Linguistic geography in Scotland; or, You can’t take the baile out of Balerno
(“Anglian” used to refer to Lowland Scots, “Cymric” or “Cumbric” to the barely recorded P-Celtic language of Strathclyde.)
Gaelic certainly was spoken in parts of the East Central Belt, at some times, but the central issue of the history of Gaelic is the anglicisation of Strathclyde. Progress in our theme would follow from understanding this process, on which history has so far declined to speak. The substance of the Gaelic speech community (on both sides of the Gaelic Sea) has been poured into Glasgow and associated towns, and it doesn’t take too much imagination to plot an alternative course in which Glasgow would have been Gaelic in 1500, Gaelic in 1800, and Gaelic in 2000, providing the gaelophone metropolis which would have supplied the business opportunities, universities, newspapers, and publishing houses, and TV stations, to keep Gaelic at the level of Welsh (if not indeed of Dutch), as a European language constantly assimilating incomers (and threatening its neighbours). Explaining why this didn’t happen calls for a greater feat of imagination, in the absence of suitable evidence. There is, to begin with, no evidence that Glasgow ever went through a Gaelic stage between the Cymric moment and the Anglian (Scots) moment. Conversely, when urbanisation has been the central fact of Scottish geography over half a millennium or so, and the towns speak Scots or English, anglicisation hitches a ride on the back of urbanisation, and doesn’t need any new and special theory to found it. The history of Gaelic in Glasgow may bear significant resemblances to the history of Gaelic in Dublin, Manchester, Liverpool, and New York. Listing these names reminds us that the Gaelic realms provided much of the human capital for the rise of commerce and colonisation, and so of the modern world economy. The condition of their entry was shedding their language.
It was in the Central Belt that the struggle of languages in Scotland took place, and it is really in the western end that the interesting and puzzling process occurred. How could you have Gaelic in Galloway, Gaelic in Argyle, and a tongue of Anglian Scots in the flat country in between them? Why did the P-Celtic of the Kingdom of Strathclyde give way to Anglian and not to Gaelic? The best information we have is that there was a rainbow trout pattern, in this western end, in the Dark Ages: there was an Anglian monastery right in the west even in the 7th C. By “rainbow trout” I mean that a large-scale map would show a whole mixture of Cumbric, Gaelic, and Anglian even within a few miles of each other. Big, smooth panes of colour showing large areas of “Cumbric”, etc., are just not right. Poor communications kept this position stable, and of course good communications wiped it away, leaving one language the winner. The small size of the linguistic patches may be what the placename evidence is trying to tell us: not successive strata, but adjacent strata, with Cumbric, Pictish, Gaelic, Anglian (and sometimes Norse, maybe even “Iro-Norse”) in a tight “rainbow scale” pattern. It is very unlikely that Gaelic ever displaced Anglian from the west Central Belt. This “rainbow scale” pattern points to penetration by incomers of empty land during a stage of history where there was much empty land. Was this against the wishes of their neighbours? Maybe so, but today new arrivals would mean the price of houses and land going up, and maybe 1500 years ago having more neighbours meant more people to socialise with, to sell your products to, and to supply social company or husbands and wives. An empty parish can be a lonely place. The pattern points unambiguously away from “invasion + displacement”, the kind of thing that happened in Virginia.
It would be pleasant to trace the process by which Glasgow moved from a Cumbric-speaking settlement to a large Scots-speaking town, but data which would illuminate this process are missing. We may have a pattern whereby the rural population buys very little of its supplies, so that long-distance trade is low in volume. Conversely, trade involving money is almost all long-distance. The merchants have to spend much of their life on the road or in offices in other towns; they are also very closely tied to their counterparts along the trade network, and would incline to speak the same language as these other merchant groups. It may be this which led Glasgow to take up Scots while it was not the majority language of Strathclyde. The comparison would be English-speaking mercantile towns in Ireland, and their predecessors, Danish-speaking towns in Ireland. Glasgow, with a small Scots-speaking population surrounded by Cumbric speakers, would be like Hansa towns in Poland or Lithuania, small in population but speaking Low German, recognizable to all Hansa towns. Scots law had special codes for the burghs, so they were under different laws from the surrounding countryside (but the same law for all burghs, to make contractual agreements clear and binding). The burghs were in some ways like enclaves, colonies of each other in a countryside which was quite different from them. The burgh populations may have had strong family links with each other, assisting the credit process and driving towards linguistic uniformity as a by-product. Trust implied kinship, kinship trust, in those centuries. However, even this is extremely speculative. Any evidence that would catch a snatch of this history would be extremely precious.
We do seem to know a difference between the Norman landowners and the Scots-speaking merchants. These were different strata. The Normans were not traders, and the merchants didn’t speak French.
You can’t take the baile out of Balerno, but you also can’t take the burgh out of Edinburgh. Chewing on this, you always come back to the “rainbow trout” pattern.
Older books retain a theory that the whole of Scotland once spoke Gaelic, a myth, disproved by Nicolaisen’s work on place-names, which is still circulated by the nationalists, as politically motivated fantasy. It was based on a single charter, showing a Gaelic noble holding land in the Borders. He would represent a “superstratum”, and the charter has, in the context of a gaelophone administration in Edinburgh, nothing to say about the speech of the peasantry. It is virtually certain that an Anglian speech nucleus remained in the Lothians and Central and Eastern Borders throughout. It is very likely that this nucleus also survived in the rest of the Central Belt, although only as the neighbour of other language communities. Since we can see so many Gaelic placenames in formerly Gaelic areas, their almost complete absence from this “nucleus zone” is an almost certain indicator that such places never spoke Gaelic. It is most likely that the placenames-map records the linguistic past of the whole population of each neighbourhood; Norman or Gaelic nobles were a class apart. MacDiarmid or Douglas Young believed in a completely Gaelic past, but they wrote before the detailed research into placenames.
It is probable, not only that there was never a Gaelic-speaking town (until modern times), but also that commerce and commercial production were things uniquely associated, in Scotland, with Anglian speech and the Anglian ethnic group. The Gaelic realm was, probably, always committed to autarky, subsistence farming, handicrafts done by non-specialists, and regulation of exchange by “relationship obligations”, to do with kinship or feudal loyalty, rather than money and commerce. Indeed, where we have economic specialists in Gaeldom (for example, bards, musicians, or doctors), it is a matter of great interest, and should draw our persistent attention. The position of a nobility based on rent in kind is quite different from that of a nobility based on money rents, owning land but in a context where land could easily be alienated and there was a vigorous market in land. This difference explains many of the differences between Gaelic (and Welsh) poetry and English poetry.
Finlayson’s book on the identity of the Scottish nation makes much of a village called Temple, south-east of Edinburgh, which once had a Gaelic name, and points to Gaelic speech in that part. He makes far too much of it. The older view, that one piece of evidence could be expanded to form the homogeneous picture for a hundred square miles around, was based on a complete lack of data. Those wonderful maps where you have three bits of data for the whole of Scotland and show the whole territory in three flat planes of colour, without any gaps. The Lothians are full of Anglian place names, thousands of them. There isn’t a square mile without a few placenames. It’s quite certain that there was an inrush of Gaelic speaking farmers at some point, and that names like Balerno or Dalkeith record their arrival. There is a “flag” of Gaelic place names streaming out to the west of Edinburgh; they date back to a time when the Gael were politically successful, and expanding. Where they found rich land, they became rich – any other result would be surprising.
I think nationalist thinkers actually wanted flat homogeneity. The “rainbow trout” pattern wasn’t as emotionally rewarding for them. I suspect beautiful things tend to be a little bit simpler than reality – which is fine for a poet, but not acceptable as linguistic science.
A related subject is the discontinuity between Old Scots and Middle Scots.
Middle Scots is extremely close to Northern Middle English, far more so than 700 years of separate development would have given. The suggestion is that Old Scots died and a new dialect washed in from the south. That is, the pedigree is not:
West Germanic
Old Anglian
Old Scots
Middle Scots
But rather:
Old Anglian
|
-------------
Old Scots Old Northumbrian
|
(Danish)……….....Anglo-Scandinavian
| |
Middle English Middle Scots
| |
Modern English Modern Scots
That is, all modern Anglo-Saxon dialects derive from Anglo-Scandinavian. All experienced the same ruthless levelling of case and verb endings, and most of the same sound shifts; therefore, all descend form the same source dialect, marginal and revolutionary with respect to Anglian or Saxon. There is no trace of an “orthodox” descendant of Anglo-Saxon anywhere, also not in Scotland. No-one says “heo syndon”, all say “they are” in various accents. This does not imply the disappearance of the original population, either in Herefordshire or in Kent or the Lothians. It does imply a speech continuum, where changes diffuse just as changes in Gaelic diffuse all over the Highlands and Islands. Nicolaisen’s work also implies that “Anglo-Scandinavian” did not originate in Scotland; the placenames show that there was very little Norse settlement in Scotland, and it was heavily concentrated in areas neighbouring Gaelic-speaking areas, nowhere in contact with Anglian. (for example: Lewis, Caithness, Orkney.) Consequently, if we find that Middle Scots has thousands of Scandinavian words, this means large-scale linguistic influence from England. This import has been associated with the Normans imported by David I (chiefly), and their north English retinues, speaking Scandinavianized English. (surnames like Sinclair, Campbell, Montgomery, Bruce, Comyn.) The French-speaking nobles were few, and their speech transient (although they patronized some writing in French), but their retinues were more numerous.
This import is oddly symmetrical with the theory that the differences between Irish and Scottish Gaelic are due to Norse influence.
The Anglo-Scandinavian pidgin language has been called “Angle-Mangle”, and it has been suggested that a Gaelic-Norse mixed language existed in northwest England, called “Cumble-Mongrel”. Angle-Mangle may have originated in many counties of England.
Sketch for comparative Celtic-Anglian history
Conversely, the similarities between the traditions of poetry in the Celtic realms, while the order which everyone calls classical for those countries still flourished, justify the use of the word Celtic for the whole system. We should be cautious about this term, but Todd’s work on family structure does come up with a typological group around the Eastern Atlantic seaboard, for which no better word than Celtic exists, and which locates a basis for these symbolic acts in elementary and long-lived social structures. The type in question is that of the stem family, in an authoritarian variant in which one brother is the chief heir, and brothers reside together in the parental home. Todd also shows, then, a dividing line between “nuclear family” and “stem family” which is also the line between “English” and “Celtic”, as known to historians.
There is a functional similarity between the gift poem and the marwnad, namely that both are related to the transfer of property. The death of someone eminent does involve the transfer of significant landed property, as well as other rights, titles, and status. The whole funeral process involves considerable transfers of resources, of which the bestowal of grave goods, in the earlier period, and the payment to the court poet for the solemn commemoration of the event, are some part. The link between the property transfer and the creation of the poem does give us a hint why a social system with a different array of rules for transferring property might produce a different range of poetic genres, in which panegyric and praise scarcely feature. The essential feature of a money-based economy might be that the bounden link between possessing property and social status, between receiving transfers and partaking in a customary relationship, is broken; it is because these very formalised poems consist quite largely of enactments of high status, that the ceremonial poems occur only on one side of the dividing line and not on the other.
We have not mentioned the genre of religious poetry, but it is apparent that certain structures to do with praise, status assertion, submission, registration of power, etc. appear in religious poetry without losing continuity with secular psychological structures.
Examination of Middle English collections of lyrics reveals quite unmistakable praise poems (or passages thereof), but only in the guise of praise of the beloved, in the so-called aureate diction: loveliest under linen, etc. These conventions would seem to be borrowed from the poem of praise of a secular ruler, but they are cut off; there are no “panegyric to overlord” poems in English to substantiate them. The source panegyrics are not English, but something else – French, for example.