... in fact it wasn't really anything at all.
A few days ago, I wrote about hearing a voice in the middle of the night. I had a few ideas at the time what I may actually be hearing, but it was only a couple of nights ago when I had the chance to try out my theory and see whether it was what I suspected. And it was.
I actually found it quite an interesting puzzle. It's a bit like watching a magic trick and trying to work out how it's done. The first thing you have to do is draw the line on the things you're not going to consider as answers. So when trying to work out how a trick is done, this is the point at which you'd say "the magician did not really make the rabbit appear from nowhere". You have to disconnect what you actual saw from what you think you saw. And so I had to stop thinking about what I think I heard, and think about what it was that I actually heard.
I've written in the past about pareidolia. It's the thing which causes people to believe they've seen Jesus in a cupcake or a human face on the moon. It's also the thing which causes people to "hear" satanic rantings when playing 70s rock songs backwards.
The answer to the problem was - rather disappointingly - very simple. Where I was lying in bed, noise drifts up from the street. Noise from the station right next door and the normal rumblings and humming of a modern home filled with machines and electronic equipment. Silence is never totally silent anymore. I had surmised that what I was actually hearing was a combination of background noise of about the right frequency and rhythm to suggest speech. In the same way that once you've seen one of those autostereogram pictures you can persuade your eyes and brain to let the picture go, or can easily bring it back again - my brain had started to hear something which sounded like a voice and so was interpreting any sounds it heard of about the right type to be a voice - just too unclear to be made out.
So the other night I was in the flat on my own and it was quite late and as quiet as it was going to get. I lay in bed with the light off and forced myself to listen for voices. I stopped listening explicitly to the sounds I could hear, and tried to tune out and search for the sound of a voice. And I heard one. Well I didn't, but I did manage to trick my brain into being convinced that there was a indistinct male human voice just out of clear earshot. It was also pretty easy to snap back out of it into hearing just the sounds for what they were too. It felt exactly like forcing yourself to see and then unsee one of those pictures.
So there you have it. No ghosts, just an artefect of the human brain's love for finding patterns where there aren't any.
Of course, this isn't truly scientific. I had a theory about what may have caused me to hear what I heard, and I manage to recreate the same result using this theory. But I know the pedants will (quite rightly) say that doesn't prove what caused it on the original night. But given that the theory fits the evidence, and gives reproducible results I'm quite happy to live with it as the explanation.
However, unlike people who would dogmatically say this genuinely was a ghostly voice, I'm quite happy to change my view if a better answer should come along. You tend to find that those people who report these things as genuinely supernatural spend more time trying to discredit any explanation other than their preferred one than investigating alternatives. A true scientist would never have a preferred answer in the absence of evidence, and that preference would only have a lifetime which matched the lifetime of the supporting evidence. But then again, how many scientists believe in ghosts..?