Martin is particularly concerned with a) the idea that there should be a clear sense of what it is to be “at” a level, and b) the idea there is some significant sense in which being at a level represents a genuine ontological novelty.
The first of these concerns arises from the fact that the placement question cannot be (or has not been) given a satisfying answer within the mechanistic application of the metaphor. The mechanistic account of levels focuses, as Martin correctly notes, on what it means to be at different levels of organization within a mechanism. Concerning what it means to be “at a level,” one can say only that two things are “at” the same level just in case they are not at different levels; if neither thing is a component entity/activity in the behavior of the other, then the two things are not at different mechanistic levels and, in this weak sense, they are at the same level. This provides a sort of answer to the placement question, but not one that will satisfy those, such as Martin, who hanker after some additional factor (such as size or similarity) that unites the items at a level and that explains why all such items are at that level. I take the failure to answer the placement question as one of the key revisionary consequences of thinking clearly about levels of mechanisms. It is a crucial guide to understanding what’s misleading about the monolithic conception of levels.
One apparent problem for this consequence (the absence of a satisfying answer to the placement question) was first raised by Lindley Darden (personal communication): X’s -ing might be a component in (and so at a lower level than) S’s ψ-ing, and yet both X’s -ing and S’s ψ-ing might be at the same level as (i.e., not at a different level than) some altogether distinct P’s ß-ing. There is a failure of transitivity. This, it seems to me, is simply a consequence of the idea that the application of the level metaphor to levels of mechanisms breaks down when one is not talking about relations between parts and wholes. This is unproblematic; it is simply an alternative way of expressing the idea that being “at the same level” is of no additional metaphysical significance within the mechanistic application of the metaphor than simply not being at different levels. This does not prevent one, of course, from using some other ordering criterion for this expressive purpose; one might lump things together on the basis of their sizes or perhaps the instruments used to detect them. But one should be aware that at this point one has left the mechanistic application of the metaphor.
Another consequence of the mechanistic view is that one might equally correctly carve the boundaries of mechanisms in any number of ways (see e.g., Craver 2004, 2009, 2012). If so, there will not be a uniquely correct answer to the question of how many levels a given mechanism has. Our decisions to privilege some grains in the decomposition of a mechanism as an appropriate place to locate a level depends on our techniques, our theoretical background assumptions, our characterization of the phenomenon, our representational tools, and perhaps certain features of human psychology. As Simon argued, these systems are only nearly decomposable; how a system is decomposed depends, for example, on the relative strength of intra- versus extra-system causal interactions that one takes to be appropriate for carving the system at its near-joints. This is another reason I am hesitant to pin too much on the idea of being “at” a level.