cognateset

A cognate set, also called a cognate class, is the group of words in a family of related languages that descend from a single word form in a common ancestral language. Members are inherited, not borrowed, and they typically retain related core meanings across languages. The idea is to capture regular correspondences in sound and sense that reveal historical connections.

Cognate sets are identified through the comparative method. Linguists compare words with similar meanings across languages, look for systematic sound correspondences, and reconstruct a proto-form in the ancestral language. A reconstructed proto-form represents a hypothesized ancestor of the set, but it is not directly attested.

In practice, cognate sets support broader goals in historical linguistics, such as classifying languages, tracing language families, and testing hypotheses about how sound changes propagate over time. They also underpin lexicostatistics and glottochronology, where sets of core vocabulary are used to estimate divergence times between languages or groups.

Examples commonly cited include English night, German Nacht, and Dutch nacht, which form a cognate set descended from a Proto-Germanic form. Another widely cited set links English father, German Vater, Dutch vader, and Latin pater, tracing back to a Proto-Indo-European root. However, distinguishing true cognates from borrowed forms or convergent similarities requires careful analysis of regular sound correspondences and historical context.

Terminology can vary; some scholars use cognate set interchangeably with cognate class. The concept remains a foundational tool for organizing etymological data and for understanding how languages relatedly diverge from common ancestors.