Wonderfully interesting replies thus far! Sharke hit the nail on the head and then drove it into the coffin with his description:
 

sharke
I get what you're asking for (I think). You want to take a sound that's spread out over the stereo field, and do things like "boost the volume of whatever's sitting at 30% left" and stuff like that. Using an interface that's like an EQ, but instead of the horizontal axis representing frequency, it represents pan position. 

I'm not sure I could put in it in better words, but I guess I'll give it another shot based on what sharke wrote - 
an interface that looks sort of like an EQ, where the Y/vertical axis still controls the volume, but INSTEAD OF the horizontal axis controlling which frequencies are boosted and attentuated, it controls which PAN POSITIONS are booster and attentuated.
 
So that one can take, for example, a recording from a stereo pair of drum overheads, and then "carve out"/turn the volume down on, let's say, +30% and -65% to make room for a guitar part and a piano. Of course this could ONLY work on stereo recordings, otherwise it's just a pan pot.
 
So, being able to adjust the volume of a track at individual pan positions, leaving the rest of the stereo image in full volume. It definitely seems like it ought to be possible, but I guess it can only work when sounds are more or less the same volume across the entire stereo field.
 
I'm gonna have to go back and check out the replies a few more times and look at some of the posted plugins to see if anyone has something that matches the description though - but hopefully that clears something up. 
It's not a plugin that I *NEED* or even plan to use much - it was more of just a thought that came to me that could make for a really cool and useful tool that eliminates a little bit of the guesswork when it comes to making space in the mix for a large number of instruments.

The absence of a qualifier such as "I realise that stereo placement is determined by the relative levels of the left and right channels, but..." makes me think that the question is naive. Otherwise it's an interesting concept

Kev999
The absence of a qualifier such as "I realise that stereo placement is determined by the relative levels of the left and right channels, but..." makes me think that the question is naive. Otherwise it's an interesting concept

Well, yeah, that's obvious. This concept would only be effective in the case that the track/recording is stereo and has information across the entire stereo field, and not just hard-panned L/R, for example. So like a chorus or reverb effect.
 
But now that I think about it, there are only those two L/R channels that determine the stereo placement, so actually yeah, it is a bit naive. It'd actually only be a SUPER-FINE attentuator for those two channels. 
 
 But I'm just a kid learning the art in his free time 

It’s an interesting idea, but I don’t think it really makes sense. Panning isn’t really moving a sound around, it only adjusts the volume of different speakers. When you pan a sound to the left you are just turning up the left speaker and turning down the right. The placement of sound is an auditory illusion, so turning up or down something at a certain pan position doesn’t seem possible.

Of course someone could make something that functions like our human perception and splits the audio into bands based on the relative ratios between the left and right channels. I don’t know how well this would work though and it might cause some artifacts.

Now that I think about it, you might be able to do this with Meldas panorama crossover, but I’m not sure.

davdud101

Kev999
The absence of a qualifier such as "I realise that stereo placement is determined by the relative levels of the left and right channels, but..." makes me think that the question is naive. Otherwise it's an interesting concept

Well, yeah, that's obvious. This concept would only be effective in the case that the track/recording is stereo and has information across the entire stereo field, and not just hard-panned L/R, for example. So like a chorus or reverb effect.
 
But now that I think about it, there are only those two L/R channels that determine the stereo placement, so actually yeah, it is a bit naive. It'd actually only be a SUPER-FINE attentuator for those two channels. 
 
 But I'm just a kid learning the art in his free time 

Hmm, it's an interesting concept to think about. The reality of how we record and edit things is that we have (for stereo) two channels, two outputs, two speakers. In order to "pan", there's no magic going on, it's just adjusting the level of a signal to one channel relative to the other.
 
But our ears and auditory systems perceive location information using a variety of methods, such as timing differences, frequency differences, and loudness.
In order to achieve more "weird" panning effects, we need to therefore recreate this by adjusting the relative timing (phase), frequency and loudness of signals in each channel.
 
I know of one plugin (Pan Noir) which is designed to pan individual channels of a classical concert-style recording, so that they match the timing delays of the main stereo pair used. Other than that, mix engineers use a few funny little phase and EQ tricks to place particular effects in the mix.
 
So, yeah, whilst the original question may possibly be naive in it's assumption of stereo information, it definitely raises some interesting ideas about unusual panning processing.

After giving it a moment of thought, I understand why it wouldn't work.
 
You guys are right, there is no stereo field, it's an illusion - there's only how much or how little a sound is heard in one channel vs the other.
 
Otherwise an application that's been described here would need to be capable of constantly determining which sounds in the left and right channel correspond to one another (based on present frequencies, timing, volume, etc), possibly create some sort of "spacial map" based on how loud certain frequencies are in each channel, make adjustments to the correct frequencies, AND still playback live with no latency.
 
I'm pretty terrible at explaining things as most of you guys have seen lately, but this is my way of putting it. It's also a bit tough to explain because the words I wanna use aren't on my tongue and the concepts I mean to explain are still a bit foreign for my pea-sized brain.
 
But it's probably not doable.
 
Heck, it's definitely 100 times easier sticking with a pan pot and a volume slider in this case.

http://wavearts.com/products/plugins/panorama/
 
Regards,
Dan

I like Cableguys PanShaper2:
 
http://www.cableguys.com/panshaper.html

There is a lot more acoustics to this than meets the eye. On the transmission end, sound is monophonic (per source), so the only way to get such a distribution is from multiple sources, but the result will still produce a net result due to how it propagates. On the receiving end, pan is determined from the time delay at the receivers (again the net result), and you cannot affect human capability beyond what it is. Two receivers also "dumbs down" the resolution capability, so a wider separation is necessary to determine multiple sources. Even with hundreds of detectors (a real SONAR), it only provides better accuracy to each monophonic source (and the ability to resolve them).
 
Just food for thought... engineers cherry pick situations to suit argument... with two speakers playing the same thing, you get a "center pan"... however... turn your head 90 degrees and now it pans to the ear closest to both (the net result) sources. The position of the listener plays as much into things as the source. Get far enough from 2 speakers, and they will collapse to mono (loss of resolution).

mettelus
Just food for thought... engineers cherry pick situations to suit argument... with two speakers playing the same thing, you get a "center pan"... however... turn your head 90 degrees and now it pans to the ear closest to both (the net result) sources. The position of the listener plays as much into things as the source. Get far enough from 2 speakers, and they will collapse to mono (loss of resolution).

That's not cherry-picking situations to suit an argument, that's just discussing the correct bleeding context. FFS.