[2. Place-Names and Settlements in Northwestern England.]
Cumberland and Westmoreland, together covering an area equal to about two-thirds that of Yorkshire, have 300 Scandinavian place-names. Yorkshire has 407 according to Worsaae's table. The character of these names in Cumberland and Westmoreland is different from that of those in the rest of England. It seems that these counties were settled predominantly by Norsemen and also perhaps at a later date than that which we accept for the settlements in York and Lincolnshire. We know that as early as 795 Norse vikings began their visits to Ireland; that they settled and occupied the Western Isles about that time; that in 825 the Faroes were first colonized by Norsemen, partly from the Isles. After 870 Iceland was settled by Norsemen from Norway, but in part also from the Western Isles and Ireland. The 'Austmen' in Ireland, especially Dublin, seem frequently to have visited the opposite shore. It seems probable that Northwestern England was settled chiefly by Norsemen from Ireland, Man, and the Isles on the west. It is not likely that any settlements took place before 900. It seems more probable that they belong rather to the second quarter of the 10th Century or even later, when the Irish began successfully to assert themselves against the Norse kings in Dublin and Waterford. Perhaps some may have taken place even as late as the end of the 10th Century.
[3. Scandinavian Settlements in Southern Scotland.]
In Southern Scotland, Dumfriesshire, Eastern Kircudbright and Western Roxburgh seem to have formed the center of Scandinavian settlements; so, at any rate, the larger number of place-names would indicate. The dialect spoken here is in many respects very similar to that of Northwestern England, D. 31 in Ellis, and the general character of the place-names is the same. These are, however, far fewer than in Northwestern England. Worsaae gives a list of about 30. This list is not exhaustive. From additional sources, rather incomplete, I have been able to add about 80 more Scandinavian place-names that occur in Southern Scotland, most of them of the same general character as those in Northwestern England. Among them: Applegarth, Cogarth, Auldgirth, Hartsgarth, Dalsgairth, Tundergarth, Stonegarthside, Helbeck, Thornythwaite, Twathwaite, Robiethwaite, Murraythwaite, Lockerby, Alby, Denbie, Middlebie, Dunnabie, Wysebie, Perceby, Newby, Milby, Warmanbie, Sorbie, Canoby, Begbie, Sterby, Crosby, Bushby, Magby, Pockby, Humbie, Begbie, Dinlaybyre, Maybole, Carnbo, Gateside, Glenholm, Broomholm, Twynholm, Yetholm, Smailholm, Langholm, Cogar, Prestwick, Fenwick, Howgate, Bowland, Arbigland, Berwick, Southwick, Corstorphine, Rowantree, Eggerness, Southerness, Boness, etc. There are in all about 110 such place-names, with a number of others that may be either English or Scandinavian. The number of Scandinavian elements in Southern Scotch is, however, very great and indicates larger settlements than can be inferred from place-names alone. In the case of early settlements these will generally represent fairly well the extent of settlement. But where they have taken place comparatively late, or where they have been of a more peaceful nature, the number of new names of places that result from them may not at all indicate their extent. The Scandinavians that settled in Southern Scotland probably at no time exceeded in number the native population. The place-names would then for the most part remain unchanged. The loanwords found in Southern Scotch and the names of places resemble those of Northwestern England. The same Northern race that located in Cumberland and Westmoreland also located in Scotland. It is probable, as Worsaae believed, that it is a second migration, chiefly from Cumberland. Dumfriesshire, at any rate, may have been settled in this way. The settlers of Kircudbright and Wigtown were probably largely from the Isles on the west. Other independent settlements were made in Lothian and the region about the Forth. That these are all later than those of Cumberland and Westmoreland is probable. According to what has been said above, the settlements in Dumfries, which seem to have been the earliest, could not have taken place before about the second quarter of the 10th Century, and probably were made later. The other settlements in Southern Scotland may extend even into the 11th Century. The name Dingwall (O. N. Ðingvöllr) in Dumfries, the place where the laws were announced annually, indicates a rather extensive settlement in Dumfries, and the dialect of Dumfries is also characterized by a larger number of Scandinavian elements than the rest of the Southern counties.
[4. Settlements in England, Norse or Danish? The Place-Name Test.]
That the Danes were more numerous than the Norsemen in Central and Eastern England from Northumberland down to the Thames there can be no doubt. The distinctive Norse names fell, tarn and force do not occur at all, while thorpe and toft, which are as distinctively Danish, are confined almost exclusively to this section. In Northumberland, Durham, Cumberland, Westmoreland and Lancashire thorpe is comparatively rare, while toft is not found at all. On the other hand, fell, dale, force, haugh, and tarn (O. N. fjall, dalr, foss and fors, haugr, tjörn) occur in large numbers in Northwestern England. Beck may be either Danish or Norse, occurs, however, chiefly in the North. Thwaite Worsaae regarded as Danish "because it occurs generally along with the Danish by." We find, however, that this is not exactly the case. In Lincolnshire there are 212 by's, in Leicestershire 66, in Northampton 26; thwaite does not occur at all. In Yorkshire there are 167 names in by and only 8 in thwaite, and 6 of these are in West Riding. It is only in Cumberland and Westmoreland that the proportions are nearly the same, but on by see below § 5. Tveit is far more common in Norway than tved in Denmark. The form of the word in place-names in England is, furthermore, more Norse than Danish. In the earliest Scandinavian settlements in England, those of Lincolnshire, for instance, thwaite might be Danish if it occurred, for monophthongation of æi to e did not take place in Danish before about the end of the 9th Century; by about 900 this was complete (see § 6). The Scandinavian settlements in Northwestern England, however, did not take place so early, consequently if these names were Danish and not Norse we should expect to find thwet, or thweet (tweet), in place of thwaite. It is then to be regarded as Norse and not Danish. Thwaite occurs almost exclusively in Northwestern England—43 times in Cumberland as against 3 in the rest of England south of Yorkshire. Garth (O. N. garðr, O. Dan. gardh, later gaard), occurs very often in Cumberland. With, ness, holm, land, and how, do not occur very often. How reminds one of the Jutish höw in Modern Danish dialect. The rest of these may be either Danish or Norse. In Yorkshire we find a mixed condition of affairs. East Riding, as we should expect, has predominantly Danish names. Thorpe, which occurs 63 times in Lincolnshire, is found 48 times in East Riding. Fell, tarn and haugh do not occur. Force is found twice, and thwaite once. Dale, however, occurs 12 times. West Riding was probably settled by Danes from the East and by Norsemen from the West. Thorpe occurs 29 times, with 8, toft 2, beck 4, fell 15, thwaite 6, dale 12, and tarn 2. In North Riding thorpe occurs 18 times. Force, fell, and tarn together 12. The large number of names in dale in North Riding is rather striking (40 in all), as compared with 52 for Westmoreland and Cumberland. While dale is predominantly Norse, it may perfectly well be Danish, and it is not rare in Denmark. Furthermore, the greater number of dales in Norway as compared with Denmark is largely accounted for by the nature of the country. No conclusions can be drawn from names in force in Yorkshire, Cumberland and Westmoreland, as it is of too infrequent occurrence. Fell occurs 22 times in York, as against 57 in Cumberland and Westmoreland (42 in Westmoreland alone), but in York occurs predominantly in West Riding, where everything points to a mixed settlement. The distribution of tarn is interesting. Tarn is as distinctively Norse as thorpe is Danish. It occurs 24 times in Cumberland and Westmoreland, 3 in North Riding, and is not found at all south of Westmoreland and York.
[5. By in Place-Names. Conclusions as to this Test.]
By has been regarded as a sign of Danish settlement for the following reasons: (1) O. N. bör would have given bo. The O. Dan. form býr becomes by. (2) By is peculiar to Denmark, rare in Norway. (3) Bö or bo is the form found in Insular Scotland, in the Faroes and other Norse settlements. First, the form bỳr is not exclusively O. Dan. It occurs several times in Old Norse sagas in the form býr and bý—in "Flateyarbók," III., 290, in "Fagrskinna" 41, several times in the "Heimskringla," as well as elsewhere. Again, J. Vibe (see Nordisk Tidskrift, 1884, 535, and Norsk Historisk Tidskrift, 2 Række, 5 Bind), has shown that by is not peculiar to Denmark and rare in Norway. It occurs 600-700 times in Denmark and Skåne, and 450 times in Norway. Finally, by is often found in Norse settlements in Scotland and elsewhere—in Iceland, Shetland, Orkney, Man, and in the Western Isles. In fact, by seems to be the more common form outside of Iceland. All we can say then is that by is more Danish than Norse, but may also be Norse. Where names in by are numerous it indicates that the settlements are rather Danish, but they may also be Norse. We have, then, the following results: Predominantly Danish settlements: Essex, Bedford, Buckingham, Suffolk, Norfolk, Northampton, Leicester, Rutland, Lincoln, Nottingham, Derby, East Riding. Mixed Norse and Danish settlements: North Riding, West Riding, Durham, part of Cheshire, and Southern Lancashire. Norse settlements: Cumberland, Westmoreland, North Lancashire, part of Cheshire, and parts of Northumberland. The number of Scandinavian place-names in Northumberland is not large, only 22 in Worsaae's list. North of the Cheviot Hills the names are again predominantly Norse.
[6. Characteristics of Old Northern, or Old Scandinavian. Early Dialectal Differentiations.]
On the characteristics of primitive Northern and the changes that had taken place in the language before the Viking period, see Noreen, P. G.2 I, 521-526. On pp. 523-526 are summarized the characteristics of General Northern. Until 800 the Northern tongue was unitary throughout the Scandinavian North. In the Viking age dialectal differentiations began to appear, especially in O. Dan. These are as follows (from Noreen):