"For early vernacular works (whether oral or written in origin), the transmitting manuscript does not merely ensure the survival of the work as a text through the operation of a technology of preservation; it actually determines conditions for the reception and transmission of the work" (O'Keeffe 1990, 5). This statement raises the critical issue that forms the focus of this discussion. The way in which we apprehend that which we call "text" when it is written down, is primarily governed by the manuscript versions in which it appears. This is particularly true for poetry, both because it frequently remains in only one copy, and because it has traditionally been respatialized into half-line pairs, emended to conform to our perception of alliteration rules, and in general "cleaned up" by editors throughout the twentieth century. Many of these editors have been inclined to disregard the physical evidence contained in the grubby, fire-damaged, ink-smudged, or scribally-imperfect page, and have instead sought to provide the scholarly world with the poems "as they should have appeared."