The proposed caste enumeration in the census of 2001 will have deep consequ
ences. The matter should be analysed at two levels. In order to understand
the reality we have to ascertain the origin, evolution, and effects of cast
e. And to change the system we have to devise a strategy. For redressing th
e caste bias, a caste census is not necessary. A village presumably represe
nts a small number (maybe, two or three) of adjacent castes. Then, the Huma
n Development Index (HDI), computed village by village, can mimic the caste
profile of the population. Such an index can help development-planning to
eradicate, first, caste bias and then caste itself. A village-level version
of the HDI can be used for an indirect estimate of caste bias, making the
proposed caste census of 2001 redundant. The census instead should collect
data on health, education, and earnings of the people.