Heads up for anyone in the future who finds this thread via Google or something, I tweeted this question to the book author (Phil Sturgeon) and here was his response:
"add a link to /authors with whatever query string items need to be there. " https://twitter.com/philsturgeon/status/732222938825560064
Basically, the sub-resource should have it's own pagination URL that would be called if the user tried to paginate the sub-resource. This new request would only return the sub-resource results. Pretty simple really, was over complicating it in my head thinking that for some reason you would keep hitting the original resource endpoint for more sub-resources.