SCREAMING
Matthew Fuller
Screaming is the first form of speech. All utterances subsequent to removal from the
womb are a simple modulation of this basic condition. Opening the cavity of the throat,
opening it fully, with direct realistic cognisance of what is before you, should often
result in a scream. The lungs lurch compressed air out of the body for the atmosphere
to bear the violent imprint of their spasming alveoli. The scream is air tearing through
air at a speed that makes it shudder to a standstill. As a fire devours oxygen from the
gases that surround it, so a scream commands the attention of the consciousnesses
attached to nearby ears. The imprint of your shrieking lungs upon the air demonstrates
their size (a surface area larger than that of any residential premises you will ever be
allowed to lock yourself into) and their capacity for high-speed expulsion, but that is
not all.
Screams weave themselves into speech, provide grounds for vocalisations that
bring them to the fore, and in their variation in kind mark raw controversies, tensions,
or ruptures. The singing-screaming of Matana Roberts, the roaring singing of Diamanda
Galas—the voice in such conditions is a rack upon which it wracks itself. Fusing horror
and erotics come Screamin’ Jay Hawkins, James Brown, and Little Richard amongst
a medley of whimpers, howls, panting, and singing where throat-slicing caresses run
through the howl. Here, the rigorous disciplined voice overcomes itself in the dynamics
of screams which are themselves both outside and in its range.
The sonic complexity of screaming is what makes it powerful from other perspectives. As one forensic researcher puts it, ‘The exact acoustical mechanisms vary and
can be quite complex, including the effect of large scale temporal patterns, turbulence
and nonlinear acoustic effects, and complex spectral patterns including harmonic and
44
inharmonic components’.1 Screams gain part of their power from their fierce incorporation of all of these features, but also from the complex movement between them in a
F uller : S creaming
compressed time. That us to say, screams exist across, but are irreducible to, multiple
regimes of quantisation. Some recent biological research proposes a quality it calls
‘roughness’, a high range of variation within a short amount of time, with screams often
moving abruptly between pitches of 30 and 150 Hz.2 This quality allows screams to
penetrate background noise, and suggests that they correlate to a few other sounds
with rapid variations in range—such as alarms. Both sounds light up the amygdala like
a meat Christmas tree of fear.
The algebra of screams is thus ready for its incorporation into wider economies of
fear and libido. The power of the scream makes it the money-shot of the horror film and
the thriller.3 Since it shows that we are being thrilled, the scream is difficult in relation
to the question of meaning. As a ‘complex of natural cries and moans’, it is, for literary
philosopher Mikhail Bakhtin, devoid of any ‘linguistic (signifying) repeatability’.4 The
scream fails against this test since it does not correspond to alphanumeric characters
that are repeatable as the self-same and can be measured against one other. In order
to be entered into a system of measure, the scream must be made numerical.
Traditional coercive societies rely on high levels of human monitoring of other
humans. The absence of camera surveillance in such environments makes this extremely
labour intensive, requiring that people have to be kept in sight by other people. Means of
auditory surveillance, record keeping, and the encouragement of commonality of belief
and culture, alongside systems of punishment for infraction, are means of ameliorating
such costs. Whilst systems for speech recognition built into tablets and mobile phones
tend to find the presentation of a scream incomprehensible,5 that is to say without
either prior encoding or sufficient opportunity for learned response, progress has been
made in developing surveillance-oriented systems that identify screams, gunshots, and
1.
D.R. Begault, ‘Forensic Analysis of the Audibility of Female Screams’, paper delivered at AES 33rd International
Conference, Denver, CO, USA, June 5–7 2008.
2.
L.H. Arnal, A. Flinker, A. Kleinschmidt, A.-L. Giraud, and D. Poeppel, ‘Human Screams Occupy a Privileged Niche in
the Communication Soundscape’, Current Biology 25:15 (2015), 2051–56.
3.
E.g., Blowout (dir. Brian De Palma, 1981).
4.
M. Bakhtin, ‘The Problem of the Text in Linguistics, Philology, and the Human Sciences: An Experiment in
Philosophical Analysis’, in Speech Genres and Other Late Essays (Austin: University of Texas Press, 1986), 105.
5.
For the purposes of this article, rigorous tests were carried out with Apple’s Siri, the Android Assistant, and
Microsoft’s Cortana.
explosions against background noise.6 For screaming not to occur requires training,
45
the passage from infanthood to adulthood. The self-invocation of order may now
the difference between a scream and the proper management of social repeatability.
As well as monitoring the environment for screams, screams may also be broadcast
into inhabited locations in order to extract experimental results. Primatologists may be
observed playing recorded scream sounds to a group of wild chimpanzees and auditing
the effects, differentiating the resulting responses into agonistic screams and tantrum
screams.7 Screams thus become part of a general regime of input-output responses
enhanced by media recording, playback, monitoring, and analysis systems that in turn
elicit the first forms of speech and condition all those that follow it.
6.
L. Gerosa et al., ‘Scream and Gunshot Detection in Noisy Environments’, paper delivered at the 15th European
Signal Processing Conference (EUSIPCO-07), Poznan, Poland. 3–7 September 2007; S. Ntalampiras, I. Potamitis, and
N. Fakotakis, ‘On Acoustic Surveillance of Hazardous Situations’ and C.-F. Chan and E.W.M. Yu, ‘An Abnormal Sound
Detection and Classification System for Surveillance Applications’, papers delivered at the 18th European Signal Processing
Conference (EUSIPCO 2010), Aalborg, Denmark, August 23–27 2010.
7.
K.E. Slocombe, S.W. Townsend, K. Zuberbühler, ‘Wild Chimpanzees (Pan Troglodytes Schweinfurthii) Distinguish
Between Different Scream Types: Evidence from a Payback Study’, Animal Cognition 12:3 (2009), 441.
F uller : S creaming
move to rewarding partnerships with ubiquitous monitoring systems that recognise