Towns, gardens and fences

Last week I went to Denbigh, a small town in the north east of Wales, to sing in a concert. On the way there there was some discussion about the origins and meaning of the name Denbigh. So I thought I’d find out more. The English name of the town doesn’t mean anything, but the Welsh name, Dinbych, means ‘small fortress’ – din is an old word for fort or castle related to the word dinas (fort; refuge; city), and bych is a variant form of bach (small). Related words include dinasfraint (citzenship), dinasol (civic, municipal), and dinaswr (citizen).

Din comes from the same root as the Irish dún (fort), the Scottish Gaelic dùn (fortress, heap), the Manx doon (fort, fastness, stronghold, bastion, earth fort, dun, fortified rock), and din (fortress) in Breton and Cornish – the Proto-Celtic *dūnom (stronghold) [source], which is cognate with the Proto-Germanic *tunaz/*tunan (fortified place), the root of the Old English tuun/tūn (an enclosure; farmstead; village; estate), from which we get the word town; and of the Dutch tuin (garden), and the German Zaun (fence, hedge).

The element -dunum in Gaulish/Latin places names, such as Lugdunum (Lyon) and Acitodunum (Ahun), comes from the same root, as does the element -ton in English places names such as Workington, Ulverston, Dalton and Warton.

The root of all these words is the PIE *dhu-no- (enclosed, fortified place, hill-fort), from *dheue- (to close, finish, come full circle) [source].

Here’s a Glossary of Welsh Place-Name Elements, and a Key to English places names.

7 thoughts on “Towns, gardens and fences

  1. A timely post, Simon. Only a week ago I was musing over the large number of Welsh placenames with the elements caer and castell (‘fort’, ‘castle’), both of Latin origin, yet I couldn’t think of any Welsh equivalent to the Gaelic element dún (e.g. Dundee, Dun Laoghaire). Dinbych (which I like to refer to as Dinbych y Sglodion, to distinguish it from D. y Pysgod) had occured to me as a possibility, but I was on a car journey at the time and had forotten about it by the time I got home. Thank you for confirming this.

    I’m not sure why the connection with dinas had never occured to me. Perhaps I was distracted by the similarity with dynasty (which, itself, could almost be a Welsh word, dinas + ty – fanciful, I know). There is a common semantic thread there – could the words be connected?

  2. Ha, I like Dinbych y Sglodion for Denbigh! There is a famous early Welsh poem about Tenby.
    Yes, I think dinas is a suffixed derivative related to din (dinas is masculine in older forms of Welsh). Din turns up in various place-names, such as the older form of the name for Edinburgh Din Eidin (with din subsequently being displaced by Caer in the modern Welsh form of the name). Bych is not so much a ‘variant form’ of bach as an ‘older’ form, preserved by being fossilized in a place-name (Proto-Celtic *bikk-, = Old Irish bec, Modern Irish beag), the usual Welsh form bach is actually from a variant, so to speak.
    You sometimes see ‘old’ forms of words or names being preserved by fossilization in placenames; e.g. old case endings, such as Cardiff, where Middle Welsh Kaer Dyf means ‘Fort of the Taff’, Tyf being a genitive of Taf (the Modern form Caerdydd is a later development).

  3. “Bych is not so much a ‘variant form’ of bach as an ‘older’ form, preserved by being fossilized in a place-name”

    There is also bychan/fechan, a common element in placenames in Wales and parts of Scotland that have come under Brythonic influence: Llanfairfechan, Ecclefechan (Scot.), Buchanty (Scot.).

  4. “Tyf being a genitive of Taf (the Modern form Caerdydd is a later development).”

    So Welsh once had cases marked by internal vowel changes, similar to Irish?

  5. David – dynasty has no connection with dinas. Instead it comes from the French dynastie, from the late Latin dynastīa, from the Greek δυναστεία (power, lordship, domination).

  6. @David, yes, in British and early Welsh changes such as i-affection occurred in certain cases, caused as in other languages by endings which subsequently were lost. For example, British *bardi ‘poets’ > Welsh beirdd, where the original plural ending was lost and the only sign of it is the i-affection of the vowel.
    As for singular genitives of nouns where this would have occurred, in most cases the singular forms have been levelled by analogy with the simple nominative. So it is suggested that, at least for a while, there was a genitive singular form **meib < mapi 'of the son', but with the development of a more fixed Old Welsh genitive word order the distinctive form was unnecessary, and so levelled to mab throughout the singular (and later on, the plural form meib got the ending -ion to doubly mark it as plural).
    Since, for o-stem nouns, this supposed genitive singular would have been identical in form with the nominative plural, in place names it is hard to determine whether you have a genuine preserved genitive form, or simply the plural. Some scholars, such as John Koch, suggest that such fossilized genitives are very rare, and this phase of the language's history is hardly represented by the surviving evidence. For example, does Dinbrain mean 'Fort of crows' or 'Fort of a crow'? It could be either. Similarly for Pentyrch, 'hill of a boar' or 'hill of boars'? Hard to tell. In the case of Ceinmeirch, it has been pointed out that the name is very likely describing the particular shape of the ridge, thus 'horse's back' rather than 'back of horses', but it is impossible to be definitive. In names like Kaer Tyf though, we can be a bit more confident because we have early references to the name of the river, which is not a common noun, and it would take some wiggling to suggest the form is plural, unless you think Cardiff means "fort of Taffs" 😉

Leave a Reply

Your email address will not be published. Required fields are marked *