110902-0952 EDT
The answer is probably yes. Depends upon the test equipment you have available, and the wiring.
An assumption is that the power flow on all phases is from a generating source to a consuming load. This means that at the load there is no generating means on one phase.
You can use a clamp-on current sensor if the current carrying wire is available. Then do a phase comparison between said sensor and the CT output. Means you need to define an orientation for said sensor based on some reference CT that you test on the bench. Easiest to do all this with a scope.
Without the extra current sensor, but using a scope, compare the CT output phase relationship to the voltage of that phase. Do this on each phase and see where the disparity is.
If you need a more detailed description I can provide that if you can define the limitations. Limitations might be the line in question is too high a voltage to work with directly, can not put a current sensor around the line the CT is located on, etc.
.