fc. ‘THE BRETON PLACE-NAME SURVEY’ (general editor with Gary German)

‘The Breton Place-Name Survey’ (or ‘Hanoiou-Lec’h Breiz Izel’ in Breton) is an ambitious project which aims to publish 42 volumes containing the local indigenous form of the 52,000 or so place-names of Western Brittany (Fr. La Basse Bretagne) along with cartographical and historical data. It is anticipated that the first volumes in the series will see the light of day in 2016. The Survey’s priority is the fieldwork and collection of the pronunciations of place-names in Western Brittany where the Breton language is still spoken, but only by a dwindling community of elderly native speakers.

The importance of collecting the traditional local oral forms of place-names may not seem so evident to toponymists studying English and French place-names, used as they are to the pre-eminence given to medieval written forms in studying etymologies. However, as in the other Celtic countries, the importance of ascertaining the pronunciation of place-names is more important in Brittany where the original Breton form is often disguised and distorted by official written forms which are anachronistic and frenchified to varying degrees. As elsewhere, historical forms are being assiduously collected by individuals and will, over time, lead to comprehensive collections but there is no urgency to this work. However, the same cannot be said of the traditional local oral forms of place-names which are more often than not ignored by toponymists, through lack of experience and expertise.

Many insights on early medieval Breton society can be gained from a study of surviving place-names. To give two examples: (1) the fairly numerous examples of Merdi ‘steward-house’ (comparable to the Maerdy of Wales); and (2) in the place-name Crec’h-metern (Saint-Fiacre, 22) we get confirmation that the well-known Old Breton term machtiern ‘local leader’ had an affected vocalism in Breton just as its cognates Cornish mytern and Welsh mechdeyrn.

A selection of contexts given here should suffice to illustrate the justification for the collecting of traditional local oral forms in Western Brittany.

1. Breton form obfuscated by an official form.

Saint-Vital (Plounévézel, 29) is known locally in Breton as Zant Waren, a saint referred to as Guarhen in the Life of Saint Corentin, and associated with the semi-legendary early medieval king Gratlon. He is found again in a frenchified form as Saint-Voirin (Le Cloître-Pleyben, 29). Official forms (even if Breton) sometimes have no relationship to the forms used in the locality, thus locally Roc’h-ar-burtul (Mael-Carhaix, 22) is Ribourti, and likewise Restgoaler (Spézet, 29) is Rest-ar-gwenteg.

2. Official forms are mere translations.

La Villeneuve (Plestin-lès-Grèves, 29) is Ar Gêr-neve locally. Vieux-Tronc (Plouyé, 29) is Ar Hef-koz locally.

3. Official forms are partial translations.

Bois-Château (Canihuel, 22) for Koad-ar-hesten ‘chestnut wood’. La Butte-du-cheval (Motreff, 29) for Lost Krec’henn-ar-marc’h.

4. Ambiguity of official forms.

Guermeur and Guerderrien (Glomel, 22) are, respectively, Ar Ger-veur and Gwerderien locally, in one representing kêr ‘settlement’ the other gwern ‘marsh’. In another parish Guermeur (La Chapelle-Neuve, 22) represents Ar Ver-veur ‘the great marsh’.

5. Official forms are medieval Breton.

Many of the official forms of Breton place-names go back to the late medieval period and this is reflected in the preservation of unmutated consonants as was the convention in the writing of Middle Breton. Examples are: Kermarzin (Plounévézel, 29) is Kervarzin; Les Mais (Callac, 22) is Lezvêz; Trémalvézen (Glomel, 22) is Tralvênn; Kerdaffrec (Spézet, 29) locally Kerzaoreg.

6. The Breton forms of French names obfuscated.

Thus Richemont (Cléden-Poher, 29) and La Fonderie (Poullaouen, 29) disguise an actual Breton Richimont and Ar Vondiri which are differently accentuated, differently pronounced and represent both a Breton phonological adaptation and a fossilised trace of a prior French pronunciations.

7. ‘Official’ Breton form is faulty.

Following a timid start in the early 1960s, Breton versions of place-names have become commoner on roadsigns, along with the sometimes rather desultory introduction of Modern Breton orthographical conventions for place-names resulting in hybrids. Examples of such ‘bretonisation’ are far from ideal. A common example is the failure to modernise the Middle Breton definite article am to ar in Kerampuilh for Kerampuil (Carhaix, 29) or in Keramborn for Keramborgne (Le Vieux-Marché, 22). In Modern Breton and in the mouths of native speakers in the respective localities these are Ker (ar) Puilh, Ker (ar) Born. What is more, since 1999 ‘Kerampuilh’, under that erroneous form, has been the site of the first Breton-medium upper high-school (lycée) and is the headquarters of Ofis (Publik) ar Brezhoneg, an officially-backed agency which gives a lead in the promotion of Breton in Brittany, not least in giving advice on the Breton form of place-names.

8. Places not noted on maps or located in lists.

As elsewhere in the world, there are many Breton place-names which do not appear on maps or even on official lists. Not all of these can be dismissed as microtoponyms.

The examples given above far from exhaust the way in which a knowledge of the local pronunciation is vital to appreciate the meaning as well as the culture attached to Breton place-names. Individuals with some familiarity with the official written forms of Welsh and Highland place-names in Britain will not readily appreciate the intensity of the frenchification operated on Breton toponymy in historical, cartographical and official documents since late medieval times.

Apart from the strictly linguistic justification for collecting oral forms, methodical place-name studies have generally suffered from being restricted to documentary investigations in libraries and archives at the expense of field investigations. This neglect of fieldwork often leads place-name specialists to commit gross errors in their interpretations, errors which can so easily be avoided by the simple precaution of consulting with local populations and benefitting from their knowledge.

Even though priority is accorded to the fieldwork campaign in unprospected areas (see below) it is proposed to immediately prepare the data assembled for publication by districts roughly equivalent to two-to-three cantons giving a 100-to-150-page volume (a canton is a well-established administrative area composed of anything between six and fifteen parishes). Apart from including IPA transcriptions of local place-name pronunciations, the place-name lists will include documentary forms from the Cassini map of the late eighteenth century, parochial cadastral maps dating from the early nineteenth century, twentieth-century official maps, as well as – as far as possible – recently bretonised forms on signposts. Where there are readily available sources for older historical forms these will be incorporated. The place-names will be placed in alphabetical order under the relevant parish and will be located to within less than 100 metres according to the officially-sanctioned UTM kilometre grid.

As the map below shows, something like 33 per cent of Breton-speaking area has been collected, but something like 66 per cent remains unprospected. The areas on the map correspond to the parishes (administratively termed communes).

The blue areas are parochial collections which have been published (Madeg; Denez; Rouz). The green areas are unpublished (Humphreys; Goyat; German; Cheveau) – it is Mikael Madeg, of course, who was the pioneer who completed the work of collecting place-name pronunciations in the north-western province of Leon and had them published. It is envisaged that these collections will be published alongside the results of the proposed fieldwork campaign described here (Madeg’s work will be re-arranged according to parish and within the Survey’s parameters). The large extent of green in central Brittany is the result of fieldwork carried out by my father Humphrey Lloyd Humpreys in the 1970s and 1980s, which includes both recordings and transcriptions. As a result of a stroke he suffered some years ago he cannot complete publication of his own personal archive. I intend to complete this.

Since 2001, much field-work has been carried out by the Ofis Publik ar Brezhoneg, completing collections of pronunciations for 55 communes (mostly in the south-western fringe of Cornouaille and around Brest) followed by an immense effort which has seen all the Vannetais sujected to eqnuries concerning place-name pronunciations (Loïc Cheveau has also succeeded in recording the pronunciation of the place-names in over half the parishes of the western Vannetais).

The urgency of a fieldwork campaign may be stated in blunt demographic terms. Due to the cessation of the transmission of the language in favour of French in the 1950s, native Breton speakers in Western Brittany have been declining at the vertiginous rate of at least 10 per cent each decade:

• 1950 77%
• 1960 66%
• 1970 55%
• 1980 44%
• 1990 33%
• 2000 22%
• 2010 11%

There are no official census language figures in France, so the figures above in italic font are retrocalculations based on an estimate for 1952 and on surveys carried out in 1999 and 2009 (see Fanch Broudic 2009 Parler breton au XXIe siècle (Brest: Emgleo Breiz): 33, 133). What this means in practical terms is that there is hardly anyone under 55 who speaks native Breton and that in a rural parish of 1,000 people a place-name researcher can only hope for more than 150 informants, and in a rural parish of 300 people not more than 45 informants. Neither will all the present-day Breton speakers have the same fluency in Breton; in general, those aged 80 and over (born in 1930 or earlier) will, obviously, be much better informants than those aged 60 and under (born in 1950 or later), since French was becoming the dominant vernacular all over Western Brittany by the late 1950s; and we find, in many cases, that even the oldest speakers have been increasingly using French as their immediate entourage becomes less Breton. The continuing and inevitable decline of native speakers of Breton, underlines the urgency of a fieldwork campaign to collect the Breton versions of Breton place-names and justifies the dramatic statement which follows:

By 2015 it will be very difficult and by 2020 it will be well-nigh impossible to assemble a comprehensive collection of Breton place-names in their original native form, and forever afterwards there shall be no resolution of this desideratum for Breton place-names…