Until the establishment of 'women in development' (WID) approaches in the early 1970s, much of the emphasis on grassroots development and indigenous agro-ecological knowledge was strongly male-biased; primarily because of the relative invisibility of women's work to outsiders. Thereafter, increasing attention was drawn to the unequal impacts of development on women and the equality- (and later efficiency-) related consequences of excluding women from the development process (Moser, 1993; Braidotti et al, 1994). 1 In particular, WID researchers highlighted women's role as major subsistence producers with well developed systems of agro-ecological knowledge. They also increased awareness of prevailing gender divisions of labour in the South that often gave women a double or even triple work burden consisting of reproductive (childbearing and rearing) work, subsistence production (often including unpaid community management duties) and, for many women, paid employment (Boserup, 1970). Further research during the UN Decade for Women (1976-85) highlighted persistent gender inequalities in sex ratios, divisions of labour, participation and remuneration in the labour force, inheritance systems, residence patterns and control over important subsistence resources (including land), plus major intra-household inequalities in the allocation of income, food, health care, education and other basic resources (Boserup, 1970; Harriss and Watson, 1987; Sen and Grown, 1987; Moser, 1993; Braidotti et al, 1994; Wieringa, 1994; Agarwal, 1992; 1997; Razavi, 1997).