Scribal Errors in Ancient Greek and Latin Manuscript Traditions: Collection, Aetiological Researches and Digital Representation

The scribal errors in the transmitted witnesses of the ancient Greek and Latin texts have been studied by philologists at least since the XV century, being at the core of the stemmatic method. Nevertheless, most of these errors are left out of critical editions, as the removal of the copyists’ mistakes – and not their analysis – has always been the ultimate goal of textual criticism. Moreover, the small minority of demonstrably wrong variants that make it into the apparatuses has often not been studied in dedicated spaces. And yet, as pointed out by many scholars in different ages, the readings that come from errors of transcription can play a major role in the philological discipline. The more we study patterns of errors, the more will we understand the transmission of texts, the scribes’ original contributions, and, finally, the reasons and mechanisms that cause written slips In my dissertation I will firstly analyse the different phenomena that go under the name of variant and the opinions that some philological traditions had of both the reasons for the variants to arise and the attitude to have towards them. I will then talk about what has been called error, how it has been perceived, how to define it and how to study it, pointing out - in the footsteps of several authors - the necessity of a precise terminological distinction between ‘error’, ‘variant’ and ‘reading’. After a look at the linguistic state of the art in terms of the study of the performance errors that are known as slips, I will argue that a psycholinguistic approach to the copyists’ activity – with special regard to their own slips – would provide a better understanding of the transcription process, of the copyists’ minds involvement in it, and of the recurring typologies of slips. To find them, I scrutinized some Greek and Latin manuscript traditions (from critical apparatuses, collations and digitized manuscripts) purposely looking for erroneous, minority readings whose genesis and aetiology could be guessed by comparison with other instances. The data will be ascribed to semantic categories, with no claim of completeness, but as a suggestion for a larger system of classification. The digital revolution in the field of the Humanities can prove extremely useful in modelling data and finding patterns. Therefore, in the second section of my dissertation I will use the data collected in the first part for a Digital Humanities project. In a first review of the existing bibliography, I will show that the manuscript variants are a common subject within the debate on the Scholarly Digital Editions and their critical apparatuses. For the purposes of this work, though, one needs to collect and study exemplary cases taken from different traditions and manuscripts. The collected data of a repertoire are non complete and non continuous. Thus, peculiar solutions need to be figured out. I will then analyze the tools and methods provided by the Digital Philology debate on apparatuses and critical texts, that can be used or re-used for this kind of desultory data as well. I will finally describe a digital archive of readings, that takes its materials from the traditional archive that is displayed in the first part. This digital archive is specifically designed to be compliant with the texts in the Perseus Library, and is thought of as an addition to the enhanced reading environment that is currently under developments within the Scaife Viewer project. Furthermore, I will reproduce the semantic complexity of the traditional variants archive by developing an ontology that is specifically designed to describe the reading types. From this, I will briefly elaborate on how my model applies to two case studies: the review of critical editions and the case of the “complex tradition”, namely a textual tradition where the text is not fixed and may vary greatly from witness to witness. In conclusion I will claim that, on one hand, the study of the erroneous variants can bring improvements to the ‘traditional’ philological studies. Not only can it, on occasions, allows a glimpse of microhistory that would be otherwise lost. But more importantly, it could help in some interpretative remarks, such as telling voluntary changes apart from slips and from gloss intrusions, or assessing which errors are more likely to happen twice independently, and how probable it is that the copyists’ life influences their writing. On the other hand, the digital modelling of a list of readings enables scholars to use the material more effectively, both in terms of searchability and comparisons, and as an irreplaceable component of an enhanced reading environment for ancient Greek and Latin texts.