The accepted rule that brothers divide the household after their father's death avoids the need for a build up of pressures to explosion point as, for example, Srinivas describes for joint households in an Indian village (Srinivas (1952)). The great insistence on absolute equality between recognised heirs prevents very much jockeying for advantage among brothers, unless exceptional circumstances give one brother an opportunity. Nevertheless, as households grow, the pressure to fission increases. Where a father lives to an old age and leaves more than one mature family of grandchildren in his house, division is likely to be much more rapid than with a less mature family, and in a few rare cases fission may begin before the old man has passed on. Men blame quarrels of this kind on the women. Srinivas's arguments (1952, p. 30) could be applied here. The tensions between one elementary family and another inside the households probably reflect as much the rivalry of the brothers as the incompatibility of the women. Disputes are very often at basis disagreements about relative rank; not so much about the pecking order - seniority settles this - as about how hard the pecking should be. Relative rank among the household women turns very largely on the relative rank of the brothers. The closeness of brothers, their mutual dependence and their opportunities for avoiding each other socially when angry, may enable them to avoid among themselves the open quarrels of which they are the underlying cause among their wives.
Srinivas's analysis is so plausible that it cannot be altogether untrue. Yet it is also plausible to argue that the tension between the women of a joint household, often born strangers to each other and brought together by chance, must be in some cases the result of their personal incompatibilities. It would be impossible to prove either explanation false; and it is also impossible to state in general any precise weighting of the two factors.