I've done a little more thinking about this - I was curious about the relationship between the torque applied on the axle nut and how if affects how the axle assembly is tensioned. Based on this, I've come around more to the thinking of EricV and Dallara (and others) that using the pinch bolts to hold the axle, even if you back off and re-tighten later, is probably not a very good idea.
Just as a brief background, the main reason for tightening the axle nut is to take an assemblage of left diff housing, inner bearing races, rear brake carrier, and spacers, and make it behave as a single unified assembly fixed to the axle and around which the outer bearing races rotate without side load. When you tighten the axle nut, you're tensioning the axle by pulling the right collar of the axle (that is eventually clamped into the right swingarm) against the races/spacers, etc. You want all this axial tension to be in the axle assembly and thus want the right side of the axle unfixed in the swingarm - if it is fixed, then some of this tension can potentially be absorbed by the squeezing of the swingarm.
While it can get more complicated, there is a basic rule-of-thumb equation that relates bolt/nut torque to tension axially along the bolt (or in our case the axle):
T (in-lb) = F (lb) x D (in) x 0.2
Where T is the torque on the axle nut, F is the axial tension, and D is the bolt/nut diameter. Plug in the 1080 in-lb spec for nut torque and assume a nominal 1" or so nut diameter and you end up with about 200-lbs of axial tension in the axle after the nut is torqued to spec. This number is surprisingly low to me - I would have expected that a high torque like this would have put several thousand lbs of tension in the axle.
Assuming this 200-lbs of tension is correct, then the question is whether having the swingarm fixed on the right by clamping it with the pinch bolt might cause a significant reduction in this tension as the swingarm takes the tension rather than the axle. On this, I suspect it very much matters how tight you have the axle nut before you tighten the pinch bolt. If you just loosely thread the axle nut on the axle, then immediately clamp the axle and start tightening the axle nut, I bet the squeezing/deflection of the swingarm is very significant and it might absorb a large portion of the tension before the axle started being tensioned. Alternatively if you get the axle nut good and snug before clamping, I bet the deflection is rather small and the swingarm isn't going to take much tension away from the axle.
But I'm just waving my arms in this last paragraph based on mechanic's intuition - I really don't know. It would be interesting to perform some tests. But I must say that given the rather low axial load generated by the axle nut, I don't think it is a very good idea to use the pinch bolt as a means to hold the axle stationary while torquing the axle nut to spec. If you're in a terrible bind and don't have an axle wrench and must use the pinch bolt to hold the axle, then I'd really try and get the axle nut as tight as possible before clamping. And I'd also suggest doing a few extra steps to re-tighten the axle nut a time or two in cycles to allow any swingarm tension to be relieved incrementally.
As others have said, the correct approach is simply to use the appropriate axle wrench to hold the axle when tightening the axle nut.
- Mark