The first is that of synthesizer - someone who brings together a variety of things (pieces of evidence, ideas, other historians' works) and forges them into a new pattern. Again, this involves making choices (what to include, what to leave out, what to emphasize, what to diminish), and, again, these choices are unavoidable.The role of synthesizer is essential. Without it, there would be no history as we understand it - only piles and piles of records, lacking shape or coherence.
'... every historian is, in some fashion, a story teller.'
In addition, every historian is, in some fashion, a story teller. In producing a synthesis, the historian has to make it available to a wider audience - whether through writing, the radio or television.
The history that we see as a public is the final production (after the historian has completed all his or her other roles), and it is necessarily a kind of narrative. Sometimes the story elements are very clear, for example the short account of John Hogsflesh with which we began is presented as a little tale.
Sometimes, the 'story' is less obviously that. For example, if we had started with a table displaying some statistics for people convicted of religious deviance in England, that would look less like a tale and more like a piece of maths. It would still, however, be a story, or part of a story. As soon as the historian starts to analyse and explain, we are led, as readers, from a beginning to a middle to an end. If the table has a point to make ('this shows that ...'), it is telling us a story.
This is not to suggest that history (the thing produced from the roles of the historian) is not true. Unless the historian has cheated or been lazy or sloppy, the story is true in that it has fidelity to the sources. But it is to note that it is a story, and that had the historian adopted a different role, the tale would have appeared a little differently.