The Logic of Filtering's main argument can be summarised in a somewhat simple statement: despite the ‘myth of perfect fidelity’, any recorded sound brings up the ‘noise of sound media’, a ‘defining feature of all technologically (re)produced music’ (p. 3). The book, therefore, is an effort to describe and problematise how the so-called ‘myth’ of sonic fidelity to an original sound has constituted every phase of technological research and the development of recorded sound. From the first mechanical attempts to the digital era, ‘noise reduction’ has always been a defining trace of recording techniques and equipment.
Kromhout argues that the ‘fight against noise’ is based on a series of artificial splits between noise and information, noise and distortion, noise and ‘signal’, which separate the ‘things that you want’ to listen to from ‘things that you don't want’ to hear. As a result, he sustains that ‘the purity claimed by noise reduction is relative and precarious, for noise is internal and inherent to all communication systems’ (p. 58).
According to the author, the idealisation of the perfect fidelity between input and output in the recording process can be traced back to the 19th century with the mathematical efforts of authors such as Joseph Fourier, Georg Simon Ohm and Hermann von Helmholtz to represent sound waves. ‘With Ohm's application of Fourier's analysis of sound, and Helmholtz's expansion and experimental verification of its principles, the sine wave came to be defined as the elemental tone: a pure frequency with no overtones and no timbral characteristics of its own’ (p. 88). Hence, the ‘sine wave’ operated as an analytical symbol, constructed upon a ‘conceptual act of noise reduction that suppresses all reference to its material carriers (transmission channels)’ (p. 89).
Pursuing this desired purity of the sine waves, all noise reduction interfaces and devices have been created as a means to achieve a perfect ideal output that should avoid the channel interference. Replacing the ‘myth of perfect fidelity’ with ‘the logic of filtering’, as Kromhout proposes, is an attempt to shift the conceptual guideline of noise reduction, highlighting that media is an active agency that also produces sound, not necessarily undesired.
A very interesting discussion of the contradiction that lies in the radical opposition between noise and ‘desired’ sound or silence is developed in the passages about the ‘dithering practice’, that ‘introduces small amounts of random noise as a means to reducing errors in sound digitalization’ (p. 13). With this, ‘the artefacts of quantization error are entirely eliminated’ (p. 39). Dithering means applying ‘desired noise’ to avoid digital errors.
In a spiralised rhetoric, the author's argument includes in each chapter more and more elements and dimensions, always highlighting the importance of the process in sound reproduction. In the eponymous fifth chapter, the concept is particularly illuminated. For Kromhout, ‘the logic of filtering … stresses that it is neither sender nor receiver, but the operations of a parasitic third (the channel itself, administering a physical cut) that produce the rich, complex, nonperiodic, and random sounds of technological media’ (p. 126).
The book thus oscillates between philosophical debates and detailed technical discussions, passing through mathematics, acoustic and communication theory. Despite recognising the author's careful explanation for non-experts, some passages in which highly specific characteristics of acoustic sounds and technical sound design are described may be hard to follow, especially for (socio)culturalist readers (like me).
For this reason, it would have been helpful to include debates about listeners and their perception of sound. Throughout the book, we are convinced that the ‘myth of perfect fidelity’ is a major concern for recording engineers and all the staff involved in the process. However, little is said about the reproduction of sound. The book's main premise that technological mediations produce a particular ‘noise’ that is constitutive of the final result is highly convincing. However, we could also take into account the conditions under which the sound would run all the way back from digital codes to physical sound waves. In this reverse movement, there is another filtering operation processed by the reproduction devices. In other words, low-bass car radios or cheap earphones are also filters that shape both the sound and the listening activity in such a way that all the filtering done during recording and mixing processes may sometimes be less important than the sound heard in the final device itself.
Moreover, it would be interesting to further develop how all these filtering operations work in live music. As we know, live music nowadays is always mediated by microphones, pickups, cables, filters, equalisation devices and sound boxes. Some of the author's findings are easily applicable to this apparatus, but live music has some particularities that may deepen the debate concerning the presence of noise in mediated music.
By the end, Kromhout's book contributes to research on the active role of the recording media in communication systems. In his words, ‘technological media are not simply tools that allow us to control and shape the world as we see fit. In many ways, it is they that control us’ (p. 155). His main achievement is to include noise reduction attempts and the ‘logic of filtering’ in this debate. For this, his book is from now on a very important reference.
