Abstract
In this paper, I discuss two aspects of top-down causation in George Ellis’s compelling account of this complex concept. I first examine whether mathematical structures can enter into any causal relations, and argue that it is unclear that they can. I highlight instead the role played by the interpretation of mathematical theories in the context of applied mathematics. Second, I consider the role played by top-down causation in measurement, especially in quantum mechanics but also elsewhere, and identify the importance of the observer in this context. I am in large agreement with Ellis’s overall message, but we may understand certain details a bit differently. I articulate the message within a broadly empiricist setting.