Drawing the Legal Family Tree: An Empirical Comparative Study of 170 Dimensions of Property Law in 129 Jurisdictions

Abstract

Traditional comparative private law scholars have a firm grasp of laws in several countries, but rarely of those in more than one hundred countries. Quantitative comparative private law scholars have placed dozens of countries into a legal family genealogy, but not based on a systematic understanding of legal substance around the world. Using a unique, hand-coded data set on 108 property doctrines (transformed into 170 binary variables) in 129 jurisdictions, we ran supervised and unsupervised machine-learning algorithms. Some of our findings confirm the conventional wisdom: French and German property laws are influential; mixed jurisdictions like South Africa and Scotland are one of a kind; common law jurisdictions form a group of their own; and a handful of formerly socialist countries, led by Russia, cluster together. Unlike the prior literature, however, we do not find that East Asian jurisdictions warrant a category of their own; but belong to distant groups. Spain and many Latin American countries form a separate group. Rather than finding a clear-cut common versus civil law division, we observe that the France-inspired group is one supercluster, separate from other jurisdictions.

Keywords: Legal substance, common law, civil law, legal origin, average-linkage clustering, sparse linear discriminant analysis (LDA)

JEL Classification: K11, K00