Noise as undifferentiated sound. This is the starting point: 'We might assume that it is possible to distinguish between musical sound and noise in acoustic terms: musical sound results from regular, periodic vibrations; noise results from non-periodic vibrations ... The distinction is based on the opposition between 'pure and simple sounds' on one hand and 'complex sounds' on the other' (Nattiez, p.45). Nattiez comments that the French language even employs an official physical definition of noise: noise is an erratic, intermittent or statistically random vibration. A sound has been considered as noise for a long time if its originating frequency was non-periodic and therefore of no determinate pitch. (Theodore Gracyk speaks of the doctrine that certain timbres are inappropriate for music making. He refers to the 19th century ideal of 'purity in articulation' where any richness of the contributing overtones was considered unwelcome noise (cf. Gracyk, p.114).) Non-periodic vibrations. Complex sounds. I call it noise as undifferentiated sound. As a starting point.
 Futurist Luigi Russolo was one of the first in the early 20th century to put the institutionalized division between music (Russolo mostly uses the term 'sounds' instead of 'music') and noises on the agenda. Russolo's essays on modern music all revolve around the main statement of his 1913 L'Arte dei Rumori. Manifesto Futuriste [The Art of Noises: Futurist Manifesto]: 'We must break out of this limited circle of sounds and conquer the infinite variety of noise-sounds' (Russolo, p.25). His main and simple question: If music is sound, then why does not music employ all the varieties that sound has to offer? According to Russolo, the traditional division between music and noise is based on the notion that music is a succession of regular and periodic vibrations. Noise, in contrast, seems irregular and fragmentary. Russolo questions this sharp distinction between music and noise and provides two arguments to support his position. The first concerns duration. 'The production of a sound requires not only that a body vibrates regularly but also that these vibrations be rapid enough to make the sensation of the first vibration persist in the auditory nerve until the following vibration has arrived, so that the periodic vibrations blend to form a continuous musical sound. At least sixteen vibrations a second are needed for this. Now, if I succeed in producing a noise with this speed, I will get a sound made up of the totality of so many noises - or better, a noise whose successive repetitions will be sufficiently rapid to give a sensation of continuity like that of sound' (Russolo, p.37). Here, Russolo tries to overcome the idea that musical sounds are continuous while noise is discontinuous. He undermines the sonorological difference between music and noise with respect to duration. By doing so, he is effectively introducing noise into the musical realm. The emancipation of noise has assumed an acoustic legitimization where noise is understood as a special kind of musical sound.
His second argument concerns timbre. Each sound is made up of a number of sounds (secondary vibrations, various harmonic sounds, overtones). Now Russolo states that noise is produced when 'the secondary vibrations are more numerous than those that usually produce a sound' (Russolo, p.39). His conclusion: the difference between musical sounds and noise must be only gradual. 'The real and fundamental difference between sound and noise can be reduced to this alone: noise is generally much richer in harmonics than sound' (Russolo, p.39). And he immediately continues: 'But, since these harmonic sounds always accompany a pre-dominant fundamental tone, every noise has a pitch'. However, giving pitch to noises does not mean depriving them of all irregular movements and vibrations of time and intensity. It rather assigns a degree or pitch to the strongest and most prominent of these vibrations. Harmonics are generally understood as sounds above the fundamental tone. They are produced by other vibrations, faster and shorter, which exist along with the principal vibration. With musical sounds, the fundamental tone is the lowest tone. This is not necessarily the case with noise, where the 'fundamental' tone is rather the loudest tone. There may exist lower (but softer) tones. The tone that characterizes the pitch in noise could therefore be an overtone of a weaker, lower fundamental tone (cf. Russolo, p.27 and p.79).
Russolo goes one step further here than in his first argument where noise could still be considered a separate case, a special sound within the musical domain. This second argument turns the relation between noise and musical sounds around. Here, musical sound is regarded as a special kind of noise, a noise where the lowest tone is at once the fundamental tone. While this could also apply to noise, it is by no means essential. Music or musical sound thus becomes a special phenomenon within the domain of noise. Music is no longer the domain within which noise needs to secure its place. Rather, noise becomes the framework in which music takes up a special (and still privileged) place. Arche-noise.
A futurist manifesto and other essays. Writings on music. Discursive expressions. However, Russolo's disruption and subversion (a deconstruction) of the boundary between music and noise is first of all manifested in his music. Unfortunately, the bulk of his work, in addition to musical instruments that he specially designed, was lost during the First World War. His ideas about a new music, a noise music, required a new method of notation as well as new instruments. In order to enrich the musical domain with new sounds, timbres and microtones, Russolo created the first musical synthesizers, the so-called intonarumori, or noise instruments that consisted of 'howlers', 'roarers', 'cracklers', 'gurglers', etc. His deconstruction of the relationship between music and noise articulates itself within music; it is a deconstruction through music.
 Noise as undifferentiated sound. The idea that musical sounds result from regular, periodic vibrations and noise from non-periodic vibrations has remained widespread until long after Russolo's death in 1947 (to date?). However, 20th century electro-acoustic research reveals that the spectrum of most musical sounds is non-periodic. This means that noise may well have the same acoustic structure as musical sounds (cf. Nattiez, p.45-6). Furthermore, there is a reciprocal relationship between volume and sound quality. Increasing the amplitude of a sound wave alters its characteristic pattern and thus its timbre. It exposes more overtones, both harmonic and non-harmonic; the sounds become more complex (cf. Gracyk, p.109). Musical sounds become noise (from 'simple' to 'complex' sounds) when the volume increases. The alleged opposition between music and noise based on physical differences becomes less clear; it becomes less of an opposition. When we attempt to exclude noise from the realm of music on physical grounds, when we try to assess the essential difference between the two as accurately as possible, the difference dissolves and disappears. The boundary between music and noise becomes uncertain and undecidable. This also happens when we try to restrict ourselves to a single tone. The tone, or, rather, the sinusoidal tone, is the smallest musical unity to which all sounding phenomena can be retraced. However, we cannot perceive sinusoidal tones. Unavoidable, there are always overtones present due to the space in which the tone necessarily resounds, and because of the way our ear functions. (A tone constantly fluctuates; it continuously differs from itself. Therefore, in its vibrating quality, a tone is nothing but a changing quality, Derrida explains in D'un ton apocalyptique adopté naguère en philosophie, p.18-26.)
In Positionen, Frank Hilberg distinguishes three problems where a physical distinction between musical sounds and noise is made. The first problem occurs with ultra short sounds. We hear tones or sounds that last too short as a click without a real pitch (i.e. noise), even where the sound structure is periodic and the sound spectrum is set up harmonically. A second problem arises when we talk about timbre, the basis upon which we are able to distinguish between instruments. It is precisely the non-stationary, discontinuous and aperiodic sound progression (noise) that determines the characteristics of an instrument. Or rather, it is the combined play of continuity and discontinuity that leads to a recognition of a specific instrument. The third problem is that every sound, be it a musical sound or noise, can be broken down into sinusoidal curves with the consequence that the distinction between musical sounds and noise can no longer be called qualitative; there is only a difference in degree (cf. Hilberg in Nauck, p.37).
 I stop here. No more examples although the list could be extended. The boundary based on a physical difference between music and noise is constantly transgressed. Each sound immediately proves to be pervaded by noise; noise is at all times a part of every musical sound. If noise is the outside and music the inside, we must conclude that the outside is always already on the inside. And this conclusion (which is not a conclusion in the conventional meaning of the word, not the final word, the definitive outcome) goes beyond the idea of the emancipation of noise, the extension of the concept of 'music' as an acceptance of sounds that were previously rejected as noise, beyond the idea that no sound can any longer be excluded from the domain of music. Noise was and is always already part of musical sounds. That's a difference between deconstruction and emancipation.