Can psychopathy be identified in pre-adolescent children? This question has both theoretical and practical implications. So far, attempts to answer this question have been made by making downward extensions of symptoms of the disorder as found in adults. We argue that for the construct of psychopathy to be viable as a construct in childhood it is necessary to demonstrate that a coherent syndrome of symptoms can be identified during childhood. To achieve this it is necessary first to use a strong measurement model, and second to use developmentally informed conceptualizations of key constructs. In order to progress from the measurement perspective, it is necessary to demonstrate (i) configural invariance of symptoms across key ages, (ii) similar performance of age cognate symptoms across key ages, and (iii) stability of relative standing on the latent trait across age. Novel techniques including item response methods and structural equation modelling techniques are required. In this paper, we review the relevant literature and evaluate the strength, or otherwise, of the evidence that key traits of psychopathy can be observed in childhood. Copyright © 2004 John Wiley & Sons, Ltd.