Ambiguous meanings.

one must take into account that tendency to magnify the importance of localities and individuals which is so common in all rural districts. All nomenclature is comparative, and when the field of comparison is limited, undue value is bestowed upon degrees of excellence which would be scarcely perceptible in a wider field.

The unconscious pride which, among Celtic tribes, exalted the chief into a righ, or king, may be traced in other terms of Celtic speech. This righ, for example, would naturally choose the best spot for his dwelling, and in our latitude the best spot is that which receives most sunshine. Hence grianán (greenan), a sunny place, from grian (green), the sun, is described by O'Brien as a royal seat or palace" and this," says Dr Joyce, " is unquestionably its meaning when it occurs in topographical names." But, in truth, it often has a much humbler origin; and Greenan in Ayrshire and Bute, Grennan, Argrennan, and Bargrennan in Galloway and Dumfriesshire, though perhaps commemorative of a chief's abode, may also bear the interpretation assigned to grianán in modern Gaelic dictionaries-a dryingplace for anything, particularly peats.

Furthermore, there is the difficulty arising from ambiguity. Many meanings are often attached to the same word either simultaneously or by successive generations. The syllable “ark” is a very frequent suffix in place-names, and no doubt it often represents the Gaelic word earc; but even when that

origin has been arrived at, one is still left in doubt as to the real meaning, for in O'Reilly's Irish dictionary that word is interpreted—“ water; the sun; any beast of the cow kind; a salmon; a bee; honey; a tax; heaven; a rainbow; red; speckled.”

seem.

More than this, even of those names which admit Names not always of intelligible explanation, many must be rendered what they as if followed by a note of interrogation in brackets. I can best illustrate this by an example from Irish topography. There is a townland near Ennis called Clonroad, and no objection could have been taken to explaining it as cluain ród, the meadow by the roadside, for that is precisely the form which those words would assume in composition. But it so happens that, in the Annals, Ennis is usually called Inis cluana-ramhfhoda-that is, the inch or pasture of the meadow of the long rowing. Here the original name has been divided between two places, Ennis representing inis, the pasture, and Clonroad the cluan ramhfhoda, the meadow of the long rowing or boatIn this compound ramhfhoda, the m and ƒ are silenced by so-called aspiration, and the result is the sound “roada.”

race.

There is no key provided to the analysis of Scottish place-names as there is in Ireland by a plentiful early literature, so it is well to bear in mind this example of the necessity for rejecting a simple and obvious explanation for a complicated and obscure one. But it would be unpardonable to take this course except upon clear documentary evidence.

It may, perhaps, be thought that I have devoted too much time to pointing out errors and dwelling on difficulties; but one of the first tasks to be undertaken by the student of place-names is the detection and demolition of fictitious etymologies: one of the last lessons he can hope to convey is that where no certain evidence-documentary, oral, or physicalcan be had as to the origin of a name, the only right thing to do is to leave it unexplained.

27

LECTURE II.

THE LANGUAGES OF SCOTTISH PLACE-NAMES.

TRACES OF PRE-CELTIC SPEECH-THE IVERIAN OR SILURIAN RACE-THE FIRBOLG OF THE IRISH ANNALISTS-THE ERNAITHE TWO MAIN BRANCHES OF CELTIC SPEECH-OBSOLETE WORDS-THE OPERATION OF UMLAUT-LINGUISTIC CHANGEEFFECTS OF ASPIRATION AND ECLIPSE-DIFFERENCE BETWEEN GAELIC AND WELSH-Q CELTS AND P CELTS-TEST WORDSSIMILARITY OF GAELIC AND WELSH-GHOST-NAMES.

JAVING dwelt in the first lecture on the general principles to be observed in the study of place-names, and pointed out some of the chief snares to be guarded against in the endeavour to read their true meaning, attention may now be given to the different languages in which such names are found in Scotland.

Leaving out of account those framed in modern English or that form of Old Northern English which survives in Broad Scots, which generally explain themselves, the rest may be assumed to have been conferred by people speaking one of the following languages or dialects :

2. Celtic, either Brythonic, Cymric, or Welsh.
Pictish.

3. Old Norse.

4. Anglo-Saxon.

Besides these there are a few, but very few, names altered from the Latin of the Roman conquerors. Considering that the Roman occupation of Southern Scotland lasted for more than three centuries, it may be matter for wonder that they failed to impress their language upon the nomenclature of that country, especially when the extent to which the Norsemen have done so is taken into account. But the fact is that, although Latin was the official language of the Romans, the legions were latterly recruited mainly from nations whose speech was not Latin. The Second and Sixth Legions, which remained longest in the northern province, were drawn principally from Gaul and Spain; hence almost the only names which commemorate them are military technical terms, such as castrum, a camp, which occurs as Chester and Chesters in the counties of Dumfries, Dumbarton, Roxburgh, Berwick, Mid and East Lothian, and Fife.

Christian missionaries, of course, introduced a number of Latin ecclesiastical terms, which became part of the Gaelic or Welsh languages, such as Gaelic eaglais, Welsh eglwys, from ecclesia, a church, which gives the name to Eccles, near Coldstream,