📄 Academic News: Aaron Michael Butts, new Professor of Semitic Studies with a focus on Ethiopian and Eritrean studies at Universität Hamburg, with the project BeInf—Beyond Influence: The Connected Histories of Ethiopic and Syriac Christianity

Summary

This short Personalia notice in Aethiopica 25 is not a research paper and reports no data or results. It announces that Universität Hamburg has given Aaron Michael Butts a W3 professorship in Semitic studies centred on Ethiopian and Eritrean studies. It also introduces his ERC-funded project BeInf, which began on 1 September 2022 at the Hiob Ludolf Centre.

The project's starting point is an old thesis from Ignazio Guidi and Carlo Conti Rossini. They held that Syriac-speaking missionaries, escaping the Chalcedonian empire, reshaped Aksumite Christianity: they brought monasticism, translated the Bible and caused a "second christianization." Later scholarship has disputed every argument behind that thesis.

BeInf has two aims:

• to determine what kind of contact actually existed between Ethiopic and Syriac Christianity, and how much;
• to build a framework that no longer relies on "influence" as its main analytical lens.

It will pursue both through five complementary case studies that combine linguistics, philology, manuscript studies, art history, textual studies and history. The project is meant to serve as a model for nuanced studies of connected histories in the humanities.

Key points

• Funding is an ERC Consolidator Grant of almost €2 million over five years, hosted at the Hiob Ludolf Centre, Universität Hamburg.
• Case study 1 analyses Aramaic loanwords in Ethiopic from a contact-linguistics angle, testing which ones, if any, can be proven specifically Syriac. Case study 2 asks what the sixth-century Naǧrān episode, in which the Aksumite ruler Kaleb intervened, contributes to the question.
• Case study 3 places the illuminations of the Ǝnda Abba Gärima gospels within an eastern Mediterranean setting, including the Syriac Rabbula Gospels.
• Case study 4 builds on Antonella Brita's finding that the Nine Saints hagiographies are medieval or later. It asks what role these texts played when they were composed.
• Case study 5 examines how Syriac texts reached Ethiopic through Arabic, with preliminary results in Butts 2020 ("From Syriac to Arabic to Ethiopic"). Earlier critiques of the influence thesis cited include Polotsky (1964), Marrassini, Witakowski and Heide.

Interesting ideas

• I1. A loanword that fits a whole language family (Aramaic) is weaker evidence than one that points to a single source (Syriac), so attribution should require evidence that singles out that one source.
• I2. The project replaces the vague, one-directional label "influence" with explicit types and degrees of contact, which makes claims testable.
• I3. A text should first be dated, and then read for what it did in its own time, rather than treated as a direct witness to the events it narrates.
• I4. Transmission through an intermediary (Syriac → Arabic → Ethiopic) is studied hop by hop to find where changes entered.
• I5. Several separate case studies, each using a different discipline's method, are aimed at one shared question so their conclusions can confirm or check each other.

For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

• I1 (learn): when crediting a silver label to a specific witness parser, count only decisions that witness made uniquely, not ones all witnesses shared.
• I2 (frames): instead of a generic "affects" relation between frame participants, require a typed relation (cause, transfer, imitation, contact) drawn from VerbNet roles, so each link can be explained.
• I3 (world): store a tale's claims about the past as the narrator's assertions made at narration time, kept separate from event order inside the story world.
• I4 (back): in the text → graph → text compressor and in translation through the meta-world, log the differences at each hop to locate exactly where meaning drifts.
• I5 (learn): treat agreement among methodologically different witnesses (rule-based, perceptron, gparse) as stronger evidence than agreement among similar ones.