LoveMarks - Statistically Wrong?

Just when I was beginning to partially like the second half of the LoveMarks book, Kev hits me with a doozy,

"I have always had an uneasy relationship with stat flacks.  My gut instinct is to protect intuition from measurement.  To nurture ideas away from the glare of metrics.  To feel about, rather than factor in."

So far so good Kev, I'm with you all the way.  But then he goes the complete opposite way,

"Peter and his colleague John Pawle created a set of sophisticated questions to measure emotion.  The results of the first tests from QualiQuant?  They exceeded everything we had hoped for."

So we didn't like stats but now we do!
But here's the classic....

"And they showed that Love and Respect are what matter.  The correlation between the two is 0.60, if that's the way your mind works."

Now correct me if I'm wrong, but doesn't that correlation undo the whole LoveMarks theory?
Kev suggests that LoveMarks consist of two axis, Love, and Respect, giving 4 quadrants,

High Love and High Respect (LoveMark)
High Love and Low Respect ((Fads)
High Respect and Low Love (Brands)
Low Respect and Low Love (Commodities)

If High Love and High Respect is a strong correlation, then doesn't this suggest that Low Love and Low Respect are equally correlated along the same straight line, whereby Love and Respect are proportional to each other along the same straight line?  With me so far?
So doesn't this mean that according to Kev's stats, it would be rare to have High Love and Low Respect, and High Respect and Low Love, because all the stats fall along the single line of Love and Respect?
So the 4 quadrants don't exist?  Just two quadrants of High High and Low Low!!
Can someone confirm this, and put this LoveMarks thing to bed once and for all!
Meanwhile, I'll continue reading. 
So far some good bits, some contradictory bits, and some really bad bits.