Listen Listen
London, 05.2009
Self published in an edition of 40.

The following text is taken from the book Listen Listen and makes frequent reference to the design of this book, which also included an audio CD. Though the text is presented here on its own, it was originally written to function as one part of three mutually dependent elements – constituting the design of the book, the accompanying audio CD and the text itself.

The Tone(s) of a Text: Thoughts on well tuned pianos and the frequencies of the Univers

We have eyes as well as ears, and it is our business while we are alive to use them.’ John Cage

In Breakfast of Champions, Kurt Vonnegut sometimes begins paragraphs of prose with the word ‘Listen’ followed by an arrow pointing to the beginning of the next sentence. Most of the time, this is because he has gone off on a tangent from the main narrative and wants to focus the reader again, on the ‘action’ about to be described. Aside from the humorous element to it, there is also something distinctly funny – ‘funny peculiar’ – about it when it happens. Each time you are asked you to listen, there is nothing to hear; there are only more words to read.

This seems like a good point at which to begin discussing the piece of work that you are currently holding, firstly because it is an experiment that is concerned with interpreting type through sound – in effect, inviting the reader to ‘listen’ to text – and secondly because the notion of attempting to find something to hear in a piece of text is indeed a peculiar one. There are many questions one might reasonably ask about undertaking such an activity, not least question why. So, lets begin, and hopefully some answers will reveal themselves along the way.

There is a point somewhere in the first ten or fifteen minutes of La Monte Young’s The Well Tuned Piano where the notes being played on the piano begin to merge and resonate together into one continuous sound. Although the tempo of this piece has sped up along the way, this development is not as much to do with the increase in the speed at which the notes are being played, but more the result of how the pitches of the notes themselves are tuned, and the way that these pitches relate to each other. For this piece the piano has been tuned so that all the pitches are selected according to the harmonic relationship between the notes, and only notes that have an overtone relationship to each other are used. This effectively means that all the notes contain some ‘part’ of the others, and when heard quickly in succession, they become hard to distinguish from each other – and they blend into one gigantic sound, like magic.

It’s all about proportion. The notes interact with each other in the way they do because of the proportional relationship that their pitches and the overtones derived from them share. When heard together, these relationships are expressed through a sound that blurs from many separate notes into one, creating a rich and even sonic texture.

In typographic terms, texture, or the density of texture of a printed block of text, is known as its ‘colour’. As Robert Bringhurst writes, ‘This has nothing to do with red or green ink; it refers only to the darkness or blackness of the letterforms in mass.’ This ‘colour’ plays a role in defining the character of the page. In the same way that differently painted rooms create a different atmosphere, so the colour of a text block can create different ‘atmospheres’ within a page. Many factors influence how the texture of a block of text appears, but the primary ones relate to the design of the type used, the spacing between the letters, the spacing between the words and the spacing between the lines.

Within the realm of the first of these factors – the design of the type – a basic consideration is its weight. Increasing the weight of type, from light to bold etc., will cause the texture of the type block to darken; a heavier weight of a font has a thicker stroke width and thus prints more black on the page. This basic ratio of black to white that exists in any given font, between the black of the glyph and the white space enclosed in its surrounding bounding box, is a part of what will contribute to the ‘colouring’ of a block of text. Spacing variables, type size, the shape of the letters and even language will all play their parts, but taken as a starting point, the weight of a font is a significant factor in defining the tonal colour value of the text.

As mentioned, the notes in Young’s ‘well tuned piano’ are tuned in such a way that they all sound a little bit the same, containing small ‘pieces’ of one another that when heard together, amass to make one overall tone. In much the same way, type designers design type so that many of the letters contain small parts of each other. Of course, a ‘c’ must be recognisable as a ‘c’ and not an ‘e’, but the two characters will invariably be made from the same building blocks – the same proportions, stroke widths, curves and so on – in order that harmony is achieved when they appear side by side and furthermore, so that when seen en masse, like the notes in Young’s piano, the letters merge into an even texture.

Adrian Frutiger’s Univers, published in 1957, is widely credited as the first typeface to be designed from the outset as a complete system, whereby each variation in the typeface family is compatible with all the others, and will always be recognisable as part of that same family. It could be seen as the first type design that really considered how the proportional relationships between various weights in a typeface could be developed, in order that designers might be able to expand their use of contrast in typography without the loss of unity or coherence. According to type historian Friedrich Friedl, when Frutiger began work on Univers, one of the ‘first considerations for the realisation of these systematised fonts started with the problem of the adequate and reasonable boldness of the character lines and their relation with the white space’.

It is this relationship between ‘boldness of line and white space’ that helps to produce the visual tone, or colour, of the text. Tone describes the lightness or darkness of a colour. A grey tone for example, is made up of a mixture of black and white, as shown in the grey tone you will see if you squint your eyes at this text. When type is set in a block, it is the combination the black type with the surrounding white space that helps to define the colour that you see.

In sonic terms, though it’s slightly more complicated, defining tone is similar in essence. So say, when a note is played on a piano, the sound heard is a combination of tones: the loudest one, known as the fundamental tone, and many overtones, known as partials. All of these tones correspond to sound wave frequencies, and the combination of these various frequencies produces the sound heard when the note is played.

In both cases, on a basic level, it is the combination of elements – black and white for type, various frequencies for sound – that create tone, both visually and sonically. How the proportions in these combinations are balanced is what characterises them.

The basic premise of this experiment is in many ways quite a literal one. There are tones for your eyes and tones for your ears, and this is an attempt to exchange one tone for another; to use what information can be gathered about the tonal value of a typeface as the basis for developing a method with which to create a corresponding auditory tone for any text set in that typeface.

Given Univers’ reputation as the original systematically proportioned typeface, it seemed like an appropriate typeface to begin the experiment with. The increase in weight of each font in the ‘basic’ range of the Univers family, from Ultra Light to Extra Black, is systematic and proportionally consistent, and the variety between the weights has the capability of providing a good range of tonal texture on a page.

One process I have used involves measuring each glyph of these 8 Univers weights, determining the black to white ratio that each glyph has within its own bounding box, and then applying that ratio to sine wave frequencies that express the same ratio. A sine wave with a single frequency is known as a pure tone. Sines waves, or sine tones with more than one frequency are known as complex tones. So the ratio of black to white in a glyph produces a relatively ‘simple’ complex tone – made of two frequencies, one for black, one for white. These tones, when combined and layered with the other letter/tones, are then able to ‘spell’ blocks of type through sound – creating sonic interpretations of text. Because of the proportional relationships that each weight of the typeface has, each one produces its own distinct sonic tonal range, as the sound frequencies mirror the black/white proportions contained in the design.

Another method I have worked with involves taking the overall black : white ratio of entire pages of text and converting these into sine tones so that the black within the text block on a page constitutes one tone and the white another. When combined this produces a much ‘cleaner’, simpler sound. The advantage of this method is that it considers spacing concerns more accurately – which the first method was unable to without making complicated adjustments to the ratios – and texts with an average amount of leading produce a very different sound to texts with a lot. It is also easier to notice particular ‘beating’ effects (when the sound ‘wobbles’ instead of playing ‘flat’), which are produced when the ratios of the frequencies are close to each other, which again, is indicative of some visual attributes of the type design.

This booklet itself demonstrates some results of both methods. As you see, it is the same text twice – with the body copy set in Univers Bold going in one direction, and in Univers Light the other. The tracks on the CD play the interpretations of indicated pages with the running order playing through back and forth from one page of text to the text sitting opposite it in that spread and so on. The running footer percentages give an indication of the black percentage in the text blocks on each page. The ‘noisier’ tracks use the first method, the quieter ones the second.

Applying sets of figures garnered from the realm of the visual to the realm of the auditory has thrown up some technical and conceptual problems – primarily because there is no real, concrete reason why a visual object should have a corresponding auditory value. Even if the proportions are the same, should a figure of 17.2 % black constitute a frequency of 17.2 Hz? Or 172 Hz? Or 172000 Hz? Allowing for a certain level of arbitrariness as a given, the best thing that can be done is to establish a framework and parameters to work within – and then to be consistent. The tests carried out within that system acquire any meaning by comparison with other work carried out in that same system with the same parameters. It’s ‘all relative’, as they say. To this end I have worked with the figure of 6Hz with which to multiply the black : white ratios, in order to get them up to a frequency that would render them audible. This figure was simply selected for the reason than I felt it produced a sound that was neither too high nor low overall, and allowed for good comparison between the ‘black’ and ‘white’ sounds. So for instance, the letter ‘o’ in Univers Light contains a black : white ratio of 13.63 : 86.37, which when multiplied by 6Hz produces a sound made of the frequencies 81.78Hz and 518.22Hz.

The higher the contrast between the black : white in a font, the higher the contrast between the tones in it; where the sound is high pitched but still contains a lot of bass. The lower the contrast is in a font, such as in a heavy weight, the ‘closer’ together the tones are in frequency. Some further consideration might also include attention to the relative amplitudes of the frequencies as this has an impact upon the sound significantly, yet visually speaking, volume does not seem to have as naturally a corresponding equivalent proportionally as black/white contrast. Type size? ‘Volume’ of text? Some questions answered perhaps, many still remain...

I won’t go into any more of the details of the work, hopefully this has been enough to provide an insight into it, and if you have made it this far, then I would say thanks for reading, and for listening too. This booklet was borne more out of a desire to give the experiment a tangible form, and document a stage of its development - rather than to provide any particular last word on it. I am reminded here of quote by Victor Papanek who, talking about how the design process can never yield one ‘right’ answer; says that instead, ‘it will always provide an infinite number of answers, some “righter” and some “wronger”.’ It’s interesting to note how by adding the ‘er’ to those two words, he takes away the concrete meanings of ‘right’ and ‘wrong’, and puts them somewhere just outside of definition. We know what the words mean, but can’t explain them exactly. I feel this is an appropriate thought to end on for now, as it sums up the process so far quite well. It has been one of trial and error, seeing which interpretations are ‘righter’ and which ones are ‘wronger’, but never finalising a definitive method. Type, after all, is not sound. Neither you nor I will ever look at a page and hear anything. This is just about asking what we might hear if we one day did.

JPH, London, May 2009

Many thanks to Maeve O' Neill for her assistance editing this piece.