Can Bayesian phylogeography reconstruct migrations and expansions in linguistic evolution?

Area of Science:

• Historical Linguistics
• Computational Phylogenetics
• Language Evolution

Background:

• Bayesian phylogeography is applied to reconstruct language family origins and spread.
• The accuracy of these reconstructions, particularly distinguishing migration from expansion, is not well understood.

Purpose of the Study:

• To evaluate the reliability of Bayesian phylogeographic methods in distinguishing between migration and expansion processes in language evolution.
• To develop methods for identifying which spatial process best explains observed language distributions.

Main Methods:

• Simulation study differentiating migration (discontinuous spread) and expansion (gradual spread).
• Evaluation of state-of-the-art phylogeographic methods under varying directional trends.
• Development of descriptive statistics to identify migration vs. expansion patterns.

Main Results:

• Phylogeographic methods fail to accurately reconstruct migration events.
• These methods perform well in reconstructing expansion events, even with strong directional biases.
• Migration and expansion exhibit distinct phylogenetic and spatial signatures.

Conclusions:

• Current Bayesian phylogeographic tools are more reliable for studying language expansions than migrations.
• Proposed statistics can aid researchers in selecting appropriate models for spatial language evolution.
• Recommendations are provided for assessing model adequacy in phylogeographic studies.