I thought that was an actual issue dealing with "loudness" and the dynamic range of the sound. ie, newer media is formatted so that it's so loud, it will clip parts off before it you can hear the softer bits in it. I could be way off though, I'm not really and audio buff in that sense...
It was more a case of hard clipping vs soft clipping. You could get tube ( valve ) amplifiers moving high tension volts way over spec at the cost of lots of third harmonic ( mostly ) distortion by providing things like DC amplifiers. Semiconductor amplifiers would simply turn into a digital switch with enough current. Fast enough current and they're more or less flip flops.
Normalising within 16 bits has led to loudness wars. That's the medium. Analogue chains all overflow towards a chaotic end. Some people prefer that chain of events.
You're confusing a couple of things here. DR is substantially increased with digital recording; so is SNR. Behaviour at the upper end of the DR differs between digital and analog media. Whereas tape, for instance, saturates (which may induce pleasant distortion!), digital audio clips (which induces headaches). However, that doesn't really matter in practice if the audio engineer in question is worth his salt. Levels at every point of the digital chain need to be adjusted accordingly: the old strategy of keeping everything close-to-maximum doesn't work anymore. One keeps everything well below clipping point instead.
Overeager compression (specifically, limiting) is a whole different problem, especially with modern pop production.
"Loudness" is an orthogonal debate revolving around mixing (pop music). There's nothing about the various media that requires or prohibits one from mixing 'loud'.
vinyl vs CD, tube vs solid-state/digital -- those are more similar to film's 24fps vs higher frame rates: straightforward arguments against higher-fidelity technologies because we've grown emotionally attached to the particular artifacts of legacy technologies.
