Why is an onion cold but a piece of garlic hot? This assignment of thermal values is popular in the folk classification of the intrinsic quality of onion and garlic in many parts of South India. At first glance, the assignment seems arbitrary and most likely based on ``myth'' and ``superstition'' perpetrated by local vaidyas (doctors) to exercise power over the rest of the population. Consider the following method of preventing Sunburn. I learnt of this prevention technique from an old acquaintance, Manikkam, a vegetable vendor and part-time vaidya in Madras.
On a hot day in Madras, when I complained about becoming soft in the US and finding myself unable to tolerate the midday sun, Manikkam advised me to take several onions and tie them into a small necklace. He then asked me to wear this fragrant necklace while plying the streets of Madras, and assured me that the ``coolness'' of onions would keep any chance of contracting Sunburn remote. Being an old friend he offered this advice to me free of charge. At the time, my only response to this friendly advice was to become inundated by images of crowded streets filled with the pungent smell of human sweat mixed with onions. The only consolation was that garlic (being hot) could not be a prescribed remedy for Sunburn.
If the intrigued observer plunges deeper into the same folk classification system, he will find a bewildering number of food items, medicinal products, diseases, individual and class/caste dispositions (kunams), labeled as possessing heating or cooling qualities. He may also find that far from being arbitrary, the assignment of hot and cold qualities taps into a large body of indigenous ``knowledge''that pertains to local adaptations of a philosophical world view that integrates materials and power relations (dharma) into multiple interacting strands (gunas), humors (dosas), elements (bhutas) and dispositions (kunams).
Early western anthropological studies investigating the use and mention of hot and cold concepts have confined themselves to a single culture/language and have generalized a single data set into a Pan-Indian and non-local characteristic. Brenda Beck [Beck] first attempted to classify food items as belonging to hot or cold categories. While such attempts provide original sources of systematic study, other authors [Daniel] have pointed out their limitations in explaining the structure and complexity of the conceptual categories themselves (ex. there is a continuum from hot to cold along which substances may be placed, rather than the classical categories used by Beck and others). These researchers have also commented on the amount of variation and diversity (across geographical regions, caste/class status, as well as individuals) that makes any attempt at categorization suspect.
This project attempts to investigate the conceptual categories of hot and cold in the following novel ways:
In this preliminary report, I hypothesize that the two ideas above can form the beginning of a systematic delineation of the semantic property space of the concepts hot and cold. It is interesting to note that the ideas of radial categories and metonymic clusters of multiple variables is also used extensively by McKim Marriott and his students [Marriott] in proposing a general ethnosociology of India. While being less ambitious and much more preliminary, the work here shares some of Marriott's critical insights and wherever possible, I will borrow terms and definitions from his considerable body of work. My theoretical analysis also intersects with Lakoff's [Lakoff] work in cognitive linguistics and in the structure of conceptual mappings (conventionalized) metaphors. I also share with Lakoff, the guiding intuition that the body is a central site in grounding interpretations of the world including those that pertain to relations of power and dominance.
This research borrows from field work by different anthropologists. The work on Javanese [Samson] [Geertz], work on Tamil [Beck], [Daniel], work on Punjabi/Sindhi [Kurin], and on Zimmerman's interpretation of the Ayurvedic texts [Zimmerman], are used as data sources for much of the material developed here. I addition to these textual sources, I will also rely on my own knowledge and experience of different Indian cultures and languages.
The general approach is divided into three parts. First, I will try to present the important aspects from data gathered from the sources mentioned above. A complete listing of the raw data can be found in the appendix. I will then use the suggested interpretation framework to explain some common practices in three different cultures.