In many industrialized countries the use of land is impeded by soil pollution from a variety of sources. Decisions on clean-up, management or set-aside of contaminated land are based on various considerations, including human health risks, but ecological arguments do not have a strong position in such assessments. This paper analyses why this should be so, and what ecotoxicology and theoretical ecology can improve on the situation. It seems that soil assessment suffers from a fundamental weakness, which relates to the absence of a commonly accepted framework that may act as a reference. Soil contamination can be assessed both from a functional perspective and a structural perspective. The relationship between structure and function in ecosystems is a fundamental question of ecology which receives a lot of attention in recent literature, however, a general concept that may guide ecotoxicological assessments has not yet arisen. On the experimental side, a good deal of progress has been made in the development and standardized use of terrestrial model ecosystems (TME). In such systems, usually consisting of intact soil columns incubated in the laboratory under conditions allowing plant growth and drainage of water, a compromise is sought between field relevance and experimental manageability. A great variety of measurements can be made on such systems, including microbiological processes and activities, but also activities of the decomposer soil fauna. I propose that these TMEs can be useful instruments in ecological soil quality assessments. In addition a "bioinformatics approach" to the analysis of data obtained in TME experiments is proposed. Soil function should be considered as a multidimensional concept and the various measurements can be considered as indicators, whose combined values define the "normal operating range" of the system. Deviations from the normal operating range indicate that the system is in a condition of stress. It is hoped that more work along this line will improve the prospects for ecological arguments in soil quality assessment.