When I write a longish review, I put most of my work into having the piece work well as reader's guide, keeping my own views to myself as much as possible until the end. But since I've signed up for an e-mail subscription to the Notre Dame Philosophical Reviews (just go here to get it), I've noticed that this practice is by no means uniform. Consider the beginning of Bruce Russell's recent review of James Sterba's From Rationality to Equality:
In this book James Sterba sets out to answer two central questions in moral philosophy: (1) Why be moral? and (2) What does morality require? (see Sterba's Preface and Introduction, esp., pp. 1-2). He understands (1) to be the question (1a): Are we always rationally required to do what we are morally required to do? However, I think the fundamental question is (1b): Is there always most reason to do what we are morally required to do? I may be rationally required to take a pill that I have overwhelming epistemic reason to believe will save my life when I am really not dying and the pill will actually kill me (imagine some doctor wants me dead and has presented me with conclusive evidence to make me believe that I am dying and that the pill is my only hope for survival). But in this case what there is most reason to do is to refrain from taking the pill even though the rational thing to do is to take the pill. I think what we want to know when we ask, "Why be moral?" is whether there is always most reason to do what is morally required. We want to know whether it is necessarily true that there ismost reason to do what is morally obligatory, not whether it's necessarily true that we are justified in believing that there is most reason to be moral.
I worked hard to try to figure out what Russell meant here, and then was frustrated to find that the phrase "most reason" doesn't occur in the review itself.