1 Decomposition of plant litter, a key component of the global carbon budget, is hierarchically controlled by the triad: climate > litter quality > soil organisms. Given the sensitivity of decomposition to temperature, especially in cold biomes, it has been hypothesized that global warming will lead to increased litter decomposition rates, both through direct temperature effects and through indirect effects on litter quality and soil organisms. 2 A meta-analysis of experimental warming studies in cold biomes (34 site-species combinations) showed that warming resulted in slightly increased decomposition rates. However, this response was strongly dependent on the method used: open top chambers reduced decomposition rates, whereas heating lamps stimulated decomposition rates. The low responsiveness was mainly due to moisture-limited decomposition rates in the warming treatments, especially at mesic and xeric sites. This control of litter decomposition by both temperature and moisture was corroborated by natural gradient studies. 3 Interspecific differences in litter quality and decomposability are substantially larger than warming-induced phenotypic responses. Thus, the changes in the species composition and structure of plant communities that have been observed in medium-term warming studies in cold biomes will have a considerably greater impact on ecosystem litter decomposition than phenotypic responses. 4 Soil fauna communities in cold biomes are responsive to climate warming. Moreover, temperature-driven migration of the, hitherto absent, large comminuters to high-latitude sites may significantly increase decomposition rates. However, we do not know how far-reaching the consequences of changes in the species composition and structure of the soil community are for litter decomposition, as there is a lack of data on functional species redundancy and the species' dispersal ability. 5 Global warming will lead to increased litter decomposition rates only if there is sufficient soil moisture. Hence, climate scenario and experimental studies should focus more on both factors and their interaction. As interspecific differences in potential decomposability and litter chemistry are substantially larger than phenotypic responses to warming, the focus of future research should be on the former. In addition, more light should be shed on the below-ground 'darkness' to evaluate the ecological significance of warming-induced soil fauna community changes for litter decomposition processes in cold biomes. © 2006 British Ecological Society.