Firstly 144x - sorry, reading back through it I realise that my thoughts were not in line with my typing. I was just trying to make the point that even if you scrape some of the graphite off the cable whilst laying, you will still have quite a bit of graphite embedded in the sheath, making a sheath test still valid and a good indication of damage whilst laying. If you terminate the cable with a gland isolated bushing (for protective purposes), you will physically have to scrape a millimetre or two (depth) of the sheath away (all around the cable) to get rid of the graphite (hence the conducting path), otherwise you will be shorting out the isolation.
The semicon layer that I think paged and vector9 are describing is for stress control and corona minimisation which can be caused by the non circular nature of copper stranded conductors. This is described as the conductor screen, and is usually extruded over the conductors (in extruded cable technology) to provide a circular equipotential surface. The same applies to the insulation screen, where a semicon layer is applied over the bulk insulation (under the earth screen), again to provide a stress free surface. The graphite goes on the outside of the cable, outside the metallic earth screen, hence it is outside the insulation electric field.
Hope this helps