Answers are grounded in this paper and cite their sources.
A recipe is not the same thing as the knowledge needed to cook it. Written recipes preserveingredients, quantities and sequences, yet cooks also judge readiness through touch, sound,smell and long practice, and they learn this judgement from other people. This paper develops aconceptual framework for studying how culinary knowledge is carried without being writtendown, and for asking what may be lost when it is converted into digital data and AI-generatedtext. Drawing on cognitive neuroscience, the anthropology of skill and food, and scholarship onBengal and Bangladesh, it distinguishes four dimensions of culinary knowledge: recorded,embodied, sensory and social. It argues that memory is reconstructive and selective rather thana store of exact records, that forgetting can be adaptive, and that these features sit comfortablywith practices in which cooks reproduce a dish from judgement rather than from exactmeasurements. It then considers current limits of food-related AI, while rejecting the claim thatAI can never represent sensory information. The paper concludes with a proposed fieldworkdesign centred on a single Bangladeshi dish, intended to test the framework withconsent-based, community-controlled methods.
No takes yet. Share an insight, caveat, or question.
ARFATUL ISLAM (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: