Actually, I'll respectfully quibble with your quibble. /ubbthreads/images/graemlins/grinser2.gif
In my observation, Axis locks and lockbacks often DO have detents - created by shaping the outer edge of the pivot area to create differential tension against the locking bar.
My Benchmade RST's are a great example - there's a fairly deep divot in the base of the blade where the Axis locking bar rests when closed, with a curved hump at the end of the divot.
To open, you need to apply enough pressure for the hump to push the locking bar back through about 75% of it's total travel, where it rides until the blade fully opens and locks- if that's not a detent, I don't know what is.
Older Spyderco, Gerber LST, and others show what happens when this is left circular - equal and smooth tension throughout the arc, until locking. By 'camming' the pivot, the behavior of the blade is altered.
The 'native closing bias' is nothing of the sort - you can make any of those systems have no such thing, simply by leaving the pivot circular, as was the norm for many years.
Slip joint and some of the more 'exotic' locking systems are a bit different story, but liner lock, frame lock, lockback, and axis lock all have a choice to use a detent or not, either through the ball-and-pit method, or by 'camming' the pivot.
While I'm the first to admit it's not conclusive, it sure seems to me that this feature has become vastly more common in the last decade or so, at least in parallel with, if not necessarily because of, the rise in thumb-stud and smooth-action features.