There is currently a considerable effort being devoted to the development of anthropomorphic test devices for the measurement of thoracic side impact response. Both the SID and EUROSID have been proposed as viable candidates for this test device. In addition, the thorax of the three year old Fart 572 has been shown to be useful in simulating side impact while used in the frontal orientation. This apparent anomaly suggests that the intuitive differences between the frontal and side geometries of the thorax may not be significant. To date, all useable thoracic models have been unidirectional. For the most part, these have been frontal models. This paper discusses some of the difficulties inherent in the development of a two dimensional thoracic model and ways these difficulties can be addressed. Based on these considerations, a single thoracic impact model is proposed for simulation of both frontal and lateral impact without adjustment of model parameters for impact direction. The model response is compared with two dimensional thoracic impact data obtained from cadaveric specimens as reported in the literature. The merits of the model as well as its deficiencies are discussed.