¹⁷I emphasize that there is nothing contradictory or impossible in such training: in particular, there is nothing contradictory in “wanting to relax all desires”. What is needed is that the agent’s information-processing system creates a desire to relax all other desires, and then takes as its goal the state where all other desires are relaxed. When the agent has relaxed all other desires, this (meta-level) desire for relaxation goes away by itself, just like any desire disappears after its goal has been reached. Thus, in the end, all desires, including the meta-level desire, have vanished. Any contradiction is avoided because this meta-level desire to relax desires is only directed at other desires, not at itself, and because desires go away automatically when their goal is reached. This is also my interpretation of Samyutta Nikaya 51.15, where the paradox of wanting not to want is resolved by explaining how “[a mendicant who is perfected] formerly had the desire to attain perfection, but when they attained perfection the corresponding desire faded away” (Trans. Bhikkhu Suhato). A complementary approach to resolving this paradox is to consider how the meditation practice changes over a time span of many years. Initially, meditation is based on the desire to reduce suffering, and makes use of the desire to let go or relax. But ultimately, you let go of even the desire to be happy, and, paradoxically, of the desire to let go (or relax). This is possible since you let go of letting go only after a long practice, so the attitudes and habits required for letting go or relaxation are now automated in your neural networks and need no effort or explicit desire to operate anymore. You just relax and let go automatically, without desire or planning to do so. Note that this is clearly related to the problem of “aversion towards aversion” that we considered above in connection with acceptance. In a similar vein, Striker (2004) emphasizes that Pyrrhonian Skeptics did not (actively and purposefully) suspend judgement, as it is sometimes claimed, but rather were unable to arrive at any judgement and gave up any such attempt; see also Herman (1979) for further analysis of the Buddhist case.