sign in sign up
<<
Home , Caithness, Cape Finisterre, Cape Wrath, Dornoch Firth, Dunnet Head, Geography of Scotland

AND IN ANCIENT GEOGRAPHY. 79

III.

SUTHERLAND AND CAITHNESS IN ANCIENT GEOGRAPHY AND MAPS. BY REV. ANGUS MACKAY, M.A., CORK,. MEM. S.A. SCOT., WKSTERDAIE, CAITHNESS.

Ptolem f Alexandriao y o flourishe 0 wh A.D.,14 . ,c dcompile a d geograph e theth nf o yknow n worl n eighi d t books, whic s suci h n ha improvement upon earlier attempt a simila f o s r kind tha t continuei t d n usi e until afte e revivath r f learnin o e l 15tth h n i g centurys Hi . longitudes were calculated fro Canariese mpoina th n i t , supposee b o dt the westernmost part of the world; but he reckoned it as only 2J° west of St Vincent, whereas the real distance is over 9°. Partly owing to this miscalculation, some countries are thrown considerably out of place when his data are reduced to map form; and this is especially true of , which fro e Clydm th Fortd ean h northwar s twistei d e du d east, making thence a right angle with . Notwithstanding this glaring defect, however, his geography of the north of Scotland cannot fail to interest the antiquary in search of light upon the place-names and people of that part. Starting from the VOLSAS SINUS in the W., which he sets down in 30'whic° d latd an 60 ,.an long ° h 29 scholar. generallw sno y understand to indicate Kyle, the names of the places given by him, workin e easge th nort th roun ts follows o a t side hy e b dar , , with Ptolemaic long latd .an , respectively rivee Th r 30'° .Nabarus 60 ; x ° 30 , e promontorth y Tarvedu r Capmo e Orkas 20'° ° 15'e 31 pro,x 60 th ; - montory Virvedrum, 31° x 60°; the promontory Verubium, 30° 30' x 59° 40'; the Ila, 30° x 59° 40'; and Alta Eipa, 29° x 59° 40', the bank Oykelle oth f , between. Ros Sutherlandd an s districe th calle w n I no td . Caithness he locates the Cornavii; the Caereni he places in the country, arid the Lugi in South Sutherland. The above too meagr el thadatal ts i a Ptolemy give r tha fo sshale t w part d l an ; 80 PROCEEDINGS OF THE SOCIETY, JANUARY 13, 1903.

now e variou looth t ka s name n detaili s , endeavourin o extract g t from them their story.1 CORNAVII.—This was the name of the tribe occupying .Caithness, and means "peopl e Horn,th f o e " fro a mCelti c root appearin n Gaelii g c corn, Cymric corn, Breton korn, all cognate with cornu, a horn. The people occupying-the peninsula between the Dee and the Mersey in England were also designated Coruavi y Ptolemyb i ; whil o thit e s have w ey Cornwalda l wit s Cornishit h d alsan ,o Cornouaillais a larg, e district in Brittany to the south of Cape Finisterre, running out into a"sharp point. Thu e Cornavith s f Caithneso i t theigo s r name froe hornshapemth d territory which they e otheroccupiedth d sdi s a , above mentioned t whe e NorsemaBu nth . n mad appearancs ehi e th n i e north, instead of giving a brand-new name to the land, he translated the old Gaelic corn into his own ness, a snout or nose, by which he often designates this part, althoug s moshi h t common forKatanesa,s mwa e th Nes Kataf so name th , e tha s stili t l used d thian ,s latte distinguiso t r h it from Gatuv, Catland, which in his hands became Sudrland, now Sutherland, meaning "Southland." CA.BRENI.—These e Strathnavedwelth n i t r countr d awaan y y west towards e wor Th s understoodi . Latinisea e b o dt de th for f mo Gaelic caora, r sheepAlexandeD s a , r Macbai d othean n r well-known authorities acknowledge meand an , s e shee"th p people. rooe s i Th t " also to be found in the Cymric caeriwrch, roebuck, and seems to be common to the Celtic languages. About one thousand years after the time of Ptolemy such a name well suited the people of these rugged parts, for this is how Matthew of Paris describes them in his quaint . c 1250ma f o p, "Eegio, montuos t e nemorosaa , gentem incultam generan t pastoraleme s , propter mariscu t mharuudinetume mounA "( - tainous and woody region, producing a people rude and pastoral, by

herp ema givee 1Th thas n i t prepare Captaiy db n Thoma papes e hi "n r Th o r fo s Ptolemaic ," in the Proceedings, vol. xi. (1875). It does not show the Lugi tribe as it should be shown, but as a place-name Logi ; and Nabasi, at the mouth of the Nauer, should be Nabaei. OCEAN ITS: DCTECALED 0NIUS

X p

OCEAN US GERMAN TC V S

EUROPAE TABULA PRLMA . ALBIONIS INSULAE BRITANNICAE SITUS. 2 8 PROCEEDING E SOCIETYTH F O S , JANUAR , 190813 Y .

reason of marshes and fens). This country, so full of straths and so suited for his favourite sheep, the Norseman dubbed Dalir, the Dales. Niale th n I Saga, translate y Dasente tolb dar d e thaw , t Earl Sigurd of held,'besides Caithness, the following lands in Scotland, viz. " Ros Morayd san , Sutherlan e Dales.th r Skened D "an d , howevern i , s Celtichi Scotland, without givin reasoy gan n whatever, jumpe conth -o st clusion that Dalir is to be found on the west of Scotland; and that it is to be identified wit Dalriade h th Irise th f hao annalists. From plaia n read- ing of the saga we should naturally expect to find the " Dales " in question not on the west coast, but in the immediate neighbourhood of Sutherland, and, further plurae th , l form dalir indicate singla t sno e dal t manyebu . Besides e Orcadesth n i ,f Torfaeuso s translatea , y Popeb d e heaw , r again of the " Dales," but this time they are associated with Caithness. Evidently by the Dalir of the we are to understand the col- lection of straths extending from the borders of Caithness and along the north of Sutherland to Assynt in the west, known at a later period s Strathnavera collectioa o T . f valleyo n s e pasturagsuiteth r fo d f o e shee e Norsemepth n were pron o givt e e suc a nameh , r exampleasfo , , shoree Dalith f Hvamsfirtn so o r n Icelandhi e wesd Dalith f an o ,t n i r —o e latteth f r Rumsdal d Gudbrandsdaan e l for a mpart . Strathnaver, that was the sheep-rearing country of the Caereni about e beginninth e Christiath f o g ne devote b d s saithaera o an t dwa ,t d e pasturtoth e 13tf flockeo th hn i scentury depopulates wa , a larg o t de exten d turnean t d into sheep-walk e earlth y n e i 19tsyear th hf o s century. It may well be called the country of the Caereni at the present day, for too much of it is under sheep and deer. LTJGI.—This tribe inhabited the coast side of Sutherland from about the river Ila at to the neighbourhood of Strathflete in the south, iu other words, wha stils i t l knowe ,th s na lowlan seae th , y db from Gaelic magh, a plain, the most fertile part of the county. An echo of this old name Lugi is still to be found in the name Loth,1 as also By Gaelic speakers Loth is pronounced Logh, with an emphatic y sound— 1 Second Stat. Account, Loth, p. 188. SUTHERLAND AND CAITHNESS IN ANCIENT GEOGRAPHY. 83

in Lotlibeg and Lothmore, flat swampy places that had to he drained about a century ago. The low, fertile stretch of rnaritime land now known as Louth, in , was in ancient times called Lug-mag,^ Lug-Held, showing a change from Lug to Louth not unlike that presented by the forms Lug and Loth in Sutherland. e namwell-knowa th f s o e wa g BunLu t Irish pagan divinity, whose dwellin s undergroundwa g d whosan , s Lugnasad,e wa festa y w da l no Lunasdainn, on the 1st of August,2 corresponding with the Saxon Lammas, originally als a opaga n feasts votivA . e offeringe th f o s first-fruits of the harvest.were rendered at Lugnasad, the god Lug was in somconnectey ewa d wit e probabilith th tillag d an s e; s wa ythai e h t a personificatio e humblth f o n e worm, from Gaelic lug, mud-worm,

showing in Sutherland the adjective form lugaideaeh, long-necked like a worm. Even so the Celtic divinity Ogma, 3 the patron of letters, whence comes ogham writing, derive s namhi s e from Gaelic , lateog , r eag, a notch. With Lugi compare Lugu-baliam, the ancient name of Carlisle, as seen in the Chronicles of the and Scots, p. 298, and also the Gaulish Lug-dunum Lyonsw no , . Both these towns e confluencstanth t a d f o e , and in a fertile neighbourhood. But to follow up : two thousand years ago the Highlanders of Strathnaver were sheepmen and the low- land ers of the Machair were earth-tillers or given to agriculture, a distinction which still holds true. NABARUS.—This rive s i undoubtedlr e Nave th a yLati n i rn dress, showing.an older form Nabar or Navar. In various charters of the 13th century it is written Nauir, Navir, and Navyr. And at the wore presen th s pronouncei dy da t e Gaelith y b cd inhabitants Nauer. thus ha s t undergonI e very little change during close upon 2000 years. When the river has run about three-quarters of its course to the sea the sacred Mo-Naire Locr o , h Mo-]^a(u)ir sometimes i t i s ,a s pronounced,

Jubainville'e D 1 s Irish Mythological Cycle, . 172p . 2 Ibid, pp. 78, 172.

Professor Auwyl's Celtic Beliyion,. 39 . p 8 4 8 PROCEEDING SOCIETYE TH F O S , JANUAR , 190813 Y .

dedicated to some one Nair or Nauir, lies within a few hundred feet of its right bank and empties itself into the stream. The name of the loch and river have come from the same root, and as the loch is the sacred object t probabli , t seldoy gavno e rivercase s s nammi th eit th es o a ,t e in Celtic Gaul. Indeed, we think it likely that the now common form Naire, instead oo folk-etymologft Nauir,e du s i y dating bac o about k middle th t f eo e 17tth h century, whe e housth n f Sutherlano e d plante n aggrana d - dising colony of Gordons in Strathnaver in the teeth of vehement opposition on the part of the native people. Hence the traditional saying that there is no virtue in the waters of this loch for a ; t basebu d upon another tradition equally general tha loce th th owes sit virtu a charme o t e d ston womana e y kepo flunb n wh i ,tg cryino m g naire, mo naire (shame, shame), as she fled with her treasure before a Gordo same robbern th n o t e a timo ye wh bent invoked ,an blessinda n go these ailinwatere th f everr go s fo y name sav er pursuer thahe f o t . Tliis folk-story ,e long-cherishe borth f no d e hatreStrathnaveth f o d r people toward e Gordonssth , account e long-drawn-outh r fo s t form Mo-Naire. r HendersonD , Lecture Celticn i r , University, maintains that Mo-Nair was a Celtic demi-goddessl to be frequently met in Irish mediaeval literature f thiI correct.s i Ifabarun ,i have sw e concealee dth name of a Celtic pagan divinity, to which the people of Strathnaver paid their devotion, suct wasi s t hLoca a , h Mo-Nai e firsth tn o Mondar f o y styled thaol d quartee e an th ,t th dow n i r withio nt e memornth e th f yo present writer. The same rites, and at the same times of the year, were observed at St John's Loch, , and at St TredwelPs Loch, Papa- . Perhaps at the two latter places the pagan cultus of JSTair was grafte Christian do n superstition vera , y common practice. TARVEDUM.—This promontory lies some e distanc th e eas f th o t o t e Naver, accordin e datth ao g t with which Ptolemy supplie . Georgus s e Buchanan surmised that it indicates one of the headlands of Strathnaver, for this is what he says in his History :—" In ea fronte tria attolluntur 1 Since writing the above we find it confirmed hy other acknowledged authorities. SUTHEKLAND AND CAITHNESS IN ANCIENT GEOGRAPHY. 85

promontoria; altissimu n mNaverniai , quod Ptolemse t Orcaes o s sive Tarvedum " (On that face three promontories jut out; the highest is in e Naveth r country, e whicOrcath r Tarvedus o i sh f mPtolemy)o . Captain Thomas vera n i y, importan t "e pape Ptolemaith n ro c Geography of Scotland," which appeared in the Proceedings of the Society of Anti- quaries of Scotland, dated 12th April 1875, is evidently so impressed hy e outstandinth g characte f Capo r e Wrath tha e identifieh t t i wits h Tarvedum, and concludes that Ptolemy made a mistake in locating it e eas th -Naver e twese o th insteat th f o to t . f do But there is no cause to lay such violent hands upon the geographer's fact thin i s alternative s th case r fo , e name Orcas indicates tha t i must t n thao e e tmainlaneastd th th par an f e ,lookeo o b t t r dfo d facing s veri t Orkneyi y probabld An . e that Tarvedu representew no s mi y b d Holborn Head, on the left side of the bay into which the river discharges Thursoitselfe Th . , indeed Norse th s i ,e Thiorsa,1 Bull Eiver, a well-known river name in . But Tarvedum comes from a Celtic root meaning "bull," as seen in Gaelic tarv, Cymric tarw, and Cornish tarow, cognates of Latin taurus, and may be compared with the Gaulish Tarve-dunu f Marcianmo , which means "Bull-fort. e probabilitTh " s yi that the old name of the river was Gaelic Tarva, that the Norsemen translated this int e Norsoth e equivalent Thiorsa, r thifo s seeme b o t s whicy ke e h th Ptolemy give n Tarvedumi s . Eve s thea n y turnee th d Gaelic corn int e Norsoth ed the ness,di yo s chang e Gaelieth c tarv into Norse th e thiorr.probabld An y they practised this syste f mtranslato - ing the native place-names of Scotland into their own language more frequently than is generally supposed. This is a point on which we woul speciay dla t thinlno emphasisko d thae s importancw tit r fo , t ye s ei sufficiently realised.2

1 The town of Thurso is Inver-hiorsa in Gaelic, showing the Norse form more purely. A good illustration of this process may be seen in Loch Long, which means 2 " Ship Loch," from Gaelic long, a ship. Ptolemy evidently indicates this inland arm Longis hi y b fluviia se e oth ostea,f givin e Celtie gth th ct e founnambu h , s it a ed Norsemen translate t dinti o Skipathein ow r Fjordr, Ship .—Johnstone's Nor- weyian Account f Ham'so Expedition,. 76 . p 6 8 PROCEEDING E SOCIETYTH F O S , JANUAR , 190813 Y .

OBKAS.—Ptolemy gives this as an alternative name to Tarvedum Head, which we have just discussed. Diodorus Siculus, who flourished B.C.0 e5 . , says that Britai s triangulani formn i r , like Sicily, terminating in three promontories, viz. Kantion (Kent) and Belerion (Land's End) in the south, and Cape Orkasl in the north. Pinkerton, in his Inquiry, sug- gests that Cap ee " h DunneOrka y ma s t Hea n Caithness-shirei d e th , most northern poin f Britaino t , frontin e Orcades.th g r AlexandeD " r Macbain 2 suggests further that Tarvedum and Orkas indicate Holborn and Dunnet promontoriee th , inteithey n so ha o e whicr th side f th ho e Thurso flows. Thes o view etw e rathe ar s r supplementary than contra- dictory, hut we believe that Pinkerton has hit the truth in equating Orkas with , for it is quite likely that Ptolemy stumbled with his facts when he gave the two names to one promontory. Ther a genera s ei l agreement that Orkas show Gaelid Ol s c ore, piga , hence sea-pi r whalego , cognate with Latin porcus, English pork, etct I . characteristia s i e Celtith f co c language droo st pradicaa l Indo-European exampler fo , as , ,p Sanskrit pitar, father, whil t remainei s pater Latinn i , becomes athair in Gaelic, and undergoes a similar change with regard to p in all the other members of the Celtic group. An exception to this rule, however, the Latin porcus remains porch in Welsh, while it becomes ore in Gaelic according to rule ; and to this extent the form Orkas shows greater affinity to Gaelic than to Welsh, although it may be objected without any proof that Welsh porch is a loan-word. But Orka closels si y connecte meaninn di g wit place-name hth e Orkney, lying

1 As Diodorus was only a compiler and not a traveller, where did he get these facts? Mr Charles Elton tells us that Pytheas, an explorer of the 4th century B.C., described Britain as " of a three-cornered shape, something like the head of a battle- axe" (Origins f English r o Rober D d tan History,; Munr ) 30 o. p say s that Pytheas wae firs th so makt t e these three promontories known (,. 3) . p Owin o distanct g e fro ma referenc e library e hav w meano , n e f verifyino s g these statements by an examination of the literary fragments of Pytheas; but if they are correct, the e place-namnth e Orkas, whic a Gaeli s hi c shal e formw ls a sho, wn i the text above, indicates that a Gadelic-speaking branch of the Celtic race was settled norte centurth h Scotlanf 4t n ho i e th y n B.Cdi . Transactions of Gaelic Society, vol. xviii. 2 SUTHERLAN D CAITHNESDAN ANCIENN SI T GEOGRAPHY7 8 .

directly to the north and not far away. The Irish annalists invariably refe o Orknet r s yInn-isa Ore, Island f Ore o sr simplo , y Ore; whilt ea presene th t momen e Gaelic-speakinth t g peopl f Caithneseo Suther-d san . land designate Orkne ye Pentlan Arcu,th d an d Firth they e knoth s wa Gaol Arcach, Narrows of Arcu. Evidently the Norsemen accepted the Gaeli cislande namth r d calleefo san d them Orte-ey islandsk Or , r ey,fo in Norse means " island " : but the old Gaol Arcach they altered into Petsland Fiordr, because they found Caithness inhabited by the Picts; e oldeth and t r ye -nam e still survive e languag th e people th n i sf o e , although it does not find a place upon our English maps. As Dunnet Head thrust t farthesou s t int e e waterschanneth oth f o l between Caithness and Orkney, which is still called Gaol Arcach, it may have for that reason got its name Orkas. In the map of 1572 given in Bishop Leslie's History, Dunnet Head is marke to-days Mercator'n da i t 159f markes o bu i , p t 2i sma d Quinicnap. The terminal is Gaelic cnap, a lump or mass, which is very descriptive of the lofty bluff headland now bearing a lighthouse on its cold brow. And the initial Quini probably shows Gaelic cumhann, narrow, for the head- land has a somewhat contracted neck. With this compare Cuinighoe, Narrow Creek, on the east side of Ardneskich, Farr, where the creek is true to its name. VIEVEDRUM.—This promontory, the next in order after Cape Orkas, indicates Duncansby e snouth , r noso tf Caithnesso e . Although broad ancomparew d lo e othestretchin o seat s th i t o t ,rt di ou leadin r gfa g northern headlands, especially on the side facing the , and completely dominated by the neighbouring Warth , 400 feet above sea-level, whence the ground slopes down for a considerable distanc e nazeth o e.t While Dunnet Head lifts commandinit s g crown abou 0 fee50 tt abov e seath e , evee easterth n r loftieso n t sidf o e Duncansby never rises abov 0 feetthed 20 an en, lies unde Warte th r h Hill. Stripped of its Latin dress, the old name Virvedrum presents the form For- or Far-fothair, for the v here is a digamma with the force of a well-know s i r d Fa Gaeli ol r e ninitiao Th cr / . worlFo d meaning 88 PROCEEDINGS OF THE SOCIETY, JANUARY 13, 1908.

"high "or "projecting," and the terminal is the Gaelic compound fo, under d tire,an , land, meaning "underland." Fothe n thii r s senss i e quite a common place-name, as in Foderty contrasted to Achterueed (which latter overlooks it and shows Gaelic uachdar, over), Stiath-fother, Inver-fother (Dingwall), Fetter-cnim, etc. Fother is compounded after e samth e fashio s Lettera n , which consist f Gaelio s c Lethd tire,an meaning "half-slope," as applied to the side of a glen. Thus Virve- drum means " the projecting underland," and should be compared with e loftth y tongu f lano e d called Farai n Durnessi d a softene, d forf mo Far-aird, " the projecting height." VERUBHJM.—This headland is most projbably what the Gaelic-speaking people of the north call the Ceann Dearg, Bed Head, of Stroma; and it e Orkneyiuga referres i th n i o dt Saga s Baudabiora g (Bed-cliffe th n i ) Pentland whicf of , e Earlhth s Thorfi Rognvald nan fierca d edha sea-fight about the year 1040. Sailing eastward along the north coast of Caithness the voyager beholds its lofty red head from afar, and for shaping a course through the treacherous narrows of the Pentland it is an indispensable landmark. Suc a hprominen t featur s sur o impreswa t e e earlth s y foreign mariners, who, greatly daring, ventured through these northern seas, and they were almost certain to make a note of it for future reference Gaelie Th .c rub, red, aki o Latit n n ruber e d rufus,th an s i roo f Verubiumo t e initiath e d Gaelib l an ;lim y c bma far, although Ptolemy does not show the double r. If our reading is correct, then it means " the projecting red," a name which exactly describes the Ceann Dearg. The Gaelic form of the principal root may be seen in e Iristh h saint name Mael-Rubha, e Tonsureth d Bed, which became Latinised into St Bufus, and even takes the form Bice. Of course, the colourocke th f so r suggeste name thougd th Headd eBe An .h Stroma is an island, yet it lies so close to the shore as to be practically a part of Caithness, which parochiall t indeeyi . dis It is but right to state, however, that Captain Thomas identifies Verubium with , near Wick, and that he bases his opinion upon the given Ptolemaic position. But too much importance should SUTHERLAN CAITHNESD DAN ANCIENN SI T GEOGRAPHY9 8 .

noattachee b t thiso dt evidentlr fo , geographee yth payins i r g particular attention to the features in the immediate neighbourhood of the dangerous Pentland, while fro Kyle m th f Lochalseo e Naveth e o h t hr ignores everything, even . ILA.—This river is the Helmsdale, in Gaelic Hi; and Helmsdale itsel s Bun-ilii f . s Ili pronouncei e Gaelis alwaydth ha Eler ci fo s, e e th soundthin i sd cas s shoran a mer,initiae a es th s a i te i l breath. n thiI s respec wore th t d differs from lie (the Gaeli ce rive th for f rmo e Tsl islann Perthshirth i a f f )o do d ,an e initiaewhic th s lha h vowel lond seemgan o comt s e fro a mdifferen t root, although some scholars, ignoring this difference, persis n derivinformi to tw se gfroth m the same source. Various derivations have been suggested for I la, from Iberian to German and Greek, but at best they seem to us but learned guesses. We venture to suggest that Ili may be connected with Gaelic li,a d thae initiaan th t, / lshow e articl th s, whosin e t e go )/, swallowed up in the following lingual according to a well-known rule. Thus rea t meani d s e flood."th " Compare with thie riveth s r Fleet, Gaelic Floid, which falls into , about twenty miles to the south of the Ili. Like the Fleet River and Fleet Street, the Sutherland Floi derives di d from Norse fljot, flood name importann a th , f eo t stream t happeni o Icelandn s i d s An thaneighbouo . tw t r river-namee th n o s east side of Sutherland may be exactly similar in meaning, but the one proclaim sGaelie itselNorse th othee th th n i c fn ed i rHi Fljot.an ALTA EIPA.—In the 1478 Latin version of Ptolemy these two words a litera e ar l translatio e originath f o Greeo ntw l k worde s th use y b d author mead e "Higan , nth h Bank f som"o e rive r firtho r . Froe mth data given High Bank lies one-thir e distancth f o d e betweed an a H n Varar (Beauly Firth), and is evidently the Ekkialsbakki, Ekkial-Bank, Norse oth f e saga , knoww whicno s Oykells a i nh , between Sutherland and Boss. The place-names Oykell; Ochill ; Glen Ogil; Ucal, near Loch ; Loch Oich, the highest of the chain of lochs forming the l derive, al etc. e ar , d fro e sammth e source cognata , f eo Gaulish uxellos, high, Welsh uchel, high, whence some maintain that 90 PROCEEDINGS OF THE SOCIETY, JANUARY 13, 1908. these forms show suc a hpronounce d affinit o Gallo-Cymrit y o t s a c indicate that Pictish belonged rather to the Cymric than to the Gaelic side of Celtic. But the radicals of the above given five forms are Oyk, Och, Og, Uc, Oich, and are related to Gaelic uchd, breast, whence uchdach, height uchdal,d an , pertaininp to heighta , o g ug t als d o ot an ; of breast, with Genitive cnaimh-uige,n uigei s a collar-bone, showin gstea m sufficiently close to Oyk or Oich. Thus Oykell shows every whit as much affinit Gaelio t y t doei s Gaulisco a st Cymricr h o whil d An e. Ptolemn yi this case took the Pictish Oykell and translated it.back into his own Greek, the Norseman took it as he found it, and stuck bakki (bank) to it. so that in the latter's hands it became High Bank or High Height.

From this cursory examinatio n namete e sth wit f o nh d whicol e th h geographer provides us, it may be concluded that in the days of Ptolemy Caithness and Sutherland were inhabited by a people speaking a dialect in close relatio e Gaelith o t nc language o intimatels d an ; y connected were they with Ireland at this date that the Irish Mo-Nair was invoked in Strathnaver, while the people of the east of Sutherland suruamed themselves by Lug,1 a name still preserved in our Lunasdainn festival, and perhaps also in Ben Lughal ( on the map), near Tongue t lonNo g . after Ptolemy's tim e inhabitantth e e nortth . hof s were know ancieno nt t writer s Pictsa s d Pictan , s they continuee b o dt when the Norsemen turned up, as witness the form Pentlands Firth, a name given by the newcomers to the channel between Orkney and Caithness Iriso T h. writers they were also know s Cruithnena whicf o , h form there still exist traces in the far north. At the south end of the townshi e f Farreas po clos d th e roadt an y ,th esidb f ,o ether e standsa circular structure, the remains of'a perhaps, called An rath Ghruifh- •neach, the circle of the Cruithne. The nam'e is also found in Assynt e immediat. Th 1 e neighbour e Lugi'tth f o e souts th oe calle ar h d Smertay b e Ptolemy ; and Smert was a Celtic goddess, whose name is often met on Gaulish votive tablets e ProfSe .. Anwyl's pape n "Ancieno r t Celtic Goddessese Th n i " Celtic Review, iii. pp. 26 seq. SUTHERLAND AND CAITHNESS IN ANCIENT GEOGRAPHY. 91

attached to some similar object, and probably in not a few other corners unknown to the present writer. In a number of tracts, etc., collected e Chroniclesth n i r Picts e Skeneod ScotsD th f an y e Pict b th , saie sar d havo t e settle Caithnesn di n earla t ya s date o havt , e intermarried with Irish, women and not with British, and to have spoken the language of their wives. This is how the matter is put in Layamon's Brut:— A certain king "gave them [the Picts hann i ] da grea t dea landf o ll al , about Caithness, where they made homes. . . . They took their messengers and sent to Ireland to the king of the land " for wives, whom he gave to them. " Through the same women, who there dwelt long, the folk began to use Ireland's speech : and ever since the usages they do in the land"—this last clause is obscure in translation. The document from which we quote is supposed to have been compiled c. 120 t 0thaa A.D, t t bu dat, e evidently r awafa d ye the bacan ,th nn ki hoary past e languag th ,e Pictisth f o eh native f Caithnesso s e mosth , t Pictish part of Scotland, because the most remote, was Gaelic or Ireland's speech d thi s jusi An s .t wha woule w t d expec e genera th fino t tf i d l opinion of scholars be correct, that the Gaelic branch of the reached Britain first, and that the Brythons, when they came afterwards, pushed the before them northwards. Though Gaelic is not now spoken by many in Caithness, it should be borne in mind that at the Revolution Settlement of 1689 Gaelic had to be preached in all the pulpits of that presbytery except Dunnet and Canisby, as the ecclesi- astical records show. It is not because Caithness was more Norse than Sutherland that it is now more an English-speaking part, but because the Sinclairs, Oliphants, etc., who settled in Caithness during the 15th century, brought along with them south country follower helpeo wh s d o spreat e Englisth d h e tonguneighbourhooth n i e f theio d r feudal superiors' seats. The Norsemen subdued the people of Caithness, but their language was not destroyed, and as soon as the foreigner took his departure the old language ruled it again in all these borders, nor is it dead yet. Another fact worth x Ptolemai si f noticyo e th s thai e f co t stations, on the coast line between the and the Firth, 2 9 PROCEEDING E SOCIETYTH F O S , JANUAR , 190813 Y .

four are headlands in the surging Penfcland Firth—to wit, Tarvedum, Orkas, Virvedrum Verubiumd an , , wit e riverth h s Nabaru d Ilo t aan s the west and south respectively. Evidently the men who provided the geographer witwholesom a s fact d hhi ha s e e neesens th f minutel df o eo y charting these wil d o makt waters s a e o thes , m navigabl e sailorth o et s who came thithe o tradet r . Dilettante explorers would roughly record outstanding features, but these men are so practical in what they note tha t look i tf ther i s sa e were some trading connection betwee norte nth h of Scotlan Mediterraneae th d dan time th Ptolemyf eo n i populae Th . r conception of the Northern Picts of the 2nd century is that they were t rudbu e savages, naked a coa sav f r paintt o thitfo e Bu utterl s si . y false, as the remains found in their fortified brochs or duns abundantly testify. The quernd yha d spindles san e formeth , o grint r d their corn and the latter to spin the wool of their flocks. They were quite in a position to export wool, skins, tallow, hides, etc., and we believe they did so, for in the 4th century B.C. the merchants of Marseilles thought it worth while to dispatch the astronomer Pytheas on an exploring expedition to Britain with a view to promote trade with these parts, and e exploreth r nortr camfa .s s ha ea 1

singulas i t I r that Ptolemy take knowledgo n s Cattie th f ,o e whence came Catuv (Sutherland d Katanessan ) 2 (Caithness). Accordino t g mediaeval Gaelic legend Pictish Scotlan s dividewa d d into seven provinces, of which the most northern was governed by Cat, son of Cruithne sais i o havdt o ewh , give tribe nams d territorth nhi e an o et y over which he ruled. In the Bodleian map of c. 1250, however, the Strathnave d Assynan r t countr s i drawy n very mountainousd an ,

1 Elton'a Origins of English History, . 13-40pp . 2 Cleasby, of Norse dictionary fame, derives Kataness from Norse kati, a small ship, and ness ; but the Norse derivation of the first limb is most improbable in view of the frequent reference in mediaeval Gaelic writings to the land of Gait. Feliree th n I (Calendar f Anguo ) e Culdeth s e (9th century) t DonnaS , s saini o dt have been commemorated " ii Cattaib," in Catland; and in the Irish additions to Historiae th Sritonum extremitiee th f Pictlano s describee dar 0 chric" s da h Chat co Foirciu," i.e. from the bounds of Cat to Forchu. Like Ore, evidently this part e countroth f knows t befory wa Ca e adven s th ne Norsemena th f A.Do t0 78 . .c , SUTHERLAN CAITHNESD DAN N ANCIENI S T GEOGRAPHY3 9 .

standing among the hills a cat-like figure is shown, with the legend "Hie habundabant lupi" (Here wolves abound). Gordo f Gordono n - stoun tell s thae u countrs th t s infestewa y d with wild cat n earli s y times, whenc e moderth e n place-name Cutuv, wit e namth h e Galtich (cats) for the inhabitants.1 But the probability is that there was a tribe of Catti, which include e Cornaviith d , Caereni Lugid an , , dwellina n gi e countrparth f o t y wher e wolwild-cad eth an f t found abundant cover; s folat a s r142 a e barone 7th f Scotlanso d were callet Ac d n upoa y nb of Parliament to extirpate all wolves found upon their lands. And besides this, not without reason did the Norsemen call the north-west shouldee countrth f o ry , " Wild-beast ness," from Norse d ness.dyran e e onlBodleiath th y p n I tow e ma northern th n l showal n n i n provinc s Wyki a espire s eha d (Wick)t buildini d an ,a castl ( g r o e church); but in that of Mathew of Paris no town at all is shown. In Mercator's map a castle is shown on the west side of the , which very probably represents Castle Varrich d whican , h was then likely inhabited, for in this map no reference is made to Borve Castle printes , wa destroye p d ima n e 1592n 1554 i dth castle s th , A .e must have been occupied up to that date at least. Bishop Leslie shows but one island off the Kyle of Tongue in the map which he gives, and this islan e callh d s Ylen Marty t nMercato ;bu r show threee d th s an , n thii s order froe westn Martinth m He , , Shyp Hand p HandHy , .

t shoulI pointee b d t thaGaelid ou dOl t c cat, Modern Gaelic cath, Welsh cad, 1 and Old Gaulish mtu-s, all mean "war"; hence the Catti (warriors), a tribe of Britons in the neighbourhood of Gloucestershire, and the Catu-riges (battle-kings), a tribe-name of ancient Gaul. Vide. Prof. Rhys' Celtic, Britain. Also from the same root comes Sucat, the original or boy name of St Patric the Briton, patron saint of Ireland compouna , f Su,o d good d cat,an , freele b whic y y hma translate d inte th o Scots vernacular "bonnie fechter." Vide Prof. Zimmer's Celtic Church, p. 38. r northerou f I n name, however d beeha , n originally derived frod GaelimOl c cat, warpresene th , t form would e rathesofteneth e rb d Cathaich, t insteabu thaf o dt i t is the hard form Cattaich, meaning " pertaining to cats" ; and the Duke of Suther- lan always di s designate n Gaelii d c Dine Cat, Duk f Catseo . Perhap totee sth f mo the tribe was the wild-cat. 94 PROCEEDINGS OF THE SOCIETY, JANUARY 13, 1908.

This is a curious mixture of names, not one of which is ever given to these islands in the papers, or in the conversation of the people — documeny an no n i , t notye r tha e havw t e e charteseen Th d stil. an rl used names of the three islands given in the same order are Island n Gail.na f strangers)(o , Rona, d Islan an n Naomh na d f saints(o r o ) Colme (Columba). Evidently Marti s misquotei n r Columbafo d o wh , had a dedication on Colme; but St Martin also had a dedication on the strath of the Naver, as the well-known holy well, Tobair Glaish Mhartain, between Ceancoille and Cnubeg, clearly shows. Bread specially prepared, accordin e followinth o t g g Gaelic recipe, promoted health when taken I'along wit e wateth h f thio r s well:—

'' Aran air f huine le connadh, Is burn a Tobair Claish Mhartain." Bread baked on brushwood, And water from the well of Martin in the Dell.