[From Manx Note Book, vol ii, 1886]


IN the valuable addition to the literature of Runic Crosses in the Isle of Mann, by Canon Taylor, the ornamentation of the Crosses is referred to two distinct styles " the pure Irish" and "the Scandinavian." "The essential difference between the two styles," he writes, "cannot be too clearly grasped: the one is essentially geometric, the other zoomorphic."-," If we assume, with Canon Taylor, that the Book of Kells and the Illuminated Gospels of St. Cuthbert and St. Chad are the tests of "pure Irish art,"-',- it will naturally follow that all the ornamentation, including the interlacing pattern termed geometric, is "pure Irish." It is, however, in my opinion, not only an assumption, but an assumption which may readily be disposed of by an appeal to the distribution of the designs in ornaments and monuments in the British Isles, and in France, Scandinavia, and Germany. We will consider these designs under two heads ; first, the scroll, spiral, and flamboyant style consisting of graceful combinations of curves; and secondly, the interlacing work more or less square and angular, the "rope" or "basket work."

(i). The first of these two styles makes its appearance in Britain, Scandinavia, France and Germany, in infinitely remote Prehistoric times in the Bronze age, and subsequently became more and more elaborate in the Prehistoric Iron age, constituting the late Celtic art of Mr. Franks.+ In illustration of this I may refer, among others, to the collections in the museum of the Royal Irish Academy, in the British Museum, in the Anthropological collections in Berlin, Copenhagen, and Stockholm, and to the Musee de St. Germains, near Paris. This art probably was ultimately derived from the centres of civilisation in South Europe, principally Greek and Etruscan, and has been clearly proved by Chantre-," to have been introduced into France from Italy. From its prevalence among the Celts of France and the British Isles, it is justly termed Celtic, and it was dominant in these regions down to the time of the retreat of the Roman legions before their Germanic foes. I do not know a single case of its association with the interlacing pattern on any weapon, ornament, or monument in Northern or Western Europe which can be proved to belong to a date before the Germanic invasion.

(2). The interlacing pattern of Class II is conspicuous by its absence from Irish art, until the days of early Irish Christianity. It may be traced far and wide over Europe, and among warriors who owe nothing to Irish art. It occurs on articles proved by the associated remains, to be Germanic or Teutonic. In Britain, it is the ruling design in Anglian and Saxon finds in cemeteries, and in barrows-such as that recently explored at Taplow. In France, it is associated with the remains of the Germanic invaders, Merovingian and others, as may be seen in the various collections, and by a reference to M. Baudot's work on the Merovingian graves of Burgundy. It has been met with both in Switzerland and in Italy, and generally on the Continent, in those regions into which the Germanic tribes penetrated. It does not occur in France or the British Isles in association with any remains of a date before the Germanic tribes had begun to move to the attack of the Roman empire. From these facts, we may conclude that it is distinctly Germanic or Teutonic, and not "Celtic," and still less "pure Irish." Whence it was ultimately derived is a question which need not be discussed in this place. The fact, however, that in the older designs, which are sometimes very complicated, it consists generally of one line variously twisted on itself seems to me to point out that it originated from a twisted rope, and not from basket work which is made of many separate twigs plaited together.

The association of the Celtic graceful spiral and flamboyant with the Germanic interlacing work in the early Irish manuscripts, and in Irish chalices and ornaments, in the VIIIth and following centuries, may readily be accounted for by the influence of the Germanic tribes (including under that head the Scandinavians), not only in Ireland but in those parts of Europe into which the Irish missionaries penetrated. It is only reasonable to suppose that the Irish missionaries who introduced Christianity into Scandinavia and Northern Germany, and founded the great abbey of St. Gall in Switzerland, should have fallen under the influence of Germanic art, and have combined the native Celtic with the foreign Germanic designs. As a matter of fact, the two styles were so combined both in Ireland, and in Scotland and England, and generally on the Continent, wherever the Celtic and Germanic peoples lived side by side. For these reasons, the "pure Irish" interlacing work of the Nial Lumgun Cross at Kirk Michael, appears to me to be of Germanic origin, although the Cross may have been put up by the disciples of the Irish missionaries. The Cross is, however, as Canon Taylor suggests, probably older than those with the interlacing pattern broken down into serpents, dragons, and other grotesque figures, which may be of Scandinavian origin. In later times, the interlacing pattern was broken down into the zoomorphic, not only in Scandinavia, but in Normandy and elsewhere. The Celtic and Germanic designs, as defined above, are found in combination in the Runic Crosses in the Isle of Mann, and elsewhere. They were used in the Isle of Mann for the ornamentation of Crosses by the Irish missionaries and their disciples.


* " Manx Note Book," No. 7, P. 103.
+ Kemble " Horae Ferales."
* L'Age du Bronze.

CANON TAYLOR'S article on THE MANX RUNES has been reviewed by most of the leading Antiquarian publications, as well as by the St. James' Gazette, Notes and Queries, the Manchester City News, the Liverpool Mercury, and the Isle of Man Examiner. The ablest and most exhaustive of these reviews is by Mr. Henry Bradley, in the Academy of the 21st of August, which we append, as it is an additional contribution of some value to this interesting subject

THE July number of The Manx Note Book contains an article by Canon Taylor on the date of the runic crosses in the isle of Mann. The writer's conclusion is that all these monuments belong to the twelfth and thirteenth centuries, except one, which he refers to the end of the eleventh century. The exception is the cross at Kirk Michael, erected by Nial Lumgun to the memory of his foster-mother, Malmuru. These views are in direct opposition to those of the great Norwegian scholar, P. A. Munch, who. if he is correctly reported by Mr. Cumming,1 ascribed some of the monuments to the ninth century; and, moreover, regarded the Nial cross as the latest, instead of the earliest, of the whole series.

With regard to the first point, it is perfectly clear that Munch was wrong. Indeed, it is somewhat difficult to understand how he can have entertained the opinion ascribed to him; although, of course, he did not possess our present materials for the comparative study of runic palaeography. The second question, however (on which, by the way, the late Prof. Worsaae is said to have agreed with Munch), seems more open to debate. No doubt Canon Taylor is right in saying that the alphabet employed in the Nial inscription is of an older type than that found on the other monuments of the island. But the two forms were for sometime used concurrently in various parts of the Scandinavian world; and it would not be easy to say a priori which of them would be likely to be the first to reach the Isle of Mann. I am myself inclined to think that the Nial cross though belonging to a comparatively early period, cannot be absolutely the oldest of the Manx crosses Perhaps it may be worth while to state the grounds on which this opinion is based.

The inscription on the Mailbrigdi cross at Kirk Michael ends with the words " Gaut made this [i.e., this cross] and all in Mann." Now, if the Nial cross be older than the Malbrigdi cross in the same churchyard, it obviously follows that the former, as well as the latter, must be the work of Gaut. But in that case how can we account for the differing Alphabets ? The newer runes, it should be observed, were not developed in the Isle of Mann itself, but were borrowed, like the older ones, from some other Scandiriavian country. It seems to me difficult to believe that Gaut adopted a new alphabet in his later works; and I prefer to regard the Nial monument as the production of another, and somewhat later, artist who came from a part of Scandinavia where the older runes were still retained. The Nial cross is the only specimen of the older alphabet known to have existed in Mann, with the exception of a stone containing the words "It is better to leave behind one a good foster son than a bad son." This stone, now lost, was in Michael churchyard. I do now know whether anyone has suggested that it must have been the foot-stone of the grave of Nial's foster-mother, but I think there can be little doubt that it was so.

The differences between the two Manx alphabets relate to the form of the characters for a, s, and t. To these Canon Taylor adds n, following the plate published in Munch's edition of the Chronica but this is not confirmed by Cumming's copies of the inscriptions There are also differences in the functions of two of the runes. The Nial inscriptions make no distinction between the two sounds of f whereas on the Manx crosses generally the soft sound is denoted by b, as in þurlabr for þórleifr, Eabr for Iöfurr; and the rune which in the Nial alphabet stands for o is in my opinion used on the other monuments of the Island to represent a and umlaut-e. Thus it often stands for the vowel of one or both of the syllables of þenna (this) On the Ballaugh Cross I propose to read Herlaibr Heriulbsunr* for þorlaibr þoriulbsunr, as read by Cumming, and on the Braddan Cross Freka (or perhaps Frakka) instead of Froga. It would be interesting to know whether the peculiar Manx rune for b was used in the Nial alphabet, but no words containing this sound occur in the inscription. Canon Taylor relies on the style of workmanship of the Nial Cross as supporting his view that it is an earlier work of Gaut, but to me this evidence rather seems to indicate a different hand.

Canon Taylor's arguments for his theory of the relative chronology of the monuments are ingenious and forcible, and (except with regard to the point dealt with above) I think he has solved the problem correctly. It would appear that the Norse sculptors received their first "commissions" for sepulchral crosses from persons belonging to the native population of the Island, and began by imitating the Celtic style of ornament. Afterwards, as Norse names gradually become predominant among the persons commemorated, the character of the decoration also approaches more and more to the purely Scandinavian type. This order of development is at all events in accord with the historical probabilities of the case. The only reserve I have to make is that the contact of Celts and Northmen was not confined to the Isle of Mann, but extended over a much wider field, so that it is possible that the explanation of the relations between Celtic and Scandinavian art may be more complex than at first sight appears. I do not think Canon Taylor is successful in his attempt to show that the "Celtic" and the "Scandinavian" crosses differ in their orthography, but he is doubtless right in regarding risti on the Manx crosses as a corrupt form of raisti (raised). In runic inscriptions elsewhere, risti means "carved," but on the Tynwald Cross the strong preterite raist is used instead.

It is to be hoped that Canon Taylor's interesting paper will lead to an exhaustive discussion of the whole subject of the Scandinavian remains of the Isle of Mann, which seem of late to have received singularly little attention, either from the artistic or the philological point of view. The article is accompanied by illustrations of two of the most remarkable of the Manx crosses, and of the similar one at Kilklespeen in Ireland. HENRY BRADLEY.

1 I have not seen Munch's paper on the Manx runes in the Memoires of the Royal Society of Antiquaries of Copenhagen. In his later discussion of the subject in the preface to the Chronica Regum Manniae, lie says nothing about either the absolute or the relative chronology of the monuments. 

2 Of course if the character on the cross is clearly a þ, this conjecture is inadmissible But the runes do not seem to be very distinct, as Munch's plate has þiutulbsunr.


Back index next


Any comments, errors or omissions gratefully received The Editor
HTML Transcription © F.Coakley , 2001