Jivari, Sawari, Rustle Noise


This week I started a half-serious project to bring my first guitar -- a cheap Kay acoustic from the 1980s -- back to life. I knew it wasn't going to be a "normal" guitar, since it was never good at being one of those and decades of poor storage have left it warped beyond reasonable repair. If I wanted an acoustic guitar, I'd be much better off dropping £100 on a Chinese one on eBay. I pulled the frets out a few years ago in an attempt to make something vaguely oud-like but that didn't work at all, and since then it's been moping around my studio getting in the way.

When I went back to it I realised the main problem was a rather common one: the top had warped up in a bulge, causing the bridge to tilt forward. There's no truss rod, and I think the adjustment would be too extreme even if there was one. So I could see I'd have to remove the bridge and replace it with something different. It now has a trapeze tailpiece and I'm experimenting with bridges. Oh and I've refretted it, or at least started the process (the fret ends need bevelling and a few high spots need to be knocked down).

The bridge is one of the most important parts of any guitar when it comes to the overall sound. While trying out different shapes and materials I stumbled on something that was first discovered thousands of years ago and have got a bit obsessed with it -- the miraculous buzzing of the sitar. I'm not trying to build a sitar-guitar, although that might be what I end up with, but I wanted to understand this sound a bit better and to think about my attraction to it.

(Fair warning: This is a long and meandering post that does have a point to make but does not include any scales, chords, tunings etc. I've included lots of videos to liven it up.)

Jivari is the Indian word for that sound. It's created by a (nearly) flat bridge on a sitar or similar instrument, which causes the string length to vary as it vibrates. "Jivari" is also, I believe, the word for the bridge itself and for the craft of making and maintaining it. Musicians and instrument-makers both seem to concern themselves with adjusting jivari -- here's an interesting look at the subject:

At the moment the bridge I have set up is made of a couple of pieces of steel I had lying around and I was amazed how much jivari, as it were, I was getting out of it. In the last few days I've learned that it should have a shallow parabolic curve (mine is almost flat) and can be tweaked using cotton threads called jeeva. When I read about this I thought it would be a micro-adjustment I wouldn't have a sophisticated enough ear to discern, but how wrong I was -- see around 9:30 in the video above. The effect is dramatic, and I had to scurry off to try it. (This video featuring the same craftsman gives a more detailed demonstration and his website is also useful.)

The jivari sound is often described in terms of a lush rainbow of overtones, with the bridge acting as the prism that draws them out of the basic movement of the string. The whole process of creating jivari is precise and belongs to an artisanal tradition that's millennia old. But for me, as a Western musician and completely unskilled instrument-modifier, there's something else about it I want to capture -- a raw, earthy rattle that responds to your touch and so connects with the physical body that's touching the strings.

In bouncing around with this topic I ran across a term coined by Robert Erickson in the 1970s: "spectral glide". Erickson was a composer -- here's one of his pieces, which I was alarmed to discover sounds very similar to a lot of what I do:

He wrote a book called Sound Structure in Music and that's where the term comes up. Spectral glide is a gradual change in the harmonic structure of a sustained sound. This is easy to do by singing a note and changing the shape of your mouth while you do so (I presume every child has experimented with this) but electronic methods can yield up very clear examples too. Here's a video demonstrating formant filters, which offer a kind of spectral glide:

and here's one demonstrating pulse width modulation, another old-school synthesis technique that demonstrates spectral glide:

While looking into jivari I discovered sawari, which is a Japanese word for a similar effect associated with the shamisen. Here's a nice basic video about that:

Here's a video that goes into more detail (in German but closed captions + YouTube's hilarious translation will be enough to anglophones to get the gist):

The effect not only sounds different from jivari but appears at first to be created in a completely different way. However, I think this paper suggests that a parabolic point of contact creates more harmonic overtones than a point or linear slope, which seems to create inharmonicity -- I can't tell for sure because I don't read Japanese but I'm interested to try some experiments with this mechanism. The significance of this is that the jivari bridge of Indian instruments is also said to have a parabolic profile precisely to bring out the overtones of the fundamental. I like inharmonicity so there are some interesting paths to explore here.

Then there are pitched instruments that have full-on inharmonic buzzing such as the mbira, where the buzz may come from loosely-fixed bottle caps:

The next thing you know we've arrived at the Delta blues, and that buzz is coming from the strings, the slide and, yes, the overdriven amplifier that makes any electric guitarist feel at home:

The Delta blues players all seem to have relished the buzz, rattle and jangle of the steel strings, the glass or metal slide and the wood underneath; resonator guitars just intensify the effect.

Right after spectral glide in Erickson's book is a section on "rustle noise". Rustle noise is characterised by many impulses -- short "clicks" -- that come randomly but with a predictable density. Here's a long video demonstrating the use of rustle noise in some modern electronic music -- there's a very exposed example around the 10 minute mark:

Rustle noise is the stuff that gives you ASMR. Since ASMR is a sensual response to sound, most videos of it tip over into being sexual, which is understandable but not the same thing; this one manages to keep it in its pants:

While lots of composers explored rustle noise in the 1960s, 70s and 80s, Helmut Lachenmann must be the most ASMR of all of them. Here are dozens of rustle noises and spectral glides performed on a solo cello:

I think there's a continuum between spectral glide and rustle noise that has jivari almost all the way down the "spectral glide" end and the jangles on an mbira almost all the way down the other. There are sweet spots on that continuum that all the videos above illustrate; I'm hoping to find one of my own in my experiments with bridges on my busted old guitar.