You have obviously put a lot of thought into this, thank you for taking the time to do this. Your thoughts are helping me.
To me that statement seems to define the special circumstance, rather than the norm, given that he goes on to say: "It has been found by experience, however, that while the shadow detail method is theoretically best for the particular type of work just mentioned, this is by no means the case for practically all other types of work."
I do agree that he is talking about a long scale subject specifically and I heartily agree with the thought that trying to straight print too long a scene scale results in dull prints.
Actually his statement "or (b), if - as is usually the case- the main interest is only at one end of the scale, to print for that end only and ignore the other end entirely." is the statement that most interests me. This idea can be applied regardless of where the interest may be on the exposure scale.
Placement of my main subject has always been my intuitive priority. In general though metering methods are not taught from that perspective.
For example with incident metering we are first taught to use the reading displayed, this is a dumbed down technical version of Dunn & Wakefield's concept, it works well for many, but it lacks the real thought that Dunn & Wakefield imply regarding the relative importance of different parts of the scene.
"But in viewing motion pictures the upper and middle tone areas - and particularly the human face - claim so much attention by virtue of motion (and sound where applicable) that the scene has normally changed before there is time for the shadows to be critically examined at all." page 33
If that "motion" doesn't describe the world we live in and the way photos are viewed today, I don't know what does.
Dunn & Wakefield seem here simply to concede to the norm of what people expect first.
My thought here actually falls at least partly outside the exposure question, in composition. If I catch the eye with the brightest spots where do I take them from there. Could be to smaller details rather than darker shadows.
In any case the shadows seem generally to hold only a supporting role
Personally I'm with Dunn on this.
For me, I prefer making those decisions at the camera and I can't remember a single shot where I got the main interest right and said to myself, "darn it I wish there was more shadow detail."
Don't get me wrong, I'm not averse to using the latitude of a film to get a shot, or to make shooting easier/faster, or to avoid underexposure where there is something in the shadows I want but I'm not fishing for more ways to interpret a scene in the darkroom.
I have made an observation about myself with regard to calibration of late.
As I gain experience and my darkroom skills grow the people at Ilford and Kodak and Fuji and Sekonic all look smarter and smarter.
The windmill diagrams you provided earlier seem to bear this out too. The +1 diagram shows a more pronounced/longer curve toe in the reproduction quadrant.