It seems to me that you don't need a decoder, but obviously it makes it easier and quicker.
I've been using the M/S function in Audition, but I think you can do it manually. All you'd have to do is the following...
OK, so after the show, you have two recorded channels...1) the MID channel (usually recorded to the left channel or channel one and 2) the SIDE channel which is usually recorded on the right channel or channel 2.
Take the Side/Fig 8 channel and insert that into both the Left and Right channels of the new mix, but reversing the phase of the right channel.
In the new mix, pan the left channel 100% left and pan the right channel 100% right.
Now mix the MID channel with the above channels and vary the percentage of the mid (volume) in the final mix to vary the amount of stereo effect you hear. In other words, if you keep mid at 100%, there will be a little stereo sound in the final mix...but if you decrease the mid to say 50 or 60 percent, you'll hear more stereo channel separation.
I'm pretty sure this is how it's done manually. Anyone with more experience, please make any corrections to the above.
Read this article for more info. and explanation, especially the part about 'mixing'...
http://www.wikirecording.org/Mid-Side_Microphone_Technique