This is a new application of the planetary aural modelling technique developed in "Carmen of the Spheres".
This is perhaps the final installment of the ongoing saga of aural planetary modelling commenced with "Carmen of the Spheres". Here, I'm applying the model to exoplanets, specifically those of Mu Arae and 55-Cancri.
In essence the two underlying principles are:
1. An orbital period is an oscillation
2. Doubling the frequency of an oscillation preserves its pitch "flavour" (at least to humans, which constitute my primary audience!)
The piece is structured into three movements:
1. A model of the Mu Arae planetary system, including the as yet unofficial "candidate planet" (so five in all). The model presents the pitches derived from the orbital periods in sequence as a melody, ie. planet b, c, d, e, "f", with the durations grouped by similarity (eg. durations of 9-13 seconds, durations of 0.4-1.1 seconds, etc.). These melodic forms are looped and layered. In addition there are two more melodic layers, derived from other arbitrary shapes in the data matrix, with the frequencies taken from two different octaves alternating and the durations drawn in sequence (see the supporting excel workbook for details of how this works).
These last two melodic elements are also subjected to an arbitrary "Doppler effect". I do not personally have the expertise to accurately draw this from the red-shift of the Mu Arae star, so as a (temporary?) compromise I simply shifted an octave, such that the melody starts and ends in the "right place" but still presents an allegory of "Doppler shifting". nb. More realistic models are WARMLY welcomed!
2. An interlude constructed of chords made out of planetary systems using just one frequency&duration for each planet. This begins with Mu Arae and ends with 55-Cancri, with the other chords being drawn from the other known systems of three planets (as of the date of publication). For details and data relating to this, see supporting documentation.
3. A model of the 55-Cancri planetary system. This works similarly to the Mu Arae model but sounds completely different due to the nature of the frequencies and durations. By a strange fluke of nature, the system sounds remarkably "tonal". This gives the overall piece a really satisfying feeling of culmination and 'resolution'.
The piece is intended as a kind of response to Jared Diamond's "Collapse", which makes the metaphor of the Earth being like a larger-scale version of Easter Island, with the possibility of rescue from outside equally unlikely.
The "SETI" in the title does not strongly refer to the "SETI" project and is not really a critique of its aims. The systems modelled are not high priorities for the search for life and are, instead, chosen simply because they have lots of planets that have been accurately measured in terms of orbital period and are thus easy and satisfying to model aurally.
Download complete via The Internet Archive
1. The method is detailed more clearly in "Carmen of the Spheres". In essence the method is this:
i. Take the planetary orbital period in days and convert to seconds.
ii. Divide by two repeatedly until you arrive at two ranges of data:
a. Durations in the usable range, ie. from about 0.1 seconds to about 15 seconds
b. Frequencies in the audible range, ie. from about 60hz to about 6000hz
It is important to note that a slight (and trivial) complication is that frequencies are drawn from very small durations by using 1/p to convert seconds to hertz. This will explain any apparent confusion as to where the frequencies "came from" in the excel documents. It is a fallacy to just grab appropriate-looking numbers (eg. 440) and "convert" to hertz simply by using the number.
2. The idea of "hearing" a planet orbiting a star is counter-intuitive but is valid. I am not proposing that there is an atmosphere in space; nor am I proposing that one can slow down time (especially by differing factors for each planet!): rather the model is like a scaled visual model, as is much more familiar, where an arbitrary scaling factor is used to retain the visual "feel" or "flavour" of reality in a usable diagram. With sound, multiple factors can produce multiple usable results. The two basic facts which form the foundation of the model are listed in the main description for this item but in brief they are the nature of an orbit as an oscillation and the acoustic "rule of octaves" which tells us that a wave of frequency F sounds "alike" to one of frequency 2F or of F/2. Therefore an octave-based model (if you like, an integer-only logarithmic transformation at base 2) is more faithful to the "sound" of the planets than a single-factor transformation which arbitrarily brings all frequencies into range (in some cases, if you're lucky!) without preserving their intervallic relationships. (What I'm saying is that a single-factor divisor changes not only the inherent pitch of each orbit but also the musical intervals between them, thus totally randomising the affective aesthetic considerations of the model.)
3. With regard to the accuracy of the model and the potential for distortion and mistakes, it is important to note that errors in the measurement of the planetary orbits are DECREASED by the process of division, rather than amplified. This has two implications:
i. Errors will not affect the final result unless they are stupendously huge
ii. A wide range of orbital periods will fall into acoustically similar bands
The second implication is due to the way the flavour of sound waves is perceived; with waves being logarithmic in nature, small (numerical) changes to high frequencies make less difference than small (numerical) changes to low frequencies. This is well-known in the case of whole-number frequencies in hertz: 400 and 500hz sound much more different than 5400 and 5500hz. However it also applies in an inverse curve to fractional frequencies in hertz (in other words to periods larger than one second) such that 1/400hz is more different to 1/500hz than 1/5400hz is to 1/5500hz. In short, our notation system is well-tailored to our range of auditory perception and the further-removed the periods being treated as potential acoustic material, the wider the numerical bands between acoustic flavours. (This becomes much clearer if you try listening to the frequencies involved!)
4. The principal source for the data regarding the orbital period of exoplanetary systems is the website: http://exoplanet.eu/
I would like to formally thank them for making public this invaluable resource.
The CD is £5 anywhere in the world. Total time: 46m36s
To order a hard copy of the CD click here to pay via paypal: