Harpoons are effective when the target is out of reach, or moving too quickly to catch. Neither is generally true when one spacecraft maneuvers near another in Earth orbit. Getting close enough in relative position and velocity to harpoon something can take a huge amount of Δv, but once there, the additional Δv to approach and contact the target spacecraft is minimal.
The BBC News article Big harpoon is 'solution to space junk' highlights AirBus' research into satellite harpooning. Is this really easier or more reliable than some kind of grappling or connecting or netting o "tying-on-to"? Surely large modern spacecraft have several sturdy things to hold on to already.
Is de-tumbling the presumably dead target spacecraft really easier if it is on the end of a harpoon? It seems much easier to de-tumble something if you are rigidly attached and can apply torque. If you only have it at the end of a line, how could one apply the correct torque to zero out all angular momentum? Wouldn't one just get wrapped around the target in a tangled mess?
The sentence "Care would be needed to avoid hitting any pressurised tanks onboard." in one of the figure captions highlights another issue with ballistic harpooning.
There's more information in the article and accompanying video RemoveDebris: Space junk mission prepares for launch.
Click small images for larger size:
below: "The miniature harpoon to be tested on the upcoming RemoveDebris mission." From BBC.
below left: "Envisat artwork: Care would be needed to avoid hitting any pressurised tanks onboard." Credit: ESA. From BBC.
below right: "The spacecraft has been assembled in the UK and will soon be packed up for launch." Credit: Max Alexander. From BBC.
below left: "The spacecraft will establish whether a net could ensnare a small satellite." Credit: RemoveDebris Mission. From BBC.