Evidence for British language through toponymy in north-east Surrey
There are a few place-names relevant here. Penge derives from the not-uncommon Brittonic compound *penn-cḝd, “wood’s end” (Gelling and Cole 2014, 211). Coates (1998, 20) has suggested the first half of the name Camberwell may derive from Latin camera, “vault, room”, perhaps a relict Roman-period enclosed wellhead structure. It should be noted the philology is inconclusive; moreover, there are no known villas or other extra-urban substantial masonry buildings in the northerly, cemetery-free zone. Then again, Croydon (OE *croh < Latin crocus, “saffron crocus” + denu, “valley”: Gelling 2005, 80-81; see also Bird 2012) and nearby Addiscombe (OE personal name Æddi + camp < Latin campus, “field”: Gelling 2005, 74-76) show similar use of non-Germanic loan-words in place-names in the heart of early Surrey.
In contrast to many commentators (e.g. Bird 2004, 172-73, and indeed Green 2012, 110), I’m going to exclude wēala place-names (Wallington in the cemetery-rich zone, and to the north Walworth, waleport and walehulle in Kingston, and potentially the furlong-name Walworth in Battersea, although I’ve been advised this last one may be manorial in origin). This is because, as I spent part of last summer arguing, I don’t believe these necessarily denote surviving incumbent British communities, rather ones of non-local slaves settled in the ninth century or later. As such, they are inadmissible as testimony for the continued survival of incumbent British communities in this area.
Definitely admissible are place-names from OE wīchām, argued by Gelling to constitute a reference to a Romano-British settlement (or settlement site) known to ‘neighbouring Germanic communities’ (see Gelling 2005, 67-74). One in the Peckham area was noted by Gelling (2005, 247) and subsequently received more detailed scrutiny in a short article (Warhurst 1994). Streatham (if OE strǣthām, “homestead/village on a Roman road” (another loan-word < Latin strata), as per Mills 2010, 238) might be thought of in not-dissimilar terms.
Another wīchām name survives as West Wickham just over the county boundary in what was historically Kent (Mills 2010, 268). This is much further south, effectively within the cemetery belt, and intimates similar interaction between incoming and native populations (or at least Roman settlement remains – and remember that clipped siliqua found hereabouts?). Whitford in Mitcham can also be noted in this regard, provided Coates’ cautious entertainment of certain place-names in wīc having ‘Roman archaeological significance’ is upheld (Coates 1999, 111, tempering the opinion of Margaret Gelling republished in Gelling 2005, 247).*
Overall, there are at least as many toponymic examples of possible early contact and interaction between Romano-British incumbents and Germanic/Saxon incomers within the suggested nucleus of early Surrey as there are in the hypothetical British-for-longer belt to the north.
* If you want a bit of a wīc-related laugh – and who doesn’t? – then you might like to read a new article in Archaeologia Cantiana (Durham & Goormachtigh 2015) that sets out a bizarre one-size fits-all interpretation of Kentish place-names containing the element, including West Wickham. The authors link them to lines of water-borne transport and the distribution of salt, which in several cases means navigating salt-laden boats up some tiny watercourses deep into the Wealden interior. I thought this sort of swivel-eyed place-names study was a thing of the past, at least in peer-reviewed publications, but apparently it’s alive and kicking in Kent…