These problems can be resolved by the incorporation of other manuscript readings, with the result that the finished, “fixed” edition does not represent exactly any surviving manuscript copy of the work. This is even more the case if other editorial policies are adopted: if the aim is to produce a stemmatic, recensionist, or hybrid edition, which use various techniques to attempt to deduce, from the surviving evidence, either the elusive archetype of all the manuscripts or, even more ambitiously, the original or definitive form of the text. Such editions do not only “fix” a text; they reconstruct it. Indeed, perhaps this is true of all serious editing, apart from the simplest single manuscript edition and the most literal diplomatic edition. Even hypertext and parallel-text editions, which set out to be as inclusive as possible, reconstruct the texts they represent, in the sense that they offer readers an experience of simultaneity that was neither anticipated by the text’s original author/s nor shared by its earliest readers.
Editing, again, is a techne: every stage of these processes involves methodical, precise, ordered work in which accuracy is at a premium and some knowledge of paleography, diplomatics, and historical linguistics is essential. An edition’s authority, its ability to stand in for the text it represents for the purposes of scholarly or general discussion, in large part depends on its specifically technical achievements.
But any given scholarly edition is also a theoria, in two senses. First, an edition constructs a reasoned hypothesis about the text it edits, deploying particular skills and choosing between possible editorial methodologies only in service to this hypothesis. An edition is an argument or “theory” about the text is represents, fixes, or reconstructs, one that stands or falls on the coherence and cogency of its approach.
Second, in doing this, an edition takes a stance towards both the question of what a text is and how this text can be most fully and lucidly represented, and should be clear about this stance. Editions may focus on either the search for “authorial intention,” or the activities of scribes, or the form in which a text was first consumed by readers, or again its language or dialect; it may privilege the text's composition, transmission, reception, or philological implications.
In assessing their manuscript readings, they may accept or reject literary criteria such as metre and sense, taking a conservative or radical approach to emendation. Standing with one foot in the past of their text, the other in their own present, editions may privilege respect for the manuscript original, carefully keeping the spellings of the original intact at the cost of comprehensibility; or, rather rarely in Middle English editing, they may opt for a degree of modernization or regularization, seeking to increase the text’s potential readership and hence its purchase on the present.
The rationale for making these decisions is theoretical in the sense that it rests on general ideas about texts and textuality that are at once political and philosophical. Because these decisions concern the human artifacts of the past and our continuing relationship with them, it is also true to say that editing, like other kinds of historical work, is ethical.