A florilegium (literally a gathering of flowers, from Latin flos and legere) is a medieval compilation of the most instructive passages from a larger body of work — typically the Church Fathers, classical authors, or scriptural commentaries. Produced in enormous numbers throughout the Middle Ages, florilegia served readers who lacked the time, resources, or access to consult the originals. Ann Blair's research situates the florilegium within a lineage of information-management technologies stretching from classical antiquity through the contemporary search engine. What unites the lineage is a principle of navigation rather than reduction: the florilegium did not try to shrink the supply of available text but to make a vast and unruly corpus traversable through deliberate human curation.
The florilegium's value was entirely dependent on its selectivity. A compilation that included everything would have reproduced the problem of abundance it was designed to solve. The compiler's judgment about what mattered — what to include and, equally, what to leave out — was the technology's core intellectual contribution. The florilegium was not a neutral filter but a curated artifact bearing the imprint of the compiler's theory of what was worth preserving.
Blair's historical argument traces a direct lineage from the medieval florilegium through the Renaissance commonplace book, the early modern encyclopedia, the Enlightenment bibliography, the nineteenth-century library catalog, the twentieth-century database, and the contemporary search engine. Each technology responded to an expansion of the information supply by embodying a principle of selection that reduced the total supply to a navigable subset. The medium changes; the navigational logic does not.
The florilegium reveals an asymmetry that recurs throughout the history of curation: the compiler received less credit than the original authors from whom she excerpted, even when the compilation's intellectual value depended on her judgment more than on any individual source. Blair identifies this systematic undervaluation of curatorial labor as a pattern that continues into the AI era, where the invisible curation that produces finished AI-assisted work goes largely unrecognized.
The florilegium also demonstrates that curation is a form of authorship. The same raw materials, selected and arranged by a different compiler, would produce a different florilegium — and the difference would be the measure of the compiler's creative contribution. This principle anticipates the contemporary argument about compilatory authorship in AI-collaborative work.
The practice has classical precedents in Greek and Roman anthology traditions, but the florilegium as a named medieval genre crystallized in the monastic scriptoria of the Carolingian renaissance (ninth century) and flourished through the twelfth and thirteenth centuries. Collections such as the Florilegium Angelicum and the compilations of Thomas of Ireland became widely circulated reference works in their own right.
Navigation, not reduction. The florilegium's purpose was to make abundance traversable, not to shrink it; the solution to too much to read was better ways to read, not less to read.
Exclusion as intellectual act. What the compiler left out mattered as much as what she included; the leaving-out was the mechanism of conversion from raw supply to usable resource.
Curation as authorship. The compiler's arrangement of excerpted material constituted an original intellectual contribution, even though none of the raw text was hers.
Lineage to the search engine. Every subsequent curatorial technology — encyclopedia, index, database, search engine, AI filter — belongs to the same structural family as the florilegium.