The greater part of the islands have Gaelic names, but some of them, even of the larger, have Norwegian ones, and no Gaelic besides; that is to say, their original form, still shining through the alteration of later days, is Norwegian. To begin from the North, Lewis is only an alteration of the Norwegian Ljóðhás (the sounding house), wherefore also the island is called Leodus in our chronicle. Northuist, Harris, and Southuist, were called Ivist, which means simply dwelling, habitation, intimating, perhaps, that this was the chief abode of the Norwegians in that quarter of the group. The names Fladda, Sanda, Watersa, Eriska, are no doubt originally Flatey (flat isle), Sandey, Vainsey, Eireksey; Paasa and Rona were called Rauneyjar, the latter consequently Rauney (from raun, experiment); Skye was called Skið (piece of board), Ulva, near Mull, seems to have been Ulfey (wolf's isle), Gometra, Gudmundarey (Gudmund's Island), Staffa, Stafey (the island of staves, because of the basaltic pillars), Jura, Diura, perhaps Dýrey (island of deer).
Now it must be that our ancestors, like most people in former times, often changed names, belonging to other languages, into forms as similar as possible in their own language, giving a kind of plausible signification and that there fore, Ljóðhus, Ivist, Skið etc., may only be substitutes for Gaelic names of a similar sound, but quite different signification; yet this does not alter the fact that the islands mentioned had got Norwegian names, and therefore were probably inhabited by Norwegians; nor is it to be denied that at least the final syllable a or ay, in so many of the names, even in those, the chief part of which are of Gaelic origin, is the Norwegian ey (island), as, besides the aforesaid ones, which are thoroughly Norwegian, Pabba (Papey). Oronsa (Oransey), Colonsa (Kolnsey), etc.; even the name calve (kálfr), calf, always affixed to the name of a larger adjacent island, of which that in question is said to be a calf, is found here in Mýlarkálfr - the Calf of Mull, near Tobermory. Also some Norwegian names of single places are yet existing, and there might on more accurate examination be found to be more than expected, as Cornbust, Kirkbust, in Benbecula, originally Korn bolstaðr or Kornbustaðr, Kirkjubustaðr (corn abode, church abode), of which names a great quantity exists in Norway, generally abridged in quite the same manner.
In Man, likewise, although the language is a dialect of the Gaelic, and the greater part of the population therefore always must have been of Celtic origin, many of the local names are Norwegian, as Snaefell (snow-mountain), the name of the highest peak, Dalby (Dalabaer), Fleshwich (Fleskvik), Wardfell (Varðfell), Perwick (Petrsvik), Strandhall (Strandarholl), Ronaldsway (Ragnvaldsvað), Langnes (Langanes), Laxi (Laxa), Egness (Eggjanes), Kennay (Kinna), Sulby (Sulabaer), Maelar, Ayre (Eyri), Jurby (Jorabyr, Dyrabyr ?), Mirescog (Myraskogr), Ramsay (Hrafnsa), etc. - several places having even double names, or rather having had them in former times, as when, from the record of boundaries, written in the same codex with the Chronicle, we learn that Kirk-Michael was also called villa Thorkel i.e. Þorkelsbaer. The best evidence, however, of the Norwegian population having been tolerably strong at Man, are the numerous runic inscriptions in the Norse language still preserved, but even these show by their numerous grammatical errors, that the language spoken here had lost much of its purity, no doubt owing to the strong mixture of the Norwegians with Celtic elements; and, indeed, of the names occurring in the inscriptions themselves perhaps the one half are Gaelic.
The transition to Old Irish, the language of the earliest sources in the Latin alphabet, takes place in about the 6th century. Since Ogham inscriptions consist almost exclusively of personal names, linguistic information that may be glimpsed from the Primitive Irish period is mostly restricted to phonological developments. Roughly 380 inscriptions are known in total is very close to the number of known inscriptions in the contemporary Elder Futhark.
These circumstances here explained account amply for the less tenacity of the Norwegian language, manners, and institutions in the Isle of Man and the western Islands, than in those north of Scotland, which were entirely Norwegianised, and it is rather to be wondered at, that the connection with and dependence on Norway could be kept up so long, as till A.D. 1266, and in ecclesiastical respects for more than a century longer. It was owing, as it appears to us, partly to the naval strength of Norway in the Middle Ages; partly, and perhaps chiefly, to the spirit of independence, prevailing among the Insular chieftains themselves, which made them prefer an allegiance which, owing to the distance, could scarcely be more than nominal, to the Norwegian kings, rather than to the dominion of the Scottish king, who being always at hand, could exercise it with greater force. It is very likely that in this respect the expedition of King Hácon, had it even been crowned with complete success, would have soon had quite different effects from what might be expected, as the chieftains, seeing that now also the Norwegian king had the mind and power to make their dependency a real, not a nominal one, they would lose their interest in keeping it up against their national sympathies, and therefore sooner or later would have thrown it off