How should climate change adaptation research be performed? Two principles are widely accepted and articulated in current adaptation-related research programmes such as the global Future Earth programme, the European Horizon 2020 programme and the Dutch Knowledge for Climate programme, just to name a few examples. The first one is that adaptation research should be transdisciplinary and solution-oriented (also called problem-oriented) in that it aims at contributing to ‘‘real-world’’ problem solving rather than purely advancing research in its own right (Cash et al. 2003; Gibbons et al. 1994; Moss et al. 2013). Achieving this aim entails co-designing and co-producing research together with stakeholders (Swart et al. 2014). The second principle is that adaptation research should be interdisciplinary in that it integrates knowledge from a range of natural and social science disciplines. And indeed, a wide variety of methods are currently applied in adaptation research, including participatory, experimental, decision analysis, behavioural analysis, institutional analysis and climate and impact simulation methods (Hinkel and Bisaro 2014). Less consensus can be found on how exactly those two principles shall be put into practice. Regarding transdisciplinarity, there is a substantial literature that has proposed various definitions and models for transdisciplinary research, including ‘‘mode-2’’ knowledge production (Gibbons et al. 1994), post-normal science (Funtowicz and Ravetz 1993) and action research (Lewin 1946). Regarding interdisciplinarity, there is hardly any dedicated literature exploring how to realise the integration of knowledge from diverse disciplines and beyond (Hinkel 2008). Nevertheless, there is a widespread model on how to combine methods, often called top-down approach (Dessai and Hulme 2004), which has dominated adaptation research from its beginnings. Climate models are run to produce climate scenarios, which are then downscaled and results are put into various sectoral climate impact models, sometimes followed by valuation methods to value projected climate impacts. Finally, the results may be fed into macroeconomic models in order to estimate economy-wide implication and/or decision analysis methods. But is the top-down model the only meaningful model of interdisciplinarity for climate adaptation? It certainly was a meaningful model in the early days of climate change research when the main question addressed was by how much adaptation could offset climate impacts at an aggregate level in order to prevent ‘‘dangerous anthropogenic interference with the climate system’’, which is the ultimate objective of the United Nations Framework Convention on Climate Change (UNFCCC 1992, Article 2). In recent years, adaptation has become a practical necessity, and the field has progressed from focusing on questions concerning climate impacts and costs and benefits of adaptation at an international scale to a much wider array of questions at all scales, related to, e.g., human perception, institutional change, equity, development and barriers to & Jochen Hinkel hinkel@globalclimateforum.org