1. Background and definitions

Precursors of artificial sound generation can be considered not only engineers of early electronic instruments such as the theremin but also the forefathers of applied algorithmic research that led to this achievement. Noteworthy is the research into arc discharge oscillators, whose typical practicality can be seen in voltaic illumination devices. Its origin, as well as the field of electrochemistry, can be originally attributed to key developments in the production of sulfur-enabled friction-based electrostatic generation devices, consequently stemming from electromagnetic experimentation and the rejection of scholastic as well as Aristotelian methods of inquiry into natural phenomena by physicists and inventors. Rejection and revolution are crucial and indispensable aspects of scientific advancement, similar to social systems.

From a cultural perspective, unexpected results yielded by purely theoretical approaches have been occasionally known to emerge as fruitful means of artistic expression in various fields, creating new and original lexemes to be further built upon into various aesthetic systems, much like how quite a large portion of physics or other exact sciences has been chiselled from what at first revolved around purely speculative hokum, adapted to reality through the assiduous work of countless largely unknown scientists.

Regarding sound: Fourier analysis shows us that any continuous signal of arbitrary complexity can be deconstructed into a sum of oscillatory components, enabling the encoding and further resynthesis of a said signal by means of adding together oscillations. Thus the size of a set of possible signals can be easily calculated based on several parameters: duration of said signal, minimum perceivable difference between frequencies, as well as the inferior and superior limits of auditory perception. Frequencies beyond human hearing are often seen as a proverbial tree falling in the woods with no one to hear it, although some arguments exist in favour of the enrichment of overall perception and the role of non or peri-auditory elements in a primarily auditory-targeted activity. It is noteworthy to mention that the minimum duration of time at which a difference is perceivable can also be determined from the higher threshold of hearing, as a perceived change in frequency cannot happen in a shorter period of time than one correlated to the wavelength of the highest perceivable frequency.

From this finite set of possible signals, the vast majority is considered perceptually identical and can be classified under the umbrella term of “white noise”. So far, studies have not yet shown the ability of humans to differentiate between or identify two or more samples of white noise, this likely being a neurological limitation. Statistical tests based on autocorrelation can determine whether a given signal falls under this category. The concept of asynthesis derives from this idea, in which generating a new signal does not take oscillatory sequences as a base (which includes subtractive synthesis by filtering a noise function), but instead sculpts out the perceptually identical or irrelevant elements from a set of all possible sounds. In the example above, a simple mathematical procedure can take care of excluding a large amount of data, however for more complex patterns, tools such as psychoacoustic analysis might be necessary to distinguish between asynthetically relevant and irrelevant categories of sounds.

2. Principles

We could further establish a set of principles that formulate a basis for asynthesis:

I. Any sound that can be obtained through asynthesis must be obtainable only through asynthesis.

II. Any other perceivable sound, which can be obtained neither through conventional means nor through asynthesis, cannot and therefore is forbidden to exist.

III. Any sound that is derived through asynthesis will be considered used, discarding it from the pool of possible sounds to be further generated. Note that this is a direct consequence of principle I, as the method of creating said sound would be a replication of a previously derived sound, and not asynthesis itself.

IV. A sound that has been used according to principle III can be reintroduced into the pool, as long as no reference to said sound of its derivation can be reproduced by any means other than asynthesis.

V. The pool of possible solutions to an asynthetic search may grow or shrink according to the evolution of perception.

VI. Any transgression of these principles constitutes a corruption of the concept of asynthesis and must be treated unsparingly.

Note that a sound and the procedure that was used in order to derive it are considered interchangeable. After a successful asynthetic derivation, the author may choose to keep or discard the resulting sound and/or the algorithm used. Keeping the algorithm in any form subjects it to principle III. Discarding the sound in order to effect principle IV can only be done by ensuring its validity. A few caveats to consider when attempting this:

• Even one’s own mind constitutes a potential way of reaching the algorithm again and thus must be subjected to erasure.

• One must consider the possibility of further advancements in technology that could retrieve information from the past, therefore one must prevent such advancements at all costs (beware of paradoxes).

• Theoretical physicists have concluded that some information collected within a black hole can be retrieved upon its evaporation, therefore this does not constitute a “quick and dirty” solution for discarding asynthesis algorithms.

Considering all the above, it is widely encouraged not to practically attempt reinsertion under principle IV. Rather than fail, it is preferable not to try at all.

3. Implementation

The trivial task of deriving a working methodology for asynthesis is left as an exercise for the reader. However, for the sake of completeness, we will give a number of starting guidelines for an aspiring asynthesist.

Asynthesis may be implemented in either digital or analogue mediums, as from a perception standpoint, both are equally existent. It is however recommended to also design different media, as long as they follow the same equivalence of existence within auditory perception.

Methods of implementing asynthesis can be divided into direct, indirect, formulaic, and undisclosed. Likewise, the taxonomy of this growing field is still somewhat flexible, allowing for further expansions.

• Direct implementation implies creating a sound by designing the entirety of a pressure pattern, without resorting to mathematics-assisted methods.

• Indirect implementation represents the creation of a pressure pattern as a consequence of one’s actions, but not through direct interaction with the wave. It is commonplace for second actors to be constrained into a creative process, as long as they are given either the illusion of meaning emerging from their actions or a minimal means for continuity.

• Formulaic implementation is the design of asynthetic algorithms that sculpt through the set of auditory possibilities.

• Undisclosed implementations represent something that must not be discussed. Please refrain from asking questions regarding undisclosed implementations; in fact, it would be desirable if any asynthesist lacking significant experience in the field forgot about their existence altogether. 

The reader is encouraged to do so in a timely manner, as a means to avoid any potential consequences.