Ethnopharmacological relevance: Research on the folk categorization of nature in preliterate societies in Europe is complicated due to the fragmentation of the information available and is rarely undertaken. Yet the data is valuable and may provide, in certain circumstances, important insights, if not into novel medicines, then into the historical logic of selection and memorisation of plants useful from a medicinal perspective. Aims of the study: We aim to understand the ethnobotany of a preliterate society by analysing the emic (derived from people) perspective on nature-related culture of one of Europe's smaller nations, whose written language and culture was shaped in the 18th-19th centuries by other, larger nations of Europe, and thus from the etic (academic) perspective. We attempt to identify how folk categorization is reflected in the relationships between plant names and uses and to map the structure of those relationships. Data and methods: We base our analysis on one of the oldest ethnobotanical manuscripts and herbaria of the Baltic governorates, compiled in 1831 by an amateur botanist, Baltic German Pastor Johann Heinrich Rosenplanter (1782-1846), which was derived from conversations with his parishioners from the tiny Parnu parish. The historical dataset was critically analysed from an ethnobotanical perspective in light of recent identifications of the herbarium specimens. Results: Although the Rosenplkner collection is fragmentary, the logic of plant categorization by non-literate peasants at that time is clearly seen in the data. Plants preserved in the herbarium were predominantly used for ethno-medicinal, food or ethno-veterinary purposes, such as treating chronic skin and joint diseases as well as severe acute diseases in humans and animals. Among 129 folk taxa analysed, more than one third had apparent purpose-related plant names providing clear links to their use, whereas a few multifunctional plants had several names reflecting diverse uses. For example, Hypericum spp., which was used in three different ways, had three semantically distinct names. However, among the plants that Rosenplkner collected, there were also some that were simply named and described by people but lacked any usability data (e.g., Trollius europaeus), meaning that use as such was not the primary criterion for recognising a plant. The web-like structure of preliterate thinking in plant-related knowledge reveals a deep relationship with the environment and the interpretation of new elements through familiar natural objects. Conclusion: Our findings demonstrate that historical ethnobotanical data, if thoughtfully analysed, can be used not only for comparative purposes, but also for understanding the logic of preliterate thinking. We encourage future in-depth studies of historical ethnobotanical data in Europe in order to understand the relationship between nature and culture of native European populations.
