and 'realistic', it will probably have to be fantastic, extreme. One might also notice, even m the matter of'realism', the critical and intrinsic ambivalence of colonial imports, deriving no doubt from the nature of the power relationships of domination and subservience which facilitate the transfer. These imports, whether of constitutionalism or cotton goods or 'realism', are at once enabling and repressive. They endow the beneficiaries with a certain limited liberation in return for absorbing them in the dominant power relationships; they also work to repress the indigenously available alternatives, of narrative technique as well as of political action and social transformation. Mukherjee's discussion of Sarat Chandra Chattopadhyaya contains, in addition to thoughtful and sensitive analysis, an interesting speculation as to the secret of Sarat's popularity: he embodies/enacts the moral dilemmas of his time. She is also drawn to making large cultural generalisations about "a bedrock of commonality" which, she contends, "exists in most parts of the subcontinent". This kind of speculation is tempting, troublesome, inescapable, necessary and treacherous. She might, nevertheless, be near an answer when she refers to the key function ofkaruna rasa—in Sarat, in Bengali culture, and obscurely mutatis mutandi, in India. It is difficult to summarise complex ideas en passant but somewhere at the root of the national fondness for self-indulgent karuna, in Sarat as in the Bombay film, there is a controlledly hysterical affirmation of the existent structures of society. The hysteria is crucial: it legitimises the pain that the structures irtflict. Jt would be ironic if exploitation alone were able to provide an adequate foundation for our 'national' identity. The latter part of Mukherjee's book consists of detailed analyses of three texts: Pother Panchali Godan, Samskara. Her observations on the originality of Father Panchali, and the formal confusion in Godan, are both interesting and powerful, as is her perceptive account of Samskara. However, it is occasionally difficult to relate the details of specific analyses to the larger argument of the book itself. In the event, one ends up wishing that the book were longer—so that it might make the connections whose possibility remains, as yet, hypothetical. For it must be said, finally, that Meenakshi Mukherjee starts many more trains of thought than she is able to pursue—or, if I may switch metaphors somewhat alarmingly, take. That can't be bad. 48 Number 9