ONlwSG

asaileag f. [ˈas̪ilaɡ̊], also aisileag [ˈaʃilaɡ̊], gen. -leig -[læɡ̊ʲ], -[lɛɡ̊ʲ] ‘storm petrel, Hydrobates pelagicus’. This word is given as assilag (Martin 1698, 108; Buchan 1727, 21), ashilag (Macaulay 1764, 160), aslag mor 

?For aslag-mhara, with the lenited genitive of muir ‘sea’; mòr ‘large’ seems unlikely to be intended here.

(Ross 1890), aisleag (Mackenzie 1911, 46: Procellaria pelagica 

Now Hydrobates pelagicus.

), aisileag (Faclan bhon t-Sluagh^˄: (Skye) ‘stormy petrel’).

McDonald (2009, 339: likely; 2015, 124–25: uncertain) derives aslag from ON áslákr, a poetic word for ‘cockerel’ and a form of the man’s name Ásleikr m., but the long stressed vowel in Old Norse militates against this. For aisileag, Henderson (1910, 123–24) suggests a connection with Scots asel ‘storm, cold sharp blast’ – as a harbinger of bad weather – (which Jakobsen (1928) relates to Norw. as in the sense ‘storm’ and the related verb ase), which on the face of it cannot be entirely ruled out, although the Old Norse antecedent is unknown. Mackenzie translates aisleag as ‘the little ferryman’, but this is a folk etymology based on SG aiseag m. ‘ferry’. Sommerfelt (1952a, 230 

So also McDonald (2009, 360).

) confidently states the word is based upon ON hafs, gen. of haf nt ‘ocean’ + SG faoileag f. ‘gull’, noting that ON haf was borrowed into Gaelic (see abh); however, a compound such as EG *af-fhaílóc would seem more likely, which would not suffice here. Citing the form asaileag, Lockwood (1976, 276–77) tentatively suggests a derivation from ON *haf-svala, which he translates literally as ‘sea swallow’ (although Norw. havsvale in fact means ‘storm petrel’), noting that ON *haf-svala > SG *asala + the Gaelic suffix -ag is feasible but unprovable.

One of the strategies for dealing with initial ON h- in Gaelic was to discard it, e.g. ON halsa ⇒ SG abhsadh, q.v., and loss of f before s in Gaelic reflexes of Old Norse loans does not appear to cause compensatory lengthening, e.g. ON *Sef-sætr > SG Seiseadar [ˈʃeˌʃad̪̥əɾ] (Cox 2022, 864–66 + fn 674), so that ON *haf-svala could be expected to yield EG *[ˈas̪əɫə] in the first instance, as Lockwood supposes, which might yield both SG *[ˈas̪əɫ̪ə] and *[ˈas̪ələ], 

With lenited broad /l/ yielding unlenited broad /ʟ/ and lenited slender /l’/, respectively.

which, with the addition of the Gaelic suffix -ag (EG -óc), might yield the syncopated forms aslag and aisleag (assuming they are authentic), which in turn could yield asaileag and aisileag through epenthesis. However, ON *haf-svala would be unlikely to yield palatal medial [ʃ] in Gaelic and, if the syncopated forms aslag and aisleag did not arise first, we would have to accept direct phonemic interchange between non-palatal /s/ and palatal /s’/ for the development of aisleag and aisileag to take place; cf. the similar doublet SG (f)annlag and (f)ainnleag 

With the variant spellings (f)amhlag and (f)aimhleag; cf. Ir. (Omeath) amhlóg ‘species of sea-gull’ (Dinneen 1947).

‘swallow, marten, storm petrel’ (< EG fannall and ainnel(l) ‘swallow’).

See also Coates 1988.