Since I'm right 97% of the time, I feel entitled to continue the discussion even in a degenerate thread like this one without worrying about the other 6%, so:

It is my observation that there generally is little modulation of dynamic range in informal group playing, such as in most sessions, since (in sessions I go to often enough to observe) the eleven participants with the big Martin guitars will force the other three with musical instruments to play as loud as possible to be heard at all. Listening to the "Irish" sessions at our recently completed festival suggests a similar effect with respect to tempo due to the tendency for the seventeen hammered dulcimers to force everyone to play faster than appropriate.

Since "session playing" is a principal learning method for a large percentage of us, we do "learn" that most music should be played too loud and (Irish music) is supposed to be played too fast.

The effective use of dynamic range and modulation of tempo is, in practice, currently confined to solo performances by "misfits" who ignore the rules.

A small professional group can effectively produce a solo performance (a group deliberately "acting as one") if they practice doing so, and if they perceive that it's a good thing to do.

Most "classical" music, regardless of the size of the orchestra, is a solo performance by the conductor, who, by mutual agreement of the members, dictates variations in tempo and dynamic range thereby "playing the entire orchestra" as as his/her "solo" instrument.

The differences in dynamic range and tempo modulation heard in music, and especially in broadcast music, are not generally characteristics of the music genres. They represent primarily the "ignorance" (or perhaps simply the uninformed(?) preferences) of the performers. Note that the "preferences of the performers" may accurately cater to the "uniformed habits of intended audiences(?)."

The "compression" already mentioned also contributes, but the real reason for this compression has not be explicitly stated. Compression of dynamic range was necessary, especially with the advent of "LP" recording, because the recording media had insufficient ability to record large dynamic range without distortion and "cross-talk" between stereo channels. Similar limitations persist in newer media, pickups, amps, and speakers, and compression "on the disk" persists (almost) universally.

The compression can be according to a variety of rules, but is always non-linear. In effect, the rate at which the amplitude changes on the recording affects how much the amplitude increases when decompressed in playback. Since a change in pitch is a similar "rate" change, there sometimes is significant cross-distortion, so compression and decompression are theoretically imperfect; but the result generally is preferable to the distortion that would result with uncompressed recordings.

Ideally, all the decompression should be done at the single point where the electrons are converted to air blips; but some broadcasters do a decompression from the recording media to the "radio signal" that gets broadcast. This may result in changes in signal amplitude sufficient to cause "drops" in reception in fringe areas when the "music goes quiet," so some broadcasters only "partially decompress." Some amount of decompression is usually done because not all "receivers" include the expected decompression capability (although there probably are few now that don't).

The bottom line is that dynamic range (in broadcast music) is only "slightly" and "sometimes" a characteristic of a particular music genre. It's affected by the recording and broadcast methods, but mostly depends on the (largely ignorant?) expectations of the consumer segment for which it is produced.

John