I recorded into an .aup file @ 9600/24 - if I export to a .wav file @ 4800/24, is the music being compressed?
It’s being resampled. Compression in the general idea is leaving out portions of the music that you’re not going to miss anyway. Both 96/24 and 48/24 are “perfect” formats. Nothing gets left out. Formats like that are useful if you’re going to do extensive production, filtering and effects. If you’re not, then 44100, 16-bit Stereo (Music CD quality) works just fine. If you go much lower than that, say 32000, that’s about the quality of FM radio and yes, many people can hear that. There aren’t enough bits there to carry high quality sound.
But that’s still not compressed. That’s just bad quality.
Koz
I recorded into an .aup file @ 9600/24 - if I export to a .wav file @ 4800/24, is the music being compressed?
You left out a zero… 96,000 = 96 kHz. 
OK… You ARE throwing-away data and resolution, but as Koz says downsampling (sample rate or bit-depth) is not considered compression… You can downsample to “telephone quality” (8kHz/8-bit) and you’ll loose quality and resolution, but that’s not considered compression. (Cell phones do use lossy data-compression)
The guys at [u]HydrogenAudio[/u] who do scientific-blind ABX listening tests “everyday”, will tell you nobody can hear the difference between a high-resolution original and a copy down-sampled to 44.1kHz/16-bits (CD quality). It turns-out that 44.1/16 is better than human hearing. Often, you won’t hear the difference between a high-resolution original and a high-bitrate MP3 (in a proper blind test).
There are two categories of file-compression. Lossless compression (FLAC or ALAC) can compress an audio file to around 60% of it’s original size. It’s similar to making a compressed ZIP file… When you decompress it, you get the exact-original data back.
Lossy compression (MP3, AAC, etc) tries to throw-away “details” that you can’t hear or that you are least-likely to hear (imagine someone whispering next to you at a rock concert… You can’t hear it anyway, so it doesn’t matter if it get’s thrown-out of the recording). A good-quality (high-bitrate) lossy audio file can be about 1/5th (or less) of the original file-size. Hopefully, it will sound identical (or almost identical) to the original but you can NEVER get the original data/bytes back. (JPEG is lossy compression for images.)
Sometimes people confuse dynamic compression with file compression. Dynamic compression reduces (compresses) the dynamic range by making quiet parts louder and/or loud parts quieter. Most popular music is highly dynamically-compressed for that modern constantly-loud sound. This has NOTHING to do with the MP3 or CD format. You can have a very dynamic classical recording in MP3 format.
imagine someone whispering next to you at a rock concert
It’s sneakier than that. Good compression can turn an expensive violin into a cheap violin and most people don’t notice. That’s why performers hate it.
Koz