 1845. So the creature reaches print once, from one collector, quoting one unnamed informant, forty years after he heard it. Dennison is his own best sceptic. He does not say the old man met the Nuckelavee; he says the man "was credited with" having met it. And in the first line of the narrative he names the model himself: Tam o' Shanter. Burns's poem of 1790 ends with Tam saved by the fact that the witches cannot follow him over running water — which is, exactly, how Tammie is saved. Whatever else the story is, its escape is a device already famous in Scotland a century before Dennison wrote it down. The kelp gives the belief a floor it cannot slip below. Kelp-burning was introduced to Orkney in 1722 by James Fea of Whitehall in Stronsay, and Tudor quotes George Barry's Statistical Account of Kirkwall on how the islanders received it: they were sure it would drive the fish from the coast, "certain it would destroy both the corn and the grass", and afraid it might stop their women having children. The Nuckelavee's rage at the kelp fires is that documented grievance with a face on it, and it cannot be older than 1722. The disease is real too: Mortasheen is glanders, and the Scottish National Dictionary records it in Orkney in the Stenness Kirk Session records of 1768 — a hundred and twenty-three years before Dennison printed the story that explains it. The rest of what circulates today was added after him, and can be dated. The etymology first. Dennison declines to give one: "Without speculating on the derivation of this name, which will be pretty obvious to those acquainted with northern mythology, it may be said that in plain English the name means Devil of the Sea." The word "knoggelvi", which hundreds of pages now attribute to him, appears nowhere in his article, nor anywhere in volumes V to VII of the Scottish Antiquary, which contain his whole series — against one occurrence of "Devil of the Sea", the control. It comes from the Orkney website Orkneyjar, run by the Orcadian journalist Sigurd Towrie, which states without a source that "Nuckelavee is a later corruption of the dialect name 'knoggelvi'". It entered the English Wikipedia on 17 September 2008, cited to that page — a page which today returns a 404, so that the article's only authority for its etymology is a dead link. The Shetland name mukkelevi comes from the same website. Samuel Hibbert is quoted for a second etymology, and here the direction of the argument has been reversed. Hibbert (1822) is discussing the water-god of the Celts and Teutons, "Nocka, Nickur, or Necker, answering to the Latin necare" — he is not analysing the name Nuckelavee, which did not exist in print for another sixty-nine years. The sixteenth century is offered as a precursor and does not hold either. Jo Ben's Descriptio Insularum Orchadiarum of 1529 does describe a sea creature molesting a woman on Stronsay, and Hibbert printed the Latin himself: the being is "indutus … algis marinis, toto corpore, similis est pullo equino" — clothed in seaweed over the whole body, like a colt, and in Hibbert's text with matted hair. Jo Ben calls such beings Troicis, trows. Hairy and clothed in weed is the precise opposite of Dennison's Nuckelavee, whose whole horror is that he has neither hair nor skin; and Dennison, who wrote a separate instalment on the Sea Trows, would have filed it there. The Wikipedia editor who added the Jo Ben paragraph in June 2014 left the honest edit summary "I'm not sure whether this is a reliable source?" And one loss, rather than an addition. The version that reached the twentieth century is not Dennison's. Sir George Douglas reprinted the tale in Scottish Fairy and Folk Tales (Walter Scott, first published 1893, reprinted 1900, 1901 and 1903) — the edition James Torrance illustrated, and the source of the picture on this page. Douglas credits Dennison in a one-line footnote and then cuts, in silence, the kelp, Stronsay, Mortasheen, the Mither o' the Sea, the protective prayer and the gloss "Devil of the Sea". His own introduction reduces the creature to a centaur, "shaped like a man above and like a horse below". For a century that abridgement, not the original, was the Nuckelavee most readers met.