Page images



UNDOUBTEDLY, Roseberry Topping is the most striking natural feature that Cleveland possesses. Standing like a circular fort at the north-western angle of the Cleveland Hills, and apparently towering above the neighbouring table-land, although, in reality, a few feet lower than the highest point on the Guisborough Moor, it cannot fail to have early impressed itself upon the imaginations of the inhabitants. It is sad to think that recent mining operations have shattered its frame, and that in a few years its glory will have departed. From time to time attempts have been made to delve out as much of its history as is hidden in its name.

The name appears in many forms in ancient documents. Canon Atkinson gives the following list :-" Otneberch, Ohtnebercg, Othenbruche, Othenbrugh, Othenesbergh, Ornbach, Onesbergh, Ohtneberg, Ounsberry, Theuerbrugh, Theuerbrught, Hensberg, and Hogtenburg." The list might be considerably increased, but many of these forms are obviously corrupt, and the true readings not very difficult to replace.

It is interesting to note that the form Roseberry does not occur until the seventeenth century, and this fact adds to the difficulties.

Graves' Cleveland, page 213n, gives us an early etymology which, though impossible to accept, may serve as an introduction. "Rosebury is a word of British origin, denoting a fortified hill; and was probably used as an exploratory station; it was so named of the British ross, a heath or common; and the Saxon bury or berg, a castrum or fortress. Oonsbury is a word of similar import, from oon, which signifies a hill. The word topping, which is frequently annexed, is evidently Danish, from toppen, an apex or point, as descriptive of the peaked summit of the mountain, which is now used as a beacon to give an alarm to the country in times of public danger. Othenburgh, the name which sometimes occurs in ancient records, is conjectured to be derived from Othan or Odin, the same as Woden, which signifies fire, and by our Saxon ancestors esteemed 1 Atkinson, Cleveland, i, 95.

and honoured for their God of battle as the Romans did their god, Mars. But this conjecture will be considered by some etymologists, perhaps, as more ingenious than conclusive."

On the other hand, it is possible that other etymologists may consider the conjecture in the last paragraph less inconclusive than the authoritative statements in the earlier paragraphs, although these are generally followed by writers of popular local histories.

Let us next turn to a historian of Cleveland, whose word is entitled to more weight, and from whom we cannot differ with a light heart.

At page 96 of his Cleveland, Ancient and Modern, Canon Atkinson shows a decided leaning towards the view taken by Mr. Haigh (Anglo-Saxon Sagas, p. 45), namely, that Roseberry is the Hreosnabeorh of the poem of Beowulf. In particular he calls attention to the fact that the meaning of the AngloSaxon adjective, hreose, is rushing on, violently approaching, and, therefore, that in point of signification "Hreose beorh," or "Hreoses beorh," would be an exact equivalent of Odinberg or Othenesberg, meaning equally the "hill of the rager, or rusher."

Canon Atkinson suggests the probability of the notion that the name Othenberg or Othenesberg was simply a translation of an old, and Anglian (or English), name, and that the present form is merely a reassertion by that older English name of its ancient rights over its mediaval Danish supplanter.

At page 166n of his Whitby Cartulary (Surtees Society, vol. Ixix) he goes a little further. Finding it difficult to account for the change from Othenesbergh, Oonsbergh, etc., to Roseberry on any supposition, he thinks that not the least probable may be that there is in the latter the popularly preserved recollection of the old Anglian name.

The idea that Cleveland is the scene of one of the oldest English poems is no doubt peculiarly fascinating as well as flattering to our local pride. I find it, however, equally difficult to believe that an old Anglian name like Hreosnabeorh, as that a Celtic name like Ros, could have lain hidden for six, seven, or more hundred years, no trace of it during that period ever having come to light, and then suddenly be reproduced, in the seventeenth century, by a people to whom Celtic, Anglian, and Danish were alike sealed languages. It is, therefore, unnecessary to consider Mr. Haigh's theory, which is not followed by modern

commentators on Beowulf. Nor, with great respect to the learned Canon, do I find the transition from Othenesburgh to Roseberry quite so inexplicable as he does. On the contrary, I venture to think that by a close attention to date the transition is fairly apparent.

In arranging any series of names found in ancient documents in order of date, it is true that as most of the documents are known to us by means of copies only, we can never be sure whether the form of the name is that existing at the date of the original or of the copy, or what the copyist believed to be the form at the date of the original. On the other hand, unless we assume one or the other, we can have no theory to work upon, and I shall assume that the deed is correctly copied unless the contrary appears.

Now, as a rule, a natural feature such as a mountain or a river is only mentioned in legal documents in connection with boundaries, in other words, is very rarely mentioned. Fortunately for us, the nearest village to Roseberry contains so common a name that it is known either as Newton-underRoseberry or Newton-in-Cleveland, whenever there is the slightest possibility of a misunderstanding otherwise occurring. The full phrase is not used in the common speech of the country; to those living near it, it is Newton, and Newton only.

This is the first record that we have of it; in Domesday it is called Neueton, and we can only identify it and distinguish it from the many other places of the same name by its collocation and by tracing its history. Nor have I been fortunate enough to find the fuller name in any of the Pipe Rolls.

The earliest notice of our subject appears to be contained at pages 2 and 4 of the Guisborough Cartulary (Surtees Society, vol. lxxxvi), and is there attributed to the year 1119. These are the two foundation charters of the Priory. In what is probably the earlier, certainly the less liberal, the form is Othenesberg; in the other it is Ohensberg; the omission of a "t" by the copyist would explain the difference, but the difficulties do not end here. A contemporaneous form of the same date, if we can trust the recital in a fine of 1239, was Outhenesberg.

Possibly the most convenient form in which to consider the gradual change of name is in the list on pages 47 and 48. I have arranged the form of the name in three columns for reasons which will appear later.

The date in the first column is obviously approximate only; but I venture to think is never fifty years out, and a closer approximation is immaterial.

With regard to abbreviations, Guis. Cart. stands for Guisborough Cartulary, Whit. Cart. that of Whitby, Riev. Cart. that of Rievaulx, all published by the Surtees Society, whose other volumes are referred to as S.S. N.R. Records are the first series of the North Riding Records. The others speak for themselves.

The collocation of these names in order of date leads one to the view that the earliest form was Othensberg (O.N., Odin). The omission of the genitive termination seems to have been so general at all times in Yorkshire that Othenberg is practically contemporaneous. Hensberg, which Canon Atkinson gives as the form in the Guisborough foundation charter, is not found in Mr. William Brown's edition of the Guisborough Cartulary, and is probably due to the learned Canon having followed Dugdale's misreading in his Mon. Angl.1, without referring to the original. Theuerberght, the form that appears in Kirkby's Inquest, is clearly corrupt; it was, no doubt, copied from rough notes taken at the time, and if we assume that an initial o had been made so small (not uncommon at that period) as to be lost in the termination of the preceding word, and that a long s had been mistaken for a long r, as there are very few words where we can distinguish u and n, we eventually get Othenesberght, or only a supernumerary t.

The alternative form, Onesberg (Guis. Cart., i, 168), in the year 1231 does seem a little premature, probably the compiler of the Cartulary having first correctly copied the form in the Charter (Utheneberg), when the word occurred a second time put the form then coming into use. If so, the Cartulary would. be written in the fourteenth and not the thirteenth century.

Apart from these difficulties, there seems to be a fairly uniform list of practically similar names up to the end of the thirteenth century. The vowel sounds are not absolutely identical, but they do not vary, with one exception, from o, ou, and u. The one exception, in 1288, is Hoyphensberg, and in addition to the unusual diphthong sound, we see that the th has changed into ph.

In the fourteenth century we appear to have three perfectly distinct lines of development.

1 Vol. vi, p. 267.

First, the local form. The th has changed into ph, thence into v, which is indistinguishable from u; so we get Othensberg, Ophensberg, Ovensberg, Ouensberg, Ounsberg, with occasional returns to the long vowel o, as in Onesberg.

Where we get ou to the north of the Cleveland Hills, we often find au to the south. Dowson is the usual Cleveland and Bilsdale form of the name, which elsewhere is Dawson; again, Broughton is pronounced Browton to the north, Brawton to the south.

It would not, therefore, be surprising if we should find Ounberg transformed by the monks of Rievaulx into Aunberg, and thence Aumberg, the diphthong au being more labial than ou, and therefore assisting, rather than preventing, the transition of n into m before b. The actual written form is Auhmberghe, and in later times about Thirsk, Malton, and Helmsley (see North Riding Records), the same word occurs under the forms Ormesbrough, Ormesburghe, Ormsburie, and Oram, probably a contraction for Oramburgh.

But there is a third form, and it is one that I venture to suggest was that of the Normans. To them th was always a stumbling block; in an early Pipe Roll the township of Lythe appears as Liz Comitis ; it is not a very violent presumption to suggest that Othenberg had gone through a similar transformation, and was known to the Norman portion of the population as Osenbergh.

A double form of the same name, the one used by the better educated and the other by the worse educated class, is certainly not uncommon. Stokesley and Stowsley, Roxby and Rowsby, are two pairs of instances that are found in the immediate neighbourhood.

The earliest form of Osenbergh that I have been able to find is in a Fine of 1340; but it is not improbable that further search may be more successful. No less than four several instances occur from 1340 to 1424, and they are confined to Fines and Lay Subsidies. Where a local jury have to give their verdict, the name of Ounesbergh is used. The Patent Roll of 3 Henry VI (see p. 48) affords a good example. The name occurs four times: twice in the recital of a fine, when Osenbergh is used; twice in the recital of an inquisition, and

1 And see 28 Hen. II, rot. 4, m. 1 (Pipe Roll Series, vol. xxxi, p. 37). "Idem vicecomes reddit comp. de xxijli et xjs de

exitu de Liz et de Eggeton quas Comes de Albemar' tenuit."

« PreviousContinue »