C 2 BERGNE, La - Undistinguished spot, woodland, by side of back‑road from St. Pierre‑des.Tripiers to La Bourgarie. The graph with French ‑gn‑, ('n mouille' = /ñ/), points to Lang -nh‑; and thus to Lang bernhe 'alder‑tree', Alnus sp. This is discussed by Paul Fabre, 169; cf. Les Bergnes, Herault. The common Lozèrien surname 'Vernhet' or 'Vernhes' /berñe:/ presumably refers. The root is a VL loan from Gaul verna 'alder', cf. Ir fern, and CoBret gwern. Places so named are often damp or even swampy.

D 3 BIERENHES ‑ subsequently forms like 'Vernhes', influenced by above. The remarkable cluster of half‑a‑dozen massively walled dwellings at the base of the uppermost, 50‑ms‑plus, corniche, opposite, and a little SW from, Chapelle St‑Gervais.

Maurice Patras, SCM 294, was correct to identify this as the medieval mansus de Bierenhes, FG i.43, of 1306 or 1307. It comes in a list of properties, subject to an enquiry by dominus commissarius concerning royal or episcopal proprietorship. The list has: mansus de Rouzeries / m. de Fornello / et cum manso de BIERENHES / et cum tenemento Sancti Gervasii / et de las Bastidas / et cum lo Capinas de Sem‑Peyre (= remains at Arcs de St‑Pierre). Other early 14th‑cent, forms, FG, are: mansus vocatus de  BERENHES, i.48, territorium de BERENHES ii.2, 252. Grottes and smaller caves adjoining this, along the same shelf, are variously named by Patras in SCM: la Bauma de Vernhes ??, Bauma de Vernhieres or Vergnes. These variants may represent late 20th‑cent. spoken forms collected by Patras from the few locals aware of them.

The name certainly has nothing to do with verna /berñ‑/ 'an alder‑tree', and the locality itself, dry, precipitous and stony, would never support a distinctive alder‑grove. The possibility is that of a name with the root *bern-; the original spoken 'B' not the Lang /b/ for initial 'V' at all ‑ being palatalised in the medieval period, /bj/, realised in writing as 'Bi‑'. Cf. elsewhere on C. Mejean the doublet Bindous Biendoux possibly for an obscure 0cc. name for a wild cherry‑tree. The -es suffix of Bierenhes, Vergnes, etc., may be no more than a rather late and secondary plural, thought appropriate to a multiple site. One‑interesting, if tentative, explanation suggests itself. Gaulish nouns for substantial topographic features were subsumed in spoken VL and by the last part of the 1st millennium AD had passed into early. 0cc toponymy, if not into everyday speech; clear examples are NANT, from nant‑u 'valley', all the 'Combe, Coumbo' names, from *cumb‑a 'small rounded valley', etc. and the prefix seen in Lanuejols and other names, from lan‑on 'flat area'.

Common Celtic had a noun *bern‑a birn‑a represented by Old Irish bern "cleft, ravine, hole in the landscape". Though this is not attested in WCoBret as the daughter‑languages of British, it was argued by Jackson, LHEB 701‑5 (and this has been generally accepted that the early North British kingdom‑name, Bernicia, rests on the same stem in British; an adjectival bernaco‑, birnaco' full of gaps, full of mountain‑passes', giving a *Bernaccia 'land of mountain‑passes' ("which is a very good description of the Pennines", wrote Jackson). Later, in my Christianity in Roman Britain to AD 500, 312‑3, I argued that the 5th‑century British-Latin place‑name of St Patrick's father's villula along the Roman Wall, Bannauenta Berniae contains a known form 'Bannaventa' which is distinguished as 'of‑Bernia', and that this is latinised Brit *bernia = the Greenhead pass linking the North Tyne and the Irthing.

In favour of allowing a cognate Gaul word *bern-, passing into early 0cc via spoken VL as *bern, *bernha or the like, we have:

1. The fact that 1307 BIERNHES, representing an older *bernhe- /bjern /, is apparently the only instance of this word as a locality‑name on C,Mejean, and moreover occurs simplex, i.e. is not further distinguished by an adjective or proper name. In other words, Mas Biernhes was named solely because of where it was, whatever the date of the first buildings there.

2. The fact that the whole S‑facing edge of C.Mejean, along the r.Jonte, from Le Rozier east to Meyrueis and indeed on to Col de Perjuret,with many small 'ravins' Or seasonal stream-gullies, has only one proper gorge or ravine of any size. This is the present 'Ravin des Bastides', from Les Douzes and St‑Gervais, running N and up to the causse, some miles long and rising from 494 m on the D996 to 870 m at Campdoulens. As a feature, this must have had its own name.

It is suggested, then, that by the fifth‑sixth cents AD this major way up to the Causse, from Le Maynial and Les Douzes (not to say Le Rozier) to all the summit grazing, was called by a word of Gallo‑Roman origin, *berna a spoken noun taken into early 0cc; it clearly did not survive as a common noun, being eventually ousted by the widespread OFr ravine ‑ originally 'a violent water‑course' or simply by 'valley' words; but an early 0cc *bernhe < *berna was perpetuated by the establishment of a mansus on its W.flank. One supposes this could have been several centuries before Les Bastides, the (Le Rozier‑inspired ?) creation of ?Cl3, placed to control the defile or pass to the causse half‑way up.

Michelin road‑atlas index presumably Gaulish *berna enters into other place‑names of the Bern kind, but one would have to go through all the recorded cases. North French names in Biern are likely to represent Germanic personal‑names cognate with A‑S Beorn and Norse Björn This Causse Mejean instance, if it has been correctly identified and analysed, takes us back toponymically and archaeologically to a stage well before AD 1000.

(Additional note, June 2001  insert at p.42)

B 12 LA BASTIDE farm, 1911 pop.4

D 4 LES BASTIDES vill., 1911 ho.5 pop.9

The origin of this name is known, and peculiar. Latin speakers in Gaul from the 5th cent. inherited two common verbs for 'to build', construere ‑ whence our construction and facere 'to do, to make', widespread in Gaul and Britain on building‑stones of the type TEMPLVM A SOLO FECIT ‘he (name) built the temple from scratch, from the ground upwards'. However in the N, under Germanic influence (Franks), a verb bastjan 'to build' was borrowed into VL as *bastire whence later Anglo‑Norman bastir 'to construct, make', and modern Fr bâtir 'build'. In the S, where the same verb must have been current, the medieval survival was a past‑participial noun la bastida <Bastit‑a, 'a building, that which has been built'.

The broad explanation for French occurrences of La Bastide, simplex or in compounds, is ‑ so A. Vincent in his Toponymie de la France ‑ "a newly constructed village, fortified, enjoying certain privileges; mainly small villages founded from the 13th to 15^th cents. to replace abandoned localities or those in decay to restore life to waste areas. They were organised to bring together folk from various places, to whom were offered individual freedom, a home, land, even certain guarantees." This is too specific for the Midi where, as Fabre points out, the name is widespread, at least 28 in Gard alone, another 32 in Aude ‑ thes are minimum figures ‑ and where a bastide is often pretty humble.

LA BASTIDE stands on the very edge of the causse, above and a bit S from Florac; extremely isolated, the only farm on the long track going S from La Pradal to Villeneuve. The place, until c.1950 run from inland Mativet by the Puel family, is deserted and none of the few buildings looks older than c.l800. It is however mentioned, end of the 13th cent., in proceedings of a commissarius visiting Meyrueis to ascertain the tenancies and incomes of various C.Mejean properties, then claimed as part of le domaine royal. Raymund de Rayssocieriis was habitator of mansus de la Bastida which lay in the parish of Vebron and was grouped with mansi del Aulanos de Rodoneilo another de Rocapertus and the public highway from Fraxineto Fraissinet de Fourques, up to Florac (Floriaco) M del Aulanos may have been a vanished settlement NE of Mativet, NW of Valbelle, at present 'les Aulagnets' ‑ Lang aulanier <*aulana 'hazel‑tree'.

There may also have been a, now forgotten, 'bastide' somewhere on the La Parade ‑ Caussignac ‑ Coperlac NS road across the plaine, the Gallo‑Roman farming basin. Stephen & William de Valgalerio held a mansus seu nemus seu territorium a farm or grove or estat called de Ribalta ‑ Rivalte, generic name for this farming belt 'sive de la Bastida', "or otherwise (mansus) of La Bastida". It was in the same area as' Carnaco Avenx and nemus de la Savatge flow Carnac, Avens and Le Sauvage, the woods N of Avens, and the via publica going from mansus Aldebal Masdeval; we have a lost Bastide somewhere between present Carnac and present Mas St‑Chely.

More important is a reference, again around the end of the 13th cent., to a mansus de la Bastidas its association with other identifiable names ‑ m.de Bierenhes = Bauma de Vernhes, walled grotto: de Rouzerias = Mas Rouzieras, another one: lo Capinas de Sem‑Peyre = a house down by Arcs de St‑Pierre; de la Villa = La Viale: and tenementum Sancti Gervasii = Chapelle de St‑Gervais makes it certain that this is Les Bastides and that it was well established as a settlement by 1307.

A separate statement, same general date, records that one Berengarius de Ruppe, Or de Rippa, domicellus (a young gentleman, a junior squire!) , had acquired the mansus of Berenhes, whatever stood down by Bauma de Vernhes, from Bertrand of Capluc, knight; that Berenhes was in the parish of S Petrus de Stirpia (St.Pierre des‑Tripiers), and was confruntatus 'faced with, more or less opposite', mansus de la Villa, La Viale, and de las Bostadas a slip for ..las Bastidas, Les Bastides.

Though royal‑rights were being widely claimed on both Causse Mejean and Causse Noir in these records, at a local level there must be an implication that Les Bastides was founded, or renewed, at a strategic position and as part of a general re‑establishment of income‑producing farms along the SE part of Causse Mejean. The stimulus would either be secular, local lords of Capluc or of the predecessor of the chateau at Peyreleau, or religious, the priory at Le Rozier ‑ de Roserio, de Sancti Salvatoris de Rosserio 13th cent., In theory this process of exploitation could go back to around 1100. The name of St Pierre‑des‑Tripiers is recorded as de Stirpia before 1061, cartulary of Aniane, and before 1070, cartulary of Gellone or St. Guilhem le Desert, and still in 1298. Latin stirps means "the lower part of the trunk of a tree; the bole and roots; what is, left when a tree is felled or cut down".  The genitive plural of stirps is stirpium as in locus stirpium "a place of cut‑down trees, tree‑stumps". Some medieval Latin writer took stirpium as a new neuter noun, plural stirpia hence ecclesia S.Petri de stirpia “of all the tree‑stumps". Clearance of much of the surface of the causse here in the 11th cent., the date of the original church, probably started a process whereby the older dwellings, many down the ravines or below the couronne and making use of walled‑up grottes, were replaced with new built homes nearer the cleared upper areas; hence all the mansi (mansus Lang mas, is simply 'homestead of those exploiting the land around') in this part, like de Cassanas 1298, Cassagnes: de Elzacio Elzas 1307, Hielzas 'the heights': de la Vila, de Villa 1298, La Viale.

Les Bastides is thus quite likely to represent more than one actual building, starting in the latter part of the l3th century; conceivably a revival of a site used in Gallo‑Roman times, as many of these mansi were, because it possessed a suitable source of water and SW Or SE facing terraced plots, and was on a useable track down to the valley. Inspection, June 2001, suggests that the two‑storey building with arched doorways ‑ a building of at least two phases on the uphill side of the road opposite chez Waygood is the oldest surviving structure; whether, e.g., its back wall goes back to c.1300 is a difficult question to answer but internally it appears to have medieval features.