Philosophers prove that digital recording is better than vinyl

Microphone and headphones

RedShark replay: There is an argument about digital audio versus vinyl that doesn’t seem to want to go away. But I think it should, and there’s a good reason why: it's a Category Mistake.

It’s often said that digital audio is inferior to analogue because of the sampling process. Absolutely by definition, sampling doesn’t record the entirety of an audio waveform. With sampling, you record important blobs, with absolute regularity, but you don’t record anything between the blobs.

So, the argument goes, you’re missing out important information.

But, digital audio theory tells us that you’re not missing out anything. If you sample often enough, then all the information you need to reproduce frequencies that you can actually hear, will be present. Some audiophiles think that there are important harmonics above the human hearing range that interact with each other to create beat frequencies that are back in the audio spectrum but it’s always possible to to counter this by increasing the sample rate.

To me, this debate is a false one. We’re not comparing like with like. We’re not even comparing anything.

For around a century now, we’ve known that atoms make up the stuff that the real world is made of. Tables, fruit and even you and me: we’re all made of atoms and their constituent parts. These atomic and sub-atomic components of matter are incredibly small, and they have - in relative terms - enormous spaces between them. So much so that you could almost argue that there’s virtually nothing there. When we look at matter, we’re looking at largely empty space.

But that doesn’t mean we can store a bookcase inside a sofa, nor that we can walk through wall. Solid matter is solid, without any doubts, both before and after we realise that matter is largely made of nothing at all.

This knowledge will absolutely not save you if you jump out of an aircraft without a parachute, of if you hit your thumb with a hammer. Both are going to hurt. The empty space within matter is not space in the conventional sense. Matter is 100% matter.

So why the confusion?

It’s called a Category Mistake. This is a term introduced by the UK philosopher Gilbert Ryle in his seminal 1949 book “The Concept of Mind. I first read it when I was at university studying philosophy. Essentially Ryle argued that the “mind" is not conscious but is instead a collective term for all observable behaviours and dispositions. This, says the philosopher, is an instance of a Category Mistake. One example he gives is of the student who visits a university and says “I can see buildings - lecture theatres and libraries etc - all around me, but I can’t see a university anywhere”. The mistake is that the university is not a thing in itself that you can stick a label to; it’s more a term that encompasses the buildings, the activities within them, and the overall purpose of the university.

So far, so tenuous.

Here’s where I think the Category Mistake is in the argument about analogue vs digital.

Sampling is designed to capture sufficient information to reproduce audio accurately. Yes, it’s possible to sample audio inadequately, and it’s arguable that even CD quality audio isn’t enough (although with a potential dynamic range of 96 dB and a frequency range up to 22.05kHz it’s way more capable than vinyl, at least in terms of specifications). But it’s easily possible to sample more frequently — routinely up to 192 kHz - and with a greater bit depth — 24 bits is commonplace and internal processing can be at a staggering 32 bit floating point; virtually enough to record the Big Bang.

And yet there are those who would say that even at high sample rates and extreme bit depths, digital audio is an incomplete facsimile of the original analogue signal. “There are gaps between the samples” says the argument.

Does this start to sound a little bit familiar? Is’t this a bit like the notion that matter doesn’t really exist because there are spaces between the atoms?

This is as clear a Category Mistake as I have ever seen. Those gaps are meaningless. You could sample at X-Ray frequencies and it wouldn’t sound any better.

Are there reasons to prefer analogue audio over digital? Yes, absolutely. But none of these have anything to do with accuracy.

Mic graphic by shutterstock.com