22

I am not sure where to post this, but I hope it fits here.

If you have ever recorded music, and play the track backwards, you'll notice it sounds quite strange, almost demonic. I have always wondered, why doesn't music sound good when played in reverse? It just doesn't seem at all intuitive to me, if you had never heard classical music recorded and played backwards you might expect it to still sound ok. There is so much structure in the sound, so why does it seemingly disappear in reverse?

Why does the harmony of the notes break down, and become (what appears to sound) so chaotic. This doesn't work the same way if I just play the notes in reverse on a piano. Scales are a perfect example, and sound identical in either direction.

I hope this is relevant, it is something I have always found curious.

2
  • 2
    The solo of Hendrix's "Castle made of sand" was pretty cool..doesn't sound bad at all. May be if you reverse something bad or random you'll end up with something cool sounding.
    – KMC
    Commented Feb 2, 2012 at 8:54
  • 1
    Just a hint, you might try listening to some music without drums backwards, it suddenly gets a lot more pleasant. This has to do with the 'volume envelope' of sound, which is very harsh with drums, and makes a sound you wouldn't hear naturally. Tonal sounds which fade in slowly can be quite pleasant, without that harsh ending (the 'attack' of the sound when it's played forward, but in reverse, the 'decay). Give it a try. Also, singing is usually unsettling reversed, again because it's not such a natural sound.
    – AJFaraday
    Commented Jan 16, 2017 at 16:21

10 Answers 10

28

The reason why music sounds so odd when played backwards has little to do with musical theory per se. It's all about the our psychological expectations of the physical sound phenomena.

Our brain expects instruments to sound a certain way, it expects a piano to begin with the attack, a quick rise in intensity, then stabilize somewhere along the way and then fade away smoothly. When we hear a reversed piano all these aspects are now reversed and it becomes a sound that begins quietly and ends with a powerful attack, something we're not familiar with because it's a rare thing to hear in nature (it might not even exist naturally).

The same goes for all instruments. Our brain uses the wave shape and variations in volume and other aspects of a sound to extract musical information. When it's listened in reverse the lack of familiarity with the sound structure makes it sound "odd", "scary", "demonic". This is even used as a compositional element, as you might be aware of artists recording reversed vocal tracks or instruments.

To further illustrate what I mean, listen to these sound samples generated with Timidity and the Unison soundfont. A little melody followed by some chords is played and in sequence it's played in reverse:

See how the piano sounds weird, but in turn the flute and strings are "passable"? We might notice there's something wrong but it isn't as much as the piano. Try opening these files in a program like audacity and observing the shape of the sound waves.

3
  • 1
    +1 This is a good point as well. I'm not sure it alone can explain the phenomena, especially since some instruments have a more "symmetrical" sound than others, but it certainly has a big effect in many cases. I can't imagine playing a recording of my trombone playing backwards… my attacks and releases are occasionally shoddy enough as it is! Commented Jan 30, 2012 at 16:47
  • This makes sense to me, nothing in nature smoothly amplifies to reach a sharp noise and then stop altogether. Of the mandmade things in our society, all I can think of that would do this sound terrible anyway. (If you use powertools, I think certain cutting noises will do it, but I believe there is concensus that powertools sound really bad.)
    – user1864
    Commented Jan 31, 2012 at 10:58
  • Case study: The ending of Pendulum-Granite, when reversed, is the beginning of Pendulum-Granite. Except all of the sounds are backwards, making it sound really weird.
    – Edward
    Commented Jan 31, 2023 at 1:50
14

Eric,

Classical music, and really, most kinds of music, sound odd when played backwards because the harmonic progression does not lead where you expect it to. Without going to deep into music theory, I'll outline the basics.

A typical cadence, or ending of a chord progression, is a IV chord to V to a I chord. (Roman numerals signify chords built out of a major scale, starting on that degree of a major scale, so V is the fifth scale degree.) This cadence is a very typical one, and you'll hear it everywhere. What that cadence does, is go from the chord with the most tension in the key… to the tonic, or home chord of the key. In this way, you have tension and relaxation, which is the foundation of harmony.

A more advanced chord progression might go I-vi-ii-IV-V-I. This progression sounds different, but it still ends with that characteristic IV-V-I, also called a Perfect Authentic Cadence.

Now getting into one of the reasons music played backwards sounds so odd: Our ears expect that Perfect Authentic Cadence. (Simplifying here, there are other pleasing cadences.) Sometimes, a composer will do a cadence like: IV-V-vi. This cadence is called a Deceptive Cadence, because your ear will expect the IV and V to inevitably lead to I. When it goes to the vi you get a denial of expectations. While I'm simplifying, these basic ideas of tension and relaxation are what gives music it's sound, feeling, attitude, soul, etc.

When these characteristic cadences and progressions get reversed, it essentially throws that whole system out the window. Now you're going I-V-IV-ii-vi-I and not only does that defy expectations, your brain likely is just going WTF?!? For lack of a better term…

To hear these cadences for yourself play this on the piano: enter image description here Courtesy this Wikipedia article:

To hear a recording of just these three chords, go here.

Hopefully this makes sense!

4

The above are all good points. Another thing I want to point out is that many instruments are in a sense defined by their attack note. I hypothesize that by cutting off the attacks in a recording of many different instruments playing the same pitch, the tones are largely indistinguishable.
So, if the attack is really as important as I think, 'backwards-tape' playing of a song/piece will loose much of its characteristics, as the attacks do not define the note or pitch, but only briefly punctuate them. The strange-sound effect is much more overwhelming this way.

1
  • 1
    Yes indeed, this seems to be the case many times. In fact the brain uses the clues given by the attack and the different volumes of the sound and its harmonics to perceive musical information. I know there's some papers analysing this somewhere but I'm not finding them now.
    – lfzawacki
    Commented Jan 31, 2012 at 21:13
3

First, it doesn't across the board sound "bad", whatever that's supposed to mean, or else backmasking wouldn't be used.

You could say it sounds "odd" or "unnatural" - without the vague word "bad" muddying the waters - and make these two points about the nature of music:

  • Playing a sound backward reverses the sound envelope
  • Playing a sound backward reverses the 'syntax' of common harmonic progression

The sound envelope is basically how an instrument's sound changes over time. Reversing the envelope will be most noticeable for instruments with a strong attack such as plucked strings and percussion. It's easy to whack a cymbal with a stick, get a big crashing attack and slow sizzling decay. But it would be very hard to make a cymbal start to lightly vibrate louder and louder until a sudden, very high vibration that instantly stops. Reversing a recording allows you to artificially achieve that reversal. It sounds strange, because it defies the physics of how acoustic instruments normally work.

Reversing musical syntax is a bit trickier to explain. There is a technical terms for reserving the time order of musical elements, usually melody, called retrograde. In many cases melodies can be reverse with no problem. But when chord progressions are considered, they don't really work in reverse. Specifically, they don't have the same effect.

A fundamental harmonic progression would be I V I and obviously there is "reversal" happening, I goes to V and then that V just goes right back "in reverse" to I. But if we think of those two as harmonic gestures in terms of progression we can say I V is a kind of "starting", an antecedent, but the reversal V I is a "ending", a consequent. The oddness of reversing harmony comes from starting with the ending and ending with the starting

It may be a bit of a stretch, but you can compare this to the idea in physics of the "arrow of time" and entropy. Basically time moves in one direction only, forward. Play a recording backwards gives the illusion of something going backwards. We can listen to a recording of a cymbal crash played backward, but the cymbal did not actually vibrate in reverse.

7
  • Because harmonic syntax is a syntax. Reverse "Jim ate chicken" and... things get weird. Various languages use various orderings of components to construct meaning, and so does the western tonal system. Commented Jan 28, 2022 at 21:27
  • @AndyBonner ..."chicken ate Jim" Yikes! :-) Commented Jan 28, 2022 at 21:31
  • I think a whip or something breaking the sound barrier might produce a smooth sound ending with an attack. Not very relaxing sound though...
    – Emil
    Commented Feb 1, 2022 at 1:53
  • @Emil only in an anechoic chamber would it come close.
    – Edward
    Commented Jan 31, 2023 at 1:52
  • @Edward what do you mean? It first makes a whooshing noise when in the air, end then it makes a crackling sound when it breaks the soundbarrier. I saw it on mythbusters a long time ago.
    – Emil
    Commented Jan 31, 2023 at 6:25
2

The answers herein all make great points, but chord progressions, ADSR envelopes, and expectations aside, the notion of music "sounding bad" is purely subjective to begin with. (Although the percentage of agreement may be higher on some pieces than others!)

Some compositions, particularly in the electronic music domain do not follow "traditional" ADSR envelope and spectral distribution expectations to begin with. There are many aesthetics of what makes a musical piece "sound good". Rhythm, spectral distribution, harmony, and balance are all still artistic tools that can be used and exploited in reverse. Don't be afraid to experiment. You may just find something new that you otherwise wouldn't have appreciated!

2

One point that I don't think is covered in other answers is the effect on the timing of the notes.

Musicians will generally concentrate on getting the beginning of each note to fall rhythmically at the correct time. But the timing of note endings will me much less accurate and often quite arbitrary. So when the music is played backwards the rhythm will sound odd or be lost completely.

1

The synth people categorize music by four categories: attack, decay, sustain and release. Attack is the initial peak when the note starts, Decay is the drop from the attack's peak to sustain's steady state, and Release is how long the note stays around when you stop playing it.

Piano and guitar are known for pronounced attack, and as lfzawacki noted, piano sounds stranger reversed. Guitar is the canonical instrument for reversing. In general, you can strike a note, get a big note that comes down and fades, or you can blow a note, which stays fairly constant until you let go. Reverse the second type and it sounds about right. Reverse the first type and you reverse the way sound works.

1

For one thing, attack and decay are exchanged. For another, anticipation and resolution. And even for instruments with continuous tone the onset is strictly on the beat but the release position depends on articulation and is rarely on the beat. So all of the onsets in the reversed music, assuming that they are reasonably hearable, are timed fuzzily and come earlier when articulation becomes more legato and later when it gets more leggiero or staccato. But that means that tension and tranquility are no longer having a timing matching their emotion.

And take an organ: the instrument may have on/of sound production and thus appear to be one of the better reversible ones, but the reverbation comes after the sound, not before. You'd need to record in an acoustic studio rather than a church or hall in order to get read of the tell-tale reverb direction.

0

In the question several descriptive terms were used that have little to no absolute definition. However i think most of us would agree that "Bad" being a matter involving judgement to determine, and"demonic" which some may deem to be a purely imaginative term, also the word "Strange".... same thing. I would use the word "Negative" to more clearly state what is happening to the sound that is being asked about.

"Why does music sound "negative" when played backwards. This allows a much simpler explanation being "because it is "negative" The same as a ligt image reversed is a "negative" image. Looking "Bad" "Strange" and "Demonic". Nothing is created in its negative state. Music is creation. It is created. When played or listened to backwards what we are experiencing is negativity which is degenerative,we are hearing creation decay or more descriptively speakin death. We as living creatures seek life and accept life. Death, decay, pestilence, disease, all negativity is rejected and knoen to be "bad" "Strange" and even "demonic". It sounds that way because it is.

-1

It actually has nothing to do with the music itself, it's just how us humans work. We are used to things starting loud and then fading out or we expect something specific, then when it's reversed we don't have what we expect, and this it sounds way creepier. (My theory may be wrong since I'm not that good to how the human body works, if someone can tell me how it actually works in case my theory is wrong you can tell me.)

1
  • 1
    Welcome and thanks for contributing! In the future, though, if you're not very confident about your answer, it's okay to not post it (or to make a comment on the original question instead), especially when a question is as old as this one and has gotten so many lengthy answers. I think your theory is probably right, though. I wonder whether there's a source that could be found that says the same thing? Commented Jan 28, 2022 at 21:24