Except in tango, we never achieve CBMP by rotation occuring during the CBMP step - it is always a result of a rotation taken on an earlier step which has set up a trend of moving across the body.
Never say never. The Feather is a good working example of your premise, but there are examples of CBMP being the direct result of rotation on the
same step (eg natural pivots, spin turn, etc), making "not usually" more accurate than "never".
There are also situations where CBM and CBMP are concurrent (eg the forward step OP following a chasse), and in such situations it's debatable as to whether the position is a result of the rotation or not. If you're a stickler for book technique, you'd argue that your body is already set up in the position, and that the CBM is minimal and strictly of the curving variety. Those who prefer a freer and fuller swinging action will break from the book technique slightly by allowing the body to wind up prior to the OP step. When this is done CBMP becomes the result of the CBM. So like I said, it's debatable.
Still, this is all beside the point. When I speak of CBMP being the result of CBM, I'm usually teaching someone what *not* to do in normal closed position situations. From a teaching standpoint, this is by far the most common mistake of newcomers (and even some not-so-newcomers).
Regardless of the frequency of occurence of any particular action in actual dancing, the fact remains that in terms of the basic principles of movement, CBMP can be achieved either by altering your direction, or by maintaining your direction and altering orientation (ie rotating), or a little of each. Assuming you're not deliberately stepping across your track, whether or not you end in CBMP is determined by how quickly you turn your body in relation to how quickly you travel. Beginners make the mistake of turning too quickly in relation to their progression, thus finishing in CBMP when they shouldn't. How many times have we seen the man's left side dropped out and pulled back into step 1 of the basic reverse turn in Viennese Waltz?
Of further interest in the feather is that in the basic version there is offically no turn on any step.
"Officially" there is actually turn. It's just not stated clearly in those terms (Then again... what ever is in the ISTD technique?). The evidence of rotation is (1) the existence of CBM on the first step, which defined as rotation, (2) the second step having a "left side leading". You can't be square to your feet on step 1, then have a left side leading on 2 and *not* turn in between.
Remember that the ISTD technique specifically describes turn as being measured through the
feet. So when you read the chart and see that there is no turn on a feather step, all you're being told is that the feet don't turn. In other words, "amount of turn" speaks nothing to the rotation of the body. For that, you are required to read between the lines, and that's where concepts like "CBM", "side-leading", "body turns less" etc come into play. It's the only way in which the ISTD makes reference to body rotation independently of the feet. Convoluted though it may be in its description, body rotation on the feather step is most definitely "officially" recognized.
At any rate, regardless of what the ISTD thinks, on any good Feather there is definitely some additional rotation between 1 and 2. Without it, you'd either be starting step 3 totally squared-off to your partner and thus ill-prepared to step O.P., or you'd have to already have your left side leading on 1, which is the same as saying you'd be in CBMP.
Regards,
Jonathan