ROMANO-BRITISH NORFOLK uncertain, and hardly any one has yet professed to detect it in Norfolk.' Its name is equally alien to the eastern counties. Previous to the Nor- man Conquest, our charters mention Icknield Street only in the west — five or six times as a Berkshire road and once as a road near Prince's Risborough. Not till three centuries later do we find its name applied to roads in Hertfordshire and Cambridgeshire, while east of Newmarket we never find it at all. Yet if it were really the warpath of the Iceni, we should expect its course to be clearest and its name oldest and most frequent in the eastern counties.* Moreover the place-names like Ick- lingham, which have been sometimes quoted (as remarked above) to fortify the Icenian hypothesis, tell in reality the other way. These names are not confined to the vicinity of Icknield Street, nor was Ick- lingham originally Iceningham, or Ickleton Icenton, or Ickleford Icen- ford.' They occur in many counties, in Sussex and Kent and Middlesex and elsewhere. Their earliest forms are known from charters, and they are derived, for the most part, from well-known English personal names like Icel and Icca. Nor indeed could either the real name Icenhilde, or the supposed names Iceningham and the like, have grown out of the name Iceni. According to philological laws, Iceni would have pro- duced in English a form beginning with Itch- or Etch-. Thus the arguments for connecting Icknield Street and the Iceni break down at every point. The real etymology of the road-name must remain doubtful, but until or unless some new and different evidence be forthcoming, it will be well to separate the road and the tribe. The question possesses a greater significance than is, perhaps, always recognized. For, if the English took a name from the Iceni, in order to denote a road which stretched from Berkshire to Norfolk, the Iceni must have been still known and existing as a tribal unit at the time of the English invasion. Now it is characteristic of Roman Britain gener- ally that the Celtic tribe-names died out : the tribes appear to have lost individuality and to have merged greatly in one another under Roman rule. If, however, the Iceni could bequeath their name to a road used by the English, they must have formed a very distinct exception to this rule. And therefore it is worth while, in a description of Roman 1 Recent attempts have been made in Knowkdge, February, 1899, and by Mr. J. C. Tingey in Norfolk Archceohgy, xiv. 140, but I cannot regard either as successful. Mr. Tingey relies especially on a Hickling Way at Swainsthorpe and Stoke Holy Cross, mentioned in documents of 1592 and later. But this is merely a brief lane ; and, as I have said above, Hickling and similar names cannot be adduced to prove the Iceni or the Icknield Street. It is also impossible, for philological reasons, to connect names like Kenninghall with the Iceni.
- Guest, Origiaei Celtide, ii. 227 ; Henry Bradley, Academy, October, 1894 ; Cartularium Saxon-
icum, No. 6o3 = Stowe Charter 22. The name Icknield Street appears about the same time in the western midlands as the name of the Roman road now usually called Rycknield Street, near Alvechurch in Worcestershire, etc. (Allies, y^nrij!. of Worcestershire, ed. 2, p. 332). In both east and west the twelfth and thirteenth century antiquaries probably helped by their speculations to extend the use of the name beyond its original sphere.
- Besides the three names quoted, there are Ixworth and Ickworth in Suffolk ; Ickburgh in Norfolk ;
Ickwell, near Sandy, in Bedfordshire ; Ickenham in Middlesex ; Ickham in Kent ; Hickstead and Icklesham in East Sussex ; Ickford near Thame in Oxfordshire ; Iccomb near Stow-on-the-Wold ; Hickling in Norfolk and Notts ; Hickleton in Yorkshire ; Ixhull, Oxon ; Iceldown in Somerset ; Hicklesworth in Dorset. 287
