There is no doubt in my mind that Marten Weber is an extremely gifted writer. There is also no doubt in my mind that he takes a delight in confusing the reader for as long as he possibly could during the course of the story. Do not get me wrong - it could be extremely rewarding to try and figure out the puzzle which the writer builds up, OR your frustration will built up and you will stop reading the book and never come back.
Second option happened with two books by this writer I tried to read - I just could not deal. However I stuck with In the mirrow a monster (yes, in that book I also wanted to strangle the narrator for the first 15-20% of the story and tell him to please figure out the way he wanted to tell the story and please to get on with already), and I found it extremely rewarding.
So I decided to be persistent with this book as well and OMG I had to be persistent for a long time to get ANY kind of understanding or emotional payoff.
I could not figure out what the heck was happening with the protagonist in the first fifty - sixty percent of the story, but not in a sense that I could not understand what I was reading. I felt like I was reading random scenes popping out of nowhere. And the moment I felt I had a grasp on what was happening - boom the rug was pulled out of me again. There is a difference between teasing the reader and annoying the reader. I was annoyed,a lot.
The first moment of clarity started to happen in the 60 percentage of the story on my kindle. And the more we were moving along the more the story was coming along and nicely. Everything made sense at the end, everything came neatly together. I am just saying that if you are like me, you may feel the story did not make sense for longer than it should have, if the point was to try and make reader's work to understand.
I also did not feel that other two main characters were well fleshed out. I thought they were chess pieces to carry out the overall message of the story, which I liked (the message that is), but I still prefer more in depth characterization.