62 comments

[ 2.7 ms ] story [ 130 ms ] thread
Like the article (thankfully) mentions, the photographs really do have nothing to with the notes themselves. Intervals and chords have characters, individual notes do not.
Perhaps I missed that then because I found no mention of intervals and chords. They seem to explicitly say that they captured pure tones (220Hz for A). This sentence also made it sound as if they were generating a pure note (frequency), "As each note’s unique vibration induced its characteristic pattern into the water, Brown captured it with her camera." Then they went through the chromatic scale capturing each note.

I do wish they posted more about how they set this up. It sounds straight forward and if using pure tones should be easily reproduced...

It's the last paragraph of the article.
Indeed, this experiment says as much about the shape and size of the vessel that holds the liquid as it does about the notes. It's not true that you'd get the same result if you repeated it at home, unless your apparatus was built exactly the same.

Also, D minor is the saddest of all keys.

That was my thought as well, that the combination of note frequency and box size creates the standing waves photographed.

It's still beautiful to look at though....

Tell that to someone with perfect pitch.
Absolute pitch doesn't require you assign character to the notes that you remember. Well, I don't, so I count as a counterexample. I'm no synaesthetic that way, at least not in any conscious way.
I don't have it, but from what I've read it seems most people with absolute pitch associate some sort of character to pitches or at least keys. I'm sure that it's not a requirement.
Yeah I noticed that. But can someone explain how she got such varied patterns from individual notes?

I agree that the picture may say more about the vessel and the water than the note itself. But then why do the pictures look so different? Was there a different vessel per note?

When the tuner registered a note—reading 220 hertz, the frequency that produces an A, for instance—Louviere stopped adjusting. As each note’s unique vibration induced its characteristic pattern into the water, Brown captured it with her camera. The pair worked together to obtain a “portrait” of each of the 12 notes.

The sound waves induce a surface wave in the 2d plane of the surface of the water.

This surface wave reflected reflects of the side walls of the container.

At specific frequencies, the reflected waves will interfere with the generated waves in such a way as to cancel the amplitudes out completely at certain points, and double amplitudes at other points. When these points are stable, the pattern is known as a standing wave.

The standing waves are the patterns shown here. Imperfections in the container, the medium, and the way the sound source is transmitted to the medium make the patterns imperfect. If they were perfected, they would have a more symmetric, kalaidescopic appearance: https://www.google.com/search?q=kaleidoscope&source=lnms&tbm...

Yeah that article presents a poor understanding of cymatics.

However, you can think of the resonant body of the vessel that contains the water as one half of the interval. It's the relationship being the note and its environment that produces the unique pattern. If the vessel were to change, the image would too. It has nothing to do with the intrinsic value of the note alone.

The article is unfortunately a bit unclear, but it's not true that they have "nothing to do" with the notes. They are just a bit noisier than what we've perhaps seen before.

Would you say these pictures have "nothing to do" with the notes?

http://www.thisiscolossal.com/2013/06/the-visual-patterns-of...

Yes, it is a function of the note/frequency and the medium, not just the note, but that still means they are very much related tot he notes.

In particular you can see the different circular symmetries pretty clearly through the noise.

http://thekidshouldseethis.com/post/20902918854

Yeah I think those patterns are completely irrelevant when talking about the characteristics of notes and combinations of them.
Intervals and chords have characters, individual notes do not

Yes they do, in the forms of texture and timbre.

Notes do not have timbre, they are just named pitches (which themselves are just frequencies).
Hm? If it's about how they appear on the staff, why make the distinction?
There are different ways to assign pitches to notes. What is commonly used in modern western music is called equal temperament, but there are other systems, like just intonation, for example.
The pitches in alternate tuning systems like JI and maqam are sometimes, but not always, denoted with extra information. Quarter-flats, cent indicators, and so on, distinguish the note-pitch representations. While tuning systems can indeed be transposed or converted to unmodified standard (ET) staff notation, that's more of a convenience that has its own learning curve.

So, in JI you wouldn't simply use 'A' to represent (e.g.) 440Hz, it might be A 4, for the "A" key (or "A" staff position) having a 4 cent difference than ET A.[1]

1. http://lilypondblog.org/wp-content/uploads/2015/11/example.p...

i thi in common usage 'note' could cover both your narrow meaning and his the parent comment's wider one.
As mentioned in one of the comments, D-E-F is a triad, not a tritone.

A tritone is an interval of three whole tones (i.e. an augmented fourth). For instance, C and F# form a tritone.

D-E-F isn't that dissonant either, especially if you add an A or Bb
Well, E-F is one semi-tone. That's among the dissonant-est intervals I can think of.
yup, E-F is more dissonant than D-E-F
Yup. I think the parent to your comment is confusing frequently heard (and thus "accepted") with dissonance.

It's not uncommon to hear those notes, even in that order, if the prevalent harmony is a Dm7(9) or Bbmaj7(#11) which aren't rare chord forms, even in popular music. That doesn't make them less dissonant, but merely common-place.

I think you could make the argument that more common place actually means less dissonant.

See https://en.wikipedia.org/wiki/Tone_cluster

I don't agree. Unless you're using "dissonance" in the sense of plain English, there are more precise definitions of the term which have meaning in music. Since the comments in this thread are referring to musical elements such as specific notes, interval relationships, and the contribution of a harmonic context (albeit single chord) I default to the more technical definitions of the term.

When I am composing or improvising, I'm thinking of one of two ways (or both) when considering "dissonance": 1) the actual physiological basis of dissonance which has to do with the perception of how the waves of different sounds interfere and are ultimately perceived... an example is when you have two tone sources at the same frequency and one moves gradually to a higher or lower frequency; as they are close, but "out of tune" with each other you can hear a "beating" sound come into place that quickens with increasing distance of frequencies. This changing of interfering waves continues through the point of a single out of tune note becoming an interval between two different notes. Hard dissonances like minor seconds are considered hard because of these frequency relationships. That's probably as good as a definition as I can get there without brushing up on the subject. 2) dissonance as part of a harmonic progression where harmonic structures of relatively greater dissonance resolve to harmonic structures of lesser dissonance. This is in part how we get "V7" chords most favorably resolving to "I" chords, though optionally resolving to "VI", for example.

Now the dominant chords are perfectly acceptable and anyone listening to blues, for example, won't hear the dissonant tritone as "objectionable" even when they're never resolved. As for your link, I've written much music using tone clusters (couldn't tell why you referenced it actually) and the kind of tone cluster I'd use depends on the kind of relative dissonance I want and for what purpose. I might start using diatonic sequences of notes (the "white keys" for example) to get a moderately dissonant start, build tension by moving to increasing chromatic (white and black keys, more dissonant) clusters, and then finally release that by going back to diatonic, but perhaps open clusters... if not some triadic constructs. Managing relative dissonance is done using common practice harmony, too, just not with clusters. I think there's an argument that that's what the practice of harmony is really about.

From what I've heard about tone clusters, they sort of came about as continuations of the natural progression of more and more intervals to sound pleasant (or at least acceptable) as time has gone by. Tone clusters are like predictions of what will sound acceptable in the future.

The more we use certain intervals, the more we like hearing them. I suppose you can view that through two lenses: the intervals sound less dissonant, or our cutoff where we say that a dissonant interval is unacceptable has changed. Though tbh those seem like the same to me.

It's extremely dissonant if the two notes are only a semitone apart, but if you substitute an E an octave up, you get an interval (11 semitones in equal temperament or a 15/8 interval in just intonation) that appears in a major7 chord.
"Triad" would be an unusal term, though, for D-E-F, wouldn't it?

Reading https://en.wikipedia.org/wiki/Triad_(music) it appears, outside of "some twentieth-century theorists," the notes of a triad should not be successive within a diatonic scale ("Triads [...] are built by superimposing every other note of a diatonic scale") so that E couldn't be part of a triad which has D and F.

That wikipedia page suggests "trichord" would be the right term for a generic set of three notes (https://en.wikipedia.org/wiki/Trichord)

D-E-F by itself would probably be considered a D minor triad with a 9, and the missing 5th is implied (the 5th would exist as an overtone for most instruments too). that's kind of a stretch though.
Indeed, but not only is it not a tritone, it is (to my knowledge) not "diabolus in musica" either (that's also the tritone). The author noted the original error, but then corrected it incorrectly. :-(
The Diabolos in Musica is the tritone: their original statement and correction are both wrong. They probably meant B-D-F or something, which contains the tritone.

Disappointed in this. Nautilus's biology stuff is quite good, but this is trite. [Then again I know far more about music than biology.]

No, this is not what musical notes actually look like. At the risk of stating the painfully obvious, musical notes are sounds, not images. They don't look like anything. The images in the article are what you get if you render the information in a sound to an image in a particular way. If you render that information in a different way, you'll get a different image.
Isn't that "what things look like"? Like what you see when you see a tree isn't actually a tree but the interpretation of light waves hitting your eye. It's a rendering of the light in a particular way and we somewhat casually call this "what a tree looks like". Often when we talk about the projection of a thing into some visual form we us this kinda casual language. I think to your point, any one of those methods of rendering the information into visual form has equal claim to "what the notes look like"
Yes, that's true, but in the case of macroscopic physical objects like trees there is a "natural" rendering designed by evolution that is reasonably called what the thing "actually" looks like. In the case of sounds, there is no such a natural rendering. Turning sound into images necessarily involves some design, and so there's no basis for calling any particular rendering the "actual" appearance of a sound.
So the problem in this view is the primacy of one particular projection asserted by the word "actually"? Fair enough.
It's important to recognize pretenses of objectivity.
Yes, exactly. If the headline had been, "A neat way to turn sounds into pretty pictures" I would have had no problem with it.
But the current headline is "What Musical Notes Can Look Like", emphasis on the "Can". Your suggestion is basically just a condescending way to rephrase that.
The headline has been changed. When I posted my original comment, the headline was the original headline: "This Is What Musical Notes Actually Look Like"

    > Turning sound into images necessarily involves some design, and so there's 
    > no basis for calling any particular rendering the "actual" appearance of a sound.
Turning light into images necessarily invokes a design, too: an eye. And depending on whether this eye is that of a human, a fish, or a fly, a different image will be experienced. So here, too, there is no basis for calling any particular image, that you experience from the light that an object reflects, the actual appearance of that object.

It's just one instance of an appearance of an object in your field of experience at one particular point in time. Not even you will ever experience that object in exactly the same way again.

Furthermore, we have no way of knowing if there, some place on earth or elsewhere, exists a life form that experiences sound exactly as these two-dimensional images. If there has been an evolutionary advantage to doing so, I see nothing preventing a sensory organ from picking up these vibrations and creating this experience out of it.

> there is no basis for calling any particular image, that you experience from the light that an object reflects, the actual appearance of that object

Sure, but "the actual appearance of an object" is generally understood to mean, "the actual appearance of an object relative to a typical human visual system", at least when the phrase is uttered by one human to another.

I don't like the idea of musical notes being a different rendering of sound. Everyone plays a piece differently. Classical music is a great example of that.

I like the analogy of notes on a page being akin to source code and the different players akin to different compilers. Every "standard compliant" player (one that knows all the constructs) produces the same result (the song/program), they just do it differently from each other (different playing styles/different assembly). Newbies just aren't as "standard compliant" because they ignore certain constructs they don't understand.

In the brain, both images and sounds are just activation patterns. So there is a sense in which the premise of the article holds. There's even that experiment on mice where researchers switched the end of the optical nerve from its normal location to that of hearing. They developed sight with the hearing part of the brain. It's the same algorithm for both.
> both images and sounds are just activation patterns.

Yes, that's true.

> So there is a sense in which the premise of the article holds.

No, that does not follow. Sounds and images have fundamentally different structure. Images occupy two spatial dimensions. Sounds occupy one non-spatial dimension (pressure) and the time dimension. Sounds are inherently time-dependent. A sound with no time-dependent components is not a sound. All this derives from fundamental physics, and the different ways in which light waves and pressure waves propagate and carry useful (to biological systems) information about the world.

They look like a representation of amplitude vs. time on a graph, if they look like anything. The art presented is very nice, but it's an artist's interpretation. Not that there's anything wrong with that.
So it sounds like these are effectively the same thing as Chladni Patterns [1] using a liquid surface. It reminds of composer Stuart Mitchell [2] who believed he saw Chladni patterns carved into the Rosslyn Cathedral in Scotland. These he interpreted as a guide to a musical score, which Mitchell then recorded as the "Rosslyn Motet" [3].

[1] https://en.wikipedia.org/wiki/Ernst_Chladni#Chladni_figures

[2] https://en.wikipedia.org/wiki/Stuart_Mitchell

[3] https://www.youtube.com/watch?v=cy2Dg-ncWoY

There must be a better way of writing music down. Out of interest I Googled "alternative music notation" and every alternative is the same lines with dots BS that translates poorly to computer and is difficult for beginners to read.

Why has there never existed an alternative musical notation system thats truely different?

Is horizontal lines and abstract squiggles with relative positions really the best we have?

There are a few alternatives, though most are instrument-specific.

For piano, there's Klavarskribo. If you've ever played Guitar Hero/Rockband, it's a bit like that: https://en.wikipedia.org/wiki/Klavarskribo

For many fretted stringed instruments, there's tablature. One line per string, and instead of abstract squiggles you get numbers that represent fret positions. They stack easily for chords, and there's intuitive notation for picking styles and such. E.g. sliding from the second to the fifth fret on the A string would look like:

  E |----------
  B |----------
  G |----------
  D |----------
  A |---2/5----
  E |----------
https://en.wikipedia.org/wiki/Tablature

I see your point about a truly different system but, as music is basically sound through time, I'm not sure how else it could be notated if not linearly.

Most of the alternatives in use I can think come in genres where improvisation is far more important than playing exactly, and as such only a reduced feature set is needed. Examples include the above mentioned tablature, solfege (https://en.wikipedia.org/wiki/Solf%C3%A8ge), Roman numeral notation (https://en.wikipedia.org/wiki/Roman_numeral_analysis), chord charts (https://en.wikipedia.org/wiki/Chord_chart), or lead sheets (https://en.wikipedia.org/wiki/Lead_sheet). Full notation in the pop and jazz genres are rare.

For Western classical, honestly, considering the information included in standard notation -- rhythmic complexity, phrasings, special symbols for various afflictions, etc. -- it would be really hard to replicate this in a system that's both easy for humans and computers to read. (Think if musicians had to use something like the piano roll found in most DAWs...)

Our current system exists because the previous systems were hard for beginners to understand. Solfeg was invented to teach new musicians. Over thousands of years of refinement, our musical system has evolved (into the 20th century, for example with the Kodaly system).
The notation system evolved as an efficient way to read Western classical music, both in terms of space and legibility. It's oriented towards the fundamentals of musical performance, and is read by millions of people around the world. For what it does, and the large ground it covers, I would argue it's really tough to beat. Not that people haven't tried!

There are plenty of alternatives, particularly when it comes to translating to a computer (see e.g. Lilypond notation). For many music projects that go from instrument straight into production without having to communicate to an intermediary performer, western music notation is unnecessary - piano rolls are sufficient. Piano rolls can also be useful in limited situations for performance (see Guitar Hero et al) but things fall apart quickly if you need to perform lots of different notes with sophisticated rhythms, and/or need to look significantly far ahead of the current point in time. (These are problems which Guitar Hero happily exploits as part of the game.)

The further away music gets from the Western classical music tradition, the more the notation system is stressed. Many of those traditions, however, are places where oral transmission of musical instructions are preferred, and where written notation may not be as developed. (See e.g. oral transmission systems for taiko and pakhawaj.)

It has evolved, but to suit the needs of professional musicians, not people in general. For professional musicians, barrier to entry is more of a feature than a bug...

Mass musical literacy is possible, as the American Sacred Harp tradition shows - but they didn't use regular notation.

How about tracker music? It is top-down instead of left-to-right. It is computer native and unambiguous.
> Why has there never existed an alternative musical notation system thats truely different?

Because musicians tend to be

1. conservative, so that they're unlikely to accept one,

2. elitist/guildlike, so that they don't especially mind barriers to entry, and

3. self-aggrandizing, so that of ten musicians inclined to propose an alternative notation, every one will suggest his own and ridicule the nine other proposals.

Western notation is really as bad as it seems, though - it encourages optical illusions (notes with opposed stems seem closer or further away from each other than they actually are, for instance) and is quite brittle (add a sixth line to a staff in the middle of the page, and many experienced sight readers will just play on and not notice anything is amiss except that the music suddenly got very strange).

Mass music literacy seems out of reach with the current notation systems. Even of people who can sight-read, many (probably most) do so by associating notes with instrument finger positions instead of intervals, so they can't read an unknown melody without their instrument at hand.

I think the most successful, more rational system is the number notation they use in China (which also has been reinvented many times in Europe). Funny that China should have the awkward writing system, but the sensible music notation system.

Most successful alternative notation systems seem to come from religious instructors, who were more concerned with helping the congregation to sing. Mass music literacy mattered from them in a way it didn't for professional musicians. Shape note is an American system (although an adaption of the regular one, still squiggles) and there are number notation systems in Scandinavia developed for the same reason, teaching the peasants to sing in four-part harmony. I wonder if the Chinese system doesn't have roots in missionary work, too - they use western numerals, after all.

According to the article, "If you were to repeat this experiment, you would get the same designs."

The initial conditions of the experiment are so susceptible to minor variations, I strongly doubt the author's assertion.

I found that to be a fun article, and it could be a starting point for creative exploration, but I think any suggestion that these images relate to our actual psychoacoustic perception of notes and music is wholly unsubstantiated.

I do perceive differences in the character of music played in different "keys", but 1) I ascribe these to aspects of timbre, dissonance, tuning, and performance practice rather than fundamental aspects of the notes themselves, 2) there's very little scientific evidence that I'm aware of that these differences correspond to psychological moods.

Nonetheless, there's a significant philosophical tradition going back to at least Plato that there is some connection between tonal systems and psychology. See https://en.wikipedia.org/wiki/Doctrine_of_the_affections and Plato's Republic for a couple of starting points.

In keeping with the religious interpretations of the article, G - Demon, C# - Tree in Eden, I don't see F as the underbelly of the frog, but more a very Blakean image of Two Eagles on either side of Christ holding the Chalice with a wee spider hanging the bottom above the Sun in Christ's lap!

More so, I see G as a Teletubby rather than a Demon ;)