Thanks Mohan, and thanks for an interesting paper. I like the arguments you make. I just don't think they can establish the strong conclusion you want to draw, viz. the falsity of the claim that "natural selection causes evolutionary change". This is because, even once the myth of selection as a reified "force" is busted, there remains an ontologically innocuous sense in which the selection of traits causes the evolution of adaptive complexity.

In your post, you settle for the weaker claim that natural selection is not a cause of evolution "in the sense that interests me". From this I gather that your paper is really aimed at Elliott Sober's 1984 ontologically-heavy conception of selection as a force. My worry, however, is that any biologist reading the paper would take you to be challenging the causal role they ascribe to natural selection in explaining adaptive complexity, and I do not think your argument actually raises such a challenge. 

I think you can see this as a plus point of your general approach to natural selection. One might worry that a statisticalist, non-reifying approach robs natural selection of causal-explanatory import when it comes to explaining adaptive complexity. And one might see this as a reason to reject this approach in favour of something ontologically heavier. But on closer inspection, it turns out we can maintain that selection of traits causes the evolution of adaptive complexity without invoking a spooky tertium quid.

As regards the example I gave, it looks like we agree on these claims:

1. The increase in frequency of Ab is a cause of AB appearing.
2. The increase in frequency of Ab is an instance of natural selection.

From this it follows that:

3. An instance of natural selection is a cause of AB appearing.

I would add another claim:

4. An instance of natural selection in which Ab increases in frequency can be legitimately described as "natural selection of Ab".

Yielding:

5. Natural selection of Ab is a cause of AB appearing.

Importantly, I don't think such a claim reifies natural selection in any spooky way. Its explanatory role here is not as a tertium quid, but rather as a label for a causally relevant frequency increase. Hence, if a biologist were to remark, "natural selection of successive earlier structures caused the evolution of the eye", I would not take her to be saying anything controversial. A statisticalist picture of natural selection is not in conflict with the causal role biologists ascribe to natural selection in making claims such as this.