Ideas become durable through return. A concept encountered once can be intelligible, but recurrence gives it a public history. Its name reappears, its definition is repeated or revised, other authors cite it, applications extend it and institutions begin to classify it. Search systems find it because there are enough traces to retrieve. Readers recognize it because they have encountered its structure before. Repetition therefore does more than reproduce knowledge: it helps produce coherence, memory and visibility. Yet recurrence is epistemically ambiguous. Ten copies of the same text are not equivalent to ten independent uses of the same idea. A definition syndicated across multiple repositories can create impressive numerical visibility while adding no new corroboration. Conversely, one careful translation, critique or unexpected application may transform a concept more substantially than hundreds of identical mentions. What matters is therefore not only how often something returns but how it returns. Repetition can preserve, extend, contest, translate or detach an idea from its origin. This distinction becomes urgent in computational environments. Search engines, citation graphs and language models all respond to patterns of recurrence. Frequently repeated formulations become easier to retrieve and reproduce; statistical familiarity begins to resemble authority. Merton’s Matthew effect already described how visibility can accumulate around established actors and ideas. Digital repetition intensifies the mechanism because one originating source can produce many technically distinct records. The resulting mass may look like consensus even when it is mostly self-recurrence. The answer cannot be to eliminate repetition. Knowledge develops through repetition, teaching, citation, translation and return. Excessive deduplication can erase legitimate variation just as excessive repetition can harden a formulation. A better infrastructure would distinguish exact copies from transformed uses, one-source concentration from independent adoption, and archival redundancy from intellectual development. Recurrence would have provenance. This changes how we understand the life of a concept. An isolated idea can have meaning; an infrastructural idea has a history of coordinated recurrence. Its importance lies not in raw frequency but in the pattern of transformations through which it remains recognizable. The central task is therefore to preserve repetition without allowing frequency to become a substitute for judgment. Knowledge needs recurrence to become public, but it also needs to remember who is repeating, where, why and with what degree of transformation.
Deleuze, G. (1968) Différence et répétition.
Derrida, J. (1972) Marges de la philosophie.
Merton, R. K. (1968) ‘The Matthew Effect in Science’, Science, 159(3810), 56–63.
Star, S. L. and Ruhleder, K. (1996) ‘Steps Toward an Ecology of Infrastructure’, Information Systems Research, 7(1), 111–134.
Tarde, G. (1890) Les lois de l’imitation.