4 r D PLACE NAMES HIGHLANDS AND ISLANDS OF SCOTLAND y-V~7'*'t^ z^--*^ Q HIGHLANDS & ISLANDS OF SCOTLAND ALEX. MacBAIN, M.A.,LL.D. WITH NOTE.S AND A FOREWORD BY WILLIAM J. WATSON, MA.,LL.D. - \^' ^'--fSSSSS^.-sll^^ ENEAS MACKAY 43 MURRAY PLACE, STIRLING 1922 INTROiJUCTION. Dr Alexander Macbain's work on Names of Places deals with the Cehic names of pre-Gaehc " " he calls Pictish with Gaehc origin which ; names, ancl with names of Norse origin which have been transmitted through Gaelic. The area from which he took his materials was chiefly Inverness-shire, Sutherland, and Lewis. His views on the language spoken by the Picts are " given in his paper on Ptolemy's Geography of Scotland" (published separately), in his edition of Skene's "Highlanders of Scotland," and in several papers contained in this volume, particu- " larly that on the Place-Names of Inverness- shire." His position is that the Picts spoke that the Celtic Early British or a dialect of it, and language of early Britain was practically homo- geneous from the English Channel to the very North. He agrees with Kuno Meyer in holding " that no Gael ever set his foot on British soil save from a vessel that had put out from Ire- larid." Further, assuming that the terms Cruthen (which is the Gaelic form of Briton) and Pict are co-extensive and mutually convertible, VI. INTKODUCTION. ' ' ' he includes under Picts ' the whole of the Celtic settlers in Britain prior to the Belgae, thus ignoring the facts that the Picts are not heard of till about 300 %^, and that all old authorities (Gildas, Nennius, Bede, &c.) state that their original seat in Scotland was in the far North. To him too, as well as to othei modern writers, the Cruithne of Ireland are "Picts." These assumptions do not, how- ever, alfect the linguistic part of Macbain's argument, and his views on the language of the Picts have been generally accepted. No one nowadays would suggest, as Sir John Ehys did once, that the Picts spoke a language that was non-Aryan, and very few^ w ould hold that Pictish was other than Early British. It must be admitted at the same time that some of Macbain's "Pictish" examples are really Gaelic {e.g. Dores, Loch Oich), or at least capable of being explained from Gaelic {e.g. Feshie, Mashie, Geldie). By his treatment of Norse names, Macbam laid a firm foundation for further investigators to build on. He was the first to recognize in practice that the changes undergone by these names in the mouths of Gaelic speakers are not arbitrary, but are capable of classification, and that no derivation which ignores the current Gaelic pronunciation, or which goes against it, can be accepted as certain. INTRODUCTION. VU. In dealing with place-names in general, Dr Macbain's method was first to make sure of the actual pronunciation in Gaelic, and then to com- pare the old written forms of the names when such are available. He also paid attention to the physical characteristics in cases where there niight be more than one possible explanation. In the case of Idrigill, for instance, I remember " how he learned first that there is no gill there, and then that there is a knob-like hill at the extremity next the sea. This is, of course, the only scientific way of treating the subject. He was not always right, and in the papers that follow it will be seen that he changed his views in particular instances; but he was on the right lines. Sometimes he was misled by wrong information : this is most notable in his paper on ISutherland. The only work on Scottish place- names comparable to Macbain's, done by a man of his generation, is the late Professor Donald Mackinnon's series of eighteen articles in the Scotsman on the Place-Names and Personal Names of Argyll, which has not been reprinted. The present volume contains practically all that Macbain printed on the subject of Place- Names. It may be proper to state that before the work came into my hands, the selection and sequence of the papers had been already decided, and pp. 1-64 printed off. Vlll. INTRODUCTION. In the notes that loUow, 1 have indicated the chief points on which, as it seems to me, one might venture to differ fi'om Dr Macbain's explanations. FAOK I . Cataobh, GaUaobh, being certainly datives, are correctly spelled Cataibh, Gallaibh. 6. Ptolemy's tribes : In his paper on Ptolemy's Geography of Scotland, Macbam places " the Caereni in Western Sutherland up " " to the Naver the Cornavii occu- near ; pied Caithness, the horn or corn of " " Scotland ; neighbours to the Cornavii southwards were the Lugi, occupying Easter Sutherland. Around Loch Shin were the Smertae, and Easter Ross was ' occupied, up to the Varar estuary, by the Decantae." The name of tiie Smertae was discovered (by myself) to survive in Carn Smeart (also, sometimes, Carn Smeartach), the name of a hill in the ridge between Strathcarron and Kyle- side. Smertae is a participial form, fiom the root siner, smear; compare M. Ir. smertha. Travedum, Travedrum : read Tarvedum, Tarvedunum. 8 . Creed river—A' Ghriota : correctly Abhainn Ghride. INTRODUCTION. IX. Pittentrail : in Gaelic Baile an Traill, Thrall's Stead, which makes it post- Norse, for train is a loan from Norse thraell. Aberscross : in Gaelic Norse no abarscaig ; ; connection with aber, a confluence. 10. Eogart : in Gaelic Raghart, i.e. Ro-ghort, Ea-ghart, Big-field. xlssynt : assendi will not do, for initial a of Norse ass, ridge, is long. There is another Assynt in Ross-shire . 11. Skinaskink : a ghost-name fouud on maps for the real Sionasgaig. Clyne : as Mr CM. Robertson has pointed out, Clin, which is the Gaelic form, is an oblique case of claon, sloping, a slope. Dornoch : place of fist-stones, i.e., rounded pebbles or dornagan: the locative case is found in Dornaigh, Dornie, in Lochalsh, and elsewhere— r. Place-Names of Ross and Cromarty. Kildonan : Kelduninach c. 1230; in view of other early spellings and of the modern Gaelic, Cill Donnain, there is no doubt " " that St Donnan's Kirk is the original, and that Kelduninach (presumably inter- preted by Macbain as Kil-domnach) is either an error or refers to another place. Lairg : in Gaelic Luirg, dative-locative of lorg, shank; compare Lurgyndaspok, X. INTRODUCTION. PACK 1390, "the Bishop's Leg" (Ant. of Aberdeen and Banff); Magh-luirg, Moy- lurg, in Connacht. Reay : in Rob Donn Mioghiadh, genitive Mioghraidh (rhyming with inntinn, cinnt- in now each, gniomh sin) ; Strathy Meagh- be from ; the rath ; cannot therefore magh second part is rath; compare Dii(n)rath, Dounreay. Elvind is Embo : may be Eyvind's Stead; probably a misprint. locative Creich : in Gaehc Craoich, possibly of craohhach, tree-place; not from cnoc/i. li). Ashore: in Rob Donn, and now, Aisir (a long); anglicised on maps Old-shore. Hysbackie, in Gaelic Hei(ll)sbacaidh; the phonetics clearly indicate an original II (or possibly nn); compare Heisker. Coldbackie, in Gaelic Callbacaidh. not 14. Migdale : in Gaelic Migein, Norse, wholly at least, and to be compared with Migvie, Miggovie, Miggernie, etc. the c Keoldale : in Gaelic Cealldail; palatal is decisive against kaldr, cold. Duible, in Gaelic Daigheabul; the hrst like aoi short. syllable is sounded Leirable, in Gaelic Lireabol; not from leir, it be from Norse liri, a tern, loam ; may also a man's nickname. INTRODUCTION. XI. PAGE Eldrable, in Gaelic Eilldreabol, which can- not be from the source suggested in the text. Mudale, in uaehc Modhadul. Halladale, in Gaehc Healadul. 15. Navidale : in Gaehc Nei(mh)eadail, which may be a hybrid from 7ieimhidh, a sanc- tuary, holy place, and Norse dah, a dale. Conamheall : properly Conmheall, either or Hound con be High Lump Lump ; may the compositional form of cu, hound, or it may represent Early Celtic cimos, high. As Conmheall is the highest part of Ben More in Assynt, it probably represents Early Celtic Cunomello- . iriimisdale, in Gaelic Eumasdal; the u is almost ao short. Iti. Meall Rinidh, in Gaelic Meall Eoidhinigh or Eeidhinigh (possibly Eoithinidh, is in Gaelic Eeithinidh) ; Loch More Loch an Eei'inidh; there is also Allt an Eei'- inidh; the ei is close. My informant connected these names with reidhneach, reithneach (in the glossary to Eob Donn, 1829, reidhne), "bo sheasg," a yeld cow. Reisgill : there is another Eeisgiil in Suther- land, which is in Gaelic Eidhisgii; I am not sure of its position. Xll. INTRODUCTION. i'A<.r ill Gaelic not from k>migol, Smidl)i*;ii ; smiiga; it appears lo represent Norse " smidbju-gil," Smithy-gill. Fresgill, in Gaelic Freisgii, perhaps froui Norse fress, a tom-cat. in Ciaelic or Siiisgil, Sithisgil Sidhisgil ; " Norse seydhir," from which Macbain takes the name, means a lire-pit, cook- mgfire, roasting fire. The derivation is somewhat doubtful . Ben Loyal : in Gaelic Beinn Laghail, Norse laga-fjaU or larja-rdJlr, Law-fell or Law- field. TralagiU : Thrall's Gill, not Troll's Gill. Baligil : the a is long, therefore Norse Bale-gulley, Flame-gulley. Melness, in Gaelic Mealanais; cf. Meala- bhaig, Bent-grass Bay. Shinness : more likely Gaelic sean-innis^ old-haugh . 17. Conesaid, in Gaelic Caonasaid; the phon '" etics to a kein-" point Norse ; compare " " Thorgeir Keingr (genitive Keings) of Landnamabok, where "Keingr" is " explained as uncus," hooked. Falside, in Gaehc Feallasaid, correctly ex- plained in text. Melvich, in Gaelic Mealbhaich, Place of Seabent; not from Norse vik. INTRODUCTION. Xlii. PAGE Golval, in Gaelic Golbhal, where o, being short, cannot represent an of Norse, which would give o long in Gaehc.