Placename notes from the Newsletters

Spring 2008

A 16th century cannon.
Prof. Nicolaisen's Autumn 2007 article: Addendum.
More map resources on-line.
Places to wipe your shoes on?
'Where do they get them from?' Corner.

A 16th CENTURY CANNON.

An article on a 16th century cannon owned by the Duke of Argyll is an unlikely place to find a salutary snippet of information about the reliability of names recorded long ago by persons unfamiliar with a place or the speech of its inhabitants. The article in question, by Robert J Knecht, is in Vol. 8, No. 2 of History Scotland, and among other matters deals with the history rather than fantasy surrounding the Spanish Armada ship which famously, and with dreadful loss of life, sank off Tobermory, Mull, in late 1588 after an explosion. What is pertinent to our interest in Scottish names is that the ship came from what is now Croatia and had been commandeered with her Adriatic captain and crew by the Spanish authorities; and survivors who made their way home wrote, or informed someone who wrote, that they had anchored at an island called 'Largona' where the local lord was called 'Maelan'.

Professor Kosti who found this archive record in Croatia has identified Maelan with (in anglicised form - there are various Gaelic spellings of the surname) Lachlan Maclean of Duart, as the facts seem to demand; though Maelan would not look out of order in a list of early Gaelic personal names. 'Largona' is less penetrable as there is no island of similar name in the relevant area, but Alison McLeay has proposed that it is an attempt at what was earlier Loarn, now spelled in Gaelic Latharna (Lorne in English), the territory of an eponymous legendary founding Gael, now best known for the name of the firth between Mull and Oban.

In this instance we know approximately where those men came from who remembered the names, and from whom the names reached written record in a far away place; the events occurred little over four centuries ago; and we could, with a little research, gain a good idea of what kind of sounds would have been represented by the names as spelled in - presumably - a 16th century south Slavic dialect of the Adriatic coast; a hasty online search gives no indication that a Croat of today would find it particularly difficult to transliterate those Gaelic names, or Muile, much more accurately than his 16th century antecedents. That these names could become so distorted and one of them could be transferred to an island from a larger territory or a firth named from it, possibly directly by those who heard them or at most through a few intermediaries before being written down, is a reminder of the problems in trying to make sense of place-names in what is now Scotland, recorded 1600 and more years ago: from languages probably unfamiliar to those who recorded the names; by way of an unknown number of intermediaries and possible changes of language or dialect; at a time when there were no accurate maps and no literacy in northern Europe; and thereafter for centuries through copying and recopying which was always liable to error. That is not to say that we should not be grateful indeed for the fragments that mediaeval intermediaries have passed on to us from ancient sources.
(Bill Patterson)

PROF BILL NICOLAISEN'S ARTICLE IN AUTUMN 2007 NEWSLETTER: ADDENDUM
Since his article in the last Newsletter Prof Nicolaisen has become concerned about omissions from the list of contributors to Scottish place-name studies. We are very happy to make this addendum, as he has requested:-

Set up in the early fifties by Winifred Temple as a major section of the founded School of Scottish Studies in the University of Edinburgh, the Scottish Place-Name Survey came under my custodian ship in May 1956. I was joined in 1965 by Ian Fraser for several years of complementary and productive co-operation. After my departure for the United States in 1969, he became the sole Head of the Survey until his retirement in the year 2000, after 35 years of influential service. The Survey is currently in the capable hands of Doreen Waugh.

As is to be expected, the incumbents have been and still are at the top of the list of published scholars in Scottish place-name studies but, since 1950, a remarkable number of other individuals have contributed to the ever-growing inventory of publications on the subject, amongst them:

William M. Alexander (Aberdeenshire), Elizabeth Allan (Deeside), John Bannerman (Gaelic), G.W.S. Barrow (Scottish History), Albert Bil (Transhumance), Andrew Breeze (Early Celtic), Daphne Brooke (Galloway). Thomas Owen Clancy (Irish and Scottish Gaelic), Richard Coates (Scandinavian) Barbara Crawford (Scandinavian), Anthony Dilworth (Gaelic), David Dorward (Angus), Peter Drummond (Mountain Names), Gillian Fellows-Jensen (Scandinavian), John Ferguson (Berneray), Carol Foreman (Glasgow), Ian Fraser (Gaelic and General), James Graham-Campbell (Scandinavian), Eric P. Hamp (Celtic), Stuart Harris (Edinburgh), Carole Hough (Linguistics, Language History), K H Jackson (Celtic), John Kerr (Atholl), Jacob King (River Names), Gregor Lamb (Orkney), Arne Kruse (Scandinavian), Donald Macaulay (Gaelic), Aidan MacDonald (Irish and Scottish Gaelic), Lindsay Macgregor (Shetland), Neal MacGregor Strathspey), Donald McKillop (Berneray), John MacQueen (Galloway), Morgan Peadar (Gaelic), R. Oram (History), H. Pálsson (Norse), David Ross (Dictionary), S. Sigmundsson (Iceland), Brian Smith (Shetland), John Stewart (Shetland), A.B. Taylor (Scandinavian), William P.L. Thomson (Orkney), R.G. Wentworth (Gaelic), John Garth Wilkinson (West Lothian), May Williamson (street names).

In addition, most of the authors of recent doctoral theses have continued to publish beyond their narrow themes.
(Professor Bill Nicolaisen)

MORE MAP RESOURCES ONLINE
The National Library of Scotland has made additions to the old maps available online at http://www.nls.uk/maps/. These include John Kirk's surveys of about 1772 of farms in Golspie and Loth parishes on the east coast of Sutherland; and General William Roy's renowned mid 18th century military survey. This is available as continuous Lowlands and Highlands sheets, which can be rolled through directly with a mouse. It is worth noting that http://www.nls.uk/maps/os6inch/ will take you direct to the First Edition 1843-1882 6 inches to 1 mile maps which may be found more user-friendly than the version available at old-maps.

An example of John Kirk's work, from his survey of farmland at Navidale
(acknowledgements to NLS maps website).

PLACES TO WIPE YOUR SHOES ON?

The Guardian reported on 4/2/2008 that IKEA product names follow a system: because the company's founder, Ingvar Kamprad, is dyslexic, he found that naming products with proper names and words made them easier to identify.

Sofas, coffee tables, bookshelves, media storage and doorknobs are named after places in Sweden (Klippan, Malmö); beds, wardrobes and hall furniture after places in Norway; floor coverings after places in Denmark; and dining tables and chairs after places in Finland. Bookcases are mainly occupations (Bonde, peasant farmer; Styrman, helmsman) and bathroom stuff is named after lakes and rivers. (Pete Drummond)

More recently there was publicity (The Scotsman, 7/3/2008) for the suggestion from two Danish academics* that the Swedish firm may have deliberately encouraged its customers to wipe their footwear on cheap objects like doormats, named after Danish places. Relations between the Scandinavian rivals are a good deal more peaceable than they were in the brutal territorial wars of the 17th century, but more subtle digs are not unknown, such as the assertion from a Swedish sister-in-law that Danish is not a language, it's a disease of the throat. However, the Danes may have the last laugh when it comes to place-names: the first article in the recent Festschrift for Gillian Fellows-Jensen (Thorsten Andersson's 'Daner and Svear - tribal rivalry in prehistoric Scandinavia' - yes it goes back a long way) points out that the Swedes' name for their country, Sverige, is linguistically Danish, and it should probably have been Sverke but for the prestige of Danish forms centuries ago. (Bill Patterson)

* Not at the Institute for Place-Name Research in Copenhagen!

'WHERE DO THEY GET THEM FROM?' CORNER

There are some wonderful place-name etymologies out there, especially on websites, but this one is something of a treasure:-

Elementary, my dear Watson. Please send us your favourites and best finds of this kind! (Ed.)

Autumn 2007

Place-Name Studies in Scotland: a brief history.
The Atholl Experience.
Bilingual Road Signs.

(Further items from the Autumn 2007 Newsletter appear in Bibliography, and Fife)

Place-Name Studies in Scotland: A brief history by Professor Bill Nicolaisen.

As far as Scotland is concerned, the threshold leading from prehistoric preoccupation to historical study in the area of place names may well have been the last decade of the eighteenth century when Sir John Sinclair planned and published the twenty-one volumes of his Statistical Account of Scotland based on the responses to questionnaires which he had sent to the ministers of the 938 Scottish parishes. The very first of the 166 items in that questionnaire was ‘The name, and its origin’, or, in the words of a manuscript, possibly written by Sir John himself, in the Special Collections of Aberdeen University Library, ‘1. What is the ancient and modern name of the Parish? 2. What is the origin and etymology of the name?’ One could not be criticised for wondering what an item like this, particularly at the head of a long list, is doing in the company of queries regarding the location and extent of parishes, road conditions in them, crops grown, food production, population figures, labourers’ wages, and so on. What kind of statistics, i.e. what sort of information countable, qualifiable, and therefore manageable and useful to the government might potentially be derived from an accumulation of place-name etymologies offered by ministers of the kirk?

I would like to argue that the term ‘statistical analysis’, brand new as it was at the time, should not be too narrowly interpreted in this case, and that the etymological suggestions of the divines were perhaps rather intended as the foundation for the search of patterns of various kinds, mostly linguistic. To the best of my knowledge, and in contrast to the scholarly investigation of many of the other aspects of the responses received to the questionnaire, the suggested etymologies have never been analysed as a corpus, a task that might well be worth undertaking as part of an exercise to determine attitudes to names and naming two hundred years ago, at least among those who responded to Sir John’s query no.1. In some instances, the replies might even assist us in finding supportable etymologies for the names in question. At any rate, the Statistical Account, though not directly concerned with what might legitimately be called systematic toponymic research (the study of names), provides an extensive body and compatible inventory for such study, whether in any sense ‘statistical’ or not, and I am prepared to regard its accumulation of parish names and their purported lexical meanings as one of the first steps, if not the first step, towards the kind of place-name research which we pursue and find acceptable today, though with more sophisticated aims and means.

Naturally, the responses received by Sir John differ considerably from each other in both length and substance. There are those respondents who ignore item no.1 altogether, others who profess a lack of knowledge on the subject, and yet others who give it short shrift in one meagre sentence. Quite a few answers, however, are fairly full, sometimes presented with conviction, and others offered as the unsubstantiated speculation that they are. There is nothing particularly dramatic about any of the replies although it is perhaps worth noting how often the epithet ‘antient’ appears. The survival of the past, especially the distant past, in the present in the form of place names is obviously a fascinating facet of the chase for toponymic etymologies.

Sometimes the authors of the equivalent articles in the New Statistical Account in the 1840s in some way continue the discussion of the origin of a name. One example must stand here for the several offered in my actual paper: in the 1790s, the Rev. Mr. James Laurie tells us that Tinwald, supposed to be derived from the Gaelic, and signifying the harbour, or from the Saxon, the house in the wood, is situated in the country, synod, and presbytery of Dumfries. Trailflat, probably, too, of Gaelic extraction, and signifying a sloping wet side, was joined to Tinwald in 1650.

Fifty years later, his successor, the Rev. George Greig, Junior, quite rightly disputes this etymology of Tinwald, but having satisfactorily corrected at least part of the name Tingwall and having connected it rightly with cognate names in the Isle of Man and in Shetland, the writer in the New Statistical Account departs from his own high standards and lamely accepts his predecessor’s claim that Trailflat is probably from a Gaelic etymon, which signifies a sloping wet side.

Seemingly Gaelic, or Celtic, will do as a convenient designation for anything that cannot easily be accounted for etymologically. [1] As we now know, Trailflat which is recorded as Traverflet in the twelfth century, is a Cumbric tref-name although its second element has never been satisfactorily explained.

These stumbling beginnings of place-name study in Scotland deserve our attention not only because of the implied emphasis on matters antiquarian but particularly because both synchronically in the First Statistical Account and diachronically between the First and the New, a corpus of studiable and patternable parish names was built up which demanded recognition in its own right and which, frequently without any expert guidance, continued to influence later efforts in the field, building a bridge from the prehistory of Scottish toponymics to a more documentable, although yet nascent, historical phase.

At the other end of that bridge, though still within view of the prehistoric side of the divide which it spans, is James A. Robertson’s The Gaelic Topography of Scotland and What It Proves Explained; with much Historical, Antiquarian, and Descriptive Information, illustrated with Map. Published in 1869, this volume is, to the best of my knowledge, the first cohesive account of the Gaelic place names of Scotland, although the author had already included a section on the subject three years earlier in the second edition of his Historical Proofs on the Highlanders (1866). Robertson’s book is modern insofar as the author refers to ‘the numerous books and records which had to be searched for the ancient forms and spellings of the names of places’ (p.iii). What makes it less than modern is that it is a book with a mission, the object of which is to refute that there is a ‘Kymric Element’ in the topography of Scotland and to prove that the Highlanders are, in Robertson’s words `undoubtedly the descendants and representatives of the valiant Caledonian Gael, who were the first inhabitants of the land of Alban, now called Scotland, and were also of England (p.iv).

The question of the ‘Kymric Element’ or of the presence of p-Celts in Scotland is, of course, one which has exercised the minds of many to this day, especially in connection with the linguistic identity of the Picts, and the arguments mustered by the two major factions have changed very little. Central to the controversy has been the derivation of the generic Aber in such names as Aberdeen, Aberdour, and Abernethy which in form and meaning is, according to the proponents of p-Celtic identical with the Aber- in such place names as Abergavenny, Aberystwyth and Aberdare, a view which Robertson calls unflatteringly ‘ridiculous’ and ‘disproved’, dividing the word Aber into two Gaelic elements, ath ‘ford’ and bior ‘water’, so that the whole manufactured compound can be claimed to mean ‘waterford’. Later proponents of the Goidelic origin of Aber- (like Diack and John Fraser) cite Gaelic eabar ‘marsh’ in order to circumvent a p-Celtic etymology but the real masters in the field, like William J. Watson and Kenneth Jackson have, of course, left no doubt, in spite of some dissident opinions, that Aber- is the same as in Wales; the real controversy goes much deeper, however, and essentially concerns the linguistic affinities of the earliest Celtic settlers in Scotland or, for our specific purposes, the Celticity of the toponymic evidence, and if we link to this a resolute design to provide an etymology, a lexical origin for every name examined, to avoid even a semblance of failure, we can expect to, and do find, in the place-name studies published in the decades just preceding or immediately following the beginning of the twentieth century, a plethora of derivations which do not stand up to closer scrutiny in their attempt to turn as many names as possible into the Gaelic words they once were or were thought to have been.

Unfortunately, competence as far as a knowledge of the Gaelic language was concerned differed greatly from author to author, and even those of them who as native speakers of the language might have been expected to offer acceptable etymologies, sometimes failed to resist the pressure to etymologise at all costs and in a contemporary context, instead of humbly admitting their ignorance in instances in which, more often than not, others had also failed before them. Thus the very determination to succeed inevitably led to speculation at best. This began to exasperate those name scholars who had been academically trained in Celtic linguistics and, abandoning their usual Highland courtesy and tolerance, boiled over into some scathing, or should I say scalding, reviews, especially by Donald Mackinnon, Alexander Macbain and W.J. Watson.

These demonstrate the contrast between the still prevailing amateurish approaches and the new professionalism which was at the point of entering Scottish name studies, especially in the Highlands, at the turn of the century. It was into this changing world of scholarship, already inhabited by Donald Mackinnon and Alexander Macbain and, as Watson would have it, Dr Alexander Cameron of Brodick, of the Scottish Celtic Review, that William J. Watson stepped just over a hundred years ago when he decided to study the names of his native county, Ross and Cromarty, after many years dedicated to the classics. Through his position as rector of Inverness Royal Academy and his close association with the Gaelic Society of Inverness and the editors of The Celtic Review - one of them Professor Donald Mackinnon whom he later succeeded in Edinburgh, the other Ella Carmichael, his future wife, who provided a link with her father, the famous Gaelic scholar Alexander Carmichael - he had the status, the expertise, and the connections to translate his wide philological knowledge and educational ideas into sound onomastic scholarship and solid publications in this complex field of study. He was, at the time, the personification and synthesis of it all. One only has to allude to his magnum opus, The History of the Celtic Place-Names of Scotland (Edinburgh 1926), to make this point, although it would be falsifying the picture if we were to perceive everything written by Watson on Scottish toponymy before 1926 as a prelude to the main work and everything published afterwards as a kind of postlude. In that respect it is good news that a complete collection of his place-name papers has been available since 2002.

Although Gaelic place names have continued to receive the attention they deserve, especially when used as evidence for the gradual spread of the language in Scotland since its arrival from Ireland, it is probably in the area of Pictish Studies that the emphasis on Celtic place names research in Scotland has been strongest since Watson. The scholar most responsible for this advance is the late Professor Kenneth Jackson who occupied the same chair of Celtic Studies as Watson in the University of Edinburgh from the early fifties to the late eighties. While his involvement in the detailed study of Cumbric names south of the Forth-Clyde line set our knowledge of the presence of p-Celtic in southern Scotland on a firm footing, it was his publication in 1955 of distribution maps of place names containing Pictish elements that gave us, for the first time, visual representations of the settlement area of the Celtic-speaking Picts.[2] Particularly his map of the scatter of names containing the element Pit- as a generic (like Pittenweem, Pittodrie, and Pitcaple), though perhaps depicting more an image of immediate post-Pictish Gaelic settlement in Pictland soon after the middle of the ninth century, is a very helpful aid in that respect. Jackson’s view of a linguistic duality in the face of material unity, i.e. the simultaneous presence in different parts of archaeological Pictland of Celtic and non-Celtic speaking Picts, has some merit but has not remained unchallenged. The study of place names of Pictish origin was further enriched in 1968 by two geographers, Whittington and Soulsby, in their ‘Preliminary Report on an Investigation into Pit-place names’[3], in which they examined the special characteristics of sites bearing names beginning with Pit-, such as soil quality, slope value, exposure, altitude, and so on.

It was only to be expected that W.J. Watson also felt intrigued by a different linguistic strand in the Scottish place-nomenclature, that of names of Scandinavian origin, for parts of his native county, Ross and Cromarty, especially in the Hebridean island of Lewis which at that time was administratively part of it, echo in their place names a considerable Norse presence. Scandinavian names in Gaelic territory had also interested Captain F.W.C. Thomas who, as early as 1876, [4] asked the vexing question to which even today we do not have a completely satisfactory answer, “Did the Northmen Extirpate the Celtic Inhabitants of the Hebrides in the Ninth Century?”, and a little later, in 1910, George Henderson who included in his book-length study of The Norse Influence on Celtic Scotland, a substantial section on Norse place names. It was, however, the interest which Scandinavian, especially Norwegian, scholars themselves took in the subject that provided this field of research with the necessary expertise and rigour. One only has to remember the activities of the Bugges (father and son), Marstrander, Christiansen, Borgstrøm, Sommerfelt, and Oftedal, and the Norwegian involvement in the Linguistic Survey of the Gaelic dialects of Scotland; whereas the use of place names as evidence was usually incidental for that Survey, as part of the Gaelic texts recorded, Oftedal’s research became dominated by them, as his monograph on the “Village Names of Lewis in the Outer Hebrides”[5] shows. One should, of course, add the names of Nils Holmer and A.W. Brøgger to this list, as the former, Holmer, recorded the Gaelic pronunciation of place names during his field-work on the Gaelic of Kintyre and Arran, and the latter, Brøgger, paid particular attention to the potential regional homelands of the Scandinavians who settled in the Northern and Western Isles of Scotland.

The Norse place names of Orkney and Shetland have received the attention of both Scandinavians and Scots, foremost among the Scandinavians was the Faroese scholar, Jakob Jakobsen who extended his study of Shetland dialects to the collecting of place names. The resulting volume, first published in Danish, appeared in an English translation in 1936. His book is not easy to use but when complemented by John Stewart’s dictionary-like Shetland Place Names, published posthumously in 1982, the two volumes together form a helpful starting-point for any enquiry into the place names of Shetland, especially those of Norse origin.

Orkney produced its own eminent place-name scholar in Hugh Marwick whose Orkney Farm Names, published in 1952 but long out of print, is a classic. Marwick also published (mostly in the Proceedings of the Orkney Antiquarian Society) monographs on the place names of most of the major islands. Investigation of the Norse toponymy of both the Northern and the Western Isles continues from a number of perspectives, including their spatial and temporal distribution, and their relation to the homeland, their habitat-forming powers and the role of analogy in their creation.

The antiquarian instincts of most name scholars and the seduction of the detective work of linguistic archaeology are probably responsible for the fact that the top stratum of place names in Scotland has received the least attention. Perhaps so many of the English names (in the widest sense of that term) which form this uppermost chronological layer have wrongly appeared too obvious to warrant closer examination. Noteworthy exceptions to this general reluctance are the dissertations produced, in the format of the publications of the English Place-Name Society, in the 1940s by students at the University of Edinburgh. Unfortunately, only one of them, Angus MacDonald’s The Place-Names of West Lothian was published in 1941, whereas May Gordon Williamson’s The Non-Celtic Place-Names of the Scottish Border Counties of 1942 and Norman Dixon’s The Place-Names of Midlothian of 1947 have remained in their original typescript format (though there seems to be some promise for their belated publication). In 2003, Margaret Scott’s dissertation on the ‘Germanic Toponymicon of Southern Scotland’ (see below) has given us some hope that the balance might be redressed.

If, at the half-way point of the twentieth century, one had asked the question of coverage, the answer would have had to be that this was uneven, both geographically and qualitatively. For instance, there existed, in the early fifties, only one dictionary of Scottish place names, J.B. Johnston’s Place-Names of Scotland which, in spite of its glaring inadequacies, had received three editions between 1893 and 1954, and even in 2006 we are only a little better off in that respect, despite the publication of another two dictionaries of substance, one completely unreliable, the other, David Ross’s, on the whole sound but for many of its entries lacking the documentation necessary to substantiate its proposed etymologies. Nevertheless, the post-World War II developments in and the current picture reflecting the study of Scottish place-names gives us the right to be greatly encouraged. Undoubtedly, some of the positive facets of the modern scene are the result of the formation of a Scottish Place-Name Survey as part of the School of Studies in the University of Edinburgh in the early fifties.

It is, however, significant that in the same year in which the Scottish Place-Name Society was founded (1996), to foster toponymic research on a national level, the triennial International Congress of Onomastic Sciences was held in Aberdeen, in recognition of Scotland’s active role and progress in name studies in an international setting. It is also worth registering that, in the twenty-five years between the publication of the two editions of my book on Scottish place names (1976 and 2001),[6] well over two hundred quotable publications on Scottish place names appeared in print, far too many to cite and assess individually, and that that stream has not slowed down since then. Important in this surge is the central role that academic dissertations and theses have played in the promotion of place-name research in this country, involving particularly the four oldest Scottish universities. Following in Macdonald’s, Dixon’s and Williamson’s footsteps, these are (to the best of this writer’s knowledge):

Dissertations and Theses (chronologically)
W.F.H. Nicolaisen, Studies in Scottish Hydronymy (B.Litt. thesis, Glasgow 1956, unpublished).
Doreen Waugh, The Place-Names of Six Parishes in Caithness, Scotland (Ph.D. dissertation, Edinburgh 1985, unpublished).
Richard A.V. Cox, Place-names of the Carloway Registry, Isle of Lewis (Ph.D. thesis, Glasgow 1987; revised version published as The Gaelic Place-names of Carloway, Isle of Lewis: Their Structure and Significance. Dublin: School of Celtic Studies, Dublin Institute for Advanced Studies, 2002).
Simon Taylor, Settlement Names in Fife (Ph.D. Dissertation, Edinburgh 1995; unpublished but integrated into the four-volume The Place Names of Fife, Donington: Sean Tyas, Vol. 1, 2006).
Peder Gammeltoft, The place-name element Old Norse bolstaðr: An Interdisciplinary study of the development of, and place-names which contain the generic bolstaðr from their origins in Norway to their dissemination in the North Atlantic area and elsewhere (Ph.D. dissertation, Aarhus 1999; published as The place-name element bolstaðr in the North Atlantic area. Copenhagen: C.A. Reitzels Forlag, 2003).
Anke Beate Stahl, Place-Names of Barra in the Outer Hebrides (Ph.D. Dissertation, Edinburgh 1998, unpublished).
Berit Sandnes, Fra Starafjall til Starling Hill (Dr. art. thesis, Trondheim 2002; published as a thesis by NTNU Trondheim, Norway).
Angus Watson, Place-Names, Land and Lordship in the Medieval Earldom of Strathearn (Ph.D. dissertation, St Andrews 2002, unpublished).
Margaret R. Scott, The Germanic Toponymicon of Southern Scotland: Place-Name Elements and their contribution to the Lexicon and Onomasticon (Ph.D. dissertation, Glasgow 2003, unpublished).
Alison Grant, Scandinavian Place Names in Northern Britain as Evidence for Language Content and Interaction (Ph.D. dissertation, Glasgow 2003, unpublished).
Alan Macniven, The Norse in Islay: a settlement historical case study for medieval Scandinavian activity in Western Maritime Scotland (Ph.D. dissertation, Edinburgh 2006, unpublished).[7]

There is therefore plenty of evidence that the study of place names in Scotland is flourishing and that its many practitioners are willing to make the best use of what modern technology has to offer to advance it. It has certainly come a long way from Sir John Sinclair’s Statistical Account of just over 200 years ago.

W F H Nicolaisen

[1] See: W.F.H. Nicolaisen, ‘A Gallimaufry of Languages’. In: Astrid van Nahl, et al. (eds.), Namenwelten (Berlin 2004), 233-40.
[2] In: F.T. Wainwright, The Problem of the Picts (Edinburgh 1955).
[3] In: Scottish Geographical Magazine 84 (1968) 117-25.
[4] In: Proceedings of the Society of Antiquaries of Scotland (XI) (1876) 472-507.
[5] In: Norsk Tidsskrift for Sprogvidenskap 17 (1954) 363-408.
[6] Scottish Place-Names: Their Study and Significance (London 1976), new edition (Edinburgh 2001).
[7] We can now add Rachel Butter’s Ph.D. thesis of 2007 on cill- names in Argyll; see Bibliography (ed.).

Note from Editor:- The above is Professor Nicolaisen’s summary, very modest about his own distinguished contributions, of the talk that he gave to the Society’s tenth anniversary conference at StAndrews in May 2006. Even since then there have been important advances in Scottish place-name studies. The first volume of Dr Simon Taylor’s detailed study of all the significant place-names of Fife has appeared (and is available at advantageous terms to SPNS members – see below) and, more recently, the first issue has reached subscribers to the Journal of Scottish Name Studies, a peer-reviewed annual publication which will encompass personal names as well as place-names. The contents of the first issue are covered in the bibliography towards the back of this Newsletter. The SPNS is supporting the copying to modern formats, for easy public access, of the typescript 1940s PhD theses of May Williamson and Norman Dixon, which are still of great value, as well as the digitising of Alan James’s work on the Brittonic Language of the Old North (see below). The vigour and range of Scottish place-name studies are well shown in the Bibliography in this section.

The Atholl Experience

Simon Taylor writes: On 1 August this year, at a splendid and well-attended event at Old Blair, John Kerr launched the Atholl Experience, the result of forty years’ painstaking research on the history of Blair Atholl. Along with Steve Connelly, the Archivist at the A.K. Bell Library in Perth, and John himself, I made a short speech at the launch, extracts of which are reproduced below. From a specifically toponymic perspective, the wealth of original and unpublished material which the Atholl Experience contains has prepared the way for a full-scale place-name survey of northern Atholl.[1]

It is a great honour to have been asked by the Trustees of The Atholl Experience to speak at this celebration which sees the completion of this magnificent and important work. I am here in various capacities, both official and personal: official in that I am representing the University of Glasgow, where John Kerr holds a well-deserved honorary research fellowship at the Centre for Celtic and Scottish Studies. Also I am here in my official capacity as the convener of the Scottish Place-Name Society, of which John and Patricia, his wife, are amongst the earliest members. [2]

The work which we are here today to celebrate is important on many levels. Firstly, of course, it is important for the people of north Atholl, the area chiefly defined by the medieval parishes of Blair itself, Kilmaveonaig, Lude and Struan. John, so ably and tirelessly assisted by Patricia, has created a resource which will underpin and enrich local studies in this area for many decades if not centuries to come. However, the Atholl Experience is also important for all those who care about the history of Scotland in general - for example about the history of ordinary people and their relationship with the land: how they survived in and moved around the landscape, and how they articulated this relationship through their place-names. The Scotland-wide importance of the Atholl Experience is that it provides an inspirational model for local studies throughout the country. There can be very few areas of Scotland which are now, thanks to the Atholl Experience, so well provided for in terms of local history - I am not meaning in the survival of archival material, but rather in the careful and intelligent collecting and ordering of that material, and in making it accessible and available to researchers, both lay and academic, throughout the world. Today we are celebrating not only the end of a 40-year long Project, but also the beginning of the many new projects which will use the Atholl Experience as a foundation, a quarry, a launching-pad, and an inspiration.

As we all know, it is not that John has been sitting on all this information for decades and today is releasing it on the world for the first time. He has explored and made publicly available many aspects of Atholl history in a series of attractively produced books, such as Life in the Atholl Glens (1993), The Living Wilderness: Atholl Deer Forests (1996), and Church and Social History of Atholl (1998). But the Atholl Experience takes these local histories to a new level by presenting in a systematic, comprehensive and user-friendly way all the documentation and sources which he has drawn on for these more popular works.

There are so many aspects to the Atholl Experience that you would be here till sunset were I to enumerate them all - I can only urge you to dip in and sample for yourselves some of the richness and variety. And as you do you will immediately appreciate not only the content but also how skilfully and beautifully these 42 Volumes in 93 archival boxes have been assembled and presented. The whole assemblage is an eloquent testimony to the fantastic team that is John and Patricia Kerr. And the vision and scholarship which have informed this enterprise from the very start is now harnessing technologies which were hardly even dreamt of when John began his work on Atholl in the 1960s: while these physical volumes will be housed in the A. K. Bell Library in Perth, the material is all digitised and plans are already afoot to put the whole thing on the world-wide web.”

[1] The event received full-spread coverage in The Times of 2 August, under the head-line in broadest journalese ‘Meet John Kerr, Scotland’s walking Domesday Book’! Mention is made therein of a temporary hunting palace in Glen Tilt associated with Mary Queen of Scots in 1564. It was in fact associated with James V’s hunting expedition of 1529 (see John Kerr Life in the Atholl Glens (Perth, 1993), 75). I am sorry to say that I was the source of this error.
[2] John delivered a paper to the SPNS Conference in May 1999 entitled ‘Along an Atholl Boundary’, a summary of which can be found in SPNNews 7 (Autumn 1999), 3-4 (and, of course, also on the SPNS website). A more detailed study of this boundary appeared in his article with the same title in Nomina 13 (1990), 73-89.

Simon Taylor.

Bilingual Road Signs.

Ainmean-Àite na h-Alba has recently completed a project researching the Gaelic forms of place-names for bilingual signs on the A82 trunk road from Glasgow to Inverness. The report, written by Jake King with Simon Taylor and Peadar Morgan acting in advisory capacities, comprises a survey of some one hundred and nine names containing old forms, consultation with informants and orthographical discussion. The research is available here.

Spring 2007

Place-Names and Cognitive Linguistics
The Brittonic Language of the 'Old North'
Glens, Burns and other Lakeland rarities
The Influence of Industrialisation on the Place-Names of North Glasgow
A Place in the Curriculum
Book Reviews
Mapping a pair of Ayrshire twins
Appreciation: Terry James 1948-2007
Recent publications

Place-Names and Cognitive Linguistics

In recent years, major developments have taken place within linguistics, resulting in a substantially new branch of the discipline known as cognitive linguistics. These developments have important implications for place-name study, which has always been closely related to linguistics, as well as to other disciplines such as history, geography and archaeology. The purpose of this paper is to discuss some of the ways in which the cognitive linguistics paradigm may affect how we look at place-names, both within Scotland and beyond.

Cognitive linguistics refers to the link between language and thought: the ways in which language reflects mental processes. For instance, all languages are hierarchically structured. Broad categories such as animal include narrower ones such as dog and horse, and these in turn include still narrower categories such as dalmatian, poodle, cart-horse and race-horse. In linguistic terms, animal is the superordinate of dog and horse, while dog is the superordinate of dalmatian and poodle, and poodle is the superordinate of miniature poodle and standard poodle. A hierarchical structure appears in all semantic fields of all known languages, suggesting that this reflects the way the mind operates.

Before the emergence of cognitive linguistics, all category levels were taken to be equal. Poodle was the superordinate of standard poodle in the same way that dog was the superordinate of poodle, and animal the superordinate of dog. It is now believed that one level of the hierarchy – in this instance, dog – is more cognitively salient than others, i.e. more meaningful in terms of everyday life. This is known as the basic level category. It is usually the first term that comes to mind when naming an item.

Another difference between cognitive and traditional linguistics is the way that categories are believed to be organised in the mind. According to traditional linguistic theory, all items are either members of a particular category or not, and all members of a category are equal. As regards the category furniture, each item within a room is either a piece of furniture or it isn’t.

However, cognitive linguists argue that some items are perceived as being more furniture-like than others. When asked to write down the names of the first ten items of furniture they think of, most people put table or chair first. That suggests that tables and chairs are regarded as the ‘best examples’ of furniture – the central or prototypical members of the furniture category. Other members, such as beds and wardrobes, are still items of furniture although not prototypical; while things like telephones might or might not be regarded as furniture.

This structural model of a category with degrees of membership is very different from the traditional view of closed categories of equal members. It applies even to categories with a scientific basis. Within the bird category, zoological criteria make it possible to establish exactly what is a bird and what isn’t. Nonetheless, some birds are regarded as more bird-like than others. For most people in Britain, the prototypical bird is the robin, with the ostrich and penguin seen as less good examples on the edge of the category.

What has all this to do with place-names? Linguistic categories reflect mental categories, and linguistic prototypes reflect mental prototypes. Place-names are based on people’s perceptions of the world; and I have suggested previously that prototypes may play a more significant role in the naming process than has traditionally been recognised.(1) The prototypical members of a category are those that are cognitively the most salient – the ones that come to mind first, and are most often used – and this may be relevant to place-names in two ways. Firstly, it might help to explain the enormous amount of repetition within the place-name corpus, with common formations being duplicated in many different areas. Secondly, it might account for names such as ‘fish stream’, which cannot be taken to have designated the only streams with fish in the area, but may be explicable as prototypical instances of fish streams – the best examples in the area. Today I should like to focus on basic level categories, which are often structured around prototypes.

Basic level vocabulary is “the highest level at which category members have similarly perceived overall shapes”.(2) Place-names are often descriptions of the overall shape or appearance of a feature, in the terms that came most readily to the minds of earlier speakers, and at a level of precision sufficient to differentiate one feature from another. It may be significant that the Old English superordinate term deor ‘animal’ occurs rarely if at all in place-names, whereas the same term with the basic level meaning ‘deer’ is quite common. I wish to suggest that place-names draw predominantly on basic level vocabulary, and that they preserve evidence for basic level categories in earlier stages of language.

Most research into prototypes and basic level categories has been based on experiments with native informants, so that the data is essentially limited to modern languages. Earlier stages of English and of other languages are not open to this approach, and it has therefore been considered impossible to investigate them from this angle.(3) Place-name evidence may allow us to do so, and thus to gain a greater insight into the minds of the people who were here before us.

Much early research in cognitive linguistics was done on colour vocabulary. As a starting-point for work on place-names, I have compiled a corpus of all Old English colour adjectives in English and Scottish place-names recorded up to about 1100.(4) The corpus contains just 17 terms. All are simplexes, and there is a great deal of duplication. This is in contrast to the wide range of colour vocabulary recorded in A Thesaurus of Old English, which has four pages of closely differentiated colour terms.(5) Many are compounds, and there is a high proportion of nonce occurrences. The section for ‘white’ records separate terms for ‘whitish’, ‘white as snow’, ‘white as milk’, ‘white as linen’, ‘pure white’ and ‘brilliant white’. Within the place-name corpus, only the simplex term hwit ‘white’ appears – 39 times. This seems to establish hwit as the prototypical term within this area of the colour spectrum, or the basic level category in this area of vocabulary.

In conclusion, I suggest that the Old English colour terms represented in place-names are the colour terms basic for landscape in Old English. Through similar studies in other areas of vocabulary, it may be possible to identify basic level categories and prototypes in different semantic fields of Old English, as well as within the other historical languages that make up the toponymicon of the British Isles.

1 C. Hough, ‘Commonplace place-names’, Fifteenth Annual Conference of the Society for Name Studies in Britain and Ireland, Bristol, 7–10 April 2006.
2 G. Lakoff and M. Johnson, Philosophy in the Flesh (New York, 1999), 27.
3 See e.g. F. Ungerer and H.-J. Schmid, An Introduction to Cognitive Linguistics (Harlow, 1996), 263.
4 The corpus is presented and discussed more fully in C. Hough, ‘Colours of the landscape: Old English colour terms in place-names’, in Progress in Colour Studies. Vol. 1: Language and Culture, edited by C. P. Biggam and C. J. Kay (Amsterdam, forthcoming).
5 J. Roberts and C. Kay with L. Grundy, A Thesaurus of Old English, 2nd edn, 2 vols (Amsterdam, 2000). <http://leo.englang.arts.gla.ac.uk/oethesaurus>

Dr Carole Hough, University of Glasgow (summarising a talk to the conference at Govan, 4 November 2006)

The Brittonic Language of the 'Old North'

Report by the author of his talk to the Govan conference

Alan James explained his philological, literary and historical enthusiasms that lie behind his project of investigating the history of the Brittonic language in the ‘Old North’. That phrase, in Welsh Yr Hen Ogledd, refers in mediaeval Welsh literature to the extensive areas of northern England and southern Scotland that were remembered as having been once subject to ‘Welsh’-speaking rulers. From the English point of view, it comprises the Anglian kingdom of Northumbria along with the obstinately independent kingdom ruled from Alclud (Dumbarton) and its 10th - 11th century successor (or successors?) known as Strathclyde and Cumberland/ Cumbria. For Scots, it might be thought of as An Deas Ùr, ‘The New South’ - at least, that’s how the Kings of Scots regarded it in the 11th and 12th centuries!

In particular, Alan described his Guide to P-Celtic (and earlier) place-name elements to be found in this region, which he is preparing in a fair copy manuscript which the Scottish Place-Name Society will house on its website. This is modelled on the volumes of Place-Name Elements published by the English Place-Names Society [refs: Smith, A H, English Place-Name Elements, English Place-Name Society, vols 25-6 (1956); Padel, O J, Cornish Place-Name Elements, EPNS, combined vol 56-7 (1985); Parsons D et al, Vocabulary of English Place Names, (1997- continuing)]. It aims to include as headwords all P-Celtic words that have been proposed in scholarly publications as elements of place-names in the region between the Forth, Glen Falloch and the Rosneath peninsula in the north and the Humber, High Peak and Mersey in the south.

Alan argued that the Brittonic of this region was not homogeneous, but rather a range of dialects. Most of these were probably extinct by the early 8th century except in the Clyde Basin and the neighbouring uplands, perhaps as far south as the Solway and western end of Hadrian’s Wall. In these regions, there are phonological and lexical features to connect the dialect with Pritenic (Pictish) beyond the Forth, indeed the P-Celtic of the Clyde basin and Lothian might be characterised as ‘Brit-Prit’. Brittonic might have survived, in remote areas further south, but he questioned whether any P-Celtic names in northern England are decisive evidence of survival rather than re-introduction - until one reaches the Mersey Basin, where proto-Welsh may well have been in use at least in the 8th century.

The headwords in the Guide are presented in a phonetic script based on that used by Jackson, modified by Coates [refs: Jackson K H Language and History in Early Britain (1953), Coates R and Breeze A Celtic Voices English Places (2000)], in a form reflecting their probable pronunciation around AD 600. In some cases, this is conventional: ‘ancient’ river-name elements are included in the Guide, as they represent the earliest linguistic stratum transmitted from Brittonic to English speakers, though of course they were not still in use as items of vocabulary at that date, while a good many elements (identified with asterisks) are not recorded in any P-Celtic language in written form until the Middle Welsh period (post 1200), so their presence in the earlier Brittonic vocabulary can only be surmised. The publication on-line of the Melville Richards Archive of Welsh Place-Names, and on-going research on the important charter boundaries in the Book of Llandaf will throw much new light on the place-naming vocabulary of early mediaeval Wales and, indirectly, of the Old North.

Fairly detailed etymologies and lists of cognates in other Celtic languages, and, where relevant, Latin, English, Old Norse and other Indo-European languages reflect the author’s philological enthusiasms. References are made to scholarly discussions of the etymologies and other linguistic questions relating to the headwords. Phonological and morphological issues are discussed, including possible dialectal variants (for example the possibility that prenn in southern Scotland is a regional form for brinn, ‘a hill’ rather than ‘a tree’).

The meanings of the elements are discussed from the point of view of a semanticist keen to go beyond the ‘dictionary meaning’ to capture the senses of the words as used in place-naming: for example, Alan accepted the view that Govan is probably *wo-vann, ‘a little peak’ [refs: Clancy T O Report of the Society of Friends of Govan Old 6 (1996) and 8 (1998)], but questioned whether bann, which generally indicates something pointed, could have referred to the Doomster Hill, with its strikingly level platform-top (to judge from early engravings) - the ridge on which the church was built may have ended in a pointed headland, and this might have been the referent, though the local geography has been too much modified by shipbuilding works for any certainty.

The distribution of places showing these elements is considered, bearing in mind, of course, the very varied depth of coverage of different parts of the Old North in scholarly surveys. A constantly recurring question is whether a P-Celtic place-name is evidence of ‘survival’ or ‘reintroduction’. Alan argued that the concentration of certain elements in and around the Solway basin and in the uplands of the upper Tweed basin and the Moorfoots, along with the phonological or morphological form of some of the names involving these elements, associate them with a phase of reintroduction or revival related to the political and economic developments of the 10th-11th centuries.

The Guide includes occurrences of Brittonic elements in place-names in the region recorded in Roman-period sources: in a few cases, words that were used in toponymy in that period ceased to be so used at any later date. Names found in early mediaeval historical sources such as Gildas, Bede and the Historia Britonum are considered, along with those certainly or presumably in the Old North that occur in mediaeval Welsh poetry relating to the region, i.e. the awdlau attributed to Aneirin and Taliesin, though Alan confessed himself to be sceptical as to the value of these writings as historical sources: from the toponymist’s point of view, they are names occurring in manuscripts of the 13th and 14th centuries which can reasonably be taken to reflect forms dating from the later Old Welsh period (10th - 12th centuries), but not necessarily any earlier.

Finally, the Guide lists modern place-names which may incorporate these elements, though a health warning is needed against any assumption that such derivations are certain or even - in a good many cases - likely: the author’s doubts and reservations are regularly noted. They are classified as simplex (or originally simplex) names, monothemes with a Brittonic prefix or suffix, generics in proper compound names, generics in name-phrases, specifiers in proper compounds and specifiers in phrases.

Photocopies of the Guide (currently sections covering letters A - C) are in the hands of Simon Taylor (for the Scottish Place-Name Society), Oliver Padel (for the Society for Name Studies in Britain and Ireland) and David Parsons (for the English Place-Name Society): they have permission to show or copy any part to any members of those societies or other bona fide scholars who may find it of help or interest. The author welcomes contact from anyone interested in Celtic place-names, and in early Northumbrian names and linguistic history, in any part of the Old North. His e-mail address is

<alanatthelimes@hotmail.com> (but he warns that he is not an enthusiastic or reliable user of electronic communication!).

In the 'Old North': the Yarrow Valley, Selkirkshire. The river-name (Gierua ca.1120, Gierwa ca.1150) is problematical: there is no easy route to this from Celtic *garb, 'rough'; *argant, 'silver', is possible, but again not readily explained; a counterpoint of the Yorkshire Ure (Earw ca.1150), from ancient *Isura, 'strong' or 'swift', would require unusual sound changes; the medieval forms recall those of Yorkshire Yare (Garienno- in Ptolemy), possibly from a root *gar/ger, 'shout, talk'; and a connection with the Tyneside Jarrow, named from the people of the marsh (OE *gerw) would be semantically implausible even if an English river name here were not. (Thanks to Alan James for thoughts on this name; any errors are not his.)

Glens, Burns and other Lakeland rarities

Prof Diana Whaley summarises her talk to the Govan conference

The paper focused chiefly on names from Diana Whaley’s A Dictionary of Lake District Place-Names (Nottingham: English Place-Name Society, 2006), and most of the information below is extracted from there; see also forms and discussions in the English Place-Name Survey’s The Place-Names of Cumberland I-III (ed. A. M. Armstrong et al., 1950-2) and The Place-Names of Westmorland I-II (ed. A. H. Smith, 1967); also E. Ekwall, The Place-Names of Lancashire (Manchester: Chetham Society, 1922) and his English River-Names (Oxford: Clarendon, 1928).

A group of names in ‘Glen-’ occurs in north-east Lakeland: four habitation or valley names (Glencoyne, Glenridding, Glenamara, Glen-dowlin), two stream names (Glenderamackin, Glenderaterra) and two lost ‘Glen’ names (Glentreske pre-1184, Glencrest n. d.). All have recorded medieval spellings in Glen- except for Glenamara (Glemorye, Glenmer, late 16th century) and Glenderaterra (first recorded 1729). The early spellings of the second elements of all these names tend to be highly unstable and their etymologies disputed. The structure of the names, with generic first followed by a qualifying noun, adjective or pre-existing name, is compatible with Brittonic origins, specifically Cumbric *glinn ‘valley’, or with origins in the cognate Gaelic/Goidelic gl(e)ann. Brittonic naming would be paralleled in nearby Penrith, Penruddock and widely throughout northern Cumbria including much of the Lake District, and, if Brittonic, the structural type represented by the ‘Glen’ names (as also by Penrith and Penruddock) would place them in the period from the sixth century onwards (according to Kenneth Jackson, Language and History in Early Britain, Edinburgh: Edinburgh University Press 1953, 225-7). Gaelic influence would also be plausible in a Norse-Gaelic context from, presumably, the tenth century: local indications of such a context include Patterdale (Gaelic personal name plus ON dalr ‘valley’, in Germanic word order), Setmabanning (ON sætr ‘shieling’ plus Gaelic personal name) and probably Dalemain (ON dalr plus ON or Gaelic personal name). Although Brittonic explanations have generally been preferred for the ‘Glen’ names, they are not always convincing, and Gaelic alternatives may at least be considered. The paper did not propose solutions, but merely suggested that the question - which is of some historical importance - needs further investing-ation on both sides of what became the border (see, e. g., W. J. Watson, The History of the Celtic Place-Names of Scotland, Edinburgh: Blackwood, 1926, 140, for a similar problem in Lothian).

A later-recorded outlier, The Glen SD4596 (Glen 1836) may well be a nineteenth-century romantic imitation of Scottish valley-names, though Smith’s suggestion of ON *glenna ‘open place in a wood, grassy place among rocks’ opens up a further possibility for this and other ‘Glen’ names. Although it is difficult to prove its use in the Lake District, the word is preserved in modern Norwegian and appears in Norwegian farm-names, especially around the Oslofjord (J. Sandnes & O. Stemshaug, Norsk stadnamnleksikon, Oslo: Det norske samlaget 1990, s. n. Glenne).

Names whose forms on the modern map contain the syllable burn include some probable examples of the reflex of OE burna ‘stream’, such as Ludderburn SD4091 (Litterburne 1537, Luderburne 1605), probably with OE hlut(t)or, ME lutter ‘clear’ as specific, and Howburn NY2836 (Holborn 1867), ‘stream in a channel/hollow’. Some names, however, are definitely not from burna: Greenburn NY2902 (Grenebotne a. 1220) and Wythburn NY3213 (Withbotine c. 1280) are clearly from ON botn ‘inner valley’, and given the general rarity of burn compared with beck in Cumbria, it may be that others of the apparent burn names have other explanations. In particular, the proximity of burn names to Bronze Age cairns or similar antiquities at Burnbank Fell NY1121, Burnbanks NY5016, Burney SD2585 , Burn Moor SD1592 and Burnmoor NY1804 might lead one to wonder (despite the lack of diagnostic spellings for these late-recorded names) whether some of these might be cases of ModE borran(s) ‘cairn, tumulus’ from OE burgæsn. For one final burn name no speculation is needed, since Burnbank Farm SD4180 was named by the present owners in 1979, influenced by their part-Scottish family origins.

The value of intelligence about recent names from local informants was the theme of the final part of the paper, illustrated by Rowan’s Ground NY3409, a commemorative name given in 1988 and therefore unrelated to the early modern farm-names of southern Lakeland in ‘Ground’, and Huyton Hill NY3601, whose situation beside Windermere fits the etymology of ‘settlement by the landing-stage’ (OE hyð + OE tun), but fortuitously, since the name was transplanted from Liverpool along with a school evacuated from there early in World War II.

The Influence of Industrialisation on the Place-Names of North Glasgow

Prior to the Industrial Revolution, the lands immediately to the north of the city were largely open countryside, interspersed with the mansions of Glasgow gentlemen, together with various farms and small villages. With the arrival of the railways and other associated heavy industries, the landscape of North Glasgow disappeared under massive engine works, machine shops and factories. Hills were levelled and lochs were drained, and the farms and meadows were replaced with mass housing for the ever-growing workforce. Naturally, such an impact on the landscape would have an impact on the local nomenclature, and this was a period of considerable onomastic upheaval and change in North Glasgow.

The process of industrialisation began in the late eighteenth century, when the Monkland Canal was dug, to facilitate the movement of coal from the extensive coalfields in the Monklands area (about 12 miles east of Glasgow). Then in 1831, Glasgow’s first railway line opened. The ‘Glasgow and Garnkirk’ line was also initially constructed for the purposes of transporting coal from the Monklands into the city. The terminus

Extract from Richardson’s 1795 map (with thanks as ever to National Library of Scotland for online maps)

of the line was at St Rollox. Within a few years a rival to the Garnkirk company started work on a goods and passenger line to link Glasgow with Edinburgh. In the late 1830s there was a massive influx of workers, many of whom came from Ireland, who came at first to lay the railway track, and then from the 1840s to work in the locomotive manufacturing workshops that evolved due to the massive demand for rolling stock for the rapidly expanding National Rail Networks. What followed was a massive industrial boom in the area, caused both by the locomotive industry itself, and also the development of other heavy industries reliant on the coal and iron being brought into the area by the new railways.

Industrialisation had a significant impact on existing settlement patterns in the area. For example, a sizeable part of North Glasgow is known in the present day as Springburn, but this name is absent from the older maps, which feature instead a place called Springvale, a slightly larger settlement to the south of Springburn. In the mid nineteenth century, Springburn expanded rapidly, whilst Springvale was overshadowed and finally vanished altogether. The reason for change in fortunes is directly linked to the arrival of the railways. When the new railway line linking Glasgow with Edinburgh was built, the Glasgow end passed between Springvale and the Cowlairs mansion. The directors of the railway were denied permission to build a viaduct over the canal just to the south of this point, and their solution was to tunnel beneath the canal, but this created a steep gradient on the track which required a rope and pulley system at the top of the tunnel in order to pull the trains up from the city centre. The directors purchased the lands of Springvale Farm in order to house the steam engines required to power the pulley, and to build workshops for their locomotives. These new works were named Cowlairs rather than Springvale, and this was the birth of the Cowlairs Incline and the Cowlairs Works (see Thomas, The Springburn Story, ch.1, for further details). Subsequently, the track which had led to the Springvale Farm became known as the Cowlairs Road, and the workers houses which sprang up on either side of this road took on the district name of Cowlairs.

Extract from 1822 map by Cleland, Smith & Wood (again thanks to National Library of Scotland)

A little to the north, the tiny hamlet of Springburn had also begun to swell in size, as the navvies who came to build the railway line took lodgings there. Later, there was an overflow of workers from the two new railway works at Cowlairs and at St Rollox, and by the 1840s Springburn was sizeable enough to warrant the building of a parish church. In 1861, Walter Neilson opened another locomotive works to the north of the Cowlairs and St Rollox works, and Springburn found itself positioned on the main road (the Kirkintilloch Highway) and in the centre of a triangle of three major employers. Housing, shops and other amenities quickly sprang up at this convenient point, and as the land here was technically within the city boundaries, Springburn in effect became a new town in all but name. It starts appearing on maps from the early 1840s, and thus the birth of the railway industry quite literally put Springburn on the map. As the various works and the associated workers’ housing expanded across the landscape, there ceased to be a meaningful distinction between Cowlairs and Springvale, or between Springvale and Springburn. For a while Springvale was a part of the Cowlairs district, within the Springburn area, but without an important industrial institution to help preserve the name, it began to fade into onomastic obsolescence, and finally with the changing of a great number of street names in the area in 1922, all trace of Springvale was gone.

Similar changes were occurring across the rest of North Glasgow. For example the name Flemington appears on quite a number of old maps, but disappears altogether during the nineteenth century. Examination of old Glasgow maps reveals the land being sold off for the laying of a railway line, and for the construction of the Hyde Park and the Atlas Works, both locomotive workshops. The only evidence of Flemington in the present day is a Flemington Street on the original site. Conversely, the name St Rollox has survived to the present day, although all that originally occupied the site was a small chapel with a graveyard. In 1800, Charles Tennant opened his enormous St Rollox Chemical Works alongside the Monkland Canal, and then the arrival of the Glasgow and Garnkirk railway heralded the construction of the massive St Rollox locomotive works at its terminus, whose workshops and parallel tracks would eventually fill a significant portion of land on the eastern side of the Kirkintilloch Highway.

Patterns of Place-Name Survival

A clear pattern emerges from the data on lost and surviving names from the industrial period. Perhaps unsurprisingly, if a local district name was adopted as the name of a new industrial works, for example the ‘Garngad’ Ironworks, or the ‘St Rollox’ Chemical works, then the name is likely to survive. However, if an industrial works is given a different name to the land where it is constructed, the older name tends to be overshadowed and is often lost altogether. A significant number of field names, farm names and estate names from this area fell victim to the industrialisation process. Of course, the pattern reveals more complexity than the issue of whether or not an industrial works incorporated an existing site name, since with the works came associated housing, schools, hospitals, churches and cemeteries and so on, all of which require their own sub-nomenclature, and therefore have the potential to influence place-name survival or obsolescence. For example, the top of Sighthill was laid out as a cemetery in 1840, and about a decade later the lands of Barnhill were used for the Barony Poorhouse, and Stobhill to the north became a poor law hospital.

I was also interested in examining the reasons why some industrial enterprises chose to incorporate the existing name of the site into their company names, and others rejected them in favour of an alternative name. The pattern elicited from the data is that the new companies generally adopted the name of the land they were building on, whereas it was the already-existing companies who were moving to larger premises or opening secondary premises who retained existing and well-established company names rather than adopting local site-names.

A number of company names were introduced to North Glasgow in this manner in the second half of the nineteenth century, and it’s interesting that whilst the incorporation of a district name into a company name helped to insure the preservation of that district name, where a company premises was instead given an alternative name, these new names had a tendency to evolve into district names themselves, superseding the original names. An example of this is Saracen, which survives to the present day. The name ‘Saracen’ has rather an exotic sound to it, but its origins are in fact rather mundane. In 1754, Robert Tennant of the famous brewing family built a hotel in the Gallowgate, which he named the ‘Saracen’s Head’, reportedly after the famous London establishment of the same name (see House, The Heart of Glasgow, pp. 190-192 for further information about this hotel). The hotel was demolished in 1905, but the lane that ran along one side of the hotel became known as Saracen Lane, and in 1849 a man named Walter MacFarlane established a foundry business in the lane. As MacFarlane prospered, he moved premises, finally finding a permanent home for his foundry when he took over Possil House and much of the Possil estate in 1869. The remainder of the estate was leased for other industries and for building workers houses. At that point, Possil consisted chiefly of the mansion and the home farm with around a dozen workers. Within 20 years, by the 1890s, MacFarlane’s foundry covered 14 acres, and thousands of workers lived in the ‘Possil Park’ which was now a suburb in its own right.

Another name which was ‘imported’ to the area in similar circumstances is Hyde Park. The development of the name starts in the Anderston district, where a couple of local bleachers bought land and built a mansion which they named Hyde Park, potentially after Hyde Park in London. This gave rise to a Hyde Park Street, which was later to be the location of an engineering firm, principally controlled by Walter Neilson. By the 1850s, the firm had come to specialise in locomotive engineering, and the Anderston premises proved too cramped and too far from the railway lines, so Neilson moved the company to Springburn, taking the street name from his old works with him. The new works were built on the meadows of Flemington, and the rest of the Flemington lands were later built on by Neilson in the 1870s, when he surrendered control of his Hyde Park works to his colleague James Reid in somewhat mysterious circumstances, and started a new locomotive works just across the tracks, which he named the Clyde Locomotive Works. This venture failed, and in 1888 the premises were bought over by Sharp, Stewart and Co of the Atlas Works in Manchester, who brought the name ‘Atlas’ north with them (see Smart, Villages of Glasgow, pp. 162-163 for further information). The Atlas works were located just to the north of Petershill, and on that site today is the Atlas Industrial Estate.

From Industrial to Post-Industrial Place-Names

Names of this sort established themselves across North Glasgow, and by the turn of the twentieth century most of the open countryside had disappeared under works and factories and a mass of interconnecting railway lines. The Cowlairs works occupied a massive chunk of the land to the west of the Kirkintilloch Highway, which had become known as Springburn Road, whilst the St Rollox Chemical Works filled up most of the rest of that area. Foundries and other metal works had covered most of the Possil Estate, a huge storage facility for locomotive engines was built on the lands at Eastfield, and streets of tenements had replaced the farmsteads and weavers cottages, as the population of Springburn reached 30,000.

What happened to heavy industry over the next sixty years is well documented. The various recessions and economic difficulties, growing competition from overseas, failure to modernise, and specifically within North Glasgow the decline of the Monklands coalfields, are the factors that led to downsizing and company amalgamations and takeovers, which in the end failed to save the industrial businesses of North Glasgow. The Atlas and Hyde Park works were amalgamated, but closed permanently in 1962. The Blochairn Steelworks shut the same year. St Rollox Chemical Works were demolished between 1964 and 1965. Cowlairs closed in 1966, and the Saracen Foundry was demolished in 1967. The St Rollox railway works survived for a time as a maintenance depot, finally closing in 1987. This decline in heavy industry led inevitably to other major alterations to the landscape. The now-derelict Monkland Canal was for the most part filled in, and the M8 motorway was built along its former route. There were other alterations to the road layout in North Glasgow, which included the driving of a new expressway right through the centre of Springburn. Many of the smaller railway stations were closed, and miles and miles of track were torn up. Waterways such as the Gad Burn were covered over and incorporated into the sewage system, and there was widespread demolition of the slum tenements, which were replaced by dozens of multi-storey towerblocks.

Whether or not place-names survived these transitions is as often as not down to whether the developers chose site-related street names for the new housing schemes they were building on the sites of the former industrial works. For example, the name Atlas has survived whereas the adjacent Hyde Park has not, because the Atlas name was given to a business park, and various street names, whereas the Barratt houses on the site of Hyde Park did not continue the name, and the college which took over the remaining Hyde Park buildings did not adopt the name either. All that survives now is the former Hyde Park school, let out as offices, perched rather incongruously on the edge of Petershill Park. Where names do survive the transition, they have sometimes been shunted around the landscape somewhat, with the Sighthill High Flats built to the south of the cemetery, on land formerly occupied by the St Rollox Chemical Works and the bogs of Pinkston. Similarly, Petershill has moved to the east of Barnhill, at least according to many modern maps of the city, and St Rollox has retracted itself into a small area to the east of Springburn Road, where Tesco StRollox and the St Rollox sorting office keep the name alive.

An interesting phenomenon in twentieth century North Glasgow is the deliberate replacement of a place-name due to social stigmatisation associated with that name. A prime example of this is the Garngad. It was an area long associated with poverty, slum housing and overcrowding, as far back as the eighteenth century when there was a large influx of navvies to help cut the Monkland Canal. Due to these negative associations, there was a movement in the 1940s to change the name. Eventually the name of the area was officially changed from Garngad to Royston, and ‘Royston’ was also inserted into street names in place of ‘Garngad’. It was felt that with a new name the area could escape further stigmatisation, although it turned out that it would take a lot more than the changing of a place-name to solve problems such as poverty and social deprivation. In effect, all that has happened is that the new name, Royston, has gained similarly negative associations within North Glasgow.

Another name which was changed due to social stigmatisation was Barnhill. As I mentioned above, the land at Barnhill was used to build the Barony Parish Poorhouse in the 1850s, but due to parochial reorganisation it later became known as the Barnhill Poorhouse. A poor-law hospital was added to the site, which was later the Barnhill Hospital. From 1930 the Poor Law Amendment Act officially saw the end of the institution’s ‘poorhouse’ status, although it continued to house vagrants and geriatrics. What happened was that people remained reluctant to be admitted to Barnhill, which was still viewed by many as the Barnhill Poorhouse, and in 1945 it was decided to rename the institute as the Foresthall Home and Hospital, to remove these negative associations of the institute’s former purpose. The hospital remained in service until the 1980s, primarily caring for the elderly, and was demolished in 1988. The new housing estate built on the site retained the name Foresthall, and the only present-day link to the earlier name of Barnhill is the Barnhill Railway station, located across the road from the former Poorhouse Gates.

In conclusion, it is clear that the development of place-names in North Glasgow is inexorably linked to the rise and fall of heavy industry in the area. For names to survive they had to adapt to fit the purpose, and these purposes have changed in the last 200 years from farms and manors to factories and works and now in the present day to housing schemes and retail parks, with numerous onomastic casualties and innovations along the way.

Dr Alison Grant (based on a talk at Govan)

Select Bibliography
Durkan, J., ‘The Bishops’ Barony of Glasgow in Pre-Reformation Times’, Records of the Scottish Church History Society, 22, 1986, 277-301.
House, J., The Heart of Glasgow (London: 1965)
Maver, I., Glasgow (Edinburgh: 2000)
Smart, A., Villages of Glasgow: North of the Clyde, New Edition (Edinburgh: 2002)
Thomas, J., The Springburn Story: The History of the Scottish Railway Metropolis (London: 1964)

A Place in the Curriculum

Place-names have acquired something of a bad reputation in geography education. One does not have to search too much into the literature of geographical education for the phrase, “Geography is more than place-names ...” . In the past 20 years, innovations in academic geography have increasingly influenced content and methods of the secondary curriculum. With these changes investigating ‘Why?’ and ‘How?’ have rightly increased in prominence in the geography syllabus. Unfortunately, this has been to the detriment of knowing and exploring ‘Where?’. The study of place-names can foster a sense of place that is essential for good local and global citizenship. Everything happens some-where—to appreciate fully the significance of past and current events, we need to be able to place them on a map and in our minds. Through the medium of place-names, pupils can engage with essential concepts relating to location, patterning, change and language, and can have an opportunity to explore topical cultural and political issues in other, perhaps less familiar, parts of the world in ways that are educational and fun.

The Royal Geographical Society has awarded me an Innovative Teaching Grant of £400 to develop a web-based place-name teaching resource and a CPD [continuing professional development] workshop for secondary teachers. The website will comprise a collection of lesson plans, project ideas, maps, photographs, and other resources related to place-names. Using the web as a platform means that the resource will be available to pupils and teachers all over the UK and further afield.

The website will also host links to other place-name resources including the English and Scottish Place-Name Societies and The Online Key to English Place-Names as well as including a bibliography of major place-name related publications that pupils and teachers will find useful. An important part of the project is the creation of a bank of digital photographs which illustrate topographical aspects of place-names. The images collected will be incorporated into specific lesson plans or available separately for teachers and pupils to use for their own projects.

I am hoping that members of the Scottish Place-Name Society might be interested in contributing digital place-name related images to the website as well as ideas for activities, games and lessons. I have already incorporated place names and a bit of Gaelic language into map interpretation lessons. The children responded with interest and enthusiasm. Recently, some Advanced Higher students commented that they had never realised how interesting maps were until they started reading the place names. Other lessons have included using articles about name changes in countries such as South Africa, India and elsewhere as part of thinking skills exercises for older pupils.

There is an element of uncertainty in the future of school geography. One way of strengthening its position in the present climate of educational reform is to emphasise the uniqueness of geography’s use of the spatial dimension in understanding the world at any scale. I hope the project will also contribute by increasing, albeit in a small way, the profile of place name-study and its importance. If you would like to ask me any questions about the project or contribute ideas or materials, please email me:

<darceygillie@fastmail.fm>

Darcey Gillie

Book Reviews:

Names Through the Looking-Glass: Festschrift in Honour of Gillian Fellows-Jensen, edited by Peder Gammeltoft & Bent Jørgensen (C.A. Reitzels Forlag, Copenhagen 2006) XVIII + 350 pages.

This Festschrift celebrates the scholarship of Gillian Fellows-Jensen, on the occasion of her seventieth birthday. Gillian has devoted her career to Scandinavian studies, and has produced numerous books and articles on many different aspects of Scandinavian language and culture but is particularly known for her work on Scandinavian place-names.

The collection of essays within this volume reflects her varied and wide-ranging interests, with topics including Danish rune-stones, early Scandinavian tribal rivalries, Icelandic farm-names, English field-names, Norwegian and Dutch onomastics, Danish hydronomy, and the geographic evidence of place-names in the Icelandic Sagas. There are several contributions on English place-names, including a paper by Margaret Gelling on the Anglo-Norse place-names of Yorkshire as evidence both of linguistic interaction and settlement density, and a paper by R. I. Page warning of the dangers of identifying otherwise unrecorded personal names and bynames as place-name specifics.

Additionally, there is a warm and light-hearted dedicatory piece by Professor Nicolaisen, which is unfortunately positioned in the middle of the volume rather than at the beginning, due to the presentation of articles in alphabetical order by author’s surname. The Festschrift concludes with a very useful bibliography compiled by Bente Holmberg, which details all of Gillian’s publications (including reviews) in chronological order.

Of particular interest to Scottish readers will be the various articles on Scottish onomastic topics, such as Barbara Crawford’s study of the administrative significance of the names Houseby, Harray and Knarston in western Orkney. In a similar vein, Berit Sandnes writes on the chronology of Scandinavian settlement in Orkney, with particular focus on topographical names. James Graham-Campbell examines the significance of the distribution of Scandinavian place-names in northern Scotland, focusing particularly on Sutherland and Easter and Wester Ross. Peder Gammeltoft offers a fresh perspective on Hebridean island names, within the context of the complex linguistic interaction which occurred between Scandinavian and Gaelic speakers. Doreen Waugh contributes a useful modern survey of the -by / -bie names in Shetland, including detailed etymological and chronological information. Shetland is also the focus of an article by Tom Schmidt, whose study of sixteenth century Shetland personal names and surnames reveals a sharp contrast with Norwegian names from the same period, and whose linguistic patterns indicate considerable onomastic influence from the Scottish mainland.
The eighteen essays in this volume provide a wealth of material on a diverse range of topics, with something to interest everyone. It is a fitting tribute to the career of Gillian Fellows-Jensen.
Dr Alison Grant

A Dictionary of Lake District Place-Names, by Diana Whaley. EPNS, 2006.
What a splendid book! The beautiful wrap-around cover photo, of Great Gable in the evening sunshine, and the fine set of photos inside (all by Ian Whaley), are but an appetiser to the main course, the dictionary listing of every name within the Lake District National Park from the one inch maps. Each name is grid referenced, all the older forms listed, and an etymology discusses its meaning. And there is a dessert course too, for all the elements – the specifics and generics – that compose so many of the names, are examined in a ‘dictionary within a dictionary’ in 40 pages at the end. Many of these elements shed light for Scottish place-names study, since Old Norse (important this side of the border) was such a major onomastic player in the Lakes area too. Loan-words from Old Norse (like the ubiquitous fell) and Old English (e.g. pike and rigg) abound in southern Scotland, and Old Norse up the west coast too. The book sums itself up very accurately on the title page, thus: “... It is based on previous scholarship and fresh research, and combines detailed and authoritative commentary on the names with insights into the setting, languages and history which gave rise to them.” Exactly.
Peter Drummond
[A fuller review, by Alison Grant, will appear in the JSNS later this year.]
ISBN:0904889726. Available from SPNS conference bookshop, or Shaun Tyas, 1 High Street, Donington PE11 4TA. 424 pages. £20.

Place-Names of the Cairngorms National Park
The Cairngorms National Park Authority (Ùghdarras Pàirc Nàiseanta a’ Mhonaidh Ruaidh) has recently published a leaflet of this name, colour-printed on both sides of A2 paper and folded to A5 size for convenient handling. One side is presented as eight A5 pages, including the cover which shows a collage of sign-posts to places in the national park. On the reverse, half is taken up by a map naming main settlements, watercourses and lochs, woods and traditional routes; numbered symbols are used for 13 ‘Places of Interest’ and for 52 ‘Topographic Features’. The other half of the reverse is the related gazetteer, including a panel to help Anglophones with pronunciation of the local Gaelic of the place-names.

A first glance shows that the place-name material in this leaflet has nothing in common with a ‘misinformation board’ once in place near the western bank of Loch Lomond, which claimed that Tarbet (G tairbeart, an isthmus over which boats could be taken) was a place-name of Norse origin. The material is up-to-date in terms of place-name studies and plainly well grounded in local knowledge: ‘research and text’ are attributed to ‘Dr.Alison M G Diack with Dr James H. Grant’.

This soundness of content should be expected, in a publication from such a public body. What is more astonishing is how much information has been packed into the leaflet, without resort to clutter or tiny print. The presentation is enhanced by attractive photos and the A5 pages include sections on Place-Names in the Cairngorms, Linguistic Heritage, Recording Place-Names, Traditional Routes, Common Words, and Further Information (which includes a short bibliography).

In the A3 sized spread of the Gazetteer there is obviously no scope for historical depth in the expounding of place-names, but apart from the ‘places of interest’ and ‘topographic features’ already mentioned no less than 26 settlements, 18 rivers and burns, 15 lochs, 16 traditional routes and seven forests and woods are listed by English or Scots name or anglicised Gaelic name (whatever is in general use); pronunciation of anglicised Gaelic; Gaelic form and its pronunciation (often more than subtly different from the anglicised); and translation from Gaelic. Sometimes there is additional information, such as that Glenbuchat contains an old Gaelic personal name Buichead or Newtonmore (Baile Ùr an t-Sleibh) has often been referred to locally just as An Sliabh (‘The Moor’). The real anatomical translation of Bod an Deamhain is given, for the mountain feature listed under its bowdlerised English name ‘Devil’s Point’.

There is a helpful mention, in the panel on pronunciation, of the local tendency to drop unstressed final syllables, which must have made many place-names sound strange, and perhaps grammatically lax, to those from Gaelic-speaking areas where this was not characteristic. The section on recording place-names admits that some older place-names remain obscure and to illustrate the effects of language change gives a neat example of how Drochaid Bhruthainn became Bridge of Brown, by way of Scots Brig o’ Broon, and thus progressively further from the Gaelic vowel sound of the burn name.

Hut near Braemar, which the leaflet tells us is Bràigh Mhàrr, the Upland of Mar,
and consists of two settlements, Baile a’ Chaisteil (Castleton) and Achadh an Droighinn (Field of the Thornbush).

This leaflet sets a high standard for other such organisations to emulate. It will be of value to residents in the national park as well as to visitors who are interested in its human history and seek reliable, scholarly information rather than the fanciful and romantic. It ought to shame those who cynically exploit obviously false Gaelic etymologies to sell their products.
William Patterson

The leaflet states that a large print version is also available from the publisher:-
Cairngorms National Park Authority
14The Square
Grantown-on-Spey
PH263HG
tel: 01479 873535 Website: <www.cairngorms.co.uk> e-mail:<enquiries@cairngorms.co.uk>

Mapping a pair of Ayrshire twins

During a recent work trip in Ayrshire, on a country road some 9km ESE of central Ayr and 2½km WSW of Drongan, the names of two adjacent farms demanded attention. They were Sandhill, at National Grid reference NS412175, and Bargenoch, at NS415714. These farms, on a small ridge between burns about 1km east of Martnaham Loch, are in an area with a wonderful linguistic mix of place-names. Besides the mass-produced Burnside, Hillhead and Mossend there are the hand-crafted Chipper-lagan, Cloquhairnan and Millmannoch. As well as the predominant Scots and the ubiquitous Gaelic, a Cumbric presence declares itself in the large village of Ochiltree, 9km ENE of the twins. Somewhat more distant place-names like Prestwick and Maybole recall the period of Northumbrian overlordship.

Genoch is a name that occurs just inland from the Heads of Ayr and again at the head of Luce Bay in Galloway, where Genoch Mains is beside the vast expanse of sand of the Torrs Warren. It is a reasonable supposition that it records a south-western pronunciation of Gaelic gainmheach (fem.), ‘fine sand’. Hence the pairing of Sandhill and a name which appears to mean the same in Gaelic (with bàrr, ‘top’, extremely common in names of places on hills in the South West, as the generic), is so striking. Maps available online through the National Library of Scotland are the most accessible way of checking whether the pairing may have coexisted for centuries. (Thanks to NLS for making the following maps freely available.)

From Johan Blaeu: Coila Provincia 1654

Unfortunately there is no surviving Timothy Pont manuscript map of the area. The first published map of the area is Johan Blaeu’s mid 17th century map based on Pont’s. This has Bargannoch where we should expect it. It also has a San, between Bargannoch and Martnaham Loch. But does this represent a simplex Sand, with <d> assimilated to <n>? Or was it a case of a longer name of which only the first three letters were readable on a worn manuscript?

Andrew Armstrong’s New Map of Ayrshire (1775) has Bargannoch again, unfortunately close to the edge of a sheet. There is no sign of Sandhill on this sheet or its neighbour.

From Andrew Armstrong: New Map of Ayrshire 1775

However, from Thomson’s map of 1820 onwards, the pair of names is always side by side, though Thomson’s Gaelic twin is Bargonoch. Whether that <o> is deliberate or a transcription error for Bargenoch one cannot tell.

From John Thomson’s maps of Ayrshire 1820

Since the Ordnance Survey First Edition 1859-60 that name has remained Bargenoch and the twins have been inseparable. Recourse to a few maps has not satisfied curiosity as to whether Sandhill is a conscious translation name or an independent naming from the same geomorphological circumstances. But it has been a pleasant reminder of the artistic quality of work by the early surveyors and mapmakers, in a toponymically fascinating area.
William Patterson

Appreciation: Terry James 1948-2007

It is with sadness that I have to report that Terry James died on 21 January this year, after a long and painful illness. Archaeologist, Dyfed historian, toponymist, printer, computer expert, and honorary member of SPNS, Terry was best known to members in Scotland as the architect of the prototype Scottish Place-Name Database.

Ian Fraser, Doreen Waugh and I first met him at a Place-Name Digitisation symposium in Belfast in September 1993. At that time Terry was Information Systems Manager at the Royal Commission on the Ancient and Historical Monuments of Wales. In his spare time he had constructed a FoxPro database to record and analyse Carmarthenshire place-names (Enwau), and over the next few years Terry and I worked together on a voluntary basis adapting and expanding Enwau’s core structure for use as the basis for a Scottish Place-Name Database. Terry gave generously of his time and expertise, making several visits to Scotland to work on the development and design of the Database and to take part in discussion groups and seminars, even after ill health had forced him to take early retirement from the Royal Commission in 1998.

It was the structure of this Database which informed and inspired the Carnegie-funded Access Database Project at Edinburgh University in 2000-2001. Terry also gave a paper at the ‘Uses of Place-Names’ Conference in St Andrews in February 1995 (which saw the foundation of SPNS). This was later published in my 1998 edited volume The Uses of Place-Names as ‘Place-name Distributions and Field Archaeology in South-west Wales’ (101-19)*. Another chapter in this book (‘Gwaun Henllan - the Oldest Recorded Meadow in Wales?’, 169-79) was written by Heather James, Terry’s wife, also an archaeologist, who at that time worked for the Dyfed Archaeological Trust.

Together Terry and Heather made a huge contribution to the understanding of many aspects of the history of south-west Wales. Something of the breadth of this contribution, as well as of Terry’s IT skills, can be gleaned from their website <http://www.terra-demetarum.org.
uk>. This includes a link to the online Carmarthenshire Place-Name Database. As a tribute to Terry’s work the Carmarthenshire Antiquarian Society intends to publish a volume of essays dedicated to his memory.
Simon Taylor

* Terry was unable to attend the conference in person. His paper was read out by Dauvit Broun. A version of this paper, with updated figures and appropriate textual changes, can be found on
<http://www.terra-demetarum.org.uk/Articles/Place-name%20distributions.pdf>

Recent publications:

Coates, Richard, 2006, ‘Maiden Castle, Geoffrey of Monmouth and Harun al-Rašid’, Nomina 29, 5-60.
Higham, Mary, 2007, Of Names and Places: Selected Writings of Mary Higham, ed. Alan G. Crosby, published jointly by EPNS and SNSBI, with editor’s Introduction and an Appreciation by Margaret Gelling.
Isaac, Graham R., 2005, ‘Scotland’ in New approaches to Celtic place-names in Ptolemy’s Geography, edd. J. de Hoz, E. R. Luján and P. Sims-Williams (Madrid), 189–214.
Scott, Margaret, 2006, ‘Previck and Leckprivick: Onomastic Connections in South-West Scotland’, Nomina 29, 115-28.
Taylor, Simon, 2006, ‘The Early History and Languages of West Dunbartonshire’, in Changing identities: ancient roots - the history of West Dunbarton-shire from earliest times, ed. Ian Brown (Edinburgh University Press), 12–41.
Woolf, Alex, 2006, ‘Dún Nechtain, Fortriu, and the geography of the Picts’, Scottish Historical Review 85 (2), 182-201.
(compiled by Simon Taylor with the help of Carole Hough and Maggie Scott)

Autumn 2006

Four Cumbric Names: Barnweill, Brenego, Roderbren, and Haltree
Standardizing International Place Names
The Place-Names of Fife
Bede's Urbs Giudi: Stirling and its Context
Jakob Jakobsen Conference
A Fiendish Puzzle
The Norse in Islay - the Place-Name Evidence
'Place-Names of the Cairngorms National Park'
Appreciation: Mary Higham
Gaelic Lesson on a Sign

Andrew Breeze: Four Cumbric Names: Barnweill, Brenego, Roderbren, and Haltree

Barnweill, Brenego, Roderbren, and Haltree* are all Cumbric forms, the first three from near Ayr, the last from pre-1975 Midlothian. They have been slightly obscure. Yet comparison with other Celtic languages may help explain them.

Let us take the Ayrshire names first. Barnweill (NS 4130) is a farm in open country between Kilmarnock and Ayr. Nearby are the ruins of Barnweill Church, below a hilltop monument (with fine views) to Sir William Wallace, commemorating his burning in 1297 of the Barns of Ayr. Professor Nicolaisen in his book Scottish Place-Names gives the forms Berenbouell of between 1177 and 1204, and Brenwyfle of 1306. He explains the first element as Cumbric pren ‘tree’ but says nothing of the second. What could it mean?

‘Barnwyl’ is in the middle of this excerpt from Herman Moll’s 1745 map of Kyle and Carrick. ‘Sim’ must be short for what is now Symington and ‘Dalketh’ is now Dankeith. ‘Gaidyant’ may be a garbled version of what is now Gadgirth and ‘Torbonton’, for ‘Torbouton’ (Tarbolton), shows trouble with minims. Thanks to the Trustees of NLS for providing this on the maps website.

Some help is offered by Barnbougle (NT 1578), on the coast between Cramond and the Forth Bridge. Nicolaisen follows W. J. Watson in talking this as also Cumbric and meaning ‘tree of a herdsman’ (it would be pren bugail in Welsh). As there is no sign of g at Barnweill, the forms cannot be the same.

However, Middle Welsh and Middle Breton bual ‘wild ox’, both from a borrowing of Latin bubalus ‘ox’, may offer a solution. Barnweill and its church are located about the 400-foot contour on terrain more suited to grazing than the plough. It might thus have supported wild cattle in Celtic times. Barnweill may hence mean ‘wild ox’s tree’ or more probably (since the plural of Welsh bual is buail, and oxen are gregarious) ‘tree of wild oxen’.

Welsh bual is a commoner word than one might think, as it meant not only ‘wild ox’ but also ‘drinking horn’. These horns were highly valued. Welsh kings drank mead or ale from them, and medieval Welsh law declares that the judge of the court had a right to a wild ox’s horn (as well as a gold ring and a cushion for his chair). So these oxen had a useful role in early Britain. To this day a herd of them exists just outside Scotland at Chillingham (NU0626), north-west of Alnwick. They are small and creamy-white, with crescent-shaped horns tipped in black.

They are also unexpectedly swift and shy. The herd, recorded from 1692 onwards, seems to have been at Chillingham since 1220, when the hunting park was built and the oxen were apparently trapped within its walls.

If the present etymology is correct, a herd of wild oxen like those at Chillingham would once have grazed and lowed at Barnweill ‘tree of wild oxen’. They congregated or sheltered there long ago, at a time when Ayrshire was inhabited by British men and women who spoke a language akin to Welsh.

Five miles south of Barnweill were, formerly, Brenego and Roderbren. Though we do not know exactly where they were, they figure in records of between 1177 and 1204, and must have been close to Enterkine (NS4223) on the banks of the river Ayr. Professor Nicolaisen again sees Cumbric pren ‘tree’ in them but is silent on the other elements.

Brenego certainly looks Celtic, and may mean ‘tree of the smith’, corresponding to Welsh pren y gof. Smiths and their smithies occur in toponyms all over the world. Nicolaisen notes that Smeaton (NT 3569) to the east of Edinburgh is an English form meaning ‘smith’s homestead’. Minnigaff (NX 4166) near Kirkcudbright seems to be Cumbric for ‘bush of the smith’. In Wales, the twelfth century Book of Llandaff mentions nant i gof ‘stream of the smith’, probably close to Undy (ST 4386), near Chepstow; while modern Tonyrefail (ST 0188), south of the Rhondda in Glamorgan, is ‘grassland of the smithy (gefail, a feminine)’.

If these forms suggest Brenego means ‘tree of the smith’, it will be evidence for a man whose horseshoes, hinges, keys, pans, cauldrons, axes, adzes, mattocks, sickles, scythes, hammers, ploughshares, coulters, spears, swords, and many other items were vital to society. The smith enjoyed high status amongst the Celts. Welsh laws mention him together with the poet and the doctor, and Irish law gives him high honour-price. Perhaps the smith of Brenego was also a man of substance.

Now for Roderbren. It is hard to be certain on the basis of one attestation. But the form may be equivalent to Welsh rhyd y pren ‘ford of the tree’, which has parallels at Rhyd-y-gwern (ST 2088) ‘ford of the alder trees’ near Caerphilly in Glamorgan, or Rhydyronnen (SH 6102) ‘ford of the ash tree’, a tiny place (passed by narrow-gauge stream trains) near Tywyn in south Gwynedd. It is true we should not expect Rod- with a back vowel from a Cumbric equivalent of rhyd, with a front vowel.

Yet it possible that what linguistics call dissimilation (the changing of similar sounds to unlike ones, as with Italian pelegrino ‘pilgrim’ from Latin peregrinus) took place after a Cumbric equivalent of Welsh rhyd y pren was borrowed by English. If Roderbren does refer to a ford, it may have been on the Ayr, where Stair (from Gaelic stair ‘stepping-stones; clapper bridge’) a mile upstream from Enterkine shows another form of crossing. Or (perhaps more probably, since the Ayr is a powerful river, with a name that Professor Ifor Williams of Bangor understood as ‘war-goddess’) it lay on the stream just north of Enterkine.

Finally, Haltree/Halltree* (NT 4152), a farm on the old Midlothian’s south-east fringe (but now in the area of the Scottish Borders Council). It lies in a narrow glen by the Edinburgh-Galashiels railway (planned for re-opening), Gala Water, and the A7. The name looks English enough. Yet the forms Haltre in 1483 and Haltrie in 1587 point to older origins. The second element is Cumbric tre ‘farm, homestead’, as with Tranent ‘homestead of the streams’ in East Lothian. It is the first element that has been problematic. Nicolaisen in his Scottish Place-Names calls it doubtful. He cites Norman Dixon’s Edinburgh doctoral thesis of 1947 for the first part as from Old Norse healdr or Old English heall, producing a Germanic-Celtic hybrid. Yet he admits that, if correct, this would be a remarkable formation.

There seems a simpler solution. Dictionaries give the Welsh form hâl ‘moor, moorland’ and Cornish form hal ‘marsh, moor’. The first is found in older place-names, as at Llwch-yr-hâl (SN3342) ‘marsh of the moorland’, a valley side farm east of Newcastle Emlyn, Dyfed. The second is best known from Penhale Point, near Newquay on the Cornish coast, called after the farm of Penhale (SW 763583), now swallowed up in an army camp. Here the sense was ‘head of a marsh’, the farmhouse being at the end of a marshy valley, as Oliver Padel’s dictionary of Cornish toponyms makes clear.

This helps us with Haltree in Lothian. It is situated by the 800-foot contour on the upper part of Gala Water. On all sides there is moorland rising to over a thousand feet. So ‘moor farm, moorland homestead’ would make good sense here, especially in contrast with settlements lower down the valley, on more favoured land. If this explanation is correct, it removes the need for Dixon’s somewhat unconvincing hybrid, and reveals something of the Britons of Lothian before they lost their territory to the English, who occupied Edinburgh in 638.

Haltree by the Gala Water, seen from the A7 road. The small ‘Haltree Rings’ hill fort is to the west, on the hill mid-way across the photo. A larger ‘fort’ on OS maps is closer to the south, above the river bank.

The farm of Haltree would have a name perhaps going back thirteen centuries. Like the other place-names quoted here, it may perhaps be looked at with a degree of historical emotion or response. It would be amongst the few memorials of the vanished Britons of southern Scotland, as also to Cumbric, their, long-extinct, language.†

* Haltree is the usual spelling, though it has occurred also as Halltree, e.g. on OS Pathfinder, 1989 edition, though not on 1965 1:63,360 or 1994 1:50,000. (Ed.)

† The author is happy to acknowledge that John G Wilkinson had independently arrived at the same etymology for Haltree.

Dr Andrew Breeze, University of Navarre, Pamplona

Standardizing International Place-Names

Have you ever wondered which is correct – Burma or Myanmar, Wien or Vienna, Greenland or Kalaallit Nunaat? Getting it right may not seem to be a big issue, but how would you like it if someone spelled your name wrongly? Place names really are important. They form an essential part of every day life and are key elements in the many and varied geographical information systems that lie at the heart of modern communications networks, from maps, gazetteers and travel guides to signposts, directories and electronic databases.

Geographical names can also be expressions of cultural identity and political power. Greece is quite possessive about the name Macedonia and will only recognise its neighbour to the north as the Former Yugoslav Republic of Macedonia. The South Africa Geographical Names Council would like to rewrite history by expunging names with a colonial context and Koreans are angered when they see the name Sea of Japan on maps and atlases.

Differences in names may result from changes, either accidental or deliberate, over time, multilingual situations, the use of different orthographic conventions, or varied transcription from one writing system to another. The proliferation of alternative name forms is only natural in an ever-changing world but what is the correct name to use if you are searching a database, compiling an atlas or writing to a foreign diplomat?

If geographical names are to be an effective means of communication there needs to be a degree of consistency in their application. To avoid confusion in an ideal world no two places should have the same name and every geographical feature should have a single name. While this could never be achieved in reality, a degree of name consistency can be advanced by the national or international application of standards or norms.

The establishment of usable and consistent written forms of toponyms depends heavily on the official use of names within each country being agreed by a national names authority. There are about 50 countries in the world which have official names authorities backed by legislation. Their role is to approve new names or name changes, develop toponymic guidelines for the rendering of names in their country and develop national gazetteers that can be used by local and national government and by users throughout the world.

A strong argument in favour of a national names programme is the need to ensure that data bases are interoperable. In addition to this, money can be saved by avoiding duplication of effort by separate agencies determining the “correct” names to be applied to their maps, legal documents and other publications. Ultimately, the consistent use of names offers many benefits to local, national and international communities engaged in trade and commerce, map production, emergency relief, security strategy, administration, navigation, environmental management and the gathering of statistics.

The need to standardize names at a national level has been recognised for well over a century. In 1890 the world’s first national names board was established in the United States of America at a time when the exploration and settlement of the west was gathering pace. The US Board on Geographic Names (BGN) was established to provide a single authority to which surveyors, geologists and geographers could address questions of nomenclature and orthography.

While the United Kingdom does not have a national authority responsible for the standardization of names in this country, it does have an independent non-departmental body that advises government departments on the appropriate use of place names throughout the world. Set up in 1919, the Permanent Committee on Geographical Names for British Official Use (PCGN) establishes and applies the principles by which foreign geographical names should be written in government documents. This involves the determination of the written form of a toponym as established by the official agencies of a foreign country and romanized to an agreed system where appropriate. A staff of three toponomists, operating from rooms in the Royal Geographical Society in London, is managed by a committee which includes a representative from the Royal Scottish Geographical Society.

An important element of the work of the PCGN is the promotion of the international standardization of geographical names by representing the British government within the biennial Sessions of the United Nations Group of Experts on Geographical Names (UNGEGN) and at the United Nations Conferences on the Standardization of Geographical Names held every five years.

The question of names standardization was raised in connection with the cartographic services provided by the United Nations Economic and Social Council after its inception in the late 1940s. A six-member sub-group, formed eventually in 1960, led to the first conference in 1967 and the creation of the UNGEGN. The Group of Experts has held 23 sessions to date, its ongoing work being undertaken through 22 geographic/linguistic divisions of the world and 10 working groups dealing with individual issues such as romanization systems, toponymic terminology, toponymic data files and gazetteers and training courses. The UNGEGN has produced a number of publications, the most recent of which is a Manual for the National Standardization of Geographical Names (2006). The ultimate goal of the UNGEGN is to establish usable and consistent written forms of geographical names and encourage their application throughout the world. This is achieved, not by making decisions on place names, but by promoting the creation of national names authorities, by advancing the development of helpful principles and guidelines through international cooperation and by assisting in the dissemination of nationally approved name forms.

Political changes, such as the break-up of the former Soviet Union, coupled with expansion of global communication and an increased recognition of ethnic minorities and indigenous groups have all contributed to a heightened awareness of geographical names and the need for a greater degree of consistency. The United Kingdom faces its own issues of place name consistency, particularly in relation to Welsh names in Wales and Gaelic names in Scotland which have been the subject of the Ordnance Survey’s Gaelic Names Policy and Welsh Language Scheme approved in 2000 and 2001 respectively. Ask the average man or woman in the street what the name of this country is and see how many different answers you get.
The problem of place name consistency has been around since mankind started to use language to communicate, but it was not until the 6th century BC that the need to address the problem was suggested when Confucius said that “the first task of a true statesman is to rectify the names.” Today, an increasing number of statesmen are, in fact, taking this issue seriously.

For useful information on geographical names issues and web links to a wide range of national and international gazetteers see the PCGN website: www.pcgn.org.uk

David Munro (based on his talk to the St Andrews conference) Dr Munro is Chairman of the UK Division of PCGN and as such is part of UNGEGN. He is also Director and Secretary of the Royal Scottish Geographical Society and an SPNS committee member.

The Place-Names of Fife

It was just over ten years ago, in February 1996, following a conference called the ‘The Uses of Place-Names’ held in this very lecture theatre, that the initiative was taken to found the Scottish Place-Name Society. Since then the Society has gone from strength to strength, and, ten years on, I would like to take this opportunity to thank all those who have devoted so much time and energy to making it the success it has become. A second reason to celebrate today, especially in relation to Fife place-names, is the success of the bid made by Professor Thomas Clancy of the Department of Celtic at Glasgow University to the Arts and Humanities Research Council. The Project for which the bid was made is entitled ‘The expansion and contraction of Gaelic in medieval Scotland: the onomastic evidence’, and is the biggest investment in Scottish place-names for a long time.

This Project, which started on 1 May this year, will run for 4 years, with the first two years committed to finishing the last three volumes of the Place-Names of Fife. You will find more details of the Project in the Spring issue of Scottish Place-Name News (no. 20), as well as on the SPNS Website. To be added to that information is the fact that the Project has now appointed as its funded PhD student Peter McNiven, who will be working on aspects of the settlement-names of Menteith.

I had hoped that this May conference would be a threefold celebration, the third reason being the launch of Volume 1 of the Place-Names of Fife, but this has been delayed until later this year. I can, however, personally celebrate since it is now off my desk and outwith my control!

I want now to look at the theme of language replacement in a Fife context as seen through the prism of place-names. Wherever two or more languages co-exist, such as was the case in Fife with Gaelic (G) and Scots (Sc) during the late 12th and 13th centuries, each language has not only its own grammar and vocabulary or lexicon but also its own toponymy or place-nomenclature. These toponymies can relate to each other in different ways. They can share the same name, with the new-comer language borrowing from the older-established one, and adapting it to its own phonetic system. This is how the bulk of the Celtic names of lowland Scotland have come down to us. However, there can be full translations, part translations, and completely new coinings - this last means that each language refers to the same place in totally different ways. An example of this from further north would be Dingwall, the Older Sc form of a Norse name, which in G is known as Inbhir Pheofharain. If G had survived in Fife for a few centuries more, there would no doubt be many examples of these parallel toponymies, but it did not, so examples are few and far between. It says a lot about social and political continuity in this era that much of the place-nomenclature of the G-speaking period survived the demise of that language: for example every single medieval parish in Fife has a Celtic name. However, names of many minor natural features as well as of settlements, many smaller ones and a few larger ones, are Sc. Some of these settlements will have been created in the Sc-speaking period on marginal land, as for example the many places called Muirton or Morton. These generally derive from Sc muir ‘rough grazing land’ + toun ‘farm’, though some, such as Morton of Blebo near St Andrews, derive from Sc myre + toun ‘farm by a mire’, as we know from early forms.

However, Sc names do not just refer to minor topographic features or later, smaller settlement on marginal land. There are several large, important estates which have Sc names, such as Friarton in Forgan parish, north-east Fife, or Mastertown, Dunfermline parish, west Fife. We could guess that both these places must have existed as units of agricultural exploitation in the G-speaking period, and must therefore have had G-names which have been superseded by Sc ones. In fact, the documentation with regard to these two names is so good that we don’t have to guess, since we get a glimpse of the earlier names, just before they disappear in the late 12th and early 13th century. Presumably the older Celtic names of each place will have existed for as long as G was spoken locally, while Sc-speakers would have used the more recently coined Sc forms. When G died out, so did the older names. Both Friarton and Mastertown are recorded with both their old G names and their new Sc ones as part of their transfer to newly founded or re-founded reformed monasteries, St Andrews and Dunfermline respectively. Both of these monasteries we know were staffed initially by monks or canons from England, speaking not only French but also forms of early Middle English, which was soon to develop into Older Sc. Burghs are usually pointed to as the major engine of linguistic change in medieval lowland Scotland, but such names as Mastertown and Friarton remind us that the 12th-century reformed Church, as major land-holders and land-managers throughout Fife and beyond, also played an important role in the change from G to Sc in the general population.

In what follows I will concentrate on Friarton, since as far as I am aware, its earlier name has not hitherto been recognised. Friarton is a large estate which is now divided into three parts, Easter Friarton, Wester Friarton (formerly Meikle Friarton) and Knockhill, formerly South Friarton or Little Friarton. All told it consists of about 3 square kilometres, much of this good, arable land, with some bog and some upland.

Between 1159 and 1163 Malcolm IV gave a carucate (c.100 acres arable + pasture) of land in Naughton (i.e. in Forgan parish) by the name of Melcrether (unam carucatam terre in Adnectan nomine Melchrethre) to the priory of St Andrews (RRS i no. 228). This is last mentioned (as Melcrether) in 1228 St A. Lib. 233. The name also occurs in the early 13th century (St A. Lib. 275), when the canons’ land of Malcrether is mentioned as one of the boundaries of land granted by Alan de Lascelles (probably the son of the laird of Naughton) to St Andrews Priory. The text of this charter is as follows (translated from the Latin): Alan son of Walter de Lascelles grants to the church of St Andrew and to the canons who serve God there two acres of arable land out of my land in the parish of Naughton (Adhenauthen) viz those two acres which are nearer to (i.e. right beside) the estate (villa) of Cowbackie (Culbakin) and which stretch towards the north from the road (via) by which you go from the villa of the said Alan to Cowbackie, along with one acre of meadow measured from the land of Cowbackie towards the west between the said two acres and the land of the said canons, (i.e.) of Melcrether (et terram predictorum canonicorum de Malcrether) etc (St A. Lib. 274–5).

It is clear from this charter that Melcrether lay in the eastern part the parish of Forgan, near the boundary with the lands of Cowbackie in the adjacent parish of Leuchars. This description would well suit Friarton, first mentioned as such in the 1230s, in a charter of Richard de Lascelles, probably Alan’s successor as laird of Naughton. This charter records that Richard gave to St Andrews Priory ‘three acres on the east side of the land of Friarton (de Frereton’) and on the north side of the road which goes from Frereton’ to the church of Forgan (Forgrund) bounded by the causeway (calceti )on one side and the burn on the other, and by the pile of stones beside the road (uiam) from Inverdovat (Inuerdoueth) to St Andrews on the third’ (St A. Lib. 274). The 3 acres mentioned here might even be the same 3 acres mentioned in the earlier charter issued by Alan de Lascelles, but some field-work is needed.

My contention is that Melcrether is in fact Friarton: the co-ordinates fit, as far as they can be ascertained from the above-quoted charters; and the names are never mentioned together, with the former disappearing from the record just as the latter appears. There is one other piece of evidence, which I think clinches the matter. We know that Melcrether belonged to the canons of St Andrews; and in Alan de Lascelles’s charter of the early 13th century we have seen it described as terram predictorum canonicorum, ‘the land of the said canons’ (St A. Lib. 274–5). This is in fact an approximate Latin translation of Friarton, bearing in mind that Old Sc friar can refer to any member of a male religious order. There is no record of the important lands of Friarton coming into the possession of the canons (friars) of St Andrews, nor is there any record of their ever losing the equally important lands of Melcrether. I rest my case.

Presumably for as long as G was spoken in this part of north-east Fife the name Melcrether existed, but when the language died, so did the name. The Sc-speakers had chosen to use for this place a different name coined in their own language, a name which emphasised and encapsulated its ownership by the recently established priory of St Andrews - the name itself proclaimed its new owners, and this overrode the usually more conservative use of names by new proprietors, both lay and clerical, during this period. Why this name-change occurred is not entirely clear, but it may have had something to do with disputed ownership, or with a more thoroughgoing replacement of the agrarian population untypical for the wider locality. It is certainly a question worthy of further investigation.

Endnote: the meaning of Melcrether: perhaps G maol + G *crithir or *creth + suffix -er ‘bare shaking place, bare shaking ground’? If so, then the second element is related to G crath ‘shake, tremble’, OIr crith (gen. sing. and nom. pl. cretha) m. ‘act of shaking; a tremble’; also crithir ‘shaking, trembling’ (DIL); found in Scottish place-names such as Cray in Glen Shee PER, Craithies by Meigle PER, Crathie by Braemar ABD and Crathes KCD, and referring to boggy land cf Irish crathaidhe, creathaidhe ‘quaking bog’ (Watson CPNS, 477–8). The bog in question would be the elongated and extensive marsh, the referent also in the parish name Forgan (Forgrund ‘above or by the bog’) and in Myreside. It stretches eastwards along the burn from the lands of Friarton/Melcrether and past the old kirk (see map).

Map: Ordnance Survey One-inch “Popular” edition, Scotland, 1921-1930, from
http://www.nls.uk/digitallibrary/map/early/counties.html
courtesy of the Trustees of the NLS.

The lands of Cowbackie, Leuchars, occupy the south-east corner. The medieval kirk of Forgan is shown as an antiquity, ‘Church’. Myreside is beside the mire or bog referred to in the names Forgan and Melcrether.

Simon Taylor (from his talk at the St Andrews conference, held in the Purdie Building, University of St Andrews)

Bede's Urbs Giudi: Stirling and its Context

The discovery on the NE slope of the Gowan Hill in Stirling of unrecorded features, earth and stone banks, by Digney (1995) led him to suggest that these might be the remains of defences of an Iron Age fort, or oppidum. This discovery strengthened the case for an important Early Historic (Dark Age) settlement there and renewed interest in the discussion of whether Bede’s urbs Giudi was Stirling (Page, 2003). It is strange that the earliest record we have of the place-name Striuelin or Strivelin is in the 1120s when it appears as a royal burgh (e.g. David I Chrs. No.19). We would expect the origin of the town to be much earlier, yet we have no documentary records, and the derivation of the place-name remains obscure.

In c.654 Oswiu, king of Northumbria, succeeding his brother Oswald who had been killed by Penda in 642, came to urbs Iudeu to continue the contest with Penda and his British allies from Strathclyde. (Anderson, 1990, 15-16, quoting Historia Brittonum ). Here Oswiu at the atbret Iudeu (restoration of Iudeu) had to give back to Penda and the kings of the Britons who were with him all the treasures he had in the town. Iudeu and Giudi are evidently alternative forms of the same name. Jackson (1964, 37) explained that Bede’s Giudi ‘looks like an attempt to spell Iudeu; the G means the sound of Y in English Yes, as does I of Iudeu; the only real difference is in the -i in Bede’s form, which could be scribal corruption, or misinformation.’ Atbret Iudeu then is ‘the restitution of Stirling’ (Duncan 1975, 61). Oswiu then retreated south, apparently pursued by Penda and ‘thirty kings’. (Anderson 1990, 173, quoting Annals of Tigernach). Bede in his History of the English Church and People says ‘In the middle of it [the eastern estuary, the Firth of Forth] stands the city of Giudi’. (Penguin Ed, 1968, 51). After this (Bede completed his History in 731) we hear no more of Giudi or Iudeu.

Skene in the 19th century identified Giudi with the island of Inchkeith, on the grounds that ‘in the middle’ of the Firth of Forth meant literally in the midst of the sea. This seems an unlikely situation for an important town of that time. Graham (1959, 64) called this conclusion ‘absurd’.

Hunter Blair (1947, 27-8) suggested that rather than the identification of urbs Giudi with Inchkeith as proposed by Skene (which is still adopted in the Penguin revised edition of Bede’s History of the English Church and People, 1968, 51), it could be either Cramond or Inveresk. Graham (1959, 64) argued that Bede himself was not familiar with the Firth of Forth, but had to depend on reports of others, especially sailors. He might well have visualised the Firth to be a more or less triangular opening, and in medio sui would mean ‘at its apex’. ‘Stirling, placed as it is at the highest point that a seaman would be likely to reach, would exactly fit this meaning’.

Professor Jackson (1964, 36-8) strongly supported Graham’s view that urbs Giudi represents Stirling. The identification was accepted by the Ordnance Survey for their map Britain in the Dark Ages, 2nd Edition 1966, and by Professor Duncan (1975, 61) - ‘… urbs Iudeu, pretty certainly a fort on Stirling rock’. Jackson (1981, 1-7) returned to the question to refute the alternative suggestions of Hunter Blair, and after thoroughly discussing the problem concluded that beyond reasonable doubt Stirling was the location of Bede’s urbs Giudi. Alcock (1981, 176) cautiously agreed - ‘If Giudi was indeed located at Stirling, then clearly urbs Giudi must be Stirling Castle Rock’. Alcock reserved judgement because no traces of early fortifications have yet been found beneath the medieval and later castle. The discovery of traces of banks on the Gowan Hill, between the Castle Rock and the Mote Hill, suggests that a sizeable settlement, perhaps an oppidum, existed there. Several banks on the north-eastern slope would seem to indicate that this was not merely a cattle enclosure - that would require only a single bank and ditch, or a simple palisade. It is perhaps too much to hope that excavations on the summit of the hill would reveal traces of structures. The soil cover of the rock is very thin, and has been subject over the ages to a great deal of disturbance.

The traces of fortifications, probably iron age or dark age, perhaps both, at Stirling do not in themselves provide evidence for urbs Giudi, but they do strengthen the argument for an important settlement there beside the Mote Hill, itself a significant feature. In conjunction with consideration of the geographic situation outlined by Graham (1959, 63) the case for urbs Giudi being the precursor of Stirling is strengthened, though the evidence is by no means conclusive.

The discussion regarding Giudi and Iudeu will benefit from being placed within a wider context. It has not previously to my knowledge made a connection with Niuduera regio, in spite of the similarity of the name and the evidence of geographical proximity.

Saint Cuthbert went from Melrose ‘ad terram Pictorum ubi dicitur Niuduera regio’ (to the land of the Picts, called Niuduera); (Colgrove 1940, 82-3, from the earlier Anonymous Life of St Cuthbert). This must obviously refer to his crossing the Firth of Forth, and the Book of Lecan, fo.43 bb, quoted by Anderson (1990, 127) refers to Saint Serf possessing Culross ‘itir sliab n-Ochel acus muir n-Guidan’ - between the Ochil Hills and the Firth of Forth. N-Guidan seems very like Niuduera, bearing in mind that G was pronounced as Y in ‘yes’. The Early Welsh equivalent of Iudeu occurs in the Gododdin. Chadwick (1963, 159, note 1) points out that ‘tra merin Iodo’ from Gododdin, 1209, means ‘beyond the Firth of Forth’.

Hunter Blair (1954, 166-168) quotes Bede’s version of the story (based on the Anonymous Life) of Cuthbert’s journey ‘… ad terram Pictorum qui Niduari uocantur, nauigando peruenit,’ and shows that the strong presumption must be that Cuthbert crossed the Firth of Forth to the *Niduera regio, the land of the Picts. He refuted the suggestions placing the Niduari in Galloway, and connecting their name with the Galloway Nith.

Wainwright (1955, 42-43) came to the same conclusion, apparently independently, using similar evidence and a quotation from an eighth-century poem, Miracula Nynie Episcopi, corrected by Levison to read ‘Pictorum nationes quae *Niduarae dicuntur’ (to the nations of the Picts called *Niduarae) to show that the ‘Picts of Galloway’ were a twelfth-century myth based on confusing Niduari with ‘of Nithsdale’ merely because ‘Nid’ and ‘Nith’ look alike. Niuduera then would not be Nithsdale but would be a large area near the Firth of Forth occupied by Picts. It could include as one of its settlements ‘urbs Giudi’ along with others such as ‘Nithbren’ (Newburn by Largo) in Fife referred to in this connection by Duncan (1975, 78 note). Wainwright made no reference, however, to Giudi and Iudeu and Stirling. We may consider then that ‘urbs Giudi’ means ‘a town of the Niduera regio’, or perhaps ‘the principal town of the Forth region’. This may well be Stirling, but other possibilities remain to be investigated; for example Clackmannan, undoubtedly an important Dark Age settlement, would also be within the territory of the Niduarae, and has not yet been considered, although the arguments adduced for Stirling being urbs Giudi would also apply here.

ACKNOWLEDGEMENT

I thank Simon Taylor for his comments on the draft of this paper. He correctly points out:-

‘Not too much weight can be put on the form *Niuduera as it is the result of heavy editorial emendation, and must therefore be used with great care. It comes from the text of the late 7th-century Anonymous Life of St Cuthbert, which reads as follows: “Alio quoque tempore, de eodem monasterio quod dicitur Mailros cum duobus fratribus pergens et navigans ad terram Pictorum ubi Mudpieralegis prospere pervenit. Manserunt ibi aliquot dies”. For Mudpieralegis read *Mudwieralegis, where p is clearly the result of a mis-copying of OE wynn, the letter representing modern English w which in form closely resembles p; see also Watson 1926, 176. Mudpieralegis is what Hunter Blair renders *Niuduera regio, although the first two syllables could equally well be read Nuiduier-. Also, it should be asked why Bede should use two such different forms for the same place.’

Mudpieralegis is a reading of Niudpæralegio that may be found in two of the seven manuscripts of the Anonymous Life used by Colgrove. Nevertheless *Niuduera regio was Colgrove’s preferred reading (1940, 82) in spite of the fact that Mudpieralegis occurs in the St Omer 267 MS (late 9th or early 10th century) that was his preferred MS (1940, 45) for the Anonymous Life. Niuduera regio was accepted by Hunter Blair and by Stenton 1971 in his Anglo-Saxon England. Colgrove 1940, 51 explains why he rejected Mudpieralegis. He calls it one of the ‘extraordinary forms’ of the place-names in that MS. Bede based his Life of St Cuthbert on the Anonymous Life, but we do not know what MS he used. Could Mudpieralegis not easily be explained as scribal mis-copying? This topic requires further investigation.

Examples of OE wynn from the Liber Vitae, New Minster, Winchester, 1031. Wulfgar is in the middle of the extract; Godwine immediately above. Thanks to Anglo-Saxon Index, Trinity College, Cambridge:-
<http://trin.cam.ac.uk/sdk13/assms.html>

REFERENCES
Alcock, L 1981 ‘Early historic fortifications in Scotland’ 150-180 in G Guilbert (ed), Hill-Fort Studies. Leicester.
Anderson, A O 1922 Early Sources of Scottish History, A D 500 to 1286. 1990 edn., with corrections by M Anderson. Stamford.
Bede A History of the English Church and People Penguin Classics, 1968 ed. Translated by Sherley-Price, revised by Latham.
Chadwick, N K 1963 ‘The Conversion of Northumbria’ 138-166 in N K Chadwick (ed), Celt and Saxon; Studies in the Early British Border. Cambridge.
Colgrove, B 1940 (ed), Two Lives of Saint Cuthbert. Cambridge. Repr.1969, Greenw3ood.
Digney, S 1995 ‘Gowan Hill: Enclosure (Oppidum?)’ Discovery and Excavation in Scotland, 16.
Duncan, A A M 1975 Scotland: the Making of the Kingdom. Edinburgh.
Graham, A 1959 ‘Giudi’, Antiquity 33, 63-65.
Hunter Blair, P 1947 ‘The Origins of Northumbria’, Archaeologia Aeliana, 4th series, 25.
Hunter Blair, P 1954 ‘The Bernicians and their Northern Frontier’ in N K Chadwick (ed), Studies in Early British History.
Jackson, K H 1964 ‘On the Northern British Section in Nennius’, 20-62, in N K Chadwick (ed), Celt and Saxon: Studies in the Early British Border. Cambridge.
Jackson, K H 1981 ‘Varia: I. Bede’s urbs Giudi: Stirling or Cramond?’, Cambridge Medieval Celtic Studies, 2, 1-7.
Page, R 2003 ‘Early Historic (Dark Age) Stirling: Was the Gowan Hill Bede’s Giudi? Forth Naturalist and Historian, 26, 97-104.
Wainwright, F T 1955 ‘The Picts and the Problem’ 1-53 in F T Wainwright (ed), The Problem of the Picts.

Ron Page, ‘Kingarth’, Airthrey Road, Stirling FK9 5PH

[Editor’s Note: Look out for a forthcoming article by Dr James Fraser, Department of Scottish History, Edinburgh University, entitled ‘Bede, the Firth of Forth and the Location of Urbs Iudeu’. In it he argues strongly against Urbs Iudeu/Giudi being Stirling, and proposes a situation considerably further east - but not one of the usual suspects. More details will be given as soon as the place of publication has been confirmed.]

Jakob Jakobsen Conference

As announced in a previous Newsletter, an international conference was held in Shetland at the start of May to celebrate the life and work of the Faroese philologist, Jakob Jakobsen, whose impressive study of Shetland place-names will be known to many readers of this Newsletter. The conference attracted approximately 70 delegates from the Faroe Islands, Shetland and elsewhere. Several speakers, including Gillian Fellows-Jensen from Denmark, praised Jakobsen’s skills as an onomastician but he was, of course, a multi-talented scholar who contributed at a very high level to various academic disciplines and, in particular, to dialectology and ethnology/folklore studies, and this breadth of scholarship was reflected in the various papers read at the conference.

It was very pleasing to hear Eileen Brooke-Freeman of the Shetland Amenity Trust describe the work of the Shetland Place-Name Project and to see the results of their database work being presented in such a convincing manner. The use of GIS software in conjunction with the database is particularly impressive and it is to be hoped that the Scottish Place-Name Database can benefit from the Shetland experience. The Royal Commission on the Ancient and Historical Monuments of Scotland (RCAHMS) is currently exploring possibilities for the future development of the Scottish Place-Name Database and will be kept informed of progress made and problems encountered during the pilot study in Shetland. The Shetland Amenity Trust has plans to make the Shetland material available on its website.

I learned a great deal about Jakob Jakobsen and his work during the conference in Shetland and I enjoyed assembling some dialect and place-name material for my own paper. I left the conference with a very strong impression of the depth and extent of Jakobsen’s impact, not just on the academic world but on the lives of Shetlanders and Faroese who have carried his knowledge of language and toponymy with them into the twenty-first century through reading his major publications. The conference organisers, Turið Sigurðardóttir and Brian Smith, deserve praise and thanks, as do the many Shetland people who contributed to making the conference such a memorable experience. Newsletter readers who were unable to be present at the conference itself can look forward to publication of the conference proceedings in due course.

Doreen Waugh

A Fiendish Puzzle

While doing research for my dissertation on hill-names in southern Scotland I came across a mention, in Pennecuik’s early 18th century notes, of a list of what he called the ‘chief’ hills in Peeblesshire including one ‘Fiendsfel’. There’s no hill of that name today. I did wonder if it had any connection with one shown on Blaeu’s Atlas map of 1654 (based on Pont’s 1590s surveys) which he recorded as Filfell, and which from its position is the hill now known as Culter Fell, above the hamlet of that name.

Then, reading a description of a river north-east of the hill, in the MacFarlane Geographical Collections, the mystery of its location was solved:

“Holms Water . . . Upon the head of this fertile Water, above Glenkirk, is a mountain called Fiendsfell (before mentioned), [where] the eagle hath nestled past memory of man.” (Vol. III, p.152).
A glance at the OS map shows it must indeed be the current Culter Fell. But why was the name changed? There is another hill, in England’s northern Pennines, also formerly called Fiendsfell, but changed to Cross Fell (still its name) in the early 17th century. Clearly this was an attempt to exorcise or ‘christianise’ the hill’s name, and the same process must have taken place in Scotland too, if a century later. Perhaps the specific ‘fiend’ was shortened to ‘Fil’ by Pont (or by his local informants) as an attempt not to invoke the devil’s name - or at least that of one of his henchmen - in common with the spirit of the age.

Culter Fell is a fine hill, and I climbed it this summer again, from yet another direction, in an ascent that was pleasant and certainly not fiendish. But its modern name is rather a poor substitute from its former one – even Cross Fell has a least the effect of a photographic negative’s representation of the Fiend: how would the good people of contemporary Coulter feel about a fell with fiends?

Peter Drummond (text and photo)

The Norse in Islay - the Place-Name Evidence

While the majority of Islay place-names are Gaelic in origin, a significant minority can be considered Norse. The introduction of names such as Olistadh (from ON *Óla(fs)staðir, ‘Óli’s/Óláf’s farm), Campa (from ON *Kambr, ‘comb, crest, ridge’) and Stremnish (from ON *Straumnes ‘headland of the current’) can, of course, be traced to the Viking Age. What this represents in terms of settlement history, however, is not quite so certain.

With no detailed references to Islay or its peoples surviving from the years between c. AD 700 and c. 1200, most explanatory models have concentrated on the relative numbers of Norse and ‘native’ farm-names on the island. Since 1882, and Captain F.W.L. Thomas’ seminal article ‘On Islay Place-Names’, the relationship has been taken as 1:2. As this is considerably lower than Orkney, where almost all farm-names are Norse, or Lewis, where the ratio is greater than 4:1, Islay’s Norse nomenclature is usually attributed to ‘seasonal exploitation’ by transient ‘Vikings’ or their small-scale assimilation into an otherwise stable Gaelophone society. Where the possibility of independent Norse settlement has been entertained, this is usually assumed to have been economically or spatially peripheral to that of the Gaelic-speaking majority.

By relying so heavily on unqualified statistics, however, these models have actually obscured important diagnostic nuance in the context, typology and inter-relationship of the place-name material.

When the location and relative agricultural potential of Islay farms are examined in detail, it is clear that Norse names are no more indicative of poor-quality land than their Gaelic counterparts. Neither are they restricted to coastal, regional or any other kind of enclave. In terms of Olsen’s ‘User Group’ theory, the implantation and survival of so many Norse names in situ points to the previous existence of a stable, Norse language user group covering the whole island. An island-wide incidence of Gaelic place-names containing Norse ex nomine onomastic units – eg. Gleann Egedale (from G ‘the valley of *Egedale’ < ON Eika(r)dalr ‘oak valley’) and Eas Forsa (from G ‘the waterfall of *Forsa’ < ON *Forsá, ‘waterfall river’) – suggests, moreover, that widespread Norse language use only later gave way to Gaelic (Figure 2). By way of contrast, the absence of Norse names containing Gaelic ex nomine onomastic units points to a rather more dramatic linguistic transition at the beginning of Islay’s Norse period. Given the fundamental connection between land and status in early Gaelic society and the central place of ethnic delineation in Viking Age Europe generally, the most straightforward explanation would be a process of ‘ethnic cleansing’ by an incoming Norse-speaking population followed by their (eventual) adoption of Gaelic speech.

While this scenario is hardly supported by Thomas’ place-name ratio, it should be noted that his studies were based on the then current County Valuation Rolls. Over the millennium or so which separates this material from the period of Norse name-giving, there have been numerous historical developments with the potential to impact on the local nomenclature. Indeed, when Islay’s place-names are examined individually, there are indications that many Norse names have been translated or otherwise adapted by speakers of Gaelic (e.g. Bun Abhainne from ON *Áróss ‘River Mouth’ and Beinn Tart a’Mhill < G beinn + ON *Hartafjall, ‘Stagfell’ respectively). Others still are likely to have been replaced outright. The over-representation of diagnostically ‘late’ (i.e. post-Norse) Gaelic Baile- names, for example, may reflect a period of administrative re-organisation and re-naming following the advent of the MacSorley Kings of the Isles in the late 1100s. Post-Norse neologisms such as these could, in theory, have replaced Norse or Gaelic material. But with the changes driven by speakers of Gaelic or (Scots) English – and not Norse – the overall effect will have been a steady dilution in the Norse component of Islay’s nomenclature.

This kind of onomastic realignment will have been particularly marked in the case of nature names. While the corpus of Islay farm-names has remained relatively stable since the introduction of written fiscal practice, there was no corresponding stabilising effect on the names of topographic features until the production of the first detailed maps in the late 19th century.

While patently or partly Norse nature names now make up only a small percentage of the total, they are invariably amongst the most dominant in every part of the island that has significant topographic features. Relative to its Gaelic nature names, many more of these can be considered Bygdens navn (‘names of the district’) which have survived periods of demographic change by virtue of their widespread user groups, and further evidence, therefore, of the dominant position once held by the Norse language in Islay.

The names of less significant features, on the other hand, are more likely to have been restricted to smaller, more localised user groups and thus, as Gårdens navn (‘names of the farm’), more susceptible to change. Episodes such as the Cawdor Campbell ‘plantation’ of Islay in the mid 1600s will have led to the displacement of many long-established communities, their nomen-clature and a radical alteration in the overall ratio of Norse to Gaelic names.

That is not to suggest that no pre-Norse place-names have survived to the present day. As yet, very few candidates have been identified with certainty. But as with the name Islay itself, we should probably regard these as having survived by virtue of adoption into the Norse nomenclature. While there is some evidence for continuity in estate boundaries and even administrative districts from the 7th century into the post-Norse period, it must nevertheless be considered significant that this did not involve the unadulterated survival of their pre-Norse names or the fundamental unit in that system, the Dalriadan tech or ‘house’. The Norse it seems were interested solely in the easy distribution of their newly acquired Islay land-holdings and not the preservation of local onomastic or administrative heritage.

Distribution of farm-names on MacDougall’s Map of 1749-51 by language background

Key

G Names which are fundamentally Gaelic in origin – i.e. independent Gaelic coinages, dependent Gaelic coinages where the ex nomine onomastic unit is Gaelic and dependent coinages in Scots English where the ex nomine onomastic unit is Gaelic

ONX Dependent coinages in non-Norse languages where the ex nomine onomastic unit is Norse. While almost all of these are formally Gaelic, there are also a few Scots English examples

ON Independent Norse coinages, or dependent Norse coinages where the ex nomine onomastic unit is Norse

U Uncertain

To further understand the process of Norse settlement in Islay, we must look to the historical and archaeological records. On the basis of the place-name material, it would seem more productive to approach this from the perspective of native annihilation.

Dr Alan Macniven (drawing on research for his recent PhD)

Place-Names of the Cairngorms National Park

The Cairngorms National Park Authority (Ùghdarras Pàirc Nàiseanta a’ Mhonaidh Ruaidh) has recently published an A2 sized folded leaflet with this title. This is no run-of-the-mill ephemeral leaflet for tourists, and its discussion of place-names will be of great interest to many readers of this Newsletter. A fuller notice will be included in the next issue.

(Publisher: Cairngorms National Park Authority, 14The Square, Grantown-on-Spey PH263HG;
website: <www.cairngorms.co.uk>;
e-mail:<enquiries@cairngorms.co.uk>)

Mary Higham

Dr Mary Higham, whose death at the age of 70 in November 2005 was noted with sadness in the last issue of SPNN, was an exceptional historical geographer, toponymist and local historian. While the focus of her work was her native Lancashire, in her rigorous and multi-disciplinary approach to her subject, she had much to teach those working in these fields well beyond this geographical area. This is especially true for Lowland Scotland, which shares so much with north-west England, both topographically and linguistically. This was clear when she made one of her too rare visits north to give a lecture at the SPNS conference on names and boundaries, held at Perth in November 1998 (a summary of which appeared in SPNN 6 (Spring 1999), 2–3).

Mary’s practical, common-sense approach, coupled with her considerable scholarship and her formidable knowledge of her Country, its archaeology, its history, its landscape, its place-names and its documentary sources, informed everything she produced with worth and weight. For Mary the landscape was a book which she had learnt to read with consummate skill. She wrote over 35 articles, all of which repay study by the Scottish toponymist and medieval landscape historian. They cover such subjects as linen-making, lead-mining, cattle- and horse-rearing, deer forest management, Norse settlement, roadways, bee-keeping and harpers, and all exhibit an admirable balance between desk-based and field-based study. But Mary was not only a scholar: she was also a passionate and dedicated teacher, and, ably supported by her husband Eric, she travelled the length and breadth of her Country giving talks and holding classes, especially under the auspices of the Centre for North-West Regional Studies based at Lancaster University.

Mary completed her Ph.D. as a mature student at the University of Lancaster in 1992. Entitled ‘The effects of the Norman Conquest on North West England, with particular reference to the Honors of Hornby and Burton-in-Lonsdale’, it analysed in minute, inspired, and inspiring detail a c.200 kilometres-long medieval boundary defining land belonging to the abbey of Barrow in Furness. She clearly demonstrated that these marches defined a 7th-century unit of land which Bishop Wilfred had given to the Church, the regio Dunutinga, itself based on an even earlier British territorial unit, a classic example of the longevity of political boundaries in the landscape.

Mary was treasurer and membership secretary of the Society of Name Studies in Britain and Ireland for ten years, and the huge investment of time and energy which she and Eric put into this task contributed in large part to that Society’s now flourishing existence.

Simon Taylor

Gaelic Lesson on a Sign

This sign is beside the farm of Lorg, the highest in the valley of the Water of Ken, Kirkcudbrightshire. Public roads stop some 3km short of a connection between this scenic but little visited valley and the equally fine and remote valley of the Scaur Water in Dumfriesshire. The ‘Lorg Trail’ is sign-posted as a link over the watershed for walkers. The two words represent a dictionary translation, plainly reflecting use of the track centuries ago.

Spring 2006

Scotland's Oldest Place-names
Turning the Tide on Orsay
Obituary: Dr Mary Higham

Andrew Breeze: Scotland's Oldest Place-names

Scotland’s oldest known place-names occur in the works of Tacitus, Ptolemy, and other classical writers. The meaning of many of them has long been clear, as with Blatobulgium ‘flour-sack’, the Roman fort (with granaries) at Birrens near Lockerbie, or Deva ‘goddess’, the rivers Dee of Galloway and Aberdeenshire. Current knowledge of the subject appears in the classic Place-Names of Roman Britain by A. L. F. Rivet and Colin Smith(1979). But there is still much work to do, as many forms defy explanation, and is has been doubted if some of them are Celtic at all.
To tackle these problems linguists have two weapons in their armoury: comparison and then emendation. Comparison is with the later Celtic languages, especially Welsh, because this is the closest living relative of the British and Pictish spoken in early Scotland. Emendation here means a willingness to emend ancient texts following comparison with the later Celtic languages. Scholars too often put blind confidence in the capacity of a scribe to reproduce accurately the ancient toponyms of Scotland, even though he was faced with (to him) strange and unfamiliar terms, where, as Americans would say, corruption was an accident just waiting to happen.
Of course, some forms are accurately preserved. Examples include Ptolemy’s Abravannus and Virvedrum. Comparison of the first with the Welsh prefix afr- ‘very’ and gwann ‘weak’ provides a meaning ‘very weak one’, and lets us identify this stream as sluggish Piltanton Burn, in flatlands near Stranraer. Roman coins found at the mouth of this stream imply that Mediterranean traders knew it. Similarly, comparison of Virvedrum with Welsh gor- ‘very’ and gweir ‘bend’ offers the meaning ‘very sharp cape’. Virvedrum is known to be Duncansby Head, by John o’ Groat’s, so the explanation well suits a dramatic change of direction in north-east Caithness. It means we can rule out the explanation ‘very wet cape’ proposed by Rivet and Smith.
Other names are harder. Here emendation is called for. One instance is Ptolemy’s Verubium, which is Noss Head, by Wick. Nobody has made sense of it as it stands. But emendation to *Verudium, in the light ofWelsh rhudd ‘red’, gives a meaning ‘very red cape’, which is apt for a headland of Old Red Sandstone. A trickier instance is Mons Graupius (identified as Bennachie in Aberdeenshire), below which Agricola defeated the Caledonians in the September of AD 83. W. J. Watson long ago proposed an emendation here to *Craupius, just as the Galgacus also given by Tacitus must be emended to Calgacus ‘swordsman’ (compare Middle Irish colg ‘sword’): a fitting name for the brave leader of the Caledonians. (Similarly, the Boadicea of Tacitus must be emended to Boudica ‘victorious one’, a fitting name for the brave queen of the Iceni.) Yet we can go further. *Craupius still makes no sense. But *Mons Cripius ‘crested mountain’ (compare Welsh crib ‘bird’s crest, cock’s comb’) does, as it aptly describes the profile of Bennachie, with its profile of five conspicuous summits in a line. There is little doubt that Agricola defeated Calgacus below a mountain that the Picts called *Cripius ‘crested one’, because it looked like the comb of a cock.
Having seen what can be done with research already in print, let us look at four other names which have never been explained. These are Bodotria, Boresti, Caelis, and Taexali.
Bodotriais Tacitus’s name for the River Forth. Yet Ptolemy refers to it as Boderia, while the Ravenna Cosmography (a late source) calls it Bdora. Nobody has reconciled the forms. However, the German philologist J. K. Zeuss (d. 1856) linked Bodotria with Middle Irish búaidrid ‘disturbs; stirs up,muddies, makes turbid (of water)’ and Welsh budr ‘filthy, dirty’. The sense would be ‘dirty river’. It is a pity this was overlooked by Rivet and Smith, because with modification it provides an answer, as we shall see.
If we reconstruct the name as *Boudra ‘dirty one’ (cognate with Welsh budr filthy’), which in the first century or so AD became *Bodra, this offers a solution. The middle part of the Forth winds through Flanders Moss, once a semi-swamp and formidable obstacle that has determined Scotland’s history; while the lower reaches of the Forth have extensive muddy tidal flats. The Forth was a dirty swampy river, difficult for armies of all ages to cross. But how do we reconcile Bodotria in Tacitus, Ptolemy’s Boderia, and the Ravenna Cosmography’s Bdora? Now, a name *Boudra ‘dirty one’ would give *Bodra ‘dirty one’. With scribal metathesis, this is what the Ravenna Cosmography actually has. So corruption in Tacitus and Ptolemy probably began with Bodria, proper and other nouns in -ia being common in Latin, but unlikely for a British hydronym. Bodotria and Boderia would result from that. As regards the Forth, therefore, editors of Tacitus may now choose to read *Bodra ‘dirty river’ in their text, and can rule out Bodotria.
Nevertheless, this *Bodra has nothing to do with the modern name Forth, where there is another solution. The Welsh knew the Forth as Gweryd, which also means ‘earth, soil; clod’. In this context Gweryd makes sense as a name of Flanders Moss, made up of spongy earthy humus; and the form that gives Gweryd in Welsh would also give Forth. If so, a regional name would have became a rivername, much as Liffey, originally meaning the plain west of Dublin, became used of its main river.
The Boresti have been a mysterious people and Boresti a problematic form. But, as we know Agricola defeated the Caledonians some eighteen miles from modern Aberdeen, we can be sure they lived on the plain beyond that, north of the Grampians. Their name is surely corrupt. The error is likely to be its unCeltic initial B-. If we read *Roresti, light starts to dawn. British Ro- (sometimes raised to Re-) is a common intensive prefix. (Perhaps a scribe of Tacitus was confused
by thoughts of Boreas, the North Wind.) If we then turn to dictionaries of the Celtic languages for corresponding forms beginning with cognates of ro-, we find one only. This is the well-attested Old and Middle Irish ruirthech, translated as ‘strong-running, impetuous’. All follow Kuno Meyer (d. 1919) in deriving this from the intensive ro + reth- ‘run’. Ruirthech ‘fast runner’ was the original name of the river Liffey. The first elements of ruirthech have a Middle Welsh cognate in rhyred ‘rush, haste; excess, presumption’.
If emended *Roresti meant ‘hastening ones, eager ones, impetuous ones, those running forward (in battle)’, it would parallel other defiant and warlike appellations in the Celtic languages. The concept is shown in the Gaulish stem cingo- ‘to go, to step’, seen in Excingus ‘he who steps forward, warrior, attacker’, and in a derived sense in Vercingetorix ‘great king of warriors’. It may also be found in Durotrages, a warrior people of south Britain, who built Maiden Castle. Their name appears with this spelling on an inscription from Hadrian’s Wall, and apparently means ‘swift-footed ones of the stronghold, warriors of the fort’, with the element -trag- seen in Gallo-Latin vertragus ‘swift-footed one, greyhound’. The same concept may occur in Ptolemy’s Venicones, a people living between the Tay and the Mounth, if their name means ‘hunting hounds’, the people seeing themselves as fierce fighters, who made enemies their prey by dashing after them and rending them to pieces. The -st- of proposed *Roresti ‘impetuous ones’ offers no objection to this etymology. Professor David Ellis Evans of Oxford quoted Gaulish Atrestus ‘good runner’ as one of many forms of -ret(t)- ‘run’. So there is reason to add *Roresti to the many formations meaning ‘run’. There is also a case for taking Begesse in the corrupt Ravenna Cosmography as a corruption of *Roresti, perhaps from *Reressi. But there is no link with the town of Forres (its name of Gaelic and not Pictish origin), despite what is sometimes said.
Now for Caelis. Ptolemy’s ‘mouth of the river Caelis’ is Banff Bay, where the Deveron enters the sea. The forms and meaning of Caelis have been puzzling. Yet there appears a simple answer. Ptolemy’s original in -ai- indicates a link with Welsh coel ‘belief; omen, portent’. Caelis may hence be explained as a Celtic form in Kail- meaning ‘portentous (river), prophetic (river)’. For the Taexali, the Deveron would have been sacred. It was thus like the Devona ‘great goddess’ or Don, also in their territory, as well as the Deva ‘goddess’ or Dee just south of it. The Deveron is useless for navigation but has excellent trout and salmon fishing. If the Taexali thought it also had prognostic qualities, it would have an equivalent in the Dee of north-east Wales. The Welsh thought that, as it shifted more towards Wales or England, it showed which nation would have success in battle. The Taexali may have attributed similar powers to the Deveron. When war threatened, their leaders may have thought hard on what wise men told them of the river's movements.
So we may conclude that Caelis is correctly given by Ptolemy; that it is purely Celtic; and that it means ‘(river) providing omens, portentous (river), prophetic (river)’. There may even have been a temple or ritual site on its banks. In this context may be mentioned from Scotland the cauldrons and other items sacrified there to water deities, and from Wales the scores of bronze, iron, and wooden objects found in Anglesey at Llyn Cerrig Bach, including weapons, harness, chariot fittings, slave-gang chains, currency bars, parts of a trumpet, ornamented bronze plaques, and a shield boss. These, deliberately consigned to the lake, indicate the kinds of item that perhaps also vanished below the Deveron’s waters.
Finally, Taexali and Taexalorum Promontorium, located in the north-east of modern Aberdeenshire. Form and meaning have been obscure, but a new approach may offer a solution. There are two pieces of evidence. The first is Smertae, the name of a people on the border of Ross and Sutherland, understood as those ‘smeared’ with enemy or sacrifical blood. The other evidence is Middle Irish taesc ‘jet, spurt, flow (of blood, etc.)’, taescach ‘act of pouring, gushing’, and taescaid ‘pours out’. If the Smertae were known as smearing themselves with blood, the Taexali may have given themselves out as shedders of blood, using a proto-Pictish root with a cognate in Irish taesc ‘jet (of blood)’. Irish saga shows the force of taesc and its related forms. They occur when a prophetess sees three naked men in a vision, ‘their spurts of blood (flowing) through them’; when ‘spurts of blood’ run down warrior limbs, or blood foams and bubbles, ‘welling up in spouts’; or injury means that there ‘gushed out’ heavy, thick blood from the body of a hero. If the above comparison is valid, it vindicates the reading Taexali. The variant Taezali can be rejected. We can be sure the Taexali ‘shedders of blood, those who make blood spurt’ were Celts and not some pre-Indo-European survival. We can also be sure that their ruling class delighted in battle, their name being warlike. It was this people that Agricola fought in the September of AD 83, and, if Taexali is currectly understood as ‘Bloodshedders; Killers’, it is thus no surprise that the Romans in the end set their frontier far from the lands of this brave and ferocious people of ancient Scotland.

Dr Andrew Breeze, University of Navarre, Pamplona (a shortened version of a talk to the Dollar conference, Nov. 2005)

Alan MacNiven: Turning the Tide on Orsay

Orsay: OS grid reference NR 163 515

oilen Eorsaigh (c1380) Insula Sancte Columbe de Ilanorsa in Iley (1507) Insula Sancte Columbe (1509) Oversay (1549) Kealsa, Killaglan and Ilandoursa (1631) Oversa (1654) Elistereyrarach, Illandowrrsay, Balladalie, Corieskallag (1686) Wester Elister, Balygawly, Archally and Island Oversaw, Easter Elister (1741) Isle Noresay (1749)

From George Langlands’ map of Argyllshire, 1801

The larger of two islets lying just off the SW tip of the Rhinns of Islay is now known as Orsay. Although the majority of the early recorded forms of this name derive from a G eilean (m) or ‘island’ coinage, the initial ‘oilen / Illan/ Island’ etc. must be seen as a later addition to a preexisting ON *Orsay. While the generic here is quite clearly ON ey (f), ‘island’, derivation of the specific has been a matter of some debate. The remains of a substantial medieval chapel on the island (RCAHMS 1984:255-6) led Maceacharna (1976:78) to suggest an ON rendering of the Gaelic saint’s name Oran, with the terminal /s/ presumably representing the common ON masculine genitive morpheme. As can be seen from the Crown rentals of 1507 and 1509, however, the earliest and indeed only known dedication on Orsay was to Columba and not Oran. While it is not impossible that all traces of a previous dedication to Oran had been obliterated by the Norse, it seems more likely that the descriptive element is also intrinsically Norse.
Perhaps the best known ON etymology is Thomas’ (MS) offering of örfiri (n) –giving the compound meaning of ‘ebb or tidal island’ (cf. Nicolaisen 1977-80:119). Even so, this interpretation appears to confuse the early forms of Orsay with those of Oronsay which are also recorded in the Islay material in conjunction with priory land-holdings. It also fails to take proper account of local topography. While Oronsay off the SW coast of Colonsay is a tidal island, Orsay off the SW coast of the Rhinns is not and cannot therefore be an *Örfirisey (cf. Gillies 1906:234-5). Gillies (1906:234-5) suggests a number of poorly attested alternatives including ON oðr, ‘a wood, woody’, orr, ‘a scar, a notch’, and óron, ‘mackerel’. An even better explanation, however, is provided by the name of a settlement on the adjacent mainland.

Orsay from the north, flanked by the settlements of Port Wemyss, Portnahaven and Ballymeanach.
(Alan Macniven)

While the island of Orsay is small and relatively infertile by Islay standards, the farm-district of that name is nevertheless listed as a 16s. 8d. or Auchtenpart land in the early rentals. Considering the substantial ‘extent’ of this holding, it seems likely that at least some of it was located on the adjacent mainland. With this being the case, special attention can be drawn to the names of nearby places. While the current fishing village at Port Wemyss is comparatively recent –dating to the third decade of the 19th century (RCAHMS 1984:302) –settlement at this site is likely to be much older. Significantly, its previous Gaelic name of Bun Abhainne (Maceacharna 1976:122) means ‘mouth of the river’ –with the river in question, Abhainn Gleann na Rainich, virtually bisecting the southern end of the Rhinns. Given the local pronunciation of Orsay, with a clear vowel sound between the [r] and the [s], it is possible that the original form of the island-name was ON *Áróssey, ‘the island by the mouth of the river’, and that the name of the mainland settlement was ON *Áróss – preserved in the Gaelic translation Bun Abhainne. The importance of Abhainn Gleann na Rainich in the local nomenclature is also commemorated in the name of the neighbouring settlement, Portnahaven, from G *Port na h-Abhainne ‘port of the river’.

John Thomson’s more precise map of 1820 shows clearly the position of Orsay opposite the burn mouth.
With thanks as ever, for these extracts, to the National Library of Scotland for providing online maps.

Farm-names derived from ON *Áróss are not uncommon in Norway (Sandnes & Stemshaug 1976:358). Norske Gaardnavne lists 6 examples (2 in Akershus Amt and 1 each in Bratsberg, Buskerruds, Jarlsberg og Larviks and Stavanger amt), noting that the initial vowel is often transformed to /o/ by the operation of vowel assimilation (Rygh 1989:22). As Marwick (1952:184) observed that initial ON /ár/ becomes /or/ in Orkney, and Jakobsen (1936:18) that it becomes /or/ and /wor/ in Shetland it can be assumed that the initial /o/ in Orsay is indicative of the same phenomenon.

Alan Macniven

References:
Gillies, H.C. (1906): The Place-Names of Argyll. London, David Nutt.
Jakobsen, J. (1936): The Place-Names of Shetland. London, David Nutt.
MacEacherna, D. (1976): The Lands of the Lordship: The Romance of Islay’s Names. Port Charlotte, Argyll
Reproductions.
Marwick, H. (1952): Orkney Farm-Names. Kirkwall, W.R. Mackintosh.
Nicolaisen, W.F.H. (1977-80): ‘Early Scandinavian Naming in the Northern and Western Isles’ Northern Scotland 3
(1977-80): 105-21.
Norske Gaardnavne –See DOKPRO (Dokumentasjonsprosjektet) at http://www.dokpro.uio.no/ for online versions of Rygh’s Norske Gaardnavne; Diplomatarium Norvegicum and others.
RCAHMS (1984): Argyll: An Inventory of Monuments: Volume 5: Islay, Jura, Colonsay & Oronsay: The Royal Commission on the Ancient and Historical Monuments of Scotland, H.M.S.O.
Rygh, O. (1898): Norske Gaardnavne: Oplysinger samlede til brug ved matrikelens revision efter ofentlig foranstatning utgivne med tilføiede forklaringer af O. Rygh: Forord og Inledning. Kristiania (Oslo).
Sandnes, J. & Stemshaug, O. (1976): Norsk Stadnamnleksikon. Oslo, Det Norske Samlaget.
Thomas, F.W.L. Undated MS portfolio of notes taken during the production of:
Did the Norsemen Extirpate the Inhabitants of the Hebrides in the Ninth Century? PSAS 11 (1876): 472-507;
On Islay Place-Names PSAS 16 (1881-2): 241-76 -
kept in the library of the Society of Antiquaries of Scotland in Edinburgh.

[Editor’s Note: Alan Macniven has recently completed a PhD at the University of Edinburgh on Scandinavian settlement on Islay.]

Obituary: Dr Mary Higham