¹ The development of spoken Latin is placed firmly in its historical setting in W. D. Elcock, The Romance Languages, 1960, a most lucid and readable work. For the philological developments themselves see C. H. Grandgent, An Introduction to Vulgar Latin, 1934 (reprinted 1962) and V. Väänänen, Introduction au latin vulgaire, 1967. A useful collection of texts, especially epigraphic, is F. Slotty, Vulgärlateinisches Übungsbuch, 2nd ed., 1960 ( = Kleine Texte 143).

² Cf. Elcock, op. cit., 21, ‘…various direct sources of information concerning the nature of the spoken language are available for scrutiny. These may be broadly summarized as follows: traces of colloquial usage in the work of Latin writers; the spelling of inscriptions; the remarks of Latin grammarians; the pronunciation of words borrowed from Latin by other languages.’ On the last point note further K. Jackson, Language and History in Early Britain, 1953, 76, on the neglect by classical philologists of the material relevant to spoken Latin in the Celtic languages of Britain.

³ Cf. V. Väänänen, Le latin vulgaire des inscriptions pompéiennes, 3rd ed., 1966.

⁴ Cf. D. Greene, in Christianity in Britain, 300–700 (ed. Barley and Hanson), 1968, 75–8, esp. 77: ‘…the speech of an Icelandic trawler-man may sound very archaic to his Norwegian counterpart, but it is none the less colloquial.’

⁵ There seems to be no evidence in Britain for the prothetic i-which is not unimportant in continental development (as in ischola, Old French escole, Mod. French école). CIL vii 1270a, Isarninus, is not an example, cf. Holder s.v. Iarnos.

⁶ RIB 631, mintla, raises the interesting question of the language an individual would choose for obscenities (cf. RIB 872, 983, 2106, 2157, JRS 54, 182 no. 22). Will it not more naturally be his first language, rather than an acquired tongue, that is used? May not some of our incomprehensible British graffiti conceal obscenities written in the British language?

7 bagis, bitam would have no place in this list if the inscription on this imported sculptured group was added before it reached Britain. It is here assumed that the inscription was added in Britain.

⁸ That this may indicate Celtic influence is a tentative suggestion based on K. Jackson, Language and History in Early Britain, 1953, 536 ff.

⁹ Trupo probably derives from Thrupo, and is presumably Germanic, since the name appears here in a list of Germani. It apparently reappears in > Thruponiana, RIB 1556, which probably does not therefore derive, by transposition, from Tryphoniana.

¹⁰ Attempts were made by grammarians to counteract the pernicious influence of the spoken tongue on the written word. The so-called Appendix Probi (text in the works by Väänänen and Slotty listed in note 1 above) is entirely devoted to this effort. The attempt to correct the omission of -n- from -ns- was sometimes, however, pursued with excess of zeal, inserting -n- where it did not belong (‘hypercorrection’). The re-correction Hercules non Herculens appears precisely in Appendix Probi, 19.

¹¹ In RIB 306, the future donabit would make much better sense than the perfect donavit: b and v apparently approached more and more closely in pronunciation.

¹² An attribution to Celtic influence could be argued from Dottin, La langue gauloise, 1918, 251 s.v. devo, and from the restricted and apparently British distribution of posivit.

¹³ There is a notable absence from Britain of the common milex for miles, cf. e.g. ILS 2328, 2346, 2811, 3223.