To explain this in more detail, he argues that we determine these 'good reasons' by making assessments of the degree of coherence and fidelity within the proposition ('story'). Our assessment of coherence considers structural/argumentative coherence - does the argument hang together. We consider material coherence - comparing the 'story' with other relevant ones we can detect errors or omissions, does it fit with other 'true' accounts. And we consider characterological coherence - what do we think of the intelligence, intregrity and values of the author of the 'story'?
Turning to fidelity we consider both the reasons given, and the values conveyed. In the first case we test things such as whether the bits of the narratives claimed as facts are indeed facts, and whether the reasoning is sound. In the second instance we try to establish the explicit and implicit values in the story, and whether, for instance, they are validated by our own experience.
Apologies if that sounds a bit dry, but it's important to set the structure up before seeing how he applies it. You maybe thinking that narrative rationality may well apply in say the political field, but Wisher argues that it even applies in science. He demonstrates this by applying his model to an analysis of James Watson and Francis Crick's proposal of the double helix model of DNA.
This goes on for several pages so I'm just going to post up a chunk that gives you a good idea (apologies for typos, I am copying straight from the book):
What arguments do the authors offer to support the truthfulness of the double-helix model? Their first argument was that the structure proposed by Linus Pauling and R. B. Coring was "unsatisfactory". The underlying reason for th erejection was that the Pauling-Corey model was not truthful; it violated chemical "laws" and prior research. Their second argument was that the structure put forward by Fraser was too "ill-defined" to warrant comment. Clearly, precision is a value and lack of it is sufficent reason, a good reason, for rejecting ideas that are "ill-defined". After describing their model - verbally and in diagram - Watson and Crick present an intertwined argument to establish its conformity with the "laws" of chemistry and current research data. Here again, there is an implication of a good reason: Sound theory is in accord with prior knowledge. Each of these three lines of argument, it should be noted, is not a strict logical demonstration, either deductive or inductive. Each is, however, a proper deductive argument if one grants the premise on which it is founded: good theory is truthful (that of Pauling-Corey is not truthful; it should be rejected); good theory is precise (Fraser's theory is not precise; it should be rejected); good theory is confirmed by the best available theory and evidence (ours is; therefore, it should be accepted). The "reasons" for accepting Watson and Crick's proposal, then, are good reasons, reasons informed by by values: truthfulness, precision, conformity with the best that is known, and the promise of useful results in its application in further theory and research.
I read this and am both a bit ashamed and at the same enthused. Ashamed because thinking things through from Fisher's perspective, I can see how I have structured things in the past to ensure they have narrative rationality. I look back at a few policy papers I wrote and consider them to be collages rather than anything else. But because I think I am relatively good at structural/argumentative coherence I can make collages look more like analysis than they actually are.
On the other hand I also read this and realise that I can see people of different views (political in particular) doing exactly the same thing, and I think internalising this way of reading information can help sort the wheat from the chaff (if you believe that there is any wheat). This shouldn't be any surprise unless you believe that there is a correlation between proficiency in argument construction and political viewpoint.
It might cause me to kill some blog posts mid-composition too, as I can sometimes feel myself trying to give an argument coherence and fidelity that it may not deserve. Whether that's a good thing or not, I don't know.