As the recent lament over falling standards by the Secretary-General of the United Nations reminds us, universal literacy is today considered a necessary feature of civilized society. This may be one reason why the problem of medieval literacy, of a society where the ability to read and write was apparently confined to a clericalélitehas so intrigued modern historians. In this paper, I wish to reconsider the extent of lay literacy in England before the Conquest. But first we must be clear what we mean by literacy. The ability to read does not necessarily imply the ability to write. To take only the most famous medieval example, Charlemagne could speak Latin, and enjoy theCity of God, but he never learnt to write. What Parkes calls ‘pragmatic literacy’ may extend from the capacity to recognize, if not sign, one's own name, to the ability to write a formal document in Latin. What might be called ‘cultured literacy’ could range from reading free prose in the vernacular to composing Latin in the classical tradition. The more advanced types of pragmatic literacy might well overlap with the more basic cultured levels. But it is obvious that we cannot deduce a widespread ability to read any-thing from the fact that the names of owners or makers were some-times engraved upon Anglo-Saxon coins, weapons or memorials; or a generally high standard of lay culture from the fact that there were documents in the vernacular. If we are to describe early English society as literate in a sense that would satisfy the ancient, or later medieval, historian, we must show that the proportion of laymen able to read at least a vernacular writ or poem was socially, if not statistically, significant.