I actually put this theory to the test. I took a pickup with no magnet in it, measured the Gauss, and it was like 3 to 5 due to extremely low residual magnetism in the slugs. I hooked the pickup up to an amp, measured again, and the Guass didn't change at all. Any small quantity of voltage coming from the amp only produced microscopic amounts of electromagnetism, if any at all. I now have a test pickup that has not slugs or anything, I'll test it out tonight and see if I find the same result in that case as well. OTOH, I tried my best to demagnetize an AlNiCo bar by exposing it to the opposite polarity of a neodymium, and IIRC I couldn't get the bar to read any lower than 10 or 20 Gauss, where it would typically read around 500. So my belief is that the magnet in OP's pickup is not truly dead, and that it's producing some magnetism, even if it can't be directly felt. Another thing I discovered is that the audible output won't be linearly proportionate to the magnet strength. For example, if you take a 500 Guass AlNiCo and take it down to 250 Guass, the audible output doesn't drop in half, in only becomes a tad quieter. It's more of a change in tone than audible output, so the magnet can get pretty weak and you will still hear something. I've tested the individual strength of all six pole pieces in various Strat pickups in my guitars and much to my surprise the pickups retain good string to string audible output, even when the amount of Guass varies widely from magnet to magnet. I found that a lower charged magnet will give slightly a softer, slightly less harsh top end, and that when there were under charged magnets among fully charged ones, it was impossible to perceive any string to string variances when strumming full chords. If the whole pickup is made up of undercharged magnets, THEN you will hear that low-Gauss difference when strumming a full chord.