Backmasking
Mark Honeychurch (July 3, 2023)
When I was a teenager, I spent a lot of time hanging around at my best friend's house. His family were committed Christians, and I was a young atheist. This was a time before I converted to Christianity, as a 17 year old, and I enjoyed arguing with Christians about the age of the earth, the fossil record, etc. It's weird that so many evangelical Christians hang their hat on the idea that the earth is only 6,000 years old, despite the mountains of evidence to the contrary that was so easy to grasp that even a clueless 15 year old like me could figure it out. Anyway, I digress…
One of the many things that my friend Merryn's parents were concerned about was his choice of music. At the age of 15 I was into metal, and listened to Iron Maiden, Metallica, Guns n' Roses and many other bands that sang about darkness, malevolence and demons, music that I'm happy to say I grew out of pretty quickly! I would have understood if my friend's parents were concerned about my music choices. But Merryn listened to bands like Queen, fairly benign rock that, although it had sexual themes (Fat Bottomed Girls definitely comes to mind!), was far from being demonic.
However, Merryn's parents weren't so concerned about the lyrics in the music he listened to as they were about the possibility of there being voices embedded in the music, often recorded backwards, that only the unconscious mind could hear - subliminal messages.
My memory's patchy, but I'm pretty sure Merryn had been given an audio tape from his parents featuring someone documenting instances of music that contained these backwards subliminal messages, which were called back masking. For each piece of music the announcer would tell you what you were about to hear being said when it was played backwards, and you were then repeatedly played the fragment of music in reverse. And, sure enough, for most of the examples the tape gave the audio clearly contained these messages when played backwards.
Maybe my friend was more of a skeptic than I was, because his first thought was to gather as many vinyl records as possible that the audio tape contained so that we could try this out for ourselves. He had a record player, and it was pretty easy to slip the drive belt off the platter so that the record wouldn't spin at all when the player was turned on. Once this was done, we could use our fingers to move the record in reverse at about the same speed that it would usually move forward (33rpm, so about one revolution every two seconds).
We tried all the records we had managed to get our hands on, starting with Queen's “Another one Bites the Dust”. Merryn already owned this one, so it was an easy start. Sure enough, the chorus, where Freddie Mercury sings “another one bites the dust” over and over, kind of sounds like “it's fun to smoke marijuana” when played backwards.
One song that we had been warned about on the tape, where it was blindingly obvious that there was no demonic influence, was by Yoko Ono. On the tape we had been listening to you could very clearly hear a voice saying “six six six”. Once we had secured the album (Double Fantasy) and looked for the song, it all fell into place. Yep, the song was called “Kiss, Kiss, Kiss”. And it's no surprise that the chorus of this song, where Yoko sings “kiss, kiss, kiss, kiss me love”, sounds like “six, six, six” when it's played backwards. Maybe Yoko was in on it, and deliberately planned this, but this one was certainly not a cleverly hidden message!
For the other songs, though, it was a little weird that these messages had supposedly been embedded in them. And as a teenager, my mind was fascinated and willing to accept any rational-sounding explanation - obviously it wasn't demons, but were metal bands deliberately placing these messages in their music for their fans? I'd like to think that, now that I'm older and hopefully a little wiser, I'd have picked up on the reality of what was going on a lot sooner - but as a naive teen, bands going to the effort of adding cool sounding messages seemed plausible, even if the messages were a little weird and kooky, and there didn't seem to be much of a good reason to do this. And who knew, my undeveloped brain wondered - maybe we really could somehow subliminally decode these hidden backwards messages?
Thankfully, in this case, TV came to the rescue in the form of a very well made documentary about the backmasking craze in the US, where (as things often do) it was getting out of hand. The backmasking cases looked eerily similar to the satanic ritual abuse cases that were tearing families apart in the US - in what is now commonly known as the “satanic panic” of the 80s. For backmasking, bands were being blamed for brainwashing the youth, even convincing them via hidden messages to take their own lives.
The documentary did a really good job of explaining audio pareidolia, the way that we can perceive patterns in noise, and how priming can crank this up to 11, making it very easy to force people to interpret sounds that sound similar to speech as speech. By telling you what words you're about to hear, when listening to noises that sound somewhat like those words being spoken, your brain is very likely to discern those words. It did this on a practical level in several different ways:
Firstly, when people were given some backwards audio to listen to that supposedly contained backmasked messaging, but weren't told what words they were going to hear, it was pretty much guaranteed that they would not discern the same words that the music supposedly contained (with the obvious exceptions of anomalous tracks like “Kiss, Kiss, Kiss”) - if they could discern any words at all. My friend and I tried this one, testing each other out on music that we had found the vinyl for where we didn't let each other see what the music was or hear it being played forwards. Sure enough, without the priming, neither of us could work out what was supposed to have been said. And when we tested other friends on music where we could clearly hear the hidden backwards messages, but we hadn't told them what it was, sure enough they had no clue what they were meant to be hearing.
The second way was to show that this devil's music not only “contained” benign messages that it would make no sense for anyone - human or demonic - to embed in a piece of music. This was well illustrated by a court case in the US featuring metal band Judas Priest, who were being sued by parents over the suicide attempt of their sons - two friends who were heavily into Judas Priest's music, and had attempted suicide with a shotgun. Judas Priest's lawyers showed in court that their music not only sounded like it contained the phrases “try suicide” and “do it”, but also (and this is from memory, so it's very much paraphrased) “Maggie Thatcher's in the kitchen throwing plates”.
The third was showing that even “godly” music, choral ensembles and the like, could also contain these backmasking messages. Wherever you looked, you could find seemingly demonic messages trying to corrupt our youth.
And, of course, on top of the idea of whether these messages are intentional or accidental, there's the question of whether they can have an effect on someone. The idea that subliminal messages are able to subconsciously alter someone's behaviour, filtering up into the conscious mind, is an idea that's on shaky ground. And, in this particular case, the brain's expected to do gymnastics and reverse the audio stream in order to access the hidden message. And I have a sneaking suspicion that the audio pathway from our ear to our brain doesn't easily allow for our brain to have the signal reversed even if it wanted to.
This whole thing would be funny if it was just benign. But sadly, as is usual in the US, it was all taken a little too seriously by people who should know better. As part of the Satanic Panic, the idea of the devil using backmasking in rock music just heightened the fear that many Christians were already experiencing.
Of course, some artists have chosen to deliberately hide messages in their audio - I'm not claiming that this doesn't happen. There are some great examples, like the Beatles' single Free as a Bird, recorded in 1995 by the three then remaining members, who added the message “turned out nice again” backwards to the end of the song. Paul McCartney said of its inclusion: “We even put one of those spoof backwards recordings on the end of the single for a laugh, to give all those Beatles nuts something to do.” For more of these, including one example by Chinese musician Jay Chou (who I unsuccessfully attempted to stalk in Wellington a few weeks ago - it's a long story!), check out this list on Wikipedia.
Interestingly, some of the musicians I listen to have experimented with embedding images in audio. How do they do that, you may well ask.
A spectrogram is a way of visualising the frequencies in a piece of music - one part of a song may contain multiple frequencies from each of multiple instruments, as well as frequencies from someone's voice. A nice way to visualise those frequencies is by using a Fourier Transform to break down the audio into its component frequencies, and then plotting these frequencies over time. Here's an example of a pop song from the ‘80s (can you guess what song it might be?):
In this visualisation of the song I chose, time is from left to right, and frequencies are from low at the bottom to high at the top, and plotted logarithmically. It's not hard to imagine manipulating this, by starting with a grayscale image and converting it to the frequencies over time that would draw the image in a spectrogram. So, on Aphex Twin's Windowlicker single, artist Richard D. James played around with embedding an image at the end of the track. This is from 6:00 in the audio, and captured using FooBar2000 and its native spectrogram analyser:
This image appears to me likely to have been a test, where frequencies were chosen that would result in a double spiral. However, for the B side of the Windowlicker single, called “∆Mi-1 = -∂ E Di[n]E Fji[n-1] + F exti[n-1]]” (but affectionately known as “equation”), appears to have gone a step further and converted an image (likely a selfie of Richard) into the audio - from around 5:20:
Apparently the equation used for the name of that track is related to Fourier transforms, so I guess it's a hint as to what's inside. If you have the track in your collection, I'd recommend loading it up and having a poke inside, as there are a bunch of other little goodies in there - weird patterns, and parts of images that look like more tests.
This technique has even been improved upon by another artist I like - a Canadian artist called Aaron Funk who makes music under the name of Venetian Snares. On his Songs About My Cats album, the last track is called “Look” - probably another hint for keen-eyed listeners. The track contains a set of images that appear to be scans of cats from below using a flatbed scanner, and culminate in a finely detailed photograph of a cat:
Now, I'm a fan of unusual music (what my wife calls “noise”), so for me as a regular listener of these tracks I never suspected that there would be anything unusual about them. But for the average music fan who enjoys a bit of pop and maybe some rock, I'd imagine they'll probably sound a little odd - there's probably no good way to use frequencies to represent a grayscale image without it sounding strange. But, if you're interested in how they sound, here are some YouTube videos I found (and as a bonus, two of them show the spectrograms so you can see the image as it relates to the audio):