Since the Karl Lachmann’s edition of De rerum naturis, through Paul Maas Textkritik and further into Lachmannian method inquiries and critical assessments, The Method states that the critical edition of a given text imply the previously acquired knowledge of every source defining the extent of the textual tradition concerning the literary work that is about to be published, say a poem, a novel, as written documents of any kinds. Any lack of knowledge may imply missing of given lectiones singulares or shared erroneous ones causing an incorrect evaluation of the relations connecting the various textual sources. That’s why any lachmannian textual reconstruction process involves a special feature called recensio, aimed to collect any known, and possibly some still unknown, source of the text about to be published. In lachmannian terms, any edition based on a limited evaluation of the textual tradition simply can’t be be considered critical at all. Since the editor has to check and collate all the sources of the literary work he is about to publish, the recensio represents a primary issue of the critical process. So primary he hardly wonders what he is actually doing while he is collecting his sources.
Basically, the recensio is conceived and practiced as a very simple activity, based on the consultation of catalogues and a more accurate direct evaluation of the items most suitable to be taken into account as sources of the poem, the romance or epic narrative or whatever the editor is about to study and publish. It looks like such an empirical process actually relies on the «family resemblances» described by Wittgenstein in his Philosophische Untersuchungen. In facts, the editor applying to recensio is absolutely committed to the identification of similarities. This commitment drives to the recollection of similar objects which have «something in common», in terms that they have some parts in common. As Wittgenstein could have eventually said in such a case, the editor “feels” the similarity, he could easily “show” it, explaining that: “here it is, can’t you see it? It’s similar!”. Empirical approaches as the one suggested by such an utterance are the ones which editors actually assume while working on the critical edition of ancient and medieval literary works, say poems or narratives of various kind. Indeed, editors categorize sources as individuals belonging to the same family without even wondering how similar two or more texts have to be to be considered different versions of the same poem, novel, whatever and what is the symbolic level the “quest for similarity” applies to.
In general, similarities are scaled on the various symbolic layers of the textual encoding. Sentence-length or more extensive and consistent parts shared by two or more sources, as shared special words, like proper or geographical nouns, can trigger some valuable analogies. Even metrics and other non-linguistic textual features, as the editorial layout of medieval manuscripts, may be taken into account as control conditions, even though different sources will be never classified as exemplars of the same tradition just because they are copied on similar books or they just share a metrical pattern, as it happens so often, for example, in middle-age Provençal, French, Italian love lyrics. In general, two or more different documents are considered as belonging to the same textual tradition if and only if they share similarities in respect to parts and/or individual special clues and markers that are pretty uncommon into other items carried by the catalogs. The minimum scaling is generally set on the sentence-length similarity, but the very meaningful matches usually involve the sharing of more extensive parts, as phrases and, of course, phrase-sequences.
This bottom-up approach, based on analogy and induction, fits pretty well the basic needs of an editor, as soon as he tries to establish the full set of sources that have to be taken into account for collatio, the process leading to the identification of common errors. At the same time, the complete lack of any deeper theoretical insight causes the editor to bypass a very crucial issue, that is a tragic, huge aporia. Indeed, the critical edition of a given literary work is aimed to investigate the textual tradition in order to attain the archetypal configuration.
Besides, the recensio is supposed to identify the sources a given literary work in order to define the extent of its textual tradition, setting up the critical investigation of the archetypal text. Before manuscripts are connected into a stemma codicum, before errors are found, even before sources are simply collated, a stable configuration of the text arises leading the editor throughout the most crucial part of his work, the recollection of sources by means of similarity-pattern recognition. Basically, while collecting his sources, the editor develops a full representation of an exemplar, that is a “control text” he relies on, he worships, when it comes to check for similarities.
This huge aporia, embedded into the basics of the lachmannian method, evidences an epistemological dilemma, a too big and complex one to be solved into the disciplinary boundaries of Textkritik. That is probably why the state-of-the art on this crucial topic is a blank slate. Neither the most analytical approach to Textkritik provides appreciable hints on the subject, probably because philology can’t answer questions as: how can an editor look for something he still does not know what it really is? Or, what is he looking for when he starts his recensio of the sources that may actually be referred to the “same” literary work? How does he assesses two different documents as sources of the same literary work? Why doesn’t he simply classify them as different ones? What is the special feature triggering the process of pattern recognition that leads to the identification of some sources as “same” and others as “not same”? Where does the editor sets the «upper» limit of the simple quotation, and where does he set the «lower» one leading to the recognition of a stand-alone literary feature, substantially independent from the rest of the textual tradition he is actually inquiring? Brief, if he hasn’t still proceeded to collatio, error recognition, elaboration of the family tree, what is he comparing to what? What does he categorize as a text? And finally, what a text basically is?
A correct answer to this set of questions requires a different approach to the text, to be regarded as a plastic feature, that may be described both as the individual and the family it belongs to, that is as a category and, at the same time, as an object to be categorized into.
(to be continued)