Food distinguishes from medicine in that it is generally consumed daily, with a focus on palatable nourishment while medicines are often prescribed and taken for limited periods and specific therapeutic purposes. Food intake is guided by olfaction, gustation and chemesthesis (Goff and Klee, 2006; Yarmolinsky et al., 2009; Breslin, 2013; Palmer and Servant, 2022). The perception of taste and flavour (a combination of taste, smell and chemesthesis) have evolved matching nutritional requirements and are particularly important in omnivores for detecting palatable and rejecting toxic and deteriorated food (Rozin and Todd, 2016). However, humans have learned to use unpalatable and toxic substances for medicine (Mann, 1984; Johns, 1990; Mennella et al., 2013). It is thought that plant-based medicine is rooted in diet and the search for nutrition and that agricultural practices refined and diversified associated knowledge (Johns, 1990, 1999; Etkin, 1994; Logan and Dixon, 1994; Brown, 1985; Leonti et al., 2006). In theory all food may be regarded therapeutic by someone who's hungry (Etkin and Ross, 1991) but whether a food item is perceived only as food or also as a therapeutic or prophylactic agent depends more specifically on culture and epidemiology (Etkin and Ross, 1982, 1991; Johns, 1990; Etkin, 2008; Lindeberg, 2010).

The origin of the frequent saying “let food be thy medicine and medicine be thy food”, is unclear and not to be found in the Corpus Hippocraticum (Cardenas, 2013; Totelin, 2015). The misquotation of the Corpus Hippocraticum (6th to 2nd century BCE) in this context is probably due to the many substances described in Classical Greece that were explicitly used as food medicines (Wilkins, 2015; Totelin, 2015, 2018a). Also today, from the 219 herbal drugs listed in the European Pharmacopoeia 9.5, (2017), 75 (≥33%) are used for both, food as well as medicine (European Pharmacopoeia 9.5, 2017; Supplementary Table A). Similar values can be expected for European herbal medicine in general and probably higher ones for the Ayurvedic and Chinese pharmacopoeias (c.f.e.g., Wichtl, 2002; The Ayurvedic Pharmacopoeia of India (API), 2001–2016; Chinese Pharmacopoeia (ChP), 1997).

In Classical antiquity, the power of botanical drugs was extrapolated from their taste (Jones, 1959; Einarson and Link, 1990; Jouanna, 2012; Totelin, 2018b). According to the Greek physician Mnesitheus of Athens (4th century BCE) “all salt and sweet juices move the bowels. But acid and pungent foods stimulate urine; bitter juices are more diuretic, and some loosen bowels and astringent ones check excretion” (Baker, 2018). Theophrastus (300 BCE) reports that sweet has the capacity to smoothen, astringent the power to desiccate and solidify, pungent the capacity to cut or to separate out heat, salty the power to desiccate and irritate, and bitter the capacity to melt and irritate (Einarson and Link, 1990). Today, in Western herbal medicine, the use of chemosensory qualities is limited to quality and identity control of botanical drugs (e.g., Wichtl, 2002; Gafner et al., 2023) but are still important cues for the prescription of medicines in Ayurvedic and Traditional Chinese Medicine (Patwardhan et al., 2004; Maciocia, 2015).

In a study using tasting-panel data applying a phylogenetic approach, we have recently shown how in ancient Graeco-Roman society the use of botanical drugs was shaped by chemosensory qualities (Leonti et al., 2024). However, a study focusing on the differences of experimentally assessed chemosensory qualities and recorded therapeutic uses of food drugs and non-food drugs has never been made so far. We expect such an exploratory analysis to provide insights into the dietary or non-dietary origin of specific therapeutic knowledge possibly linked to specific chemosensory qualities. Here, we therefore compare chemosensory qualities and ancient therapeutic uses of orally applied botanical drugs that are commonly used in diet (food drugs) and chemosensory qualities and uses of botanical drugs generally only used for therapeutic purposes (non-food drugs) as recorded in Dioscorides’ De Materia Medica (DMM). We address the following questions: 1): Which chemosensory qualities augment the probability of an orally applied botanical drug to be also used for food? 2): which chemosensory qualities augment the probability of an orally applied botanical drug to be only used for medicine? and 3): Are there significant differences in the frequency of therapeutic indications between orally applied food drugs and orally applied botanical drugs used exclusively for medicinal purposes, and, if yes, how can these differences be explained?