Survey of English Place-Names

Historical Forms

• Celtan hom, æt Celtanhomme 803 BCS309 11th

• of Ciltan ham 11 Heming

• Chinteneham 1086 DB

• Chilteham 1155–1159,1195 P 13 WinchLB Hy3 Surv c.1220 GlR 1220 Fees 1223 Pat 1227 Ch 1316 WinchLB

• Schilteham c.1260 GlR

• Chelteham 1156 RBE 1218 ClR 1248 Ass 1272 Glouc 1292 Episc

• Chiltha(m)', Chyltha(m)' 1160–1204 P 1205 Cl 1226–8,1236 Fees 1240 Cl 1247 Misc 1252 Ch 1262 Pat 1287 QW

• Chiltenham c.1240 GlR 1252 Cl 1271 Abbr c.1275 For 1284 Episc 1291 Tax 1675 Ogilby

• Shiltenham 1341 Cl

• Cheltenham c.1250 GlR 1296 Episc 1322 Abbr 1327 SR 1409 Pat 1610 FF

• Cheltenham spaw 1779 Rudder

• Cheltnam 1562 FF 1599 Comm 1631 PR

• aqua de Incham in 1620 FF

Etymology

Cheltenham is a town standing in the comparatively low-lying ground on the banks of the R. Chelt some 2–3 miles west of the Cotswold scarp of Cleeve Hill; the name originates as that of a water- meadow (v. hamm ) near the Chelt, and the name of the Chelt (i, 5supra ) is usually considered to be a back-formation from the place-name; certainly in modern usage that is true, as the river flowing through Boddington, Barrow, Leigh, Uckington and Hardwicke (which is now the Chelt) was called Alr in the 11th century (v. Arle 104infra ) and aqua de Incham in 1620FF . One of the difficulties with the first el. is its actual original form, which already in OE  varies between Celtan - and Ciltan -; in ME  Chilt - is the predominant form at an early stage but Chelt - reasserts itself from the middle of the 13th century. One explanation that offers itself is that Celtan - is an Anglian i -mutated form and Ciltan - is the corresponding later West Saxon form from older WSax  Cieltan - (cf. Bülbring §§ 179, 306, Phonol. § 12); these presuppose a PrOE  base *calti - (broken in WSax  to *cealti -). It would certainly be unusual for OE  Celtan - to be a variant of an original Ciltan -; although AN  -e - is often used for -i - (Feilitzen 50), it would not account for the OE variation, and the early ME spellings which might have been so explained are comparatively rare.

Various parallels have been adduced by Zachrisson, StNPh v, 60, Tengstrand, ib vi, 91 ff, Anderson 11, Ekwall in DEPN (s.n. Chiltington, Chiltern, Chilcomb, etc.), Mawer and Stenton in Sx 174 (s.n. West Chiltington) and Reaney in Ess 123 (s.n. Childerditch).For the form Chelt - we have an unidentified Celta (BCS 87), which Reaney (l.c.) associates with Childerditch, and also the surname of William de Cheltheued 1248Ass 16d, a juror in Botloe Hundred (iii, 166infra ), some 12 or more miles from Cheltenham in the extreme north-west of the county; Cheltheued has not been identified, but Anderson would place it near Cheltenham presumably on the form of the name. There are more parallels to Chilt - in West Chiltington (Sx 174) and a local unidentified le Chilte (Sx 4), Childerditch and a stream-name in the surname of Saman de Chilteburne in Havering (Ess 123), Chilcombe (Ha), Cilta (n )cumb 9, 909 BCS 493, 620, Chiltley (Ha), Ciltelei 1086 DB, and Chilcombe (Do 246), Ciltecome 1086, Chiltecumbe 1285, Childecombe 1333. These parallels to Chilt - are all from the Saxon dialect area.

Several interpretations have been offered. The editors of the Sx volume postulated an unrecorded OE  pers.n. Cilta , which has no parallel in the Germ languages; but it could have originated as a byname meaning 'the prattler' from the PrGerm  root *kalt -, which survived in OFris  kaltia 'to prattle', MHG  kalzen , kelzen 'to chatter', and in an original side-form *klat - in OE  clatrung 'noise'. We should have to assume a WGerm  wk. form *Kaltj  , which would with i -mutation give Angl  Celta and WSax  Ci (e )lta .

Zachrisson and others have assumed an OE hill-term *celte or *cilte (which is adopted in EPN i, 88), and Ekwall has suggested that this might be related by vowel-gradation to Norw  kult 'lump, hillock' (for etymological connexions of the IE  *gel-t cf. also Pokorny 358 and A. S. C. Ross in Beiträge zur Namenforschung viii, 172–3); Ekwall also thinks that Cheltenham might contain a Brit hill-name cognate with that of the Chilterns (Hrt 7) from a supposed Brit  *celto -; but this word does not occur, according to Professor Jackson, in any of the Celtic languages. The surname Cheltheued is assumed to mean 'Chelt -hill' (v. hēafod ) and to refer to a hill in the neighbourhood of Cheltenham; but the old part of the town of Cheltenham, to which the name would presumably refer, is not particularly hilly, esp. in those parts in which the hamm would be found; even the bottom of the Cotswold escarpment is at least a mile distant.

Reaney (Ess 124), especially with the surname Chilteburne in mind, tentatively suggests that the first el. is a river-name. Although a sense 'hill' may be appropriate in some of these names, combinations with words like cumb 'valley', dīc 'ditch', hamm 'water-meadow' and burna 'stream' tend to support Reaney's view; topographically this is preferable for Cheltenham, and it would also be reasonable for Cheltheued (if it belongs to the district), for hēafod 'head' is often used of 'the source of a river'. Although Chelt as a current river- name appears to be an e.ModE  back-formation from Cheltenham, it may also have been the original name of the river before it was renamed Arle and aqua de Incham . But the origin of such a r.n. is obscure. We may therefore have to interpret Cheltenham as 'Celta's water-meadow'.