Diachronica - Volume 41, Issue 1, 2024

Divergence-time estimation is one of the most important endeavors in historical linguistics. Its importance is matched only by its difficulty. As Bayesian methods of divergence-time estimation have become more common over the past two decades, a number of critical issues have come to the fore, including model sensitivity, the dependence of root-age estimates on uncertain interior-node ages, and the relationship between ancient languages and their modern counterparts. This study addresses these issues in an investigation of a particularly fraught case within Indo-European: the diversification of Latin into the Romance languages. The results of this study support a gradualist account of their formation that most likely began after 300  CE. They also bolster the view that Classical Latin is a sampled ancestor of the Romance languages (i.e., it lies along the branch leading to the Romance languages).

Ancestral State Reconstruction (ASR) is an essential part of historical linguistics (HL). Conventional ASR in HL relies on three core principles: fewest changes on the tree, plausibility of changes and plausibility of the resulting combinations of features in proto-languages. This approach has some problems, in particular the definition of what is plausible and the disregard for branch lengths. This study compares the classic approach of ASR to computational tools (Maximum Parsimony and Maximum Likelihood), conceptually and practically. Computational models have the advantage of being more transparent, consistent and replicable, and the disadvantage of lacking nuanced knowledge and context. Using data from the structural database Grambank, I compare reconstructions of the grammar of ancestral Oceanic languages from the HL literature to those achieved by computational means. The results show that there is a high degree of agreement between manual and computational approaches, with a tendency for classical HL to ignore branch lengths. Explicitly taking branch lengths into account is more conceptually sound; as such the field of HL should engage in improving methods in this direction. A combination of computational methods and qualitative knowledge is possible in the future and would be of great benefit.

The interdependence of accusative case and object agreement has changed dramatically during the history of Ugric languages. While Proto-Ugric exhibited full interdependence (mediated by topicality), this connection has loosened in the extant Ob-Ugric languages (Mansi and Khanty), and it is severed completely in Late to Modern Hungarian. In this paper, I introduce new, hitherto unreported empirical evidence from nicknames and family names that preserve archaic syntactic features for an intermediate stage of Early Old Hungarian (which predates our earliest written records) where case assignment was still a function of topicality but object agreement was already a function of definiteness. In addition to providing insight into an unrecorded stage of Hungarian, my findings also contribute to a more thorough understanding of the connection between case, agreement and information structure in Ugric and beyond.