The very concept of ancient computers might sound bizarre, and the notion of centuries old open source computer music stranger yet, but both represent historical realities with roots far deeper than is generally recognised. At the time of writing this article The Science Museum in London is running an exhibition called The Information Age, with subject matter that ranges from personal computers and the internet, to the telegram and Congolese talking drum. Yet the ancient computers and open source code I wish to discuss are not mentioned. The problem is not that they are forgotten or ignored, but rather that they are not recognised for what they are to begin with.
Before describing the computers themselves it is important to note the context in which they arose. Throughout the European Middle Ages there existed a desire to measure time with ever greater accuracy. This led not only to the invention of the mechanical clock, but triggered a general shift towards a more mechanistic and automated civilisation. A world which once measured time by the light of the sun, agrarian rhythms, and the patterns of human events, gave way to one of objective, standardised time, abstracted from the variable, unregimented flow of nature. Furthermore, since time is concerned with events, when the focus of events shifted from the natural to the mechanical, the seed of the computer age was sown.
During and prior to this transformation, bells played an important role in organising society. In his book Revolutions in Time, author David Landes describes their significance in community life during the Middle Ages:
“Bells sounded for start of work, meal breaks, end of work, closing of gates, start of market, close of market, assembly, emergencies, council meetings, end of drink service, time for street cleaning, curfew, and so on…”
In 17-century Britain the technology of bell towers was adopted for another purpose, or to use a modern term, was ‘hacked’ (not hacked as in a computer being broken into, but the original meaning of the word: to modify technology in creative and inventive ways). The bell technology of church towers – originally intended for community timekeeping – would be put to use for music making. Various combinations of harmonically tuned bells were hung in these towers, controlled by ringers pulling on ropes lower in the building. As the technology for mounting the bells improved, ringers gained better control over their movement, allowing them to create music of greater complexity. This combination of having improved control over the bells, while nevertheless operating within the constrained framework of sounding a large bell using a rope, led to the creation of a unique form of music, composed from algorithms.
An algorithm simply refers to a series of step-by-step instructions, in this case for the purpose of exploring a vast number of bell patterns permutations, in a variety of inventive and aesthetically pleasing ways. This transformed bell towers into computers: with multiple processors (the bellringers), computing algorithms (the compositions), with hardware that output sound (the ringing of bells). This music became know as ‘method ringing’ (or ‘change ringing’), and, interestingly, shares the word ‘method’ with modern programming, in which this word is at the core of its lexicon.
Here is an example of a pattern generated by an algorithm for a three bell tower:
Each number above represents one of the bells, which are rung in the order that the numbers appear. In the above code every possible combination of three bells is rung, until the bells return to the order in which they started (i.e., the top and bottom rows, which are bold). The algorithm for computing this is relatively simple:
First switch the order of the last two bells (e.g., 123 → 132), then switch the order of the first two bells (e.g., 132 → 312), then repeat these instructions until you return to 123.
While the above algorithm is relatively simple, and its music short in duration, as more bells are hung the number of permutations increases exponentially: a tower with four bells has 24 possible permutations; a tower with five bells 120; and St Paul’s Cathedral in London – which has twelve bells – has 479,001,600 possible permutations!
Lastly, it should be noted that the above process has a binary dimension to it: a bellringer is instructed to either ring a bell, or not ring a bell. So were we to take “1” to mean that a bell is rung, and “0” that it is not rung, the previous code could be rewritten for the ringer of bell 1 as:
So we have computing, but what about open source?
In 1667, the fifth edition of a book called Campanalogia Improved was published, which documented dozens of change ringing methods. The book reads much like a modern day computer programming manual: brimming with algorithm examples, and filled with pages covered in large blocks of computed numbers.
The titles of the compositions in the book contain some fascinating revelations too, since they are frequently named after regions rather than individuals, lending them a strong sense of shared ownership e.g., “Cambridge Surprise”, “Winwick Doubles”, and “Lancashire Delight”. Furthermore, the titles demonstrate that compositions were modified and refined, often regionally. For instance there is a composition genre called “Bob”, which takes on numerous forms: “Plain Bob”, “Plain Bob Triples”, “Double Bob”, “London Double Bob”, “Oxford Double Bob”, “London New Bob”, “Royal Bob”, “Bob Major Double” etc.
The above descriptions meet opensource.org’s first three definitions of open source: there is “free distribution” which does “not require a royalty or other fee”, there is a “well-publicized means of obtaining the source code”, and there is an allowance for “modifications and derived works”.
It should also be noted, while much criticism of modern day open source culture focuses on the notion that it devalues creative work, method ringing was not built on free labour. As described on bellringing.org it was a creative economy with a source of financial sustenance:
“The cost of maintaining bells and payments to the ringers, who at the time were paid for their services, could be quite a high proportion of running the Parish. St Margaret’s, Westminster paid ringers one shilling each for ringing at the beheading of the Queen of Scots. Less than twenty years later, the same church paid ten times that for ringing ‘at the time when the Parliament House should have been blown up’.”
Method ringing continues to be practiced up to the present day, with early ringing societies such as the Lincoln Cathedral Guild claiming to date back to as early as 1612, and, according to the book Music and Mathematics (2003), there exist “over 5200 sets of bells hung for change ringing in England, another 200 in the rest of the British Isles, and about 100 elsewhere in the world.”
While the computers described in this article are neither electrical nor wholly mechanical, to ignore such an important tradition creates a blindspot in the history of computing. Strange though it may initially seem, centuries before Charles Babbage built his Analytical Engine, or Alan Turing imagined his Turing machine, there existed a thriving open source computer movement, which created, shared, and modified algorithms, for the purpose of generating new music.
Note: This article was originally published the April 2015 Issue of Contributoria. It is republished here under a Creative Commons licence. If you wish to republish it, please attribute the original article.