The title of this paper refers to the title of a book by Paul Saenger entitled Space between Words: The Origins of Silent Reading (Stanford University Press, 1997). In this book Saenger examines one of the conventions of printed and written texts that we tend to take for granted: the fact that spaces are used in writing to indicate where one word ends and the next begins. He argues convincingly that the ability to read silently and rapidly is a direct result of the introduction of this convention. Literacy was introduced to these islands by the Romans in the pre-Christian era. However, with the coming of Christianity, literacy was zealously promoted by the Church, with much emphasis being placed on the role of the written word. Aconvention that made comprehension of the written word easier to master was obviously of considerable benefit.

In the Roman world, manuscripts were generally written in unseparated text, otherwise known as scriptura continua. Saenger illustrates this from a page of the fifth-century text, Livy's History of Rome (Paris BN, lat. 5730, f. 59r: Saenger 1997, p. 5, Fig. 1). There are no spaces between words, no punctuation, and frequent run-on of words from one line to the next. It was in this sort of form that writing entered Britain under Roman occupation. Some of the inscribed stones from Roman Britain are indeed rather similar in layout, for example a dedication slab from South Shields, Co Durham, now kept in the Arbeia Roman Fort and Museum, South Shields (Collingwood and Wright 1965, no. 1060, p. 354 and Fig. 1060). The text on this stone commemorates the installation of a supply of water for the Fifth Cohort of Gaul. The words in the text are unseparated except that, by utilising the idiosyncratic abbreviation typical of Romano-British inscribed stones, almost all the lines begin with a new word.

However, not all Romano-British stone inscriptions have texts like this. There is, for example, an altar from Lanchester, Co Durham, now kept in Lanchester Parish Church, on which there is some attempt made to indicate word division by means of dots (Collingwood and Wright 1965, no. 1074, p. 358 and Fig. 1074). The use of dots, or interpuncts, to separate words had been used to some extent in manuscripts in antiquity, and this is presumably the origin of it on Romano-British stone inscriptions.