Dead spots on necks are, depending on how you want to think of things, not all that simple. A neck, if you excite it by attaching something to apply force, will resonate at a bunch of different frequencies, and in a bunch of different geometric fashions - there are bending modes (mostly what we're interested in here), torsional modes (which might play into dead spots, I'm not entirely certain about that), and modes from compressional waves in 3 axes, which probably don't affect us much in the bass world as far as necks go, but the "zing" on new round wounds probably involves some of those.
OK, so you have a bunch of different resonances, at a bunch of different frequencies. If you look at how they move, there end up being "standing waves" if you let the resonance settle from the initial excitation. A standing wave means there are places that are moving, and places that aren't moving. In the lowest frequency vibrational mode, if that is happening all by itself, the body is relatively stationary, the end of the neck is moving back and forth along with the whole rest of the neck, but the end is moving the most. That mode is usually low enough in frequency, and the most motion is at a place where strings aren't attached (the very end of the neck) that we don't typically encounter it - Fenders don't tend to have issues with that mode, but there are some headless basses, and a few basses with 2 + 2 headstocks, where if you have light enough tuners on them, the open A string or a nearby note will be a dead spot. I have a bass tuned in D standard (the "A" string on that is actually tuned to G) where I had to put heavier tuners on it to tame the dead spot at the end of the neck.
The dead spot we talk about a lot is a different mode - in that one, the neck is not all moving one direction, the end of the neck is moving one way, there's a point in the middle where the thing is sitting still, and then a section that's moving the other way. Again, the body is relatively stable (it's so much heavier than anything else in the system, so it tends to be relatively stationary). Anyway, when the placement of a fret and the tuning of a string at that fret corresponds with an antinode (a place that likes to move a lot in one of the vibrational modes), the string loses energy relatively fast - it's transferring its energy and feeding the vibration of the neck in doing so. In the case of my D tuned bass, it has no noticeable dead spot issues where we tend to look for dead spots - the D tuning means that the location of the fret where there might be a problem has moved, as it's 2 frets away from where it would be on a standard E tuned bass - that is enough in this case to make it practically immune to the dead spot typically in that area of the neck.
If you replace the string with a source of force at a fret location, where there is a dead spot, and look at the force and amplitude of neck motion, yes, there is a quadrature relationship (90 degree phase shift) between the forcing and the motion (amplitude in this case - the velocity and acceleration are all shifted from each other, so you have to specify what part of the motion you're talking about when you start to discuss phase shifts of things). We wouldn't call that phase cancellation, however - phase cancellation is used to describe systems where 2 things in antiphase (aka opposite polarity) tend to cancel one another - a humbucking pickup would be a good example of phase cancellation, where two hum signals cancel one another.
Extra credit for math wanna be geeks: Things in quadrature (a 90 degree phase shift) don't cancel, they add (but a bit funny) - if they're equal in magnitude, the result of adding 1 and 1 is not zero (which would be phase cancellation, or addition of two signals at 180 degree phase relationship) or 2 (which would be the result adding things in phase - or zero degrees phase relationship), its...the square root of 2. Those fluent in the Pythagorean theorem will understand this implicitly - a 90 degree phase shift means a right triangle, so you're calculating the hypotenuse of an equal legged right triangle. To add 2 equal things together and get less than one, you need a phase shift between 120 and 240 degrees - the closer you are to 180 degrees, the more they cancel.