How do you create an edition of texts which, according to the evidence of the medieval manuscripts, somehow belong together, but whose internal system cannot be clearly defined?
This problem is particularly acute for the Corpus Hippocraticum, which gathers central texts of ancient medicine and is associated with the name of the famous doctor Hippocrates of Kos from the second half of the fifth century BC even come from Roman times.
It is still unclear to this day when this collection (which has been modified many times over the centuries) was created for the first time. A plausible assumption is that it was not until 288 BC that Alexandrian scholars created a compilation of the (supposedly) Hippocratic writings and integrated them into the new library of Alexandria. But until recently there has been disagreement among scholars dealing with the history of medicine and philology about the composition of the corpus - the compositions vary between around 60 and 75 texts. Corresponding uncertainty also affects the most famous part of the corpus, the Hippocratic oath, the - next to the Bible - the most momentous and prominent text handed down from antiquity;the suggested dates range from the fifth or fourth century before to the earlier first century after Christ.
The n-gram analysis
Digital methods of "authorship attribution" now offer new possibilities for knowledge, which are not simply intended to replace the traditional instruments of source criticism and style analysis, but rather to supplement them with additional options. Whereas previously printed concordances and relatively undemanding surveys that scour the existing databases for individual words were used in order to postulate stylistic similarities and potential joint authorship as well as to identify text groups that belong together, a much more complex process of digital stylometry is now being practiced, such as the Leipzig ancient historian Charlotte Schubert can demonstrate convincingly using the example of the Corpus Hippocraticum (“To determine the location of the digital in the ancient sciences”, in: Gymnasium.Journal for Ancient Culture and Humanistic Education, Vol. 127, Issue 4, 2020 / Winter).
A method established in computational linguistics is used, the n-gram analysis, which is based on breaking down texts into fragments. Schubert subjected 63 works from the Corpus to a “letter trigram analysis”, a procedure based on the statistical analysis of sequences of letters made up of three letters. In this way, clusters of texts can be identified that could belong together, since they can also be viewed as "related to one another due to stylistic, content and conceptual features". With the help of a network visualization program ("Gephi"), these groups are also graphically displayed in order to illustrate connections within the clusters as well as between them. Overall, this results in findingswhich in turn provide the basis for further philological and historical considerations.
In the specific case it becomes clear that “the oath as a text is by no means next to the corpus, but rather has various connections to other, early and late writings of the corpus”, that is, it does not have a “special position” postulated again and again. He is particularly close to the treatises on the ancient art of healing, on diet and on the glands, as well as the "Letters", the "Law" ("Lex") and the "Regulations". Since the first two works mentioned, as well as the treatise “Lex”, are dated to around 400 BC, one might be inclined to also date the oath to this early phase. Schubert expressly does not draw this conclusion, but rather places the oath in a "line of tradition" in terms of content with the aforementioned texts of the cluster, because: "What all works in this cluster of the Hippocratic oath have in common isthat they claim to orient techne and dietetics to the ethical dimension of measure and the middle. "
In the ancient literary tradition, the Hippocratic oath is mentioned for the first time in the first century AD (by the doctor and grammarian Erotian, who wrote a lexicon on the Hippocratic works for Nero, and by the doctor Scribonius Largus). A connection between the Hippocratic oath and the name of the famous doctor from Kos is by no means certain. Even the new, highly profitable stylometric methods have not yet provided the new information desired in this regard. It is possible that new computer-aided methods will not help here, but only new finds of inscriptions or papyri.