Hi Caroline,
Not sure I understand you question exactly - but does this help?
- Yes, ulnar nerve innervates 2 lumbricals on the ulnar side - but an ulnar claw primarily occurs because the interossei have been paralysed. Remember that each finger has 3 muscles that primarily control MCP flexion - one lumbrical, a palmar interossei and a dorsal interossei. In an ulnar claw all 4 fingers claw - the ring and little finger most obviously because they lose all 3 muscles, but index and middle fingers will also have at least a 'latent' claw (you may only 'see' it in a power grip) because they also lose 2 (interossei) out of the 3 muscles that control MCP flexion.
In an ulnar claw the MCP hyperextension is due to unopposed extension force of EDC and the extensor mechanisms on the MCPs. All that force 'stops' at the MCPs, and the lumbricals are no longer contributing via the lateral bands to get PIP extension.
In a median nerve injury there is paralysis of index and middle finger lumbricals - but you never see overt clawing of these fingers unless ulnar nerve is also paralysedbbecause index & middle finger still have functioning interossei to control MCP flexion. Median paralysis most obviously affects palmar abduction / opposition / flexion function of the thumb. This remains obvious even in a high lesion that involves FPL - only difference may be overpronounced IP flexion in a distal lesion as FPL tries to flex/oppose/palmarly abduct the thumb, but doesn't have help to successfully do so at the first MCP joint. Obviously there is no IP flexion in a high (proximal) lesion.
So - don't forget the interossei! I think understanding their function might be key to your question about paradoxes.
Hope some of that makes sense. Does it help answer your question?
Anne