An interesting paradox underlies all our research on Old Norse-Icelandic poetry from the Viking and early Middle Ages: our knowledge of it depends absolutely upon its existence in written form, yet probably very little of it had a primarily written existence before the latter part of the twelfth century. That is to say that, except for the compositions of a number of thirteenth- and fourteenth-century Icelanders, which may well never have existed in oral form (a subject to be considered in later chapters), all medieval Norse poetry is likely to have been first composed and recited orally. This applies to skaldic as well as to eddic poetry and means that we have to be aware of the dynamics of orality when considering Norse poetics but at the same time recognise that the processes of transmission and of writing the poems down will have almost certainly had their effects on the poetic texts we know. An important question, then, is that of the nature of the effects of transforming an oral poetic tradition to a written one and of maintaining it in writing.

When Old Norse poetry did come to be written down, it was not written down on its own or for its own sake, at least not to begin with. In Chapter 1 we saw that very little of the large corpus of Old Norse poetry that has come down to us in medieval manuscripts exists outside a prose context. This statement holds true both for eddic and skaldic poetry. The fact is that, with some important exceptions, Old Norse poetry survives in the written record as quotation within a higher-order prose narrative or non-narrative context. This means that its function, within that prose context, is very often exemplificatory or evidentiary; it provides confirmation (or is made to appear to do so) of what the narrative voice of the prose text asserts. Compare, for example, Snorri Sturluson's use of examples (dæmi) of skaldic diction from the best older poets to demonstrate the use of specific kenning-types in the Skáldskaparmál section of his Edda. Compare also the use of verse citations in kings’ sagas to provide confirmation for certain points in the prose narrative.