443 30 11MB
English Pages 352 [391] Year 2020
Testing Hearing
Testing Hearing The Making of Modern Aurality Edited by
Viktoria Tkaczyk, Mara Mills, and Alexandra Hui
1
3 Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and certain other countries. Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America. © Oxford University Press 2020 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization. Inquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above. You must not circulate this work in any other form and you must impose this same condition on any acquirer. Library of Congress Cataloging-in-Publication Data Names: Tkaczyk, Viktoria, 1976– editor. | Mills, Mara, editor. | Hui, Alexandra, editor. Title: Testing hearing : the making of modern aurality /Edited by Viktoria Tkaczyk, Mara Mills and Alexandra Hui. Identifiers: LCCN 2020001510 (print) | LCCN 2020001511 (ebook) | ISBN 9780197511121 (hardback) | ISBN 9780197511138 (paperback) | ISBN 9780197511152 (epub) | ISBN 9780197511169 Subjects: LCSH: Hearing. | Audiometry. Classification: LCC GN275 .T47 2020 (print) | LCC GN275 (ebook) | DDC 617.8/075—dc23 LC record available at https://lccn.loc.gov/2020001510 LC ebook record available at https://lccn.loc.gov/2020001511 1 3 5 7 9 8 6 4 2 Paperback printed by Marquis, Canada Hardback printed by Bridgeport National Bindery, Inc., United States of America
Acknowledgments Testing Hearing results from two events: the conference Testing Hearing: Science, Art, Industry in December 2015 and a subsequent authors’ workshop in October 2016, both held at the Max Planck Institute for the History of Science, Berlin. Over the past years, most of the authors have additionally spent time at the Max Planck Research Group “Epistemes of Modern Acoustics” in Berlin to fine-tune their papers. The editors thank the Max Planck Society, Mississippi State University, New York University, and the Alexander von Humboldt Foundation for their generous support of this intense, insightful, and enjoyable collaborative process. We are also grateful to the many participants of both the conference and workshop, most especially Karin Bijsterveld, Myles W. Jackson, and Carolyn Birdsall for serving as commentators at the October 2016 meeting, and to the anonymous reviewers of this book for their likewise constructive and very helpful readings. Not all authors of this volume are native speakers of English, so our special thanks go to Kate Sturge of the Max Planck Institute for the History of Science for her careful and tireless language editing of this volume. Penelope Krumm supported us with her sharp-eyed reading of the copyedited manuscript. In addition, we owe much to Birgitta von Mallinckrodt, who helped organize and arrange every step of the process in administrative terms. A group of highly committed student assistants at the Max Planck Institute for the History of Science (Hannah Eßler, Jonathan Haid, Alina Topf, and Fabian Voigtschild) helped to set up the database “Sound and Science: Digital Histories,” which provides rich additional material for this volume. Norm Hirschy at Oxford University Press has been an excellent, open-minded, and always supportive editor. Over the past years, several of the authors became parents and/or received tenure. We grew with this volume as much as this volume grew with us, and we thank our departments and families for their care. It was a great pleasure for the three of us to co-edit this volume. It has been a big project for all of us. In recognition of this, we chose to list our names alphabetically in the introduction, while inverting the order for the cover of the book.
List of Contributors Joeri Bruyninckx, Maastricht University Lino Camprubí, Universidad de Sevilla Emily Dolan, Brown University Jennifer Hsieh, University of Michigan Alexandra Hui, Mississippi State University Sebastian Klotz, Humboldt-Universität zu Berlin Stefan Krebs, University of Luxembourg Mara Mills, New York University Trevor Pinch, Cornell University Alexander Rehding, Harvard University Hans-Jörg Rheinberger, Max Planck Institute for the History of Science Benjamin Steege, Columbia University Jonathan Sterne, McGill University Viktoria Tkaczyk, Humboldt-Universität zu Berlin Roland Wittje, Indian Institute of Technology Madras
Testing Hearing An Introduction Alexandra Hui, Mara Mills, and Viktoria Tkaczyk
Just before midnight on October 16, 2014, Times Square became a testing room, its scrolling digital billboards hijacked by Japanese sound artist Ryoji Ikeda’s flickering barcodes, pixels, scan lines, and electronic snow. test pattern [times square] was billed as a “silent concert” in “one of the busiest and craziest areas in New York.”1 Watching the twitching black-and-white screens, audience members wore wireless headphones and listened, individually, to a synchronized composition made up of the microsounds for which Ikeda is famous: chirping pulses, bursts of white noise, signal patter, circuit fry.2 The multimedia installation opened with a single tone that swung dramatically upward in pitch until it exceeded the threshold of human hearing. This gesture referenced Imaginary Landscape no. 1, the pioneering electroacoustic work in which John Cage played a Victor tone record, generally used to calibrate instruments and room acoustics, at increasing speed to skew its frequency. Flipping Cage’s conceit on its head, Ikeda took test sounds that had been turned into music (provocatively—this was Cage, after all) and turned them back into a test. Ikeda’s composition denaturalizes the homologous sounds of hearing tests and machine signaling, which have become increasingly familiar since the early twentieth century. test pattern calls the parameters of such tests into question: What sort of hearing function does a tone test? What is the boundary between “pure” tones and pulses of interference? In what ways do reference tones adjust and standardize listeners as well as calibrate equipment? Ikeda expands our understanding of when and where testing takes place, underscoring the ubiquity of the test as a modern form of knowledge- making. He reminds us that testing does not leave its subjects and objects of study unchanged—the plastic sleeve of the first test pattern album carried a warning sticker: “Caution! This CD contains specific waveform, impulse and burst data that perform a response test for loudspeakers and headphones. High volume listening of the last track may cause damage to equipment and Alexandra Hui, Mara Mills, and Viktoria Tkaczyk, An Introduction In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0001.
2 Testing Hearing
eardrums.”3 Testing hearing can lead to injury, new medical diagnoses, and reconfigured social classifications. test pattern invokes the entanglement of music, industrial sounds, and scientific disciplines that is central to Testing Hearing: The Making of Modern Aurality. The modern cultural practices of hearing and testing have emerged from a long interrelationship. Since the early nineteenth century, auditory test tools (whether organ pipes or electronic tone generators) and the results of hearing tests have fed back into instrument calibration, human training, architecture, and the creation of new musical sounds.4 Whether employed to detect impairment or skill, hearing tests proliferated alongside new sound- making and audiometric devices, as well as the professionalization of otology and deaf education in the nineteenth century. Hearing tests received a further boost around 1900 as a result of injury compensation laws and state and professional demands for aptitude testing in schools, conservatories, the military, and other fields. Applied on a large scale, tests of seemingly small measure— of auditory acuity, of hearing range—helped redefine the modern concept of hearing as such. During the twentieth and twenty-first centuries, the epistemic function of hearing expanded. Hearing took on the dual role of test object and test instrument; in the latter case, human hearing became a gauge by which to evaluate or regulate materials, nonhuman organisms, equipment, and technological systems. The present volume considers both the testing of hearing and testing with hearing to explore the co-creation of modern epistemic and auditory cultures—indeed, the creation of modern aurality.
On Tests and Testing The dual character of testing hearing derives from two overlapping etymologies that have yielded slightly different meanings in the long history of “testing.”5 The first is testifying, that is, the witnessing of a particular event. In this context, the terms “test” and “experiment” have been used interchangeably since the early modern period to indicate the testing of philosophical propositions, beliefs, and opinions. In canonical accounts of the modern scientific method, hypotheses must also be tested, either through rhetorical persuasion or through observation, data collection, or experiment.6 Here, tests cannot be performed alone. They require the testimony of eyewitnesses or the “virtual witnesses” produced with the assistance of scientific writings and illustrations, and they often need to be repeated to establish validity.7 The second etymology of testing is linked to “teste” (Lat. testum = pot, vessel). In the fourteenth century, this word was used in alchemical experiments to mean
An Introduction 3
the cupel in which gold and silver were treated at high temperatures. Since the seventeenth century, “putting something to the test” has signified the examination of a wide range of substances, technologies, and emotional, physical, and mental states. This line of testing analyzes objects to identify elements, compares attributes to predefined benchmarks and norms, or delegates test subjects to stand in for a wider heterogeneous set. While these two etymologies are entangled, it is our contention that the latter has steadily expanded in significance in the modern era. Over the last two centuries, applications of testing have circulated throughout the sciences and beyond. We find four significant shifts within the trajectory of testing as “teste.” First, tests are no longer simply applied to objects, but also produce knowledge about human and other animal subjects. Second, new forms of tests, combined with new statistical approaches and computational technologies, have facilitated an expansion of testing to broad cohorts and populations. Third, the ambit of testing has extended from the evaluation of physical properties to also include functional qualities of materials, living beings, technological systems, urban infrastructures, and outdoor spaces. In the extension of testing to functional qualities, it is evident that tests carry epistemological power. In modern electrical engineering, for instance, “black box” testing arose as a means to study the outputs of complex, massive, or unknown apparatuses based solely on inputs. The “black box” rapidly disseminated as a metaphor and tool for understanding such disparate phenomena as telephone lines, animal behavior, and human neurology.8 Here tests serve, fourth, as a means to examine areas of knowledge beyond human physical and perceptual reach. These shifts redefine what qualifies as a test subject, as well as the potential contributions of these subjects (distinct from experimental ones) to the co- construction of knowledge. Everyone and everything can now be put to the test—whether by epistemological, industrial, artistic, or other cultural forces. This volume argues that testing as such became an enduring and wide-ranging social practice in the modern period. Individuals engage with tests from the moment they are born. Tests are built into daily lives with astonishing pervasiveness. Testing is a “cultural technique,” comparable to other key techniques such as writing, reading, painting, experimenting, seafaring, and filtering.9
Tests in the History of Science and Technology Tests have until now mostly been regarded as the “boring” part of science— unchanging and without a history. Yet much as historians of science have shown for the ubiquitous “experiment,” the various forms of scientific testing
4 Testing Hearing
also transformed or consolidated the definition of the object under investigation.10 And, while recent scholarship has shown experiments on music and sound to be flexible, with the same experiments being employed in more than one context (in the sciences, instrument making, musical performances),11 tests have an even broader field of application. Accordingly, many modern tests are as significant for everyday life as they are for scientific and technological developments. (For more on the relation between tests and experiments, see the “Afterword” by Hans-Jörg Rheinberger.) This volume shows that tests prompted the development of new scientific disciplines such as psychophysics, applied psychology, audiology, food science, and more. At the same time, new techniques of testing—for example, tests of threshold, just noticeable difference, range, and reaction time—arose across multiple disciplines, including but not limited to physiology, psychology, empirical aesthetics, and neuroscience. Some researchers accentuated minute individual differences and the inevitability of human limitations (“personal equations”), whereas others incorporated statistical data into broad categorizations of the normal and the pathological. Further, in their uneven efforts to quantify perception, both the practitioners and subjects of tests participated in the development of new standards of objectivity. As astronomers and other scientists pursued mechanical objectivity, for example, they sought to understand and minimize their own individual differences in observation; for the same reason, they also began to test the wide range of timing and precision in sensory responses.12 Indeed, as several chapters in Testing Hearing: The Making of Modern Aurality show, tests contributed to new forms of scientific objectivity altogether. The testing of human and nonhuman ears, in particular, prompted the invention of numerous means to standardize test procedures and render test subjects comparable. Testing became equally important for mass production and the administration of the social welfare state. Yet when applied in new contexts, by different people, and over longer time periods, tests undergo modification. Each of the historical cases discussed in this volume reveals how hearing tests proliferated and iterated in unanticipated ways, sometimes pushing against standardization or uncovering messy subjective experiences. Much in line with recent scholarship on scientific objectivity, we may say that objectivity has been a constant but moving target in the history of testing.13 Many tests applied in everyday life originated in the sciences, and they still bear the traces of experimental investigation. Some forms of test, however, entail observation, data collection, and inference from real-world events rather than modeling or experimentation. In these cases, the world is the testbed. As new building materials, technologies, and large technical systems emerged in
An Introduction 5
the modern period, testing codeveloped with these new fields of application. Testing started to play a major role during the planning phase of the applied sciences and the expansion of the modern state. Materials testing offices were founded, alongside national bureaus of standards, to facilitate building design, urban planning, technical engineering, and the food industry. Quality control testing emerged in the early twentieth century as a statistical method for setting minimum standards in mass manufacturing, with components and products sampled to monitor the production process. Testing now occurs at every stage of technical development, from design to manufacturing, marketing, maintenance, safety, and forensic analysis. In some cases, mathematical or physical models substitute for the direct testing of the thing under investigation. In engineering, testing often entails optimizing technology in comparison to a prototype or in conjunction with a human “user.” Test images, for example, enable the calibration of television and other electronic screens according to human visual function; test tones calibrate volume and pitch response for the audio signal. Intelligence tests of humans and machines, genetic tests, clinical trials, newborn screens, test tubes, litmus tests, model organisms, and drug tests pervade the history and anthropology of science, though they have yet to receive comprehensive analysis. Everything has been rendered testable, such that tests are now central tools for planning and regulation across all possible fields. And whereas scientific experiments have largely been investigated in relation to inscription and visualization, the history of testing is decidedly multisensory.14 For instance, as engineers employ black box testing to grasp the workings of complex technological systems, psychologists apply aural and visual projective tests to similar ends with humans. Moreover, psychological tests are applied for widely disparate purposes, whether to diagnose disorders, demonstrate “occupational aptitude,” or evoke imaginative responses. This wide-ranging applicability is what allows tests to be taken from the industrial laboratory to the classroom, to the concert hall, to the radio broadcast, or from an aquarium to a nuclear submarine. As science has moved beyond the walls of the laboratory, so have tests. They are mobile, but they are not immutable. Tests are materialized networks: they are the manifestation of practices, ideas, values, and institutions. Testing therefore serves as a political technology. The rise of the social sciences around the turn of the twentieth century relied heavily on the use of testing to establish both scientific and political credibility. It can be said that testing was experimentation for the social and human sciences. This suggests—once again—that the test holds substantial political weight. Tests produce numbers and other data, setting standards for scientific, industrial,
6 Testing Hearing
and educational policies. They are attempts to make the incomparable comparable. Testing calibrates, disciplines, and normalizes individuals, groups, populations, materials, and technologies. It is also a social practice that requires skilled testers and testees. From this perspective, the historical meaning of testing as witnessing gains new significance: testing enables individuals, groups, populations, materials, setups, and technologies to generate and practice a form of self-witnessing. Furthermore, the category of the “test subject” confers cohort status on a population with one or more testable features. As such, tests gave birth to the modern individual, who can also serve as a representative of the larger unit. By declaring that the history of testing remains undertheorized, we certainly do not wish to imply that there has been no work on testing at all. We draw on several existing disciplinary clusters as a background for our project, perhaps most prominently the biopolitics of testing. In the wake of Foucault’s writings on biopolitics and governmentality, a number of studies have focused on the history of testing populations in medical, psychological, and educational institutions.15 Tests that compare individuals to statistical norms, these argue, define therapeutic resources and “create social categories (of ‘learning-disabled’ children, ‘high-risk’ employees) in order to preserve existing social arrangements.”16 In the medical sciences, where tests were originally diagnostic tools, they have over the last two centuries been used for both prognosis and prevention. New “screening” tests evaluate seemingly healthy subjects to estimate risk, detect presymptomatic disorders, or establish carrier status for genetic conditions. Governments screen their citizens for a wide variety of abilities and impairments. Cancer testing has differentiated into the stages of screening, diagnosis, carrier testing, and monitoring. Tests are also “gatekeepers, controlling access to employment, hospitals, and schools, and allowing organizations to shape their clients as a projection of their own economic and administrative needs.”17 Clinical trials in medicine, another site where testing and experimenting converge, entail rigorous protocols to weigh the efficacy and safety of substances in relation to living bodies; the accurate interpretation of such tests also requires the existence of a non-tested randomized control group. These trials are often performed on a global scale, where test subjects in postcolonial settings assume bodily risks as delegates for health care consumers elsewhere.18 Additionally, several publications in science and technology studies (STS) have addressed the testing of technology since the 1980s (for a review of the STS literature on testing, see Trevor Pinch in this volume). Rather than emphasizing the power of tests to identify, classify, confirm or falsify, and regulate, these studies have largely equated testing systems with experimental
An Introduction 7
systems. STS scholars have surveyed “testing traditions.”19 They have revealed entanglements between test and context, coined the term “projective testing” to describe comparison tests of new technologies being introduced to the market, and examined the historical formation of test juries and test settings.20 In our view, this literature needs to be expanded through a more substantial and systematic engagement with the history of testing and its epistemic power. A closer look at the structures and agency of test systems, test sites, and transitions—from tests of objects to the testing of subjects and vice versa—provides new insight into large-scale technical control and prediction after industrialization. Reaching beyond the contrast between tests and experiments, we take tests to be not only disciplining, normalizing, or objectifying tools but also indices of complex epistemic dynamics and of disciplinary negotiations between the natural sciences, social sciences, humanities, arts, and industry.
Testing Hearing in and Beyond Audiometry The testing of hearing proves to be one of the richest sites for historical inquiry into the epistemic power of tests. The education of speaking and hearing citizens, the development of telecommunications tools, and the optimization of public spaces and multimedia arts encouraged the proliferation of hearing tests as well as the invention of new procedures such as screening. The modern scientific and popular understanding of hearing—and the practice of hearing itself, reinforced through standardization and training—codeveloped with such testing. Hearing no longer exists without audiometry. Most obviously related to this volume’s theme is thus the historical arc of audiology, a topic on which otologists have produced a handful of internal studies.21 Testing Hearing: The Making of Modern Aurality builds on—and complicates—this literature. We extend the canonical narrative of the long history of audiometry into the twentieth and twenty-first centuries, querying rather than accepting as transparent and progressive the role of testing itself. (Most of the tests mentioned in this brief survey, and throughout the volume, can be accessed on the database “Sound & Science: Digital Histories,” https://acoustics.mpiwg- berlin.mpg.de/sets/clusters/testing-hearing.) In Western medical history, diagnoses of hearing impairments go back at least to Hippocrates, although the earliest documented hearing tests only commence in the sixteenth century. Particularly prominent in the early period were tests of hearing through bone conduction, which made use of a rod held between the teeth and attached to a sound source.22 Through bone
8 Testing Hearing
conduction, it was possible to distinguish disorders of the eardrum from those of the nerves. Experiments on bone conduction hearing using tuning forks were later perfected and theorized by nineteenth-century physiologists, most notably Ernst Heinrich Weber in 1834 with his “Weber test” (using a fork placed on the forehead) and Heinrich Adolf Rinne in 1852 with the “Rinne test” (using a fork held beside the ear).23 After 1864, testers used electric tuning forks that produced a pure tone series.24 From the eighteenth century onward, hearing tests also addressed auditory acuity (the level of loudness of standard frequencies). This was initially done by placing ticking clocks at different distances from the listener; subsequent testers employed instruments including the “acoumeter” designed by Christian Heinrich Wolke, a language educator from northern Germany. Physicists and physiologists in this period also began to perform long series of experiments on the range of hearing (pitch levels) and the perception of minimal tone differences. These started with experiments using organ pipes and vibrating strings and were followed by experiments with loosely stretched strings; tuning forks, pipes, chirping crickets, and singing sparrows; sounding rods and the toothed wheel; a whistle adjusted by a micrometer screw; and, at the end of the nineteenth century, the first sets combining tuning forks, closed pipes, and Galton whistles to measure “continuous scales.”25 Forming a feedback loop of sorts, acoustic properties—the power and frequency of individual tones, or loudness and pitch, measured and controlled using tuning forks, whistles, sirens, and sounding rods— dominated scientific understanding of the function of hearing in the late nineteenth century. The same period also saw the development of instruments and methods to test more specific hearing abilities, such as the hearing of speech—that is, the perception of vowels and consonants. These included the Helmholtz tuning forks, von Kempelen speaking machines, and Appunn overtone apparatus.26 Speech sounds also began to be analyzed for their suitability as testing tools. Drawing on the results, European physicians in the nineteenth century tried a number of schemes to measure hearing for speech: comparing the recognition of vowels and consonants; arranging phonemes by pitch, with fricatives in the highest range; and employing whispered speech to minimize the unique fundamental frequency of a speaker’s voice and thus pinpoint the “invariables” of each phoneme. By the late twentieth century, the extent of the challenge would become apparent to engineers developing machine listening systems and automatic speech recognition technology.27 These tests of auditory acuity (loudness level), hearing range (pitch level), and speech perception grew out of an increasing interest among physiologists
An Introduction 9
and physicists in the capacities of human hearing, possible medical cures for disorders of the ear, and the development of measuring technologies and hearing aids. Over the course of the nineteenth century, studies of the perception of sound shifted from physiological and acoustic approaches to psychological frameworks— hearing tests reinforced existing epistemologies but also generated new ones. Pertinent topics of study included tests of “perfect pitch,” sensitivity to loudness or frequency, and the determination of cross-perceptual thresholds (e.g., the audiotactile interface). In turn, those tests relied on particular modes of defining and representing hearing ranges, from descriptive accounts to statistical graphics. If human hearing had long been conceptualized in terms of subjective qualities such as loudness or timbre, the establishment of psychophysics and related subfields in the nineteenth century spurred the quantification of this subjective experience, via the correlation of physical stimuli to sensation; moreover, hearing began to be approached as a continuum.28 That shift marks the beginning of the period covered in Testing Hearing: The Making of Modern Aurality, one accompanied by the scientific and humanistic design of increasingly specific hearing tests for the arts, education and communication, colonial and military applications, or sociopolitical and industrial endeavors. Key among these contexts was music (see Alexander Rehding, Emily Dolan, Viktoria Tkaczyk, Sebastian Klotz, and Benjamin Steege in this volume). In the eighteenth and nineteenth centuries, audiology embodied the aesthetics of bourgeois Western musical culture.29 For one thing, the principal tools for testing hearing perception came from music and musical instrument making: violin strings, tuning forks, metronomes, whistles, overtone apparatuses, and all sorts of musical devices.30 New forms of cooperation between instrument builders and physiologists facilitated the design of increasingly specific testing tools. Here, the tools themselves were as much test objects as was the human ear. Most significantly, hearing tests required not only test subjects but also skilled testers whose competence in making and using particular musical instruments determined the validity of the test results.31 Structurally, testing procedures were thus initially derived from both musical instrument construction and physiological experimental practice: the capacity of the human ear was determined experimentally, after which laboratory experiments were reconstructed and applied to larger numbers of subjects in the form of medical tests outside the laboratory. Musical technologies informed the early parameters by which hearing was defined, the boundaries of the ear’s capacities, and the paradigm of “normal hearing.” Tests of hearing capacities also served as a basis of comparison in cross- cultural anthropological studies, most prominently in colonial expeditions.
10 Testing Hearing
In their pursuit of applications for hearing tests, investigators in the late nineteenth century recognized that their findings depended on the choice of test instruments, test subjects, testers, test procedures, and test locations. As a result, hearing tests were now considered valid only if applied to larger groups, in batteries of tests, and in different localities so as to objectively collect statistics and determine averages. In the same period, the rise of national and cross-cultural statistical surveys changed the stakes of these tests, which were conducted at larger scales to differentiate and reveal the distribution of human hearing capacities, sorting broad cohorts and populations as well as individuals. To some extent, the history of hearing tests follows the seemingly ineluctable Foucauldian narrative of testing: testing establishes thresholds and statistical norms, which are incorporated into clinical or classroom tests and used to diagnose individuals. People found to be deviant from the norm are then treated or otherwise “disciplined.”32 Yet the ear has been more than merely the dependent variable in the history of hearing tests, the object being assessed. The ear and its technological delegates are also themselves test tools for the evaluation of objects, technologies, spaces, and large-scale systems. Building materials are evaluated based on models of human hearing; noise meters for managing public spaces incorporate standards based on loudness thresholds. Acoustic equipment is now constantly tested via calibration: in the laboratory, in the field, in the recording studio, and on the performance stage. The components of this equipment, if mass-produced, undergo quality testing at each stage of the manufacturing process (see Mara Mills and Roland Wittje in this volume). Such procedures are situated within the narrative of quality control more generally. In the twentieth century, the field of acoustics underwent a paradigm shift, with electrical tools and metaphors supplanting musical ones.33 Industrial research, largely conducted in telecommunications laboratories, reworked musical approaches with precision electronic instruments for pure-tone audiometry, substituted electrical metaphors for the anatomy and physiology of the ear, and asked new questions about hearing related to the ideals of transmission. The electroacoustic industry further gave rise to new tools and components such as the microphone, the vacuum tube amplifier, and the electrical filter, created and applied in an entangled context of science, music, and industry. At the same time, standards—measuring units, average thresholds, maximum allowable amplification—codified, embodied, or defined musical, scientific, and industrial values (see Jennifer Hsieh and Joeri Bruyninckx in this volume). Data from hearing tests are thus fed back not only to manage bodies but also to design objects, most obviously in the fields of music and
An Introduction 11
the electroacoustic industry, where hearing and listening preferences are used strategically in the creation of new buildings and environments, sounds and devices, and public policy. Hearing norms are built into all manner of audio apparatus, from telephones to stereo speakers (see Stefan Krebs and Jonathan Sterne in this volume). Beyond audiometry, this anthology maps proliferating modes of testing across the twentieth century. As new testing procedures have appeared in the sciences and social sciences—the double-blind, the questionnaire, the software simulation, the market factors poll—they have quickly been deployed to test hearing. In some cases, hearing tests led the way in the development of these procedures. Classroom hearing tests, as an example, were among the first large-scale “screens” for medical impairments. Musical listening surveys and aptitude tests evaluated taste and learning as well as capacity; these tested music itself as well as the ears of workers, soldiers, marine biologists, and many other professionals (see Lino Camprubí and Alexandra Hui in this volume). New forms of tests, combined with new statistical approaches and large arrays of testing devices, facilitated an expansion of testing to particular musical skills, provoking aesthetic debates in academia and in musical praxis. In turn, human hearing capacities were tested via the arts, most especially experimental music, avant-garde art, sound film, and sound design.
A Guide to Testing Hearing: The Making of Modern Aurality This volume amplifies the fundamental features, or problematics, of the double function of testing hearing—the human ear as periphery, measured by musical, industrial, or military modalities, and the ear as center, the human factor used in the testing of materials or the design of objects at times not even intended for humans. It argues that both hearing tests and testing with hearing had an enormous influence on the creation, maintenance, and destruction of auditory and broader epistemic cultures in the modern period. Testing Hearing: The Making of Modern Aurality grew out of two conferences convened in 2015 and 2016 at the Max Planck Institute for the History of Science in Berlin. The essays are grouped thematically rather than chronologically to highlight submerged continuities and overt conflicts. Each chapter brings in a substantial case study (building on expertise in musicology, history of science, media studies, STS, and anthropology) to illuminate one of the four themes of the volume: “Testing the Culturally Molded Ear,” “Designing Instruments, Calibrating Machines,” “Managing Sound, Assessing Space,” and
12 Testing Hearing
“Testing Beyond Human Auditory Perception.” Together, the four sections respond to the reconfiguration of test subjects in the modern period and underscore the role of hearing in that process of reconfiguration. Understanding testing as a cultural technique allows us to trace negotiations between the sciences, music, and industry, all of which have undergone rapid but not necessarily linear change in the period covered by the authors, 1870 to the present. Arranging the contributions outside of time also foregrounds place, revealing relationships between local and transnational developments. Together, the contributions trace the contours of the political power of testing hearing, its invisible (but very audible) shaping and reshaping of the soundscape of modernity. The two codas offer a synthetic overview of the contributions as well as extended discussions on the volume’s contribution to historical epistemology (Rheinberger) and to the history of the testing sciences (Pinch). If readers intend to skip and skim, we encourage them to start with these concluding pieces. Testing Hearing: The Making of Modern Aurality opens with human hearing as an object of testing. In the section “Sorting and Screening Human Hearers: Testing the Culturally Molded Ear,” contributors examine the ways in which hearing tests normalized, standardized, and obscured specific understandings of hearing and, by extension, reinforced the exclusion of specific communities. One such case is that of the resistance among some physicians and engineers to pure-tone audiometry as the principal measure of hearing function, examined by Mara Mills. In her study of researchers at American Telephone and Telegraph (AT&T) in the interwar period, Mills argues that their interest in establishing a measure for “useful” or “serviceable” hearing, which they identified with speech perception, was part of a broader quality control program for the expanding telephone system. The word and sentence lists they developed to test hearing not only became the standard for testing electroacoustic equipment but were also adopted by the medical field of speech audiometry. They were, however, partial with respect to hearing function, in both senses of the term: limited and biased. From a related perspective, Viktoria Tkaczyk’s examination of the new audiological testing procedures developed by the German tone psychologist and comparative musicologist Otto Abraham shows how musical aptitude tests originated in what was starting to be called the “nature versus nurture” debate among psychologists. Abraham’s tests marked a transition from testing hearing in the laboratory to statistical inquiries in everyday life, and from test design in the natural sciences to alternative testing practices in the emerging applied human sciences. Sebastian Klotz explores the history of cross-cultural hearing tests in the case of the Cambridge Anthropological Expedition to Torres Straits
An Introduction 13
(CAETS), a key endeavor in the formative years of British anthropology. The British experimenters’ hearing acuity tests (presumed to measure intelligence) proved inconclusive, perhaps for the first time fully exposing the problematic relationship between biological difference and cultural relativism that permeated the discipline. In “Designing Instruments, Calibrating Machines,” the contributors emphasize device testing and design, analyzing the constraints placed on musical aesthetic systems by the inertia of tradition, the limitations of human perception, and the goals of industry and materials science. Emily Dolan examines public hearing tests to distinguish between old and new violins, revealing a positive feedback loop whereby new instruments are judged against old ones—which have, in turn, been updated to meet new standards of playing. She proposes that the violin can thus be regarded as a “mendacious technology”: an instrument that lies about its own much-praised historicity. Similarly, the nineteenth-century sirens investigated by Alexander Rehding imitated the sirens of Greek mythology. The German tax collector and amateur astronomer Friedrich Wilhelm Opelt proposed an ambitious new music theory that unified major musical parameters—rhythm, pitch, and harmony—based on his redesign of Baron Charles Cagniard de la Tour’s siren. In Opelt’s hands, the sound-producing device became a test instrument that probed the limits of human hearing and is best described as “proto-digital.” A contrasting case is the “analog modeling” software for musicians discussed by Jonathan Sterne. Many music technology companies have sought to reproduce the sonic signature of analog audio devices—amplifiers, compressors, signal processors, instruments—in the software domain. This is a classic case of remediation, where a new media form attempts to represent an older one, as well as of commodity fetishism. Sterne shows that analog modeling in the digital domain is one of the latest chapters in the long history of hearing tests, for it is at the moment of the listening test that engineers and users attempt to resolve competing epistemologies of sound. The third theme addressed by the volume, “Managing Sound, Assessing Space,” complements the previous sections on music and human hearing by turning to materials science, institutions, and standards. The three contributions make it clear that the tests performed in materials development, technological innovation, and the machinations of the state are all simultaneously tests of sensory perceptual systems. Indeed, they test—and renegotiate— the very boundaries of the hearing sense according to particular value systems. The chapters illustrate the tensions that arise as tools and standards for testing hearing disseminate across fields (music, science, industry) as well as national contexts. Exemplary in this respect is the application of Western
14 Testing Hearing
technologies of noise control in Taiwan from the 1960s to the present—the period of political transition from an authoritarian regime to a democratic state—as examined by Jennifer Hsieh. Beginning as a state promise to provide the public with a quieter, more civilized way of life, noise control soon became dominated by an inflexible technorational approach to noise that did not correspond well to human-perceived loudness. As residents call civic hotlines and interrogate the definition of noise, we see how individual perceptual judgments can, through noise abatement testing, be transferred to the state. Stefan Krebs compares three interdependent German research groups that began designing dummy heads for concert hall acoustics, research on sound localization, and binaural hearing aids in 1967. All three groups struggled to find reliable and reproducible methods for measuring the subjective experience of spatial hearing. The resulting dummy-head systems (artificial hearing testees) failed to represent an average human listener, yet facilitated the design of new spatial audio technology for displaced listening scenarios such as radio and the music recording industry. A closer look at by-products of testing procedures reveals that, not unlike experiments, tests may develop a life of their own and lead to unexpected scientific insights, technological invention, social dynamics, and aesthetic change. Roland Wittje’s contribution examines the acoustic laboratory of Norges Tekniske Høgskole (NTH) in Trondheim, Norway, established in the late 1920s to test and certify construction materials. Such testing was mainly the domain of mechanical and civil engineering, but the acoustic testing of the materials was performed by physicists and electrical engineers, who built much of the measurement technology themselves to measure parameters of absorption, transmission, and reflection, and, ultimately if not entirely deliberately, produced new insights in oscillation research. The final section of the volume examines the testing of nonhuman, or posthuman, hearing. In “World as Testbed: Testing Beyond Human Auditory Perception,” the sounds are silent to human ears but exceedingly—existentially, even—meaningful. Joeri Bruyninckx asks how experimental psychologists in the 1930s and 1940s built on techniques for testing human hearing to determine the absolute upper frequencies of hearing for other mammal subjects. The ultrasonic domains of animal auditory perception (outside that of man) began to take on lives of their own. Propelled by various scientific, military, and commercial interests, fascination with the ultrasonic stirred imaginaries of sonic control. Lino Camprubí and Alexandra Hui’s contribution examines another soundscape exclusive of human auditory perception, at least until the exigencies of nuclear war fueled the development of new listening practices. In the oceanic soundscape, marine life was defined relative to Soviet vessels
An Introduction 15
and was approached through an epistemology of error that would later, through negotiation with marine bioacousticians, become scientific knowledge. The development, standardization, and classification of both sounds and hearing practices under the water’s surface thus provide a continuity between the secret search for errors and the open practice of marine biology. The case-study contributions to the volume conclude with the blast of nuclear weapons testing and a strange new sensory phenomenology. Benjamin Steege discusses anthropologist Günther Anders’s response to the devastating human and environmental impact of U.S. nuclear testing in the mid-1950s. Where the fact of worldhood was being put to the test, Anders provocatively called for a “test” of the capacity for imagination through desperate acts of aesthetic listening and sober exercises in “techniques of feeling,” an ethically naive and philosophically suggestive form of hearing as redemption. Together, the contributions to Testing Hearing: The Making of Modern Aurality illustrate the variety of tests as well as the multiplicity of meanings that surround testing. Readers will recognize the consistently central role of science (from audiology to acoustics and sound engineering, architecture and urban planning, bioacoustics and marine biology, and anthropology and critical theory), music (from instrument making to the creation and provocation of new or old aesthetics, listening skills, and habits), and industry (from fledgling form to international conglomerates engaged in materials, technology, or music manufacture) in testing hearing during the modern period. As this volume demonstrates, a single test can have multiple uses and be reapplied again and again in different contexts. To test the senses is to traffic between the physical, the physiological, and the psychological; testing hearing thus attempts to make the untestable testable and the incommensurable commensurable. A chronological reading of the stories told in this volume (as indicated in the previous section) would reveal that pivot points in the larger narrative of testing hearing occurred with the introduction of statistical thinking, recording technology, national standards, safety and quality regulations, and electrical control of sound. Some objects, anxieties, and, of course, sounds reappear across the contributions, often in unexpected places. Readers should keep their ears open for Dalton’s whistle, Seashore’s test of musical skill, the whine of the siren, the creatures of the sea. Concerns about the representation and communication of sound on paper or through other sounds also recur. The fidelity or authenticity of sound is, as always, a nagging problem for those creating sound objects and technologies. Also woven through all of the contributions is a thematic thread of the construction of norms. Forms of hearing were normalized and, at times, standardized. Tests too were
16 Testing Hearing
normalized and then applied to new and different epistemic problems, often with curious results. Hearing tests, we argue, have not only redefined modern hearing but have also altered the very meaning of “the test” and the academic fields from which they emerged. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/introduction-hui-mills-tkaczyk.
Notes 1. Lori Zimmer, “Experience Times Square.” 2. A video rendition of the test pattern [times square] can be found at Streaming Museum, http://streamingmuseum.org/ryoji-ikeda-test-pattern-times-square/. Between 2:23 and 2:25, two quick flashes of audience members wearing wireless headphones can be seen. 3. https://www.discogs.com/Ryoji-Ikeda-Test-Pattern/release/1275360. 4. Of the many works on aurality, Veit Erlmann’s Reason and Resonance and Ana María Ochoa Gautier’s Aurality cover a similar time span to our own. 5. Oxford English Dictionary, s.v. “test (n. 1),” accessed October 7, 2019, https://www.oed. com/view/Entry/199677?result=1&rskey=7nKwTY&. 6. See E. Bright Wilson, Introduction to Scientific Research. 7. Steven Shapin, “Pump and Circumstance,” 481. 8. W. Ross Ashby, Introduction to Cybernetics, ch. 6. 9. Bernhard Siegert, Cultural Techniques. 10. See Cyrus C. M. Mody and Michael Lynch, “Test Objects.” 11. Penelope Gouk, Music, Science and Natural Magic; Benjamin Wardhaugh, Music, Experiment and Mathematics; Alexander Rehding, Hugo Riemann; Myles W. Jackson, Harmonious Triads; Benjamin Steege, Helmholtz; James Q. Davis and Ellen Lockhart, Sound Knowledge. 12. Simon Schaffer, “Astronomers Mark Time.” 13. Lorraine Daston and Peter Galison, Objectivity; Flavia Padovani, Alan Richardson, and Jonathan Y. Tsou, Objectivity in Science. 14. Peter Galison, Image and Logic; Timothy Lenoir, Inscribing Science. 15. Stephen Jay Gould, Mismeasure of Man; Theodor Porter, Trust in Numbers; Rayna Rapp, Testing Women. 16. Dorothy Nelkin and Laurence R. Tancredi, Dangerous Diagnostics, 17. 17. Ibid. 18. See Adriana Petryna, When Experiments Travel. 19. Edward W. Constant, Origins of the Turbojet Revolution. 20. Donald McKenzie, “From Kwajalein to Armageddon?”; Trevor Pinch, “ ‘Testing—One, Two, Three . . . Testing!’ ”; Eefje Cleophas and Karin Bijsterveld, “Selling Sound.” 21. The earliest is a two-volume study by the Austrian otologist Adam Politzer, published in 1907–1913, followed by Karl Ludolf Schaefer’s 1914 article on investigating the acoustic
An Introduction 17 functions of the ear and Harald Feldmann’s 1959 history of audiology, translated in 1970. Politzer, Geschichte der Ohrenheilkunde; Schaefer, “Untersuchungsmethodik”; Feldmann, History of Audiology. 22. Girolamo Cardano, De subtilitate, Book 13, 387. The effect was visualized later in the frontispiece of John Bulwer’s Philocophus of 1648, https://doyle.com/sites/default/files/images/ lots/511/1137511.jpg. 23. Ernst Heinrich Weber, Annotationes; Adolf Rinne, “Beiträge.” 24. August Lucae, “Untersuchungen.” 25. See, respectively, Joseph Sauveur, Principes d’acoustique; Ernst F. F. Chladni, Die Akustik; William Hyde Wollaston, “On Sounds Inaudible”; Félix Savart, “Über die Empfindlichkeit des Gehörorgans”; Francis Galton, Inquiries, 38– 40, 375– 78; Friedrich Bezold, “Demonstration der kontinuierlichen Tonreihe.” 26. All these are illustrated in Oskar Wolf, Sprache und Ohr. 27. Richard Lyon, Human and Machine Hearing; Xiaochang Li, “Divination Engines.” 28. Myles W. Jackson, “From Scientific Instruments.” 29. Alexandra Hui, Psychophysical Ear. 30. Laura Otis, Networking, 11. 31. On the competence of testers, see MacKenzie, “From Kwajalein to Armageddon?,” 413–14. 32. Nelkin and Tancredi, Dangerous Diagnostics; Graeme Gooday and Karen Sayer, Managing the Experience of Hearing Loss; Jaipreet Virdi, Hearing Happiness; Jacques Vest, “Malingering Ear.” 33. Roland Wittje, Age of Electroacoustics.
References Ashby, W. Ross. An Introduction to Cybernetics. New York: J. Wiley, 1956. Bezold, Friedrich. “Demonstration der kontinuierlichen Tonreihe in ihrer neuen von Dr. Edelmann verbesserten Form.” Zeitschrift für Ohrenheilkunde 25, nos. 1–2 (1894): 66–67. Cardano, Girolamo. Hieronymi Cardani mediolanensis medici de subtilitate libri XXI. Basel: Ludouicum Lucium, 1554. Chladni, Ernst Florens Friedrich. Die Akustik. Leipzig: Breitkopf und Härtel, 1802. Cleophas, Eefje, and Karin Bijsterveld. “Selling Sound: Testing, Designing, and Marketing Sound in the European Car Industry.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 102–24. Oxford: Oxford University Press, 2012. Constant, Edward W. The Origins of the Turbojet Revolution. Baltimore, MD: Johns Hopkins University Press, 1980. Daston, Lorraine, and Peter Galison. Objectivity. New York: Zone Books, 2007. Davis, James Q., and Ellen Lockhart, eds. Sound Knowledge: Music and Science in London, 1789–1851. Chicago: University of Chicago Press, 2015. Erlmann, Veit. Reason and Resonance: A History of Modern Aurality. New York: Zone Books, 2014. Feldmann, Harald. A History of Audiology: A Comprehensive Report and Bibliography from the Earliest Beginnings to the Present. Chicago: Beltone Institute for Hearing Research, 1970. Galison, Peter. Image and Logic: A Material Culture of Microphysics. Chicago: University of Chicago Press, 1997. Galton, Francis. Inquiries into Human Faculty and Its Development. London: MacMillan, 1883.
18 Testing Hearing Gooday, Graeme, and Karen Sayer. Managing the Experience of Hearing Loss in Britain, 1830– 1930. London: Palgrave Macmillan, 2017. Gouk, Penelope. Music, Science and Natural Magic in Seventeenth-Century England. New Haven, CT: Yale University Press, 1999. Gould, Stephen Jay. The Mismeasure of Man. New York: W. W. Norton, 1996. Hui, Alexandra. The Psychophysical Ear: Musical Experiments, Experimental Sounds, 1840– 1910. Cambridge, MA: MIT Press, 2013. Jackson, Myles W. “From Scientific Instruments to Musical Instruments: The Tuning Fork, the Metronome, and the Siren.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 201–23. Oxford: Oxford University Press, 2011. Jackson, Myles W. Harmonious Triads: Physicists, Musicians, and Instrument Makers in Nineteenth-Century Germany. Cambridge, MA: MIT Press, 2006. Lenoir, Timothy, ed. Inscribing Science: Scientific Texts and the Materiality of Communication. Stanford, CA: Stanford University Press, 1998. Li, Xiaochang. “Divination Engines: A Media History of Text Prediction.” PhD diss., New York University, 2017. Lucae, August. “Untersuchungen über die sogenannten ‘Knochenleitungen’ und deren Verhältnis zur Schallfortpflanzung durch die Luft, im gesunden und kranken Zustande.” European Archives of Oto-Rhino-Laryngology 1, no. 1 (1864): 303–17. Lyon, Richard. Human and Machine Hearing: Extracting Meaning from Sound. Cambridge: Cambridge University Press, 2017. MacKenzie, Donald. “From Kwajalein to Armageddon? Testing and the Social Construction of Missile Accuracy.” In The Uses of Experiment: Studies in the Natural Sciences, edited by David Gooding, Trevor Pinch, and Simon Schaffer, 409–35. Cambridge: Cambridge University Press, 1989. Mody, Cyrus C. M., and Michael Lynch. “Test Objects and Other Epistemic Things: A History of a Nanoscale Object.” British Journal for the History of Science 43, no. 3 (2010): 423–58. Nelkin, Dorothy, and Laurence R. Tancredi. Dangerous Diagnostics: The Social Power of Biological Information. Chicago: University of Chicago Press, 1994. Ochoa Gautier, Ana María. Aurality: Listening and Knowledge in Nineteenth-Century Colombia. Durham, NC: Duke University Press, 2014. Otis, Laura. Networking: Communicating with Bodies and Machines in the Nineteenth Century. Ann Arbor: University of Michigan Press, 2001. Padovani, Flavia, Alan Richardson, and Jonathan Y. Tsou, eds. Objectivity in Science: New Perspectives from Science and Technology. Dordrecht: Springer, 2015. Petryna, Adriana. When Experiments Travel: Clinical Trials and the Global Search for Human Subjects. Princeton, NJ: Princeton University Press, 2009. Pinch, Trevor. “‘Testing—One, Two, Three . . . Testing!’: Toward a Sociology of Testing.” Science, Technology, & Human Values 18, no. 1 (1993): 25–41. Politzer, Adam. Geschichte der Ohrenheilkunde. 2 vols. Stuttgart: Ferdinand Enke, 1907–1913. Porter, Theodore M. Trust in Numbers. Princeton, NJ: Princeton University Press, 1996. Rapp, Rayna. Testing Women, Testing the Fetus: The Social Impact of Amniocentesis in America. New York: Routledge, 1999. Rehding, Alexander. Hugo Riemann and the Birth of Modern Musical Thought. Cambridge: Cambridge University Press, 2003. Rinne, Adolf. “Beiträge zur Physiologie des menschlichen Ohres.” Vierteljahrschrift für praktische Heilkunde 45, no. 1 (1855): 71–123, and 46, no. 2 (1855): 45–72. Sauveur, Joseph. Principes d’acoustique et de musique, ou Système général des intervalles des sons. Paris: Académie Royale des Sciences Paris, 1701. Savart, Félix. “Über die Empfindlichkeit des Gehörorgans.” Annalen der Physik 96, no. 10 (1830): 290–304.
An Introduction 19 Schaefer, Karl Ludolf. “Untersuchungsmethodik der akustischen Funktionen des Ohres.” In Handbuch der physiologischen Methodik, edited by Robert Tigerstedt, 3/1:204–396. Leipzig: Hirzel, 1914. Schaffer, Simon. “Astronomers Mark Time: Discipline and the Personal Equation.” Science in Context 2, no. 1 (1988): 115–45. Shapin, Steven. “Pump and Circumstance: Robert Boyle’s Literary Technology.” Social Studies of Science 14, no. 4 (1984): 481–520. Siegert, Bernhard. Cultural Techniques: Grids, Filters, Doors, and Other Articulations of the Real. Translated by Geoffrey Winthrop-Young. New York: Fordham University Press, 2015. Steege, Benjamin. Helmholtz and the Modern Listener. Cambridge: Cambridge University Press, 2012. Vest, Jacques. “The Malingering Ear: Audiometric Surveillance in the Early Twentieth Century United States.” Paper presented at the History of Science Society Annual Meeting, July 25, 2019. Virdi, Jaipreet. Hearing Happiness: Fakes, Frauds, and Fads in Deafness Cures. Chicago: University of Chicago Press, 2020. Wardhaugh, Benjamin. Music, Experiment and Mathematics in England, 1653–1705. Farnham, UK: Ashgate, 2005. Weber, Ernst Heinrich. Annotationes anatomicae et physiologicae, programmata collecta. Leipzig: Koehler, 1834. Wilson, E. Bright. An Introduction to Scientific Research. New York: McGraw-Hill, 1952. Wittje, Roland. The Age of Electroacoustics: Transforming Science and Sound. Cambridge, MA: MIT Press, 2016. Wolf, Oskar. Sprache und Ohr: Akustisch-physiologische und pathologische Studien. Braunschweig: Friedrich Vieweg und Sohn, 1871. Wollaston, William Hyde. “On Sounds Inaudible by Certain Ears.” Philosophical Transactions of the Royal Society of London 110 (1820): 306–14. Zimmer, Lori. “Experience Times Square in Solitude with Ryoji Ikeda’s test pattern [times square] with SOUND.” October 14, 2014. http://art-nerd.com/newyork/experience-times- square-in-solitude-with-ryoji-ikedas-test-pattern-times-square-with-sound/.
1 Testing Hearing with Speech Mara Mills
Introduction: From Qualia to Quality Control In 1890, aurist George Fiske gave a lecture to the Chicago Medical Society titled “The Phonograph in Testing Hearing.” The first phonograph parlors had recently opened in the United States, Edison’s tinfoil recordings having been replaced by sturdier wax cylinders. In the absence of other means for amplification, horns and hearing tubes condensed sound waves and channeled them to listeners’ ears. Edison had predicted, in an 1878 article on “The Phonograph and Its Future,” that the device would be used for letter writing, dictation, and recorded music; talking clocks, dolls, and books; and advertising, record- keeping, and the teaching of elocution.1 In this moment of wild interpretive flexibility, the phonograph also allowed the voice to become a tool for testing hearing—a reproducible stimulus to provoke hearing behavior and gauge the ear’s sensitivity. Like many of his contemporaries, Fiske felt that hearing tests were essential for the treatment of ear disease, allowing those in the new medical subspecialty of otology to determine “1 exactly how much the hearing is impaired as compared with normal hearing, and 2, whether there is improvement or loss, however slight, between two dates.”2 Identical printed copies of Dutch ophthalmologist Hermann Snellen’s eye charts had recently been disseminated into clinics across the United States, and Fiske’s patients demanded a similar yardstick for other sensory functions. Fiske wrote: The patient places no such confidence in the physician’s voice as in the test types of the oculist, hanging at the other end of the room, with the normal visual distance clearly marked upon them. Even with confidence on the patient’s side and perfect conscientiousness on the physician’s, it is impossible to reproduce with absolute exactness the words whispered or spoken—say one year after the first examination, and a slight mistake here might lead to a great mistake in prognosis and in treatment.3
Mara Mills, Testing Hearing with Speech In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0002.
24 Sorting and Screening Human Hearers
New watches of standardized loudness, and tuning forks with graduated frequencies, had become available to otologists a few decades earlier (see introduction in this volume). Yet Fiske believed the “definition of normal hearing” could only be established through a combination of tests with musical tones, watch ticks, and spoken words. Everyday hearing—one’s ability to perceive the timbre of speech sounds and speech as a sequence in time—should be investigated along with the minimum loudness thresholds for particular pure tone frequencies. Human speakers seemed intractable, however, to the kind of standardization required of a testing tool. As Fiske lamented: [The voice] is by far the most important of our tests, not only because it is more necessary in daily life to hear the human voice than any other sounds, but also because it is heard with more constancy through life, with less variation on account of age. But this, our most important means of testing, is also by far the most uncertain. Not only is it extremely difficult to reproduce exactly the volume and quality of a given voice, but the voices of different examiners vary so greatly that the patient’s statement that he heard two years ago Dr. So and So’s voice at such and such a distance would be in very many cases of no help to the present examiner.4
Fiske’s solution was to record his own speech on a phonograph cylinder and measure the distance at which a patient could just make out the words. (He did not explain in his lecture what those test words should be.) He then labeled the cylinder with the date, patient name, and hearing distance and stored it with the patient’s other records—to be replayed on the same machine, cranked at the same speed (120 rpm), during the person’s next appointment, as a means of assessing the course of hearing loss. The few extant histories of hearing measurement place the phonograph at the origin of modern speech audiometry, a field that today encompasses a wide variety of apparatuses and inputs.5 At the 1944 meeting of the American Otological Society, Philadelphia physician Douglas MacFarlan—who had a keen interest in engineering—suggested that the clinical use of the phonograph was initiated only in 1904 by William Sohier Bryant, an aural surgeon who practiced in Boston and New York. As the 1890 example of Fiske suggests, however, phonographic speech audiometry was a longer-standing practice than MacFarlan thought—a response to the clinical imperative to quantify “useful” hearing. Harald Feldmann, a leading German audiologist who completed his dissertation at the Free University Berlin in 1955, published an authoritative history of the field in 1960 (translated in 1970) in which he repeated MacFarlan’s erroneous origin story for “modern” audiometry. German
Testing Hearing with Speech 25
audiologists, Feldmann added, were more suspicious than their American counterparts of the quality of phonographic reproduction. Internationally, otologists agreed that tones were the easiest stimuli to standardize. By the 1920s, relatively “pure tones” with controlled parameters of amplitude and frequency could be generated, owing to the development of the wave filter and the vacuum tube in the field of electroacoustics (specifically, radio and telephone engineering).6 In 1922, researchers at the Western Electric (WE) and American Telephone and Telegraph (AT&T) branches of the Bell System produced an electronic audiometer, along with “area of sensation” charts that portrayed average curves of maximum and minimum audibility—based on tests with a few dozen subjects—for a series of tones selected from the piano scale.7 They later redesigned what they called the audiogram to document “the percent of normal hearing,” wherein the quietest tones an individual could detect were plotted in comparison to a “reference zero.”8 What sort of hearing function did this audiogram manifest? The test material—a set of individual tones—lent itself to precise electronic control, but it did not divulge such things as speech capacity, nor necessarily capture discrete “notches” due to injury, nor adequately account for losses due to the masking effects of tinnitus.9 Against the backdrop of industrialization, workers’ compensation, rehabilitation, public schooling, and new communication technologies in the workplace, otologists continued to pursue measures of “useful hearing” after the introduction of electronic audiometers.10 Pre-employment physical examinations, school tracking, and injury remuneration depended on the measurement of hearing capacity for speech, rather than music or sounds in the range of higher tones. Researchers thus endeavored to forge speech—from nonsense syllables to spondees to sentences—into an objective measuring tool. Audiologist Ira Hirsh argued in 1952 that speech audiometry required a shift “from the concept of detectability to the concept of intelligibility.”11 Beyond the control of amplitude and the repeatability of recorded sounds, “intelligibility” entailed optimizing speech itself. Across the twentieth century, electroacoustic engineers and audiologists alike composed test lists that included all the speech sounds in a particular language—arranged as syllables or words that were not unduly influenced by locale or by memory—for a listener to repeat aloud in a testing situation. This chapter takes a closer look at the second phase of phonographic speech audiometry, from its first mass application via electrical recording and playback machines to the ensuing crisis in test materials for measuring hearing- for-speech. In his History of Audiology, Feldmann placed WE at the fulcrum
26 Sorting and Screening Human Hearers
of the audiometric shift from the sheer fact of phonographic reproducibility to the content of sound recordings, noting that “seemingly trifling changes in test conditions required a complete alteration of the test material”: The large-scale screening tests conducted in the United States with the aid of the Western Electric 4A audiometer and other similar instruments had demonstrated the shortcomings of numbers when used as test words. Quite often, such test material had left high-frequency hearing losses undetected. The task boiled down to finding more suitable test words.12
WE was the manufacturing arm of the Bell System, and as Feldmann explains, the WE 4A phonograph audiometer (Figure 1.1) was employed in the first “large-scale screening tests” of hearing, on hundreds of thousands of American schoolchildren, in the 1920s.13 It was with these hearing tests, Emily Thompson argues in The Soundscape of Modernity, that “the experience of being tested . . . became a new element of aural culture for increasing numbers of people” in the United States.14 Similarly, in their contribution on audiometry in Instruments of Science: An Historical Encyclopedia, Stuart Blume and Barbara Regeer suggest that phonographic speech audiometry “reached its apogee” in 1927 with the mass marketing and application of the 4A.15
Figure 1.1 “Western Electric 4C Audiometer,” Western Electric Sales Bulletin T-1646. Western Electric Sales Bulletin vol. 14, p. 16, B04 04. Courtesy AT&T Archives and History Center.
Testing Hearing with Speech 27
Nonetheless, the validity of the 4A was undermined as a result of its own extensive application, which revealed the difficulty of standardizing dispersed testing environments and the limitations of using numbers to test the hearing of speech. Extending Feldmann’s claims, this chapter will demonstrate that audiologists in the post–World War II period looked to the speech and hearing research group at Bell Telephone Laboratories— which descended from the WE Laboratories—for a new guide to speech audiometry, once it became clear that the popular WE 4A phonographic audiometer failed to adequately detect hearing loss and to measure speech hearing. In a separate line of research, Bell engineers employed the ear as examiner, developing “articulation tests” to assess the clarity of carbon telephone transmitters as part of a broader quality testing program; during World War II, those tests were adopted by scientists at the Harvard Psycho-Acoustic Laboratory (PAL) for measuring the intelligibility of military communications equipment in combat noise and estimating the impact of hearing injuries on speech perception. Quality testing was a key aspect of quality control, an industrial production routine that commenced in the telephone industry at the outset of the twentieth century. The testing of mass-produced telephones with spoken words emerged concurrently with the mass testing of American ears by WE tools and engineers; ears and electroacoustic equipment were simultaneously enrolled into the testing and optimization paradigms. I suggest that quality testing is one of the underexplored origins of medical screening, with which it shares an epistemology: establishing limits on variability, eliminating sources of extreme variance, and either “junking” or rehabilitating those objects or people who fall outside the limits.16 In terms of consequences, the practice of testing “presymptomatic” individuals for hearing loss ramified extensively in the categorization and education of “deaf ” and “hard of hearing” children, and in the hiring or exclusion of adults in the workplace. The quality control paradigm continues to encompass machines and humans today: similar “phonetically balanced” word and sentence lists are used in articulation testing for media technology and speech audiometry, naturalizing the ambiguous concept of intelligibility as a quantifiable variable and reinforcing biases about what counts as average conversational spoken English.
Speech Audiometry Before the Phonograph Feldmann insisted that speech audiometry is in fact as old as otology: “Any discussion between the physician and his patient may be considered a
28 Sorting and Screening Human Hearers
hearing test even if not constructed explicitly for that purpose.” But systematic attempts at speech audiometry, he noted, emerged jointly with tonal audiometry at the opening of the nineteenth century, abetted by the sciences of acoustics and phonetics and the availability of standardized sound-measuring tools.17 Deaf schools in Europe, and later the United States, were a source of demand for diagnostic tests. Physicians and deaf educators such as Jean Marc Gaspard Itard developed their own speech hearing tests, often based on the distance at which particular words could be heard. Students were classified according to their ability to understand speech sounds, for instance, as “semi- deaf ” if they could detect vowels but not higher-pitched consonants. Such classifications were particularly important in “oralist” school settings, which privileged speech communication over sign language. Experimental psychologists, especially those in the subfield of psychophysics, had also begun training their instruments on the human senses in the nineteenth century, posing challenges to philosophy and quantification alike. They attempted to treat subjective experience in an objective manner, ranging perceptions and feelings upon a number line. In an attack on psychophysicist Gustav Fechner in 1890, the philosophically minded psychologist William James insisted that physical sensations, and the inward experiences they engendered, could not be cleanly divided into attributes, each then scaled. “Our feeling of pink,” he wrote, “is surely not a portion of our feeling of scarlet; nor does the light of an electric arc seem to contain that of a tallow-candle in itself.”18 Yet psychophysicists insisted that the mind could be bypassed or inferred (in today’s parlance, “black-boxed”). Experience and feelings per se might elude quantification, but stimuli could be controlled, and then related to observable behaviors or measurable responses (a pushed button, a repeated word). The measurement of hearing, however, was fraught with unique uncertainties. Injuries to distinct parts of the auditory system—the middle ear, the auditory nerve—had different effects on hearing. The key attributes of sound had first to be enumerated before they could be turned into probes for “sounding out” the ear. It was one thing to calibrate pitch and loudness, but quite another to take on timbre and intelligibility, the very definitions of which were subject to immense debate.19 If to measure means to “assign numerals to events,” Ira Hirsh could still query as late as 1952, while his field of audiology was professionalizing in the United States, “what are the observable events in hearing?”20 The science of phonetics, also formalized in the nineteenth century, brought about new classifications of speech sounds. The abundant insights of physicists and physiologists such as Hermann von Helmholtz and Cornelius Donders regarding the frequency composition and other physical properties
Testing Hearing with Speech 29
of speech began to influence the design of hearing tests. Many physicians and deaf educators devised speech tests in analogy with tuning forks, designating certain phonemes to represent frequencies in a series, with fricatives in the highest range. Other physicians began to utilize “whisper tests” (whispering a word to a subject from an arm’s length away) to minimize the unique fundamental frequency of the speaker’s voice and thus control for differences in pitch between testers. Speech began to be analyzed for its suitability as a testing tool: How accurately could pitch and loudness be correlated to physical properties and measured, controlled, or replicated? How should the testing environment be designed to minimize noise and reverberation? Could words and sentences be used as test material for an accurate gauge of “natural speech” perception? If so, should widely recognized words, such as numbers, be selected to avoid mishearing and confusion? Or should monosyllabic and nonsense words be employed to control for guessing? Or simply phonemes, whose parameters were more easily quantified? How could these bits and pieces be added back together to measure the ability of a person to genuinely hear speech? The more that was learned about speech, the more remote the goal of bending it to the requirements of objectivity seemed to become. Each isolated detail became another factor to control; moreover, it was unclear which elements of speech mattered most to hearing. Despite many ingenious attempts to monitor volume or pitch—such as Dr. August Lucae’s phonometer (Figure 1.2)—the human voice continued to be less reproducible and more difficult to gauge than tuning forks.
Screening Tests Although internal histories point (incorrectly) to William Bryant as the founder of “modern” speech audiometry, as the first to employ a phonographic device in the clinical setting, the field owes less to the phonograph per se than to controlled vacuum tube amplification, low-cost playback machines, the improved clarity of electrical disc recording, and the widespread demand for hearing tests in Europe and the United States that was precipitated by compulsory education and workers’ compensation. A number of European physicians—Leopold Lichtwitz, Marie-Ernest Gellé, Dagobert Schwabach, Hermann Gutzmann—experimented with the phonograph soon after its invention but ultimately decided that the quality of the reproduced voice was too poor, even if it promised constant pitch.21 Bryant himself adapted an Edison machine from the Standard Phonograph Company,
30 Sorting and Screening Human Hearers
Figure 1.2 August Lucae’s phonometer, which attempted to measure expiration as an estimate of intensity. A. Lucae, “Über das Phonometer,” Archiv für Physiologie (1878): 588.
adding stethoscope listening tubes with a valve to control air flow and regulate volume (if imprecisely), thus eliminating the requirement of distance retained by Fiske’s phonograph audiometer. Bryant demonstrated his “phonographic acoumeter” (Figure 1.3) at the 1904 meeting of the American Otological Society but failed to arouse much interest. He offered cylinder reproductions of “monosyllabic words in constant use” from his master records, advocating a “universal standard” in speech audiometry. Bryant’s acoumeter additionally included a “malingerer’s valve,” which allowed sound to be switched quickly from one tube to another, to “distract” those who tried to fake deafness in one ear. This tiny component indicates the outsized anxiety provoked by the establishment of the first statewide workers’ compensation laws in the United States in 1902.22 Deaf schools—which had always included students who varied with regard to hearing and communication—began to adopt gramophone-style audiometers to assist with tracking and assessment in the 1920s, once vacuum tube amplification allowed more precise volume control. William Bristol, an inventor known for his Audiophone loudspeaker system and Bristolphone sound film system, worked with teachers at the Lexington School for the Deaf in Manhattan to adapt one of his electric playback machines for speech audiometry. The school created a test list of common children’s words— animals, numbers, days of the week—to minimize “the factors of education
Testing Hearing with Speech 31
Figure 1.3 Bryant’s phonographic acoumeter (in soundproof box for restriction of volume to tubes). The main valve is C; the malingerer’s valve is rotating unit D. William Sohier Bryant, “Phonographic Acoumeter,” Archives of Otology 33 (1904): 440.
and intelligence.”23 MacFarlan and other otolaryngologists made use of the recordings produced by the Lexington School, typifying the pedagogical and technological roots of “medicalization” in the history of audiometry. MacFarlan later composed his own list of test words for children, using monosyllables to minimize guessing, with accompanying sketches to which young children could point if they weren’t able to write down the recorded words that they heard.24 In 1925, WE released a speech “phono-audiometer” for group testing, which consisted of a modified gramophone set connected to forty to eighty telephone receivers. The capacity to simultaneously test a large group—for instance, a classroom of students—afforded mass testing, which would ultimately be conducted on a nationwide scale by WE employees in collaboration with the American Association for the Hard of Hearing and the U.S. Public Health Service. Members of the New York League for the Hard of Hearing, founded in 1909, had begun petitioning the Red Cross for hearing loss to be considered a military disability in 1918—a classification that required more precise hearing tests than yet existed. The social workers and “deafened” activists who launched the League quickly recruited prominent physicians to their membership and expanded from New York City to become a national organization. One of their members, Dr. Edward Prince Fowler, returned from
32 Sorting and Screening Human Hearers
service in the army with an interest in rehabilitation; he asked his neighbor, Alexander Nicholson of WE, for assistance with the construction of an electronic audiometer, using up-to-date wave filters and vacuum tubes.25 Fowler and Nicholson built a basic speech audiometer as their prototype, recording test sentences on a disc; however, they soon switched to tonal audiometry. In 1922, WE released the resulting “1A,” the first commercial electronic audiometer in the United States, followed quickly by the 2A, a portable version that tested human hearing at various volumes for tones from 64 to 8,192 Hz. Although WE introduced their “audiogram” along with the 1A, a formal reference zero accepted by the American Standards Association would only be produced with data from the 1936 U.S. Census, which was collected using WE tonal audiometers and the help of AT&T staff. Before that time, the frequencies chosen for testing—and the so-called normal curve of the audiogram— fluctuated considerably.26 Walter Shewhart, the “father of quality control” (also known as the father of “modern” or “statistical” quality control), was hired into WE’s telephone quality research program in 1918, specifically to work on improving the noisy carbon transmitter. In 1925, his position was transferred to the newly founded Bell Telephone Laboratories. He introduced statistical methods to communication engineering at AT&T, including the normal curve, which he imported from biometrician—and eugenicist—Karl Pearson’s work on race and human skull size.27 Although Shewhart is a neglected figure in the history of technology, Friedrich-Wilhelm Hagemeyer argues that his work prepared the way for Claude Shannon’s statistical definition of information as entropic or unpredictable.28 Shewhart’s principle, the “reduction of variation”—still a central tenet of quality control today—entailed identifying the acceptable quality range for a product; inspecting materials, tools, and processes at each stage of production; “junking or reclaiming” any items found “defective” (i.e., having a “quality characteristic falling outside [the] specified range”); and finally diagnosing “the unknown or chance causes of defectives and [trying] to remove them” (Figure 1.4).29 In her study of American design in the 1930s, Christina Cogdell traces links between streamlined aesthetics and the eugenic ideals of progress, fitness, and efficiency; electrical engineering and mass manufacture were not immune to similar cross-pollination.30 Yet if quality control drew on Pearson to define “defects” statistically, the field placed as great an emphasis on repair and reclaiming as “elimination.” At the same time, mass production demanded extraordinarily narrow standards for “acceptable variation.” The inspection stages took place through what came to be called screening, where either a mass or a sample was measured or tested for a particular “quality characteristic.” These tests could be chemical, statistical,
RU SILBBER CO K LINTTON WOEN CO OL A IRO L LE N AD CO ZINPPE NICC R AL KEL GOUMIN SILLD UM PL VER MI ATIN COCA UM SH PAL TI ELLA C ASN PH AL TU M
Testing Hearing with Speech 33
RAW MATERIAL
INSPECTION TO REDUCE COST OF PRODUCTION
PARTS
DEFECTIVE MATERIAL
PARTS
DEFECTIVE PARTS
PARTIAL ASSEMBLIES
100% INSPECTION TO PROTECT CONSUMER
FINISHED PRODUCT GOOD
DEFECTIVE PARTS
DEFECTIVE ASSEMBLIES
UNITS
TELEPHONE AND LINE
DEFECTIVE UNITS
Figure 1.4 Stages of quality control: Screening parts and products to remove “defects” and improve production processes. W. A. Shewhart, Economic Control of Quality of Manufactured Product (New York: D. Van Nostrand Company, 1931), 28. Courtesy AT&T Archives and History Center.
or comparative, depending on what quality was being assessed. The term “screening” was simultaneously adopted for hearing tests and quality tests at AT&T at some point around 1930, where it referred to preliminary, rapidly applied, rough, and comparatively inexpensive tests, applied at broad scale or to a selected subpopulation, “to sift or separate; to exclude or eliminate.”31 For those groups of people or items deemed defective after a screen, more detailed information about causes and consequences was provided by follow- up diagnostic tests. The word “screening” was not present in the memos and publications on either telephone quality or audiometry in the 1920s; it suddenly appears in both in the 1930s. In Shewhart’s own writing from the 1930s, the term is italicized,
34 Sorting and Screening Human Hearers
indicating its novelty. Existing histories of medical screening locate the practice in the 1940s, but it clearly predates that decade.32 Proto-hearing screens took place in localized settings in the late nineteenth century, for instance hearing tests of German and U.S. military personnel, but these were discrete rather than mass studies and did not involve the creation of a general baseline for hearing nor the element of repair and prevention. In today’s language, employers and the military had long “screened out” candidates, but screening tests to simultaneously segregate and rehabilitate were a later iteration.33 WE engineers returned to speech audiometry in 1925, with the production of the 4A phono-audiometer. As test material, they supplied a series of numbers, recorded by both male and female voices, played back to students in graduated steps starting at 30 dB. Older students were tested with three-digit numbers, younger ones with two digits. The results did not yield an audiogram; instead, the percentage of words correctly recognized was calculated for each intensity (Figure 1.5). The use of numbers to test hearing for speech dates back at least to the German otologist Friedrich Bezold in the late nineteenth century. Numbers were among the most highly recognizable words in a given language, reducing the likelihood of misinterpretation in the test setting, and, as MacFarlan later pointed out, their recognition seemed to be relatively impervious to differences in accent.34 With volunteers from the American League for the Hard of Hearing—who had embraced the objective of early hearing impairment detection in the hope of refining school placements and increasing rehabilitation—AT&T staff conducted a quarter-million hearing tests in the schools of twenty-three cities during 1926. They concluded that 8 to 12 percent of all students had “defective hearing.” Students thought to have hearing problems received a follow-up diagnostic test with a tonal audiometer. Hearing loss for speech was at first described as any loss over 9 “units” (later, “decibels”), but this was soon found to be too stringent—it seemed to indicate a veritable epidemic. AT&T engineers depicted the phenomenon in the emerging terms of quality control. As Harvey Fletcher, director of the AT&T Department of Speech and Hearing— and soon-to-be president of the American Association for the Hard of Hearing—explained in an article in Bell Laboratories Record on “Methods of Measuring Children’s Hearing”: Three million American schoolchildren are partially deaf. It is harder for them to learn, and later on it will be harder for them to earn. If these children were early given medical attention, social and economic loss might be saved. To pick them out, a method of testing is needed, quick enough to survey large groups in a short time, and accurate enough to segregate the deafened ones for further examination.35
Testing Hearing with Speech 35
Figure 1.5 WE 4-type audiometer hearing test chart. Courtesy: Kenneth Berger Hearing Aid Museum and Archives, Kent State University.
Phono- audiometers were installed in schools across the United States; AT&T also placed them in museums and at fairs to collect rough data on the hearing of older Americans. The audiometer was further used to measure urban noise, albeit with a different set of recordings.36 Today, the WE phono- audiometer project from the 1920s is recalled by American audiologists as the precursor to state-mandated school hearing screens in the 1960s as well as “universal” newborn hearing screening in the 1990s—nearly absolute
36 Sorting and Screening Human Hearers
extensions of biopolitics.37 It thus played a significant role in the sorting of citizens based on physical traits: along with the tonal audiometer, the phono- audiometer contributed to the expansion of the population classified as “hard of hearing” and the shrinking number of those sent to deaf oral or sign language schools. MacFarlan was involved in the first classroom test of the WE phono- audiometer, conducted with Harvey Fletcher at the Glen Mills school.38 After comparing tests taken by the same students with both the tonal and speech models, MacFarlan concluded that although the 4A was much faster and simpler, the results were roughly the same. He warned, however, that speech audiometry revealed nothing about hearing losses above 3,000 Hz—the upper threshold of the voice—nor could it specify the type or “degree” of impairment. As a “complex of pitches,” a recorded word or phoneme might be reproducible, but it was nonetheless an “uncontrollable, undependable, and imponderable” stimulus.39 Within a year, MacFarlan additionally criticized the mass-marketed WE 4A for being overly expensive. He published an article in Archives of Otolaryngology in 1928, offering guidelines to audiologists for building inexpensive models of their own with “parts available in any metropolis.” In addition to a buzzer audiometer made of either clockwork or electric bell parts, he recommended affixing to a typical record player a new Bristolphone—an electrical tone arm and speaker-amplifier with graduated intensity control that had become available in 1926. He then advised that recordings be selected (he did not name titles) to “determine the patient’s ability to hear voices of various register or musical instruments of various quality.”40 Working in Uppsala, Ernst Bárány and Erik Sperber coauthored a similar article in 1936, recommending the construction of audiometers from radio kits as a means both to save money and to tailor the devices to clinical needs.41 A stereotype exists in much of the engineering literature of the 1920s and 1930s that otologists resisted electroacoustics and clung to “old-fashioned” tuning forks, yet examples abound in the first half of the twentieth century of physicians following in the footsteps of George Fiske to assemble their own audiometers—a forgotten species of electronic tinkering.42 MacFarlan further concluded that the 4A did not provide an accurate assessment of speech perception, because the intensity of each spoken word was controlled by the recording itself: The Western electric [sic] phonographic audiometer uses a record in which the intensities are graduated at the time of making the record. Numbers are spoken out
Testing Hearing with Speech 37 which drop down in graduated steps; but only one number is spoken at each intensity. Speech is such that there is considerable variation in the force and audibility of various words and numbers. It is more serviceable to have an instrument which can be operated at will at any intensity. When the region of deafness is reached, the hearing can be tested by any number of test words or numbers. Thus, a more exact threshold of hearing interpretation is obtained.43
At the University of South Dakota, physician Augustus Pohlman built a phonograph audiometer in 1929 that instead allowed graduated regulation of intensity by the operator.44 Nonetheless, speech remained in a double bind as a test stimulus: it only seemed suitable for rough tests of hearing loss, not for testing the hearing of speech itself. Nor did tonal audiograms seem to give an adequate account of the loss of hearing for speech—which had become the defining task of the ear in school and the workplace.45 With their complex frequencies and varying loudness, speech sounds still seemed impossible to standardize for testing purposes. Even if the loudness of different words was equalized by a playback machine, as MacFarlan noted, “words vary in their interpretability.”46 Some words—namely, familiar ones—were simply more recognizable than others, making it impossible to separate “hearing” from “interpretation.” (Scholars who research sound today draw a related distinction between hearing and listening, although this history reveals how difficult the two are to pull apart.) To create a test list that assessed the range of speech sounds in a language, and then to grade a person’s hearing loss based on that list, required the quantification of frequency content, loudness, and the nebulous factor of familiarity.
From Detectability to Intelligibility In a separate line of research, in which the human ear was tester rather than testee, telephone engineers had long investigated the “articulation” of the telephone— its limited ability to transmit intelligible speech.47 Early models of the telephone did not transmit to a great distance, and the reproduced speech was often “muffled and indistinct.” In 1876, Alexander Graham Bell himself remarked on the difficulty of holding a conversation over the phone: Familiar quotations, such as “To be, or not to be; that is the question.” “A horse, a horse, my kingdom for a horse.” “What hath God wrought,” &c., were generally understood after a few repetitions. The effects were not sufficiently distinct to admit
38 Sorting and Screening Human Hearers of sustained conversation through the wire. Indeed, as a general rule, the articulation was unintelligible, excepting when familiar sentences were employed.48
To more closely assess the capacities of the telephone, Bell spoke the English speech sounds one at a time; in this case, he found that vowels were recognizable but consonants hardly at all.49 In 1910, Bell research engineer George Campbell reported to the readers of Philosophical Magazine on his investigations of “telephonic intelligibility” at the AT&T headquarters (then in Boston).50 Whereas prior articulation testing of phones had been relatively unsystematic and had allowed for guessing and the mere detectability of speech, Campbell began to consider how best to prepare the test material, and how to analyze it statistically, to determine the average distorting effects of telephone equipment on speech sounds. Employing a list of nonsense syllables starting with an English consonant and ending with the vowel “ee,” he made thousands of tests with the assistance of two other researchers. A follow-up test included additional vowels, six callers, and four listeners. Campbell also compared the rate of telephonic distortion to that of common mistakes made in plain speech (confusion of n for m, for instance). WE established a research laboratory in New York in 1911 and launched a formal program to study telephone quality in 1916, mainly because the WE carbon electrical transmitters were noisy and inefficient to the point of causing situational hearing loss. When Shewhart was hired two years later, he joined a team that included Harvey Fletcher and Robert Wegel. Telephony represented an extreme situation where communication was reduced to speech and hearing in the absence of the other senses; moreover, the frequency band of the telephone was itself restricted, so it was imperative to limit additional factors (thermal noise, electrical interference, crosstalk) that distorted speech. To test for noisy lines and equipment, the WE team utilized a list of questions: Name a prominent millionaire of the country. Why are flagpoles surmounted by lightning rods? Tell what is meant by an Indian reservation. What are some personal characteristics of the people of Japan? What is the chief purpose of industrial strikes? A listener’s ability to answer these questions over the phone—ambiguous and fraught with cultural bias though they were—offered rough proof of that device’s “intelligibility.”51 In the 1920s, engineers at the WE lab (renamed Bell Telephone Laboratories in 1925) adopted two tongue-twisters to gauge the articulation of standard phones via the limit case of high-frequency sibilants: “Sister Susy sewing silk shirts for Southern soldiers” and “Some settlers suggest settling Southern settlements in succession.”52 The team also created two “acoustically complete” sentences for testing telephone volume. These contained the English
Testing Hearing with Speech 39
speech sounds that most contributed to loudness: “Joe took Father’s shoe bench out” and “She was waiting at my lawn.” By 1922, however, WE engineers determined that lists of consonant-vowel- consonant (CVC) monosyllables, randomly assembled and spanning all of the phonemes in English, afforded precision testing to accompany the spoken questions and numbers used in screens.53 At first, the “articulation” of an apparatus was simply rated as the percentage of CVC monosyllables correctly identified by a listener. By 1929, however, the WE team had also statistically weighted each of the monosyllables based on its frequency of occurrence in everyday English speech, which they injudiciously estimated using Godfrey Dewey’s Relativ Frequency of English Speech Sounds, a 1923 collection of word counts from published rather than oral sources.54 The intelligibility of a given telephone system could then be evaluated based on a weighted tally of correctly identified sounds, accounting (or so it seemed) for the problem of familiarity. MacFarlan closely followed these articulation studies in the telephone industry, and by 1939 he was urging other otologists to rename speech hearing exams “intelligibility tests.” Nevertheless, he felt that nonsense syllables were not suitable for testing the human ear—they were too confusing for test subjects to understand and transcribe.55
Spondees, Phonetically Balanced Word Lists, and Harvard Sentences During World War II, many American otologists—looking ahead to postwar rehabilitation— sought to approximate speech with the seemingly more objective testing tool of tonal audiometry. As a member of the Council on Physical Therapy, MacFarlan helped author a 1942 report to the American Medical Association on the topic of “useful hearing.” He and his coauthors recommended that otologists adopt a set of six frequencies, based on “the relative values of the different portions of the auditory range for intelligibility of speech,” for estimating percentage loss of hearing—rather than relying on the “curve of serviceable hearing” drawn on the WE audiogram, which was distributed with WE pure-tone audiometers.56 Another council member, Cordia C. Bunch, explained that the curve of serviceable hearing seemed to have been derived from the maximum threshold of feeling, such that total loss of hearing was defined as the inability to hear at extraordinary amplification. Bunch and MacFarlan felt that a curve representing “total loss of usable hearing,” at significantly lower decibel amounts, was more adequate for the purpose of compensation cases and for school tracking.57 This approach to
40 Sorting and Screening Human Hearers
tonal audiometry—known as “the speech banana” because of the visual shape that resulted from plotting sounds in the speech range on an audiogram— remains in place in the international clinical context today, and audiologists continue to revise the English-based chart for other languages.58 After the war, the industrial goal of quantifying “serviceable hearing” (serviceable meaning “in working order”) disseminated from a small group of telephone engineers and “mechanically minded” otologists to the wider engineering and audiological communities via the Harvard PAL.59 Known as “the largest university-based program of wartime psychological research,” the PAL was founded in 1940 with funding from the National Defense Research Committee and directed by psychologist Stanley Smith Stevens.60 The primary aim of the lab was “sound control” for military field communications. An interdisciplinary team of researchers at the PAL, including several who went on to work at Bell Labs and at deaf schools after the war, initially appropriated the Bell sentences and CVC monosyllables to test the intelligibility of military equipment under conditions of extreme noise.61 They found, however, that the sentences “vary considerably in difficulty, they frequently refer to details peculiar to New York City, and they are too much a test of knowledge and intelligence for use with some grades of listeners.”62 As for the nonsense CVC syllables, although these had the advantage of impartiality with regard to a listener’s vocabulary, they were nonetheless difficult for listeners to identify even if accurately heard. The WE 4C test (an updated version of the 4A) did not cover a very wide range of English speech sounds—just the consonants and vowels in seven spoken numbers.63 The PAL group devised new lists of words and sentences to test electroacoustic equipment and (especially after the war) hearing loss for speech in humans. The precedent of telephony thus expanded to all human–machine systems. It seemed that impaired hearing could be caused by any element, as PAL employee James P. Egan explained in 1948: “Distortion of speech sounds by inferior transducers, masking of these sounds by ambient or electrical noises, faulty enunciation, partial deafness— all these factors and many others may conspire to make communication uncertain and unreliable.”64 Over several iterations, the PAL researchers created new test lists in three categories— “phonetically balanced” (PB) monosyllabic words, disyllabic spondees, and PB sentences—some of which they made widely available as twelve-inch disc recordings. The PB lists were composed of common monosyllabic words (as opposed to nonsense sounds), representing all the English speech sounds in proportion to their relative frequencies of use. They were applied at the PAL to test equipment as well as ears.65 To approximate “relative
Testing Hearing with Speech 41
frequency of use,” the PAL scientists again borrowed from Dewey’s already outdated study. In the 1950s, the PB word lists were updated by Ira Hirsh and his research team at the Central Institute for the Deaf (where Hallowell Davis had moved from the PAL), drawing on a 1929 Bell Laboratories study of telephone speech to achieve better phonetic balance.66 These new recorded lists, known as Auditory Test No. W-22, were widely marketed to audiologists by the Central Institute for the Deaf. They remain among the most common speech hearing tests today, despite the impossibility of capturing “average” English speech (across speakers, dialects, and so on), and despite evidence that phonetic balance is not even necessary to test word recognition.67 The PAL group also created a spondee list in which each spoken syllable received the same stress and all of the words had equivalent audibility. Hothouse. Inkwell. Mousetrap. These lists were considered best for testing the minimal threshold of speech hearing (rather than speech discrimination, or how accurately speech is heard); they were additionally applied to test hearing aids and headphones.68 Updated at the Central Institute for the Deaf as Auditory Tests No. W-1 and W-2, they also circulated into broad use. Finally, the PAL group generated a new set of PB sentences, made up of familiar words and a “normal sampling of English speech sounds.”69 Egan and his colleagues believed that sentences were of limited use for testing transmission equipment, because the extra context they provided made guessing words too easy.70 Nonetheless, the “Harvard sentences” became entrenched in quality control for electroacoustic equipment in 1969, when the Institute of Electrical and Electronics Engineers released a revised set as a “Recommended Practice for Speech Quality Measurement.”71 Even today, these lists are considered the “gold standard” for testing microphones, mobile phones, cochlear implants, and countless other audio technologies.72 These days a chicken leg is a rare dish. Glue the sheet to the dark blue background. A large size in stockings is hard to sell. Feed the white mouse some flower seeds.73 Unlike the other words and phrases embalmed in test lists since the nineteenth century, the Harvard sentences have had crossover success. Picked up in technology journalism, recycled into found poems and automated poetry generators, their evocative signification obscures their standardized sound and the biases inherent in their “normal sampling.”74 Anything but fortuitous in their construction, the Harvard lists exist in the uncanny valley of engineered speech. Their novelty, to one audience, is strikingly incongruous with their tedium for those who routinely sit through auditory testing (e.g., to calibrate a hearing aid or implant). They are widely miscredited as the origin of the test sentence phenomenon, which—as this chapter has shown—instead took root in asylums for deaf students, injury
42 Sorting and Screening Human Hearers
compensation tables, industrial laboratories, quality inspections, and the mathematical fantasy of average English speech.
Conclusion: What Counts in Speech and Hearing In 1952, Ira Hirsh noted that hearing function had by then been parsed into categories such as “absolute thresholds, differential thresholds, hearing for speech, influence of noises, and even a little measurement of psychological dimensions such as loudness and pitch.”75 Each of these attributes had been scaled and, in turn, correlated to scaled quantities of physical stimuli. This history raises the question of what counts as hearing, and whether anything remains after the “observable events” and functions have been tallied and prioritized. The statistical turn in both speech and tonal audiometry means that a person’s hearing is routinely judged relative to a norm rather than to an environmental setting, a profession, or a preference. That norm, established by the telephone company with its narrowband bias, discounts most hearing above 3,000 Hz (in the case of speech audiometry) or above 8,000 Hz (for pure-tone audiometry), although speech sounds such as sibilants stretch into a higher range.76 The long history of speech audiometry also raises the question of why hearing tests are performed. Hearing is “serviceable” to school, state, and industry; tests incorporate ears into engineered systems and spoken language drills. Can a hearing test be easily and inexpensively requested, or is it more often imposed? Regarding mass screening, otologist Robert Ruben— who helped establish the U.S. National Institute on Deafness and other Communication Disorders (NIDCD)—argues: Hearing screening has historically been used to protect a third party and not for the benefit of the patient. For example, military screening is used to assess the ability of personnel to fulfill military functions; industrial screening is used to determine hearing status before a worker is exposed to sound trauma to set a baseline against potential legal suits.77
Industrial screening also screened out hearing-impaired candidates from the workplace, to preclude expensive insurance and compensation claims.78 Hearing screens are least often performed on older people, even though the aging population is most likely to experience hearing loss and benefit from therapeutic intervention.
Testing Hearing with Speech 43
The partisan deployment of mass hearing screens and the molding of speech sounds into yardsticks of useful hearing both result from the anchoring of speech audiometry in the historical-industrial context of quality control: the transfer of tools and concepts from medical diagnosis to machine calibration and back again. It is perhaps startling to find tests of hearing, a seemingly subjective experience, at the forefront of medical screening. Yet the dual entanglements of the ear with “the largest machine in the world”—the telephone—and with compulsory education ratified the notion that humans could and should be inspected in industrial fashion.79 From sorting students at deaf schools in the nineteenth century, articulation tests became universalized through the inspection, rating, and calibration of humans as components of transmission systems. In the course of this 150-year history—micromeasure by micromeasure and component by component—all communication came to be understood as “uncertain” and “unreliable.” Proposing new speech-hearing test methods in 1948, Egan commented that “all articulation scores are relative scores, contingent upon the use of specific announcers, microphones, amplifiers, earphones, noises, listeners, and test lists. Little trust can be placed in absolute statements about articulation.”80 Nonetheless, from relative scores have come determinate classifications and durable machines. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/testing-hearing-with-speech-mills.
Notes 1. Thomas A. Edison, “Phonograph and Its Future.” 2. George F. Fiske, “Phonograph in Testing Hearing,” 855. 3. Ibid., 855. 4. Ibid., 855. 5. Tonal hearing tests administered via phonograph, dating to around 1919, followed these early examples of speech phono-audiometry, as discussed later in this chapter. 6. For more on the production of “pure tones” for early twentieth-century tests of musicality, see Viktoria Tkaczyk in this volume. 7. Ira Hirsh, Measurement of Hearing, 309. 8. Edmund Prince Fowler and R. L. Wegel, “Presentation of a New Instrument”; Mara Mills, “Deafening.” 9. High-tone losses that exceed that standard audiogram often go unrecognized. Moreover, the audiogram determines other hearing losses such that they are classified according to the visual shape of their graphs: “notches,” “reverse slope,” “cookie-bite hearing loss,” etc.
44 Sorting and Screening Human Hearers 10. By the World War I era, “inability to work” had come to define the umbrella category of disability. Sarah Rose, No Right to Be Idle. 11. Hirsh, Measurement of Hearing, 127. 12. Harald Feldmann, History of Audiology, 87. 13. Ibid., 85, 87. 14. Emily Thompson, Soundscape of Modernity, 148. 15. Stuart Blume and Barbara Regeer, “Audiometer,” 40. 16. W. A. Shewhart, Economic Control, chapter 2. 17. Feldmann, History of Audiology, 77. 18. William James, Principles of Psychology, 1:545–46. 19. The measurement of loudness was hardly incidental, and eventually prompted a multiyear international committee to classify the various possible scales of measurement (nominal, ordinal, interval, ratio). S. S. Stevens, “Theory of Scales of Measurement.” 20. Hirsh, Measurement of Hearing, 5. 21. Viktoria Tkaczyk, “Nur ein Verdacht.” 22. W. Sohier Bryant, “Phonographic Acoumeter,” 442, 439. 23. Douglas MacFarlan, “Speech Hearing Tests,” 72. 24. Ibid., 115, 82, 80. 25. For a detailed history of this collaboration, see Mills, “Deafening.” 26. For more on these tonal audiometers, see ibid. 27. W. A. Shewhart, “Some Applications of Statistical Methods.” 28. Friedrich-Wilhelm Hagemeyer, “Die Entstehung von Informationskonzepten.” 29. Shewhart, Economic Control, chapter 2. 30. Christina Cogdell, Eugenic Design. 31. Oxford English Dictionary, s.v. “screen.” “Screening” is a practice that has become increasingly prevalent and is the subject of much debate in bioethics. Examples range from mammograms and other detection screens, to prenatal diagnosis, to universal newborn hearing screening. 32. Alfred Morabia and F. F. Zhang, “History of Medical Screening.” 33. On “screening out,” see Rose, No Right to Be Idle, 7, 138, 163–67, 224. 34. Writing in 1945, MacFarlan specifically argued that there was “a 10 per cent better hearing performance with numbers.” MacFarlan, “Speech Hearing Tests,” 80. 35. Harvey Fletcher, “Methods of Measuring Children’s Hearing,” 154. Emphasis added. 36. Thompson, Soundscape of Modernity, 148. 37. Jerry L. Northern and Marion P. Downs, Hearing in Children, 290–91. For the later history of speech audiometers employing magnetic wire and tape recording, see Feldmann, History of Audiology, 86–87. On biopolitics, see Foucault, Security, Territory, Population. 38. MacFarlan worked in Philadelphia; this was presumably the Glen Mills school “for juvenile delinquents” in a nearby Pennsylvania town. Douglas MacFarlan, “Voice Test of Hearing,” 27. 39. Ibid., 28. By 1945, MacFarlan had also concluded that speech audiometry tended to indicate better hearing than did tonal tests. MacFarlan, “Speech Hearing Tests,” 112. 40. Douglas MacFarlan, “Circuit Plans for Inexpensive Audiometers.” 41. Ernst Bárány and Erik Sperber, “Über den Bau von elektrischen Audiometern.” Bárány’s father Robert earned the Nobel Prize for Medicine in 1914.
Testing Hearing with Speech 45 42. On the reluctance of otologists to adopt electronic audiometers, see Fowler and Wegel, “Presentation of a New Instrument,” 119. 43. MacFarlan, “Circuit Plans for Inexpensive Audiometers,” 532. 44. Feldmann, History of Audiology, 86. 45. In 1945, MacFarlan argued, “Definitely it is impossible to determine how well a person hears speech, from studying a frequency audiogram.” MacFarlan, “Speech Hearing Tests,” 113. 46. Ibid., 105. 47. For more on the early history of telephone intelligibility testing, see Mara Mills, On the Phone. 48. A. Graham Bell, “Researches in Telephony,” 8. 49. Ibid., 9. 50. George A. Campbell, “Telephonic Intelligibility.” 51. H. Fletcher and J. C. Steinberg, “Articulation Testing Methods,” 853–54. 52. Helen Cooke and Russell Maloney, “She Was Waiting,” 12. 53. Harvey Fletcher, “Nature of Speech.” 54. Fletcher and Steinberg, “Articulation Testing Methods.” See Godfrey Dewey, Relativ Frequency. 55. Douglas MacFarlan, “History of Audiometry,” 516. 56. Council on Physical Therapy, “Tentative Standard Procedure.” 57. C. C. Bunch, “Usable Hearing.” 58. “The Speech Banana.” On the limitations of the initial Anglophone speech banana, see Nittayapa Klangpornkun, Chutamanee Onsuwan, and C. Tantibundhit, “Constructing Speech Banana.” 59. MacFarlan, “History of Audiometry,” 514. 60. James H. Capshew, Psychologists on the March, 147. 61. For instance, John Karlin went on to head the first Human Factors group at Bell Labs, and Hallowell Davis became the research director at the Central Institute for the Deaf. For more on the PAL and the military employment of articulation testing, see Paul Edwards, Closed World, 210. Edwards claims: “Communications engineering, previously concerned with technologies and their interface with people, now took over the articulation of the messages themselves as well. ‘Natural’ language was converted into a technology, a code or cipher device.” In fact, as this chapter has shown, telephone companies had already been engineering speech along with equipment and interfaces in the 1920s. 62. James P. Egan, “Articulation Testing Methods,” 967. 63. The PAL team argued that the 4C worked well enough to test the threshold (minimal level) of hearing, but not for the discrimination of speech sounds or the assessment of much equipment. C. V. Hudgins et al., “Development of Recorded Auditory Tests,” 59. 64. Egan, “Articulation Testing Methods,” 955. 65. Ibid., 957. 66. Norman French and Walter Koenig Jr., “Frequency of Occurrence of Speech Sounds.” 67. Frederick N. Martin, Craig A. Champlin, and Desirée D. Perez, “Question of Phonetic Balance.” 68. Hudgins et al., “Development of Recorded Auditory Tests,” 64–65; Egan, “Articulation Testing Methods,” 966. 69. Hudgins et al., “Development of Recorded Auditory Tests,” 53.
46 Sorting and Screening Human Hearers 70. “Under most test conditions articulation scores obtained with lists of sentences are so high that communication systems must differ considerably before substantial difference in the scores is obtained.” Egan, “Articulation Testing Methods,” 967. 71. Institute of Electrical and Electronics Engineers, IEEE Recommended Practice. This report also cites research conducted by John Karlin post-Harvard at Bell Labs. 72. Indiana University speech researcher David Pisoni, as quoted by Sarah Zhang in “The ‘Harvard Sentences.’ ” Of course, these widely used sentences are far from “secret.” 73. For the full list of Harvard sentences, see http://www.cs.cmu.edu/afs/cs.cmu.edu/project/ fgdata/OldFiles/Recorder.app/utterances/Type1/harvsents.txt. 74. The Harvard sentences have become source material for online automated poetry generators, countless other poetry experiments, postcards, installation art—and beyond. 75. Hirsh, Measurement of Hearing, 4. 76. Hudgins et al. dismiss this problem by noting that most people can understand speech over the telephone, or at 3,000 Hz and below if their hearing is not impaired. Hudgins et al., “Development of Recorded Auditory Tests,” 63. 77. Robert Ruben, “The History of Hearing Screening for Hearing Ability: Part 2 Adult,” unpublished manuscript dated February 14, 2011, quoted with permission from the author. 78. For extensive evidence of the unintended detrimental consequences of workers’ compensation policies for disabled people, see Rose, No Right to Be Idle. 79. The phrase “the largest machine in the world” is repeated across WE and other Bell System advertisements from the mid-twentieth century as a descriptor for the telephone network. It is also the title of c hapter 4 of Phil Lapsley’s Exploding the Phone. 80. Egan, “Articulation Testing Methods,” 968–69.
References Bárány, Ernst, and Erik Sperber. “Über den Bau von elektrischen Audiometern.” Acta Oto- Laryngologica 23, no. 1 (1936): 182–99. Bell, A. Graham. “Researches in Telephony.” Proceedings of the American Academy of Arts and Sciences 12 (1876–77): 1–10. Blume, Stuart, and Barbara Regeer. “Audiometer.” In Instruments of Science: An Historical Encyclopedia, edited by Robert Bud and Deborah Jean Warner, 39– 41. New York: Garland, 1998. Bryant, W. Sohier. “A Phonographic Acoumeter.” Archives of Otology 33 (1904): 438–43. Bunch, C. C. “Usable Hearing.” Annals of Otology, Rhinology, and Laryngology 49, no. 2 (1940): 359–67. Campbell, George A. “Telephonic Intelligibility.” Philosophical Magazine 19, no. 6 (1910): 152–59. Capshew, James H. Psychologists on the March: Science, Practice, and Professional Identity in America, 1929–1969. Cambridge: Cambridge University Press, 1999. Cogdell, Christina. Eugenic Design: Streamlining America in the 1930s. Philadelphia: University of Pennsylvania Press, 2004. Cooke, Helen, and Russell Maloney. “She Was Waiting.” New Yorker, February 1, 1936: 12–13. Council on Physical Therapy. “Tentative Standard Procedure for Evaluating the Percentage of Useful Hearing Loss in Medicolegal Cases.” Journal of the American Medical Association 119, no. 14 (1942): 1108–9.
Testing Hearing with Speech 47 Dewey, Godfrey. Relativ Frequency of English Speech Sounds. Cambridge, MA: Harvard University Press, 1923. Edison, Thomas A. “The Phonograph and Its Future.” North American Review 126, no. 262 (May 1878): 527–36. Edwards, Paul. The Closed World: Computers and the Politics of Discourse in Cold War America. Cambridge, MA: MIT Press, 1996. Egan, James P. “Articulation Testing Methods.” Laryngoscope 58, no. 9 (1948): 955–91. Feldmann, Harald. A History of Audiology: A Comprehensive Report and Bibliography from the Earliest Beginnings to the Present. Chicago: Beltone Institute for Hearing Research, 1970. Fiske, George F. “The Phonograph in Testing Hearing.” Journal of the American Medical Association 15, no. 24 (1890): 854–56. Fletcher, Harvey. “Methods of Measuring Children’s Hearing.” Bell Laboratories Record 2, no. 4 (1926): 154–57. Fletcher, Harvey. “The Nature of Speech and Its Interpretation.” Journal of the Franklin Institute 19 (June 1922): 729–47. Fletcher, H., and J. C. Steinberg. “Articulation Testing Methods.” Bell System Technical Journal 8, no. 4 (October 1929): 806–54. Foucault, Michel. Security, Territory, Population: Lectures at the Collège de France, 1977– 1978. Edited by Michel Senellart, translated by Graham Burchell. New York: Palgrave Macmillan, 2009. Fowler, Edmund Prince, and R. L. Wegel. “Presentation of a New Instrument for Determining the Amount and Character of Auditory Sensation.” Transactions of the American Otological Society 16 (1922): 105–23. French, Norman, and Walter Koenig Jr. “The Frequency of Occurrence of Speech Sounds in Spoken English.” Journal of the Acoustical Society of America 1, no. 34 (1929): 110–20. Hagemeyer, Friedrich- Wilhelm. “Die Entstehung von Informationskonzepten in der Nachrichtentechnik: Eine Fallstudie zur Theoriebildung in der Technik in Industrie-und Kriegsforschung.” PhD diss., Freie Universität Berlin, 1979. Hirsh, Ira. The Measurement of Hearing. New York: McGraw-Hill, 1952. Hudgins, C. V., J. E. Hawkins, J. E. Karlin, and S. S. Stevens. “The Development of Recorded Auditory Tests for Measuring Hearing Loss for Speech.” Laryngoscope 57, no. 1 (1947): 57–89. Institute of Electrical and Electronics Engineers. IEEE Recommended Practice for Speech Quality Measurements. IEEE Report No. 297. New York: Institute of Electrical and Electronics Engineers, 1969. James, William. The Principles of Psychology. 2 vols. New York: Henry Holt, 1890. Klangpornkun, Nittayapa, Chutamanee Onsuwan, and C. Tantibundhit. “Constructing Speech Banana for Thai Consonants: Some Considerations for Male and Female Voices.” Proceedings of the 18th International Congress of Phonetic Sciences, Glasgow, Scotland, 2015, https://www.internationalphoneticassociation.org/icphs-proceedings/ICPhS2015/Papers/ ICPHS1033.pdf. Lapsley, Phil. Exploding the Phone: The Untold Story of the Teenagers and Outlaws Who Hacked Ma Bell. New York: Grove Press, 2013. Lucae, August. “Über das Phonometer.” Archiv für Physiologie (1878): 588–89. MacFarlan, Douglas. “Circuit Plans for Inexpensive Audiometers.” Archives of Otolaryngology 7, no. 5 (1928): 527–32. MacFarlan, Douglas. “History of Audiometry.” Archives of Otolaryngology 29, no. 3 (1939): 514–19. MacFarlan, Douglas. “Speech Hearing Tests.” Laryngoscope 55, no. 2 (1945): 71–115. MacFarlan, Douglas. “The Voice Test of Hearing.” Archives of Otolaryngology 5, no. 1 (1927): 1–29.
48 Sorting and Screening Human Hearers Martin, Frederick N., Craig A. Champlin, and Desirée D. Perez. “The Question of Phonetic Balance in Word Recognition Testing.” Journal of the American Academy of Audiology 11, no. 9 (2000): 489–93. Mills, Mara. “Deafening: Noise and the Engineering of Communication in the Telephone System.” Grey Room 43 (2011): 118–43. Mills, Mara. On the Phone: Deafness and Communication Engineering. Durham, NC: Duke University Press, forthcoming. Morabia, Alfred, and F. F. Zhang. “History of Medical Screening: From Concepts to Action.” Postgraduate Medical Journal 80, no. 946 (2004): 463–69. Northern, Jerry L., and Marion P. Downs. Hearing in Children. 5th ed. Philadelphia: Lippincott, Williams, & Wilkins, 2002. Rose, Sarah. No Right to Be Idle: The Invention of Disability, 1840s–1930s. Chapel Hill: University of North Carolina Press, 2017. Shewhart, W. A. Economic Control of Quality of Manufactured Product. New York: D. Van Nostrand Company, 1931. Shewhart, W. A. “Some Applications of Statistical Methods to the Analysis of Physical and Engineering Data.” Bell System Technical Journal 3, no. 1 (1924): 43–87. “The Speech Banana.” Alexander Graham Bell Association for the Deaf and Hard of Hearing website, https://www.agbell.org/professionals/auditory-functioning/the-speech-banana/. Stevens, S. S. “On the Theory of Scales of Measurement.” Science 103, no. 2684 (1946): 677–80. Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the Culture of Listening in America, 1900–1933. Cambridge, MA: MIT Press, 2002. Tkaczyk, Viktoria. “Nur ein Verdacht: Hermann Gutzmann, die Phonographie und ein leerer Kasten.” In Archäographien: Aspekte einer radikalen Medienarchäologie, edited by Moritz Hiller and Stefan Höltgen, 115–26. Berlin: Schwabe Verlag, 2019. Zhang, Sarah. “The ‘Harvard Sentences’ Secretly Shaped the Development of Audio Tech.” Gizmodo, March 9, 2015, https://gizmodo.com/the-harvard-sentences-secretlyshaped-the-development-1689793568.
2 The Testing of a Hundred Listeners Otto Abraham’s Studies on “Absolute Tone Consciousness” Viktoria Tkaczyk
In his 1901 study on “absolute tone consciousness,” the Berlin physician and physicist Salomon Otto Abraham reminisced about a parrot that he had owned some time before. Abraham would always sing the same C-minor symphony to his companion. After a while, the bird astonished its master by whistling the beginning of the symphony in absolutely correct tune: “G–G– G–E-flat.” Just once, the parrot was mistaken by a semitone and struck A-flat instead of G, but it interrupted itself after the third tone and started afresh.1 What is noteworthy about this anecdote is not only that Abraham made his parrot whistle Beethoven’s Symphony No. 5 but also that he took his observation as proof that even animals were able to develop exceptional musical skills through intensive practice. Abraham’s nonhuman test subjects were only two in number (a parrot, and later a starling), but in the same study he analyzed a survey that he had distributed among a hundred human test subjects to investigate their musicality. In this undertaking, Abraham drew upon existing scholarship on what was commonly known as “absolute pitch” or “perfect pitch,” but that he preferred to call “absolute tone consciousness.” This facility involved not merely the identification of tones or sensitivity to minimal pitch differences, but a broader range of individually differing abilities—whether singing and whistling by sight or spontaneously reproducing melodies that had been heard.2 Abraham also proposed to replace the notion of “relative pitch” with the more general category of a “musical memory” for intervals, rhythm, harmony, and melody and the category of the “consciousness of musical keys.”3 He argued that all these abilities arose not so much from native genius as from a tailored musical background and education. What makes Abraham’s project so significant for the history of modern aurality is that its investigation of auditory abilities went beyond the stimulus– response tests generally conducted in audiology around 1900. As discussed in the introduction and Mara Mills’s contribution to this volume, the nineteenth Viktoria Tkaczyk, The Testing of a Hundred Listeners In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0003.
50 Sorting and Screening Human Hearers
and early twentieth centuries saw the emergence of various tests for auditory acuity (the loudness of standard frequencies), the range of hearing (pitch levels), and more specific hearing abilities such as the hearing of speech. To evaluate musical abilities, Abraham proposed a still wider range of tests, based on a detailed questionnaire. Abraham thought carefully about the choice of his test subjects and sent out his questions—nineteen in total—to one hundred of his acquaintances, “violinists, pianists, singers, the first psychologists and self-observers.”4 The questionnaire inquired into musical preferences, listening habits, and family members who had absolute pitch, and placed special emphasis on ways of training musical skills.5 In this chapter, I first outline how Abraham’s large-scale survey superseded earlier physiological and psychological methods of audio testing as well as particular studies on “perfect pitch.” I then address Abraham’s notion of “absolute tone consciousness” itself. When Abraham published his 1901 study, he was working as Carl Stumpf ’s assistant at the Institute of Psychology in Berlin. Whereas most of the research conducted at Stumpf ’s institute focused on a few test subjects only, Abraham’s use of a questionnaire helped him frame responses as easily computable statistics and turn his results into applicable data. In this respect, his study marks the beginning of a new research trend in the human sciences, one geared toward application and soon to be named “applied psychology.” Finally, I sketch the application of Abraham’s study in the fields of comparative musicology and music education in the years that followed, looking more specifically at the testing procedures carried out at the Berlin Phonogramm-Archiv, the Berlin Academy of Music, and, with U.S. psychologist Carl Seashore’s “measures of musical talent,” American schools and music academies. Some of these follow-up projects—quite at variance with Abraham’s primary intention—used the testing of perceptual discrimination as a tool to delineate cultural, social, and racial differences.
Changing Sounds: Test Tones for the Culturally Cultivated Ear From Abraham’s study, it is not clear whether everyone who filled in the questionnaire was invited to participate in further testing in Abraham’s laboratory. Those who did visit the Berlin lab underwent a whole series of tests. For the identification of minimal pitch differences, for example, each subject was presented with test tones of varying pitch levels, intensities, durations, and sequences.6 Abraham, who considered himself to have perfect pitch, unreservedly sang, whistled, and played several instruments to his test subjects.
The Testing of a Hundred Listeners 51
He found that some identified the pitch levels through the instruments’ tone color, which prompted him to invent a new testing method. Abraham used an electric motor–driven Edison phonograph, donated to Stumpf ’s lab by the Louise Bose Foundation, to test what he called “consciousness of musical keys” with “pure tones” (tones independent of the timbre that accompanies their pitch level when reproduced on particular instruments, such as violins or pianos).7 Abraham recorded violin chords and single tones, and played them back at different rotation rates, with half of the recording rate being one octave lower. He was thus able to transpose a tone and its overtones within the range of one and a half octaves. The characteristic color of a given violin was transposed as well, preventing subjects from relying on timbre for pitch recognition.8 In developing this method, Abraham asserted that the same test would not work with the recording of a human voice: different recording rates would alter the sound color of the voice due to “stable but impure side tones” of the mouth cavity.9 This harked back to the earliest use of the phonograph as a sort of tone varier, in the experiments with recorded vowels carried out by British physicists Fleeming Jenkin and J. Alfred Ewing in 1878 to test Helmholtz’s theory of vowels. Their experiments were immediately replicated by Harvard physicist Charles R. Cross and picked up again from 1898 to 1911 by the Königsberg physiologist Ludimar Hermann, from 1912 to 1914 by Wolfgang Köhler, and finally by Carl Stumpf for his study Die Sprachlaute (Speech sounds) of 1926.10 Stumpf ’s research corrected Abraham’s rather vague 1901 explanation by adding a more nuanced theory of the shifts of stable formants. But it was Abraham’s early attempts to measure the consciousness of musical keys with an Edison phonograph, and his dual interest in the recorded instrumental and vocal sounds on the one hand, and the phonograph as a research tool and test instrument on the other, that paved the way for Köhler’s and Stumpf ’s studies.11 For the present chapter, Abraham’s search for pure tone tests is crucial, because it indicates his awareness of individually variable listening habits and skills. Such variation was already becoming apparent in the 1870s, in connection with the earliest studies on perfect pitch. One key publication on the issue is the 1876 study Über die Grenzen der Tonwahrnehmung (On the limits of tone perception) by the Jena physiologist William Thierry Preyer.12 Preyer criticized the existing experiments on the limits and exactitude of auditory perception not only for ignoring the test subjects’ individual musical background but also for taking insufficient account of the particular timbres of the testing instruments used.13 He commissioned the instrument maker Georg Appunn from Hanau, near Frankfurt, to build the apparatuses he envisaged,
52 Sorting and Screening Human Hearers
including a tone differential device to investigate sensitivity to sounds with minimal pitch differences—this being, for Preyer, the main criterion for perfect pitch.14 Very few test subjects were able to identify subtle differences of 0.4 vibrations per second. Preyer thus began to look more closely at the musically gifted and realized that talent was a matter of practice: “In the case of violinists, the makers of acoustic and musical instruments and tuning forks, and also in the case of a clinician who is very practiced in auscultation and percussion and a linguist accustomed to observing dialectal differences in pronunciation with extraordinary precision, I found this astonishing sureness of judgment—not in the case of pianists.” Preyer argued, however, that this power of judgment could be acquired through persistent practice in just a few weeks.15 For him, even the range of human hearing could be improved by training, and he gave instructions on how to enhance sensitivity to the highest or lowest audible tones.16 Preyer’s findings achieved international exposure thanks to a paper by Alexander J. Ellis. In “On the Sensitiveness of the Ear to Pitch and Change of Pitch in Music” of 1876, the British musicologist, phonetician, and translator of Helmholtz agreed with Preyer that absolute pitch was affected by both music education and musical traditions.17 Ellis further stressed that international standards in pitches were still lacking, so that one could not simply say “ ‘a good ear knows when a note is in tune,’ because the meaning of ‘being in tune’ is at present unfixed both as to standard pitch and desired intervals.”18 Alongside the “musical ear,” Ellis was also interested in the “dialect ear.” In his studies of the varying pronunciation in British dialects, carried out from 1879, Ellis drew up what is still called the “Dialect Test,” which for the English- speaking area contains possible pronunciation variants for seventy- six 19 words. Clergymen in forty-two regions of England were asked to implement the tests. Unfortunately, when Ellis came to analyze the results, he realized that the questioners—themselves mostly speakers of the standard language— had not been fully capable of carrying out the test: without a dialectally cultivated ear or special training, they were unable to hear the subtle differences in pronunciation.20 It is probably no historical coincidence that this increasing awareness of varying cultures not only of tuning but also of hearing and memorizing speech and music was accompanied by efforts to standardize pitch.21 A first international standardization of performance pitch was accomplished at the International Conference on Pitch held in Vienna in 1885, where the French diapason normal (a1 = 435 Hz) was accepted as binding.22 Around the same time, the earliest training programs for absolute pitch emerged in musical
The Testing of a Hundred Listeners 53
education. An important figure here is the music theorist and music educator Hugo Riemann, who, as early as 1882, advocated the method of dictating notes designed by Ambroise Thomas, the director of the music conservatory in Paris. In this method, 1,560 different exercises taught every trainee, first, “absolute pitch”; second, “relative pitch”; and, third, an ear for rhythms and melodies.23 The aim of Preyer, Ellis, and Riemann, then, was not simply to define what we understand today by the term “absolute pitch”—a rare ability that is inherited and that, although it may be enhanced by musical education in early childhood, cannot be achieved at a later stage.24 Instead, they were interested in a sense of subtle distinctions that was not innate, but the outcome of musical and linguistic cultural practices and standards. That made it all the more important to have full information about the musical background of both test subjects and testers for studies on absolute pitch. It was in this context that a more sophisticated theoretical framework for absolute pitch began to emerge—one that would have a great impact on the work of Abraham, who agreed that the testing of absolute pitch depended on the test subject’s musical education. The only way to abstract from this musical knowledge, he argued, was to use synthetic test tones such as those produced by the Edison phonograph, and a detailed questionnaire.
Changing Epistemologies: Theories of Tone Consciousness Abraham’s study on absolute tone consciousness is extensive. The psychologist assesses the one hundred responses he received to his questionnaire over the course of eighty-six pages. The questions he posed signal a growing psychological interest in the individual hearing experiences and unconscious abilities of test subjects; most of them could not have been answered using only the conventional test methods of audiology. Were the subjects only able to reproduce the note they heard by singing or whistling, or could they also correctly name it? Could they identify individual notes, or the key of chords and melodies? How did the pitch level affect the accuracy of their judgment? How long, and how loudly, did a note have to sound for them to identify it? Did the subjects compare what they heard against an internally fixed tone (concert pitch a1)? Further questions in the survey were directed at the associations evoked by certain tones: did they conjure up the letter label of the note, a timbre, a tone characteristic, a color, moods, feelings of tension in the larynx, fingers, or veins?25
54 Sorting and Screening Human Hearers
In designing his questionnaire, Abraham was clearly influenced by the director of the Berlin Institute of Psychology, Carl Stumpf. The foundation of the institute in 1893 had been preceded by a dispute within the field of psychology at the University of Berlin that also affected Abraham’s later work. In the 1880s, the Berlin philosopher Wilhelm Dilthey had begun to build epistemological foundations for the distinct epistemic interests of the human sciences (Geisteswissenschaften) and the natural sciences (Naturwissenschaften)—a distinction today known as the “two-culture” divide.26 Controversy quickly arose around those research fields, such as psychology, that could be affiliated with either area. Whereas psychologists working with experimental methods from the natural sciences focused on the nature and laws of human perception and judgment, Dilthey’s “understanding psychology” relied on the cognitive, experiential, and interpretive capacities of the human mind and was interested in both individual and cultural variations of mental states. In 1893, this difference led Dilthey to block the appointment of an outstanding experimental psychologist, Hermann Ebbinghaus, as a professor at the University of Berlin, because he regarded Ebbinghaus’s research as an “utterly scientific radicalization of philosophy.”27 Instead, Dilthey successfully backed the candidature of Carl Stumpf, who was also appointed head of the university’s new Institute of Psychology.28 In this function, Stumpf reproduced Dilthey’s irreconcilable duality between the distinct epistemic interests of psychology but shifted the argument to what anthropologists of his time had begun to describe as the “nature–nurture debate”—a debate over whether human behavior is determined genetically (as most scholars in the natural sciences believed) or by the environment (the view of most scholars in the human sciences).29 Stumpf supported the latter view but tried to prove it by setting up a new research trend in the human sciences: he combined methodological approaches from the natural sciences, such as stimulus–response experiments, and from the humanities, such as introspection. Under the aegis of the philosophical faculty, Stumpf ’s Institute of Psychology soon expanded from a tiny three-room venue to increasingly prestigious premises in different university buildings on Dorotheenstrasse, before moving to forty-one rooms in the Berlin City Palace in 1920.30 With its numerous and soon-to-be-famous doctoral students, assistants, and associated researchers, the institute exerted great influence on several disciplines in the early twentieth century, including psychology, physiology, philosophy, aesthetics, musicology, phonetics, linguistics, physical anthropology, and ethnology.31 Retrospectively, it is celebrated as the hothouse of the Berlin school of Gestalt theory, with scholars such as Kurt Lewin, Max Wertheimer, and Wolfgang Köhler emerging from Stumpf ’s laboratory.32 The Institute of
The Testing of a Hundred Listeners 55
Psychology was also home to the Berlin school of comparative musicology, encompassing Stumpf ’s direct colleagues—Erich Moritz von Hornbostel, George Herzog, Mieczyslaw Kolinski, and Fritz Bose—and his personal network, including Béla Bartók, Jaap Kunst, and Franz Boas.33 Otto Abraham ranks among the less prominent members of these circles, and historians of science have paid little attention to his work. Abraham was trained in the natural sciences (medicine and physics) at the University of Berlin, but after being appointed at the Institute of Psychology in 1896, he began to work mainly in the field of the humanities, though, like Stumpf, adopting some of the experimental methods of the natural sciences. In this sense, Abraham was remarkably representative of Stumpf ’s institute. He published several papers on tone psychology and comparative musicology,34 and from 1900 he additionally served as administrator of the Berlin Phonogramm-Archiv, which was directed first by Stumpf and from 1905 by von Hornbostel.35 It was this wide-ranging academic expertise that led Abraham to his study on absolute tone consciousness, which may also be read as an answer to Stumpf ’s earlier work on Tonpsychologie (the psychology of tone). Stumpf had attributed absolute pitch not to the refinement of the sense of hearing, but to the perfection of the individual memory. Logically enough, he spoke not of “absolute pitch” but of “absolute tone memory.”36 Stumpf was disputing the focus of neuroanatomists, who in the 1860s had started to regard the human memory as a neuronal network in which various different “memory centers” (visual, auditory, taste, or motor centers) and the corresponding nerve pathways are interlinked and can be localized in the human brain. From the outset, this area of research was accompanied by fierce controversies as to whether such specific localization was indeed possible.37 Stumpf did not doubt that there was such a thing as an “acoustic center” in the human brain, and he supposed that scholars would soon attempt to predict the degree of individuals’ musicality from the structure of their brain.38 In addition, he argued that the human memory of tone was not confined to the brain alone. For singers and actors, in Stumpf ’s view, musical experience also inscribed itself into the body, partly in the form of a “laryngeal memory.”39 But for Stumpf, such neuroanatomical attempts at localization could only explain “where one hears and imagines, not how.” He considered it far more important to delve into the individual abilities of memory.40 For Stumpf, the function of “tone memory” was closely related to that of “consciousness of sensation” (Empfindungsbewusstsein), defined as an ongoing feedback loop between sensory perception, memory images, and sensory skills.41 This consciousness of sensation is, according to Stumpf, self- referential in that it consists of a latent awareness of one’s own experiences
56 Sorting and Screening Human Hearers
of the past and the consequent sensory dispositions. “In practice,” Stumpf explains, “we grasp the matter not only by hearing the tone but also by hearing audition itself.”42 In 1890, Stumpf added the telling term “music-infected consciousness” (musik-infiziertes Bewusstsein) to describe a mental state that develops in close dialogue with musical environments, local traditions, and individual habits.43 Finally, Stumpf postulated the existence of a culturally molded musical consciousness that stretched across generations, because “the ear and brain, in their present behavior, archivally preserve the practice of our forebears and the musical history of the millennia become flesh and spirit.”44 I will return to the Berlin Institute of Psychology’s role in early twentieth- century debates on genetics. First, though, it is important to note that Stumpf ’s somewhat dubious talk of the “archival ear” was directed at physiologists (particularly Hermann von Helmholtz’s notion of the “sensation of tone”) and experimental psychologists (specifically those associated with Wilhelm Wundt) who focused merely on the nature of human hearing and disregarded their test subjects’ musical background and habits.45 Stumpf, in contrast, was interested in the culturally nurtured ear and consciousness. He drew his insights from a combination of tone-psychological experiments on individual musical experiences and skills, a growing body of scholarship on non-Western listening habits, and the work of turn-of-the-century composers—Debussy, Ravel, Satie—who adopted non-Western music or experimented with new musical systems.46 In one of his investigations, Stumpf also tested and compared his own tone memory performance with that of three other subjects (two of them musicians) over a period of three months. He concluded that absolute tone memory could be perfected by training and was not expressed exclusively in the ability to identify tones or sight-read.47 It is impossible to miss the influence of Stumpf, in both epistemological and methodological terms, on Otto Abraham’s inquiry into absolute tone consciousness. Quite in line with Stumpf, Abraham stressed that absolute tone consciousness concerned not only perfect pitch but also a range of musical abilities. However, Stumpf ’s notion of tone consciousness designated, precisely, a conscious mental state, whereas Abraham regarded the term “absolute tone consciousness” as misleading because he was also—or rather, primarily—interested in unconscious processes of tone perception. He remarked that he had chosen the title “Absolute Tone Consciousness” for his study only because this phrase had occurred most frequently in his survey “for the relevant activities.”48 Abraham found that not only sensitivity to tone pitch but also all the unconscious associations between the sensory centers in the brain and the larynx could be strengthened to acquire what he preferred to call a “memory of absolute pitch.”49
The Testing of a Hundred Listeners 57
To underline this claim that consciousness (or unconscious perception) was trainable, Abraham referred to contemporary studies that, instead of localizing memory faculties in the human brain, focused on the processes of neural association. He cited the bell model presented by the French neuro anatomist Jean-Martin Charcot to visualize the connections of auditory, visual, and motor impressions and the corresponding associations in the human brain (Figure 2.1).50 Abraham expanded on this scheme to represent the associations between the images of tones and words in the “absolute memory” (Figure 2.2).51 He additionally addressed associations between
Figure 2.1 Charcot’s bell schema. Gilbert Ballet, Le langage intérieur et les diverses formes de l’aphasie (Paris: Félix Alcan, 1888), 7.
Mund
Sprach - u. Gesangscentr.
Hörcentr.
Wort fis
Mund
Sprach - u. Gesangscentr.
Hörcentr.
Wortvorstellung
Tonvorstellung
Centr. and rap Sinnes = vorstellung.
Otto Abraham, “Das absolute Tonbewußtsein: Psychologisch-musikalische Studie,” Sammelbände der internationalen Musikgesellschaft 3, no. 1 (1901): 71.
Ton fis
Wortvorstellung
Centr. and rap Sinnes. vorstellung
Tonvorstellung
Figure 2.2 Abraham’s schema of association.
Wort fis
Ton fis
Ohr
Ohr
The Testing of a Hundred Listeners 59
imagined tones and colors52 and built upon Stumpf ’s studies of the association between tone images and motor memory by coining the perhaps rather unwieldy term “absolute laryngeal muscle consciousness.”53 His questionnaire and reaction tests, Abraham said, had shown that memory of absolute pitch was not (as Stumpf had suggested) hereditary to any great extent, but that there were individual anatomical dispositions, and human hearing took different forms depending on the surrounding musical culture.54 For Abraham, musical cultures meant nothing but education and practice. “No child is born into the world with the ability to identify a standard A,” he stated in his study. “Other than that, the results of my survey convinced me that inheritance has no influence on absolute tone consciousness. More than half of the respondents stated that no family member had absolute tone consciousness or other musical talents.”55 Abraham was arguing clearly and explicitly against a genetic view of musical talent.
Changing Methods: Tests, Surveys, and Statistics Abraham seems to have been the first and only member of Stumpf ’s immediate circle to have challenged the methods of in-laboratory testing, working additionally with surveys and the statistical analysis of big, or at least bigger, data.56 In this respect, the trained physician may have taken inspiration from the medical practice of his day. As indicated in the introduction to this volume, medical scholars at the turn of the twentieth century were starting to question the range and credibility of certain test instruments, test subjects, and test procedures. As a result, hearing tests were applied to larger groups, as batteries of tests, and in different local settings to objectively collect statistics and determine averages. The Viennese otologist Abraham Eitelberg, for example, initiated a comparative study of one hundred patients with hearing impairments in 1886. The study, conducted in the city’s Allgemeine Poliklinik and in Eitelberg’s private practice, used a testing battery of clocks, whispered speech, and a newly designed tuning fork.57 Like Eitelberg’s work, Abraham’s project studied a hundred pairs of ears, but it is a different case again.58 Eitelberg’s survey was looking not for individual skills, but rather for average human capacities or properties. Average capacities were also of interest to experimental psychologists such as Gustav Fechner and Wilhelm Wundt, who became known for their experiments with large numbers of test subjects. Unlike Abraham, they aimed to formulate physical or psychological laws rather than identify individual differences. In 1872, for example, Fechner reported on an “album” (a sort of questionnaire)
60 Sorting and Screening Human Hearers
that he had distributed among 11,842 visitors to the famous exhibition of two competing versions of Hans Holbein the Younger’s Madonna held in Dresden in 1871. Fechner was interested not in which version the visitors considered to be the original, but in the visitors’ sense of beauty. Only 113 visitors responded to Fechner’s request, and he declared the album invalid.59 For a later study on the aesthetic perception of the golden ratio and square shapes, published in 1876, Fechner even worked with 247 test subjects—but this time, he preferred to rely on laboratory tests only, and no questionnaires were distributed.60 In both cases, Fechner was concerned with a general human sense of beauty and not with individually varying aesthetic judgments. Wilhelm Wundt’s large-scale laboratory for experimental psychology in Leipzig also guided numerous test subjects through “long series of experimental observations.”61 This was an approach exemplified by Carl Lorenz’s “Untersuchungen über die Auffassung von Tondistanzen” (Investigations on the apprehension of tone distances), in which eight test subjects underwent a total of 110,000 tests.62 Wundt strongly preferred this method of immediate experimental observation, and even in the field of comparative psychology and anthropology he found questionnaires for missionaries and travelers to be of little use.63 Wundt thought that such questioning should, if applied at all, remain subsidiary to the analysis of test subjects who were physically present. For more general psychological issues, “the system of the enquête, like any statistical survey” was useless, because “it treats duly completed forms and unreliable statements as equal.”64 It is most likely that Abraham drew on the more complex questionnaires employed in neighboring fields such as cultural anthropology to investigate the cultural nurturing of the mind. Anthropologist Francis Galton’s 1883 questionnaire on “the degree on which different persons possess the power of seeing images in their mind’s eye, and of reviving past sensations,” for example, explored auditory mental representations such as “the beat of rain against the window panes, the crack of a whip, a church bell, the hum of bees, the whistle of a railway, the clinking of tea-spoons and saucers, the slam of a door.”65 These questionnaires had been distributed through trusted informants: travelers, scholars, local public servants, teachers, clergymen, or physicians. Abraham’s survey, by contrast, directly addressed its subjects— subjects that did not represent the population as such, but a more elite musical circle.66 Unlike Galton, a strong advocate of eugenics whose questionnaires sought to prove genetic differences between populations, Abraham took his studies as evidence against a genetic interpretation of musical talent. We can only speculate about these various possible sources of inspiration for Abraham’s survey; neither the survey itself nor any of the responses or
The Testing of a Hundred Listeners 61
additional notes by Abraham survive in the relevant Berlin archives. What is certainly remarkable about his survey on absolute tone consciousness is that it broke with the nineteenth-century “traditions of testability” of the ear as such, significantly modifying and expanding the scope of methods for testing musicality.67 Whereas even most of Stumpf ’s tests in tone psychology imitated experimental settings in the laboratory sciences, Abraham’s questionnaire was additionally informed by the methods of large-scale surveys being applied in the emerging social sciences and humanities at the time. The still- new research medium of the survey created very divergent user scenarios: in medicine and experimental psychology, surveys made it possible to compute average values, while in anthropology and in Abraham’s tone psychology, they shaped a new understanding of the varying degrees of human skills. Abraham’s questionnaire thus marks the transition from the laboratory into everyday life, but also from an experimental psychology to a psychology of the unconscious that sought proximity to anthropological field research and was interested in cultural—but not racial—diversity. Abraham transferred the practice of testing hearing from the natural sciences to the applied human and social sciences.
Changing Fields of Application: Testing Otherness When Abraham claimed that children especially, but also adults and even birds, could be trained to achieve absolute tone consciousness,68 he paved the way for far more comprehensive studies on individually and culturally variable forms of musicality. Abraham’s attempts to make his parrot sing Beethoven, for example, were discussed by Stumpf in Die Anfänge der Musik (The Origins of Music) of 1911. Abraham had continued to train his parrot for several years; the bird showed impressive talent with regard to absolute pitch, but lacked an ability for musical transposition.69 Stumpf therefore reasoned that “a bird’s feeling of pleasure—if it is linked to the sounds themselves (for muscular sensations probably also contribute to this) may well be substantially different from that of human beings’ listening to human and avian music.”70 Stumpf concluded that absolute pitch is not an indication of musical ingenuity but an animalistic heritage, and even advocated adding musical transposition to Charles Darwin’s list of the criteria of human musicality.71 Stumpf ’s claims, though, were soon doubted and disproved by ornithologists, while Abraham would be regarded as a pioneer by the later field of bioacoustics.72 As well as influencing the earliest testing procedures in ornithology, Abraham’s study was applied in comparative musicology, in schools, and
62 Sorting and Screening Human Hearers
at conservatories in the years to follow. To begin with comparative musicology, the Phonogramm-Archiv in Berlin, once directed by Stumpf, von Hornbostel, and Abraham, holds a number of still-unidentified boxes of 106 original experimental cylinders (Experimentalwalzen). According to an inventory list, they contain “tests” and “tests for trials with primitives.”73 It is possible that the recordings produced by Abraham for his study of absolute tone consciousness are to be found among these “tests,” while the “tests for trials with primitives” may be related to an essay published by Abraham and Erich von Hornbostel in 1904, “Über die Bedeutung des Phonographen für vergleichende Musikwissenschaft” (On the significance of the phonograph for comparative musicology). The essay mentions acoustic tests for not otherwise specified “exotic musicians.”74 Abraham and von Hornbostel tested these musicians’ sense of intervals and their memory of tones, asked them to tune their own instruments, and confronted them with Western musical intervals and tunings.75 Though problematic in their imperialistic gesture, the tests seem to have been motivated by a critical awareness of the “delicate issue of cultural and racial characteristics” and a rather nuanced interest in finding criteria to measure and appreciate non-Western musical skills.76 Earlier hearing tests in comparative anthropology, such as those applied by Charles Myers in the Cambridge Anthropological Expedition to Torres Straits (see Sebastian Klotz in this volume), had focused on far more general hearing abilities, such as auditory acuity, hearing ranges, and just perceptible tone difference. In contrast, Abraham applied his 1901 study on individual absolute tone consciousness to his subsequent, nuanced research on the hearing of non-Western musicians, and the data collected from European test subjects in 1901 were used for comparison in later studies. Such comparisons are also mentioned in von Hornbostel’s 1910 essay “Über vergleichende akustische und musikpsychologische Untersuchungen” (On comparative studies in acoustics and music psychology).77 In this essay, Hornbostel warns against the idea of sensory racial differences and takes offense at Myers’s comparison of the hearing capacities of different people regardless of their familiarity with the laboratory-like test materials.78 Tests of the ability to repeat test tones, argues Hornbostel, say more about the test subjects’ musical habits—rhythms, melodies, polyphonic structures, and so on—than about their true musical gifts.79 A further test series is reported in Abraham’s 1922 study “Tonometrische Untersuchungen an einem deutschen Volkslied” (Tonometric studies with a German folksong), intended to provide comparative data for Abraham and von Hornbostel’s studies on non-European song.80 Abraham had asked
The Testing of a Hundred Listeners 63
twelve test subjects to sing the “Song of the Germans” (Haydn, with lyrics by von Fallersleben) to test the absolute tone pitches of the sung intervals. The test subjects were requested to sing and whistle the well-known melody by heart. The study was based on intonation tests recorded by Abraham between 1907 and 1914 on Edison cylinders; a few of these recordings have been rediscovered and digitized.81 In it, Abraham found that the intervals sung even by well-trained test subjects with absolute pitch deviated from theoretically defined intervals to a degree that exceeded all the differences (themselves contested) between just intonation and equal temperament.82 Abraham’s investigation thus disproved his own earlier claim that absolute tone consciousness could be deduced as successfully from test subjects’ tone judgments as from their singing or whistling.83 Another line of influence leads from Abraham to the “Measures of Musical Talent” developed by the American psychologist Carl Emil Seashore. After obtaining his doctorate at Yale in 1895, Seashore traveled through Europe and spent some time at Stumpf ’s Institute of Psychology in Berlin, where Abraham was also working.84 On this occasion, Seashore may have learned about Abraham’s strong conviction that musicality was a matter of education. Abraham’s 1901 study on absolute tone consciousness cites several other recently designed programs that proposed to introduce young students first to the tones of a single octave, then to chords in different keys. In addition, Abraham sketched out his own methods of training audition in children aged three to six by confronting them daily with the sound of tuning forks or whistles and inviting them to develop acoustic, visual, and motor associations with the sounds. In Abraham’s immediate circle, these suggestions were taken up by the music educator Hugo Riemann in his work on the “imagination” of tone (Tonvorstellungen). Riemann claimed that composers and listeners must be judged by their imaginative ability to contextualize any single tone within a culturally molded system of scales, tunings, and intervals.85 Whereas Abraham’s and Riemann’s comprehensive testing and training methods never gained a particularly high profile, Seashore’s 1915 manual for assessing musical talents was soon to be applied in numerous educational institutions.86 In line with Abraham, Seashore proposed testing with “pure tones,” produced by tuning fork sounds augmented by a vacuum tube resonator.87 But unlike his predecessor, who believed that almost everyone could be trained to achieve “absolute tone memory,” Seashore pared absolute pitch back to absolute accuracy in discriminating tones and postulated that very few people possessed it: “It would perhaps, be facetious to say that some persons come into my laboratory at the State University of Iowa with absolute pitch, but no one has yet been known to leave with it,” yet that would be “the
64 Sorting and Screening Human Hearers
truth.”88 When designing his talent tests, Seashore therefore first proposed to examine five abilities, which—again, unlike Abraham—he considered not trainable but innate: the sense of pitch, the sense of intensity, the sense of time, the sense of consonance, and tonal memory. In the United States, the Seashore tests were constantly modified. In 1939, the consonance test was replaced by a timbre test, and more up-to-date electronic beat frequency oscillators were used to produce the test tones.89 Seashore’s tests did not remain undisputed, however, first and foremost because his theory of inherent musical abilities was deeply rooted in the eugenics movement of his day—Seashore propagated his measurements as an aid to racial hygiene and the selective breeding of musical talents.90 Seashore’s bracketing out of his test subjects’ musical education also led to the “holism/ atomism” controversy of the early 1940s as to whether musical talent can be divided into five components (as Seashore proposed) or is an inseparable whole that needs to be tested with the aid of musical contents and musical instruments.91 Seashore’s tests were nevertheless applied well into the 1950s at numerous elementary schools, high schools, and colleges of music, as well as in military settings (see Lino Camprubí and Alexandra Hui in this volume).92 Only in the 1960s were they replaced by more sensitive tests with a greater focus on the test subjects’ musical preferences, such as Herbert D. Wing’s Standardized Tests of Musical Intelligence (1961).93 At some colleges, the Seashore test is still in use today.94 The case of Seashore shows how Abraham’s open-ended studies on musicality came to be alienated from their original intention and flowed into an attempt to demarcate racial properties.95 Much closer to Abraham’s broad-based notion of musical talent was the test developed from 1921 onward by Georg Schünemann, a music educator at the Berlin Academy of Music. In the 1920s, Schünemann was also responsible for moving the Institute of Psychology’s Phonogramm-Archiv to the Academy of Music. He was familiar with Abraham’s research.96 Schünemann developed his aptitude test—a “test for the talented,” Begabtenprüfung—at a time when young musicians from the age of thirteen were being sought for the academy’s orchestra classes, which were expanding for military reasons. He designed the psychotechnical examination in collaboration with the aptitude testing office of the Berlin association of professionals (Eignungsprüfstelle des Landesberufsamtes). It is first mentioned in the Academy of Music’s annual report of 1921.97 For the annual report of 1925, Schünemann provided copies of the test sheets, which covered intelligence, perceptiveness, grasp of tones and harmonies, grasp of rhythm, and melody, and were complemented by a series of medical tests.98
The Testing of a Hundred Listeners 65
The test clearly echoes Abraham’s very broad concept of musical talent, especially in the subsections on grasp of tone, “a.) recognizing tones, b.) singing notes that are struck, c.) striking notes that are heard, d.) octave transposition, e.) interval transposition.”99 Schünemann registered his test results at the Academy of Music in a specially established “central office for questions of musical talent” (Zentralstelle für Fragen der Musikbegabung), which, as it turned out, did not experience the anticipated demand.100 The archives of Berlin’s University of the Arts, to which the music college belongs today, provide no evidence for such an office, but the admissions criteria show that Schünemann’s general tests for musical talent were combined with specific tests for each of the music study programs.101 In the subsequent years, Schünemann refined his entrance examinations, focusing increasingly on the sense of rhythm, auditory memory, and musical creativity,102 and attempted to deflect criticisms of the “artificially constructed tests” by arguing that they were part and parcel of a new musical education that built on a strong interrelation between the testing and training of “total musical personalities.”103 Schünemann became especially interested in what he called “music of daily use” (Gebrauchsmusik) and the “new music professions” that specialized in gramophone music, radio music, film music, or electronic music.104 To promote these, in 1929 he founded the Rundfunkversuchsstelle, the radio technology and arts laboratory attached to the Academy of Music, to offer relevant courses for musicians, actors, journalists, educators, and sound engineers. Here, Schünemann continued to test and train his students’ capacities using ever newer procedures.105 Among other things, he commissioned the invention of x-ray sound-film tests to examine and improve singers’ and actors’ phonation.106 Schünemann, then, took Abraham’s understanding of musical trainability in an unexpected direction. His tests must be considered in the context of the heyday of psychophysics in the 1920s and 1930s, when sonic skills were discovered as a potential means to improve the nation’s economic performance. Illustrative cases are the training records for human engineering that were produced around the same time, and not far from the Academy of Music, by the phonetician and director of the Lautabteilung (“sound department”) of the Prussian State Library in Berlin, Wilhelm Doegen. In 1926, Doegen and the Stuttgart psychologist Fritz Giese recorded hearing tests for “psychotechnical aptitude selection,” testing the sense of rhythm and concentration of industrial workers.107 But whereas these tests were made inside and for the Lautabteilung—there is no reliable evidence that they were ever played outside the department—Schünemann’s tests for musical talent continued to be applied at the Academy of Music throughout the following two decades.
66 Sorting and Screening Human Hearers
They may even have facilitated Schünemann’s later (unsolicited, but then opportunistically pursued) career in Nazi Germany, first as director of the Academy of Music, then as director of the music department of the Prussian State Library, and subsequently as vice chairman of the Propaganda Ministry’s department of music arrangements and collaborator in Alfred Rosenberg’s ideological bureau. In one of the poignant ironies of history, Abraham’s disclosure of the cultural constructedness of contemporary physiologists’ tests of absolute pitch ultimately led to the development of Schünemann’s aptitude tests, which went beyond the question of total musical personalities to pursue a disastrous agenda of cultural upheaval.
The Politics of the Testing of Many From a present-day point of view, it might seem somewhat paradoxical that various psychologists and physiologists around 1900 chose to talk about “memory” of absolute pitch, even while arguing that there was no absolute pitch as such, only culturally defined standards of pitches and tones. Indeed, this prompted the Jena physicist Felix Auerbach to quip in 1906 that since “everything in the world is relative,” so are musical tones: they do not relate to each other absolutely either in mathematical terms, or in their production by musical instruments, or in their reproduction in the human mind.108 Auerbach argued that requiring absolute tone consciousness of a composer was as absurd as asking an architect to have an “absolute sea level consciousness,” because being perfectly acclimatized to, say, the Munich sea level would mean never being able to transpose a design for a villa in Dresden.109 In a 1907 essay, Abraham challenged Auerbach’s comment, retorting that matters of sea level hardly affect architectural design, or only if a building from the German city of Jena were to be reconstructed at the top of Mont Blanc, whereas in music, tiny differences in pitch make all the difference.110 To be sure, Abraham conceded, having absolute tone consciousness can be a hindrance for musical transposition and interfere with the “consciousness of intervals,” but in most cases these two musical abilities (today called absolute pitch and relative pitch) if anything support each other. And whereas sea levels generally refer to mean sea level, there is no such point zero in music, nor do those who have absolute tone consciousness need such a relational feature: they simply can or cannot learn to discriminate tiny differences in tone pitches with absolute accuracy.111 Evidently, the two men—one a physicist, one a psychologist—differed in their notions of absoluteness. Whereas Auerbach thought in terms of a
The Testing of a Hundred Listeners 67
universal and objectively definable absoluteness, Abraham was interested in an absoluteness that emerges from a particular musical culture, without any reference point outside that culture. What is more, Abraham did not draw a strict line between perfect pitch and relative pitch, but rather between culturally shaped abilities to make mental comparisons within particular tonalities. Abraham’s 1901 questionnaire in the form of a personality test marks, one might say, a shift of scientific interest in audiology from the nature of hearing in general (with tests deriving from stimulus–response experiments in the natural sciences) to individually varying skills of listening and the conventions of musical culture. For his questionnaire, Abraham was inspired by new qualitative survey methods as used in the human sciences and especially the cultural anthropology of his day. Examining the musical background of the test subjects, Abraham’s study is driven by the desire to take the broadest possible spectrum of a test subject’s musical background into account. In the past three decades or so, historians of science have reduced human agency in experimental settings to but one of many factors of knowledge production (alongside instruments, machines, tools, material objects, and so on).112 Yet if we shift the perspective from a history of experimenting to a history of testing, the subject’s influence on the test result—and therefore the choice of and knowledge about the test subject—becomes crucial. This, at least, is what Abraham may have had in mind when he designed a questionnaire on absolute tone consciousness that could take into account his test subjects’ musical background and skills. In this respect, Abraham’s approach drew on the understanding of “absolute pitch” to be found in the early research of Preyer, Ellis, and Riemann. Their main objective had been to spotlight not the ear itself—as was the case in quantitative tests by physiologists in the early nineteenth century—but culturally imprinted listening skill and its sensitivity to tonal distinctions. But Stumpf and Abraham propounded an even more complex notion of musicality, one that derived from a more general interest in tone consciousness (or, indeed, unconscious tone perception) and related studies on auditory memory in neuroanatomy. In his idea of an auditory memory that could be trained to reach even absolute tone consciousness—or at least a musical memory for keys, intervals, rhythm, harmony, and melody—Stumpf interpreted the neuroanatomical and physiological findings less as inherent facts than as dispositions that had been at least partly culturally nurtured. Venturing an important step further, Abraham then claimed that inheritance had little or no influence on musicality. Abraham’s work nudged the Berlin Institute of Psychology, which aimed to enhance the profile of new fields of psychological research in the human
68 Sorting and Screening Human Hearers
sciences, in a new direction. Whereas his mentor Carl Stumpf conducted pure research in tone psychology, Abraham’s 1901 study was oriented toward application. This was taken up by Abraham himself and Erich von Hornbostel in their comparative musicological studies, and by test developers in musical education, such as Seashore and Schünemann, in their sharper demarcation between absolute and relative pitch and their primary focus on testing relative pitch. Seashore’s notion of “innate” relative pitch partially departed from Abraham’s original intent, while his plea for seemingly objective musical measurements of racial differences did so to a much greater extent. Similarly, Schünemann—much in the spirit of Abraham—paved the way for testing procedures able to account for the fact that their test designs, embedded in the musical and scientific culture of their era, were far from able to paint a neutral portrait of subjects’ musical talents. Schünemann made no secret of using his tests to pursue a practical agenda, aiming to foster a “music-infected consciousness” or “total musical personalities.” But he did so in a political landscape hostile precisely to scholars such as Abraham. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/testing-one-hundred-listeners-tkaczyk.
Notes 1. Otto Abraham, “Das absolute Tonbewußtsein,” 69, available online at https://acoustics.mpiwg-berlin.mpg.de/text/das-absolute-tonbewusstsein. Here and throughout, all translations are my own unless otherwise attributed. Thanks to Kate Sturge for her input. 2. Ibid., 1–2, 5–6, 73. 3. Ibid., 80, 40–44. 4. Ibid., 2. 5. Ibid., 3–4. 6. Ibid., 7–25. 7. Ibid., 40–44, 54–55. On the long tradition of the use of violins for hearing tests, see Emily Dolan in this volume. 8. Ibid., 44–45. 9. Ibid., 45. 10. Carl Stumpf, Die Sprachlaute, 228–33. See Julia Kursell, “A Gray Box.” 11. Moreover, Stumpf ’s later study on speech sounds was based on a test series of phonographic recordings of sung and spoken vowels, tuning forks, and organ pipes that had been produced between 1914 and 1916 with Abraham’s help. They were stored at the Berlin Phonogramm-Archiv. Stumpf, Die Sprachlaute, 230. See the Phonogramm-Archiv’s interwar inventory list, reproduced in Susanne Ziegler, Die Wachszylinder, 21, and the notes to the accompanying CD, 11–15.
The Testing of a Hundred Listeners 69 12. William T. Preyer, Über die Grenzen der Tonwahrnehmung, iii. 13. Ibid., 2–4. 14. Ibid., 29. 15. Ibid., 30. 16. Ibid., 11. 17. Alexander J. Ellis, “On the Sensitiveness of the Ear,” 18, 23. 18. Ibid., 24. 19. See Alexander J. Ellis, Existing Dialectal Pronunciation, iv, Preliminary matter 8*. Ellis discusses his earlier tests in ibid., 1–3 and Preliminary matter 7* and 16*. 20. Ibid., 6–7. Ellis’s attempt to train his informant’s ear can be identified in his instructions to testers: “Try to characterise the nature of the singsong of the speech, underlining as may be, rough, smooth, thick, thin, indistinct, clear, hesitating, glib, whining, drawling, jerking, up and down in pitch, rising in pitch at end, sinking at end, monotonous. Give any singular pronunciations of words not mentioned; and any information respecting your dialect that you will have the kindness to impart.” Ibid., Preliminary matter 25*. 21. This is exemplified by the silk manufacturer Johann Heinrich Scheibler, whose 1834 “tonometer” allowed him to measure pitch and tune keyboard instruments in equal temperament with absolute precision—in the same study, Scheibler presented his mouth harp as the most reliable instrument to train and perfect the musical ear. Scheibler, Der physikalische und musikalische Tonmesser, 77. See Wolfgang Scherer, “Hör-Versuche,” 109. 22. See Myles W. Jackson, Harmonious Triads, 218–30. 23. Hugo Riemann, “Die systematische Ausbildung,” 211. 24. See Diana Deutsch, “Absolute Pitch.” 25. Abraham, “Das absolute Tonbewußstsein,” 3–4, 34–46. 26. Wilhelm Dilthey, Introduction to the Human Sciences. 27. Wilhelm Dilthey and Paul Yorck von Wartenburg, Briefwechsel, 165. 28. Bernhard Siegert, “Das Leben zählt nicht,” 180–81. 29. The first to use this couplet, in the 1870s, was Francis Galton, in “History of Twins.” 30. See Helga Sprung and Lothar Sprung, “ ‘Wir brauchen einen Mann.’ ” 31. Stumpf refers to the wide range of bordering areas (Grenzgebiete) of his institute in Carl Stumpf, “Das psychologische Institut,” 206–7. 32. Mitchell G. Ash, Gestalt Psychology. 33. Dieter Christensen, “Erich M. von Hornbostel”; Sebastian Klotz, Vom tönenden Wirbel; David Trippett, “Carl Stumpf.” 34. Carl Benedict, “Abraham, Salomon Otto.” 35. The Phonogramm-Archiv was initially based at the Institute of Psychology. In 1923, it was transferred to the Academy of Music, and parts of it were moved to the Museum für Völkerkunde (today’s Ethnologisches Museum) in 1934. On further relocations of the recordings during and after the war, see Ziegler, Die Wachszylinder, 19. 36. Carl Stumpf, Tonpsychologie, 1:280. 37. See Anne Harrington, Medicine, Mind and the Double Brain; Michael Hagner, Geist bei der Arbeit; Olaf Breidbach, Materialisierung des Ichs. 38. Stumpf, Tonpsychologie, 1:289, 1:290. 39. Ibid., 1:295. 40. Ibid., 1:290, 1:287–89. 41. Ibid., 1:12–15.
70 Sorting and Screening Human Hearers 42. Ibid., 1:13. 43. Carl Stumpf, “Über Vergleichungen von Tondistanzen,” 445. 44. Stumpf, “Tonpsychologie,” 1:341. 45. See Stumpf ’s discussion of nativism versus empiricism, ibid., 1:95–96. Stumpf ’s comments are part of a debate with the Leipzig experimental psychologist Wilhelm Wundt and his student Carl Lorenz about whether experiments on the ability to judge the middle tone between two tones—choosing between several possible tones with tiny pitch variations of 0.4 c/s—can be performed without taking account of the experimental protocol (the tone systems on which the experiments are based, and the musical abilities of the experimenter and the experimental subject). See Edwin Boring, “Psychology of Controversy”; Alexandra Hui, Psychophysiological Ear, especially 126–36. 46. Hui, Psychophysiological Ear, 125–26, 341. 47. Stumpf, “Tonpsychologie,” 1:310. 48. Abraham, “Das absolute Tonbewußstsein,” 72. 49. Ibid., 66, 73. 50. Ibid., 60. On Charcot’s bell model, see Gilbert Ballet, Le langage intérieur, 7. 51. Abraham, “Das absolute Tonbewußtsein,” 5, 46–47, 70–71. 52. He analyzed the questionnaires in which pitches were particularly frequently associated with certain notions of color. Ibid., 34, 37, and the figure at 38. 53. Ibid., 48–50. 54. Ibid., 70. 55. Ibid., 60, 70. 56. The design of “scientific” questionnaires can be traced further back to early modern traveling culture and apodemica. Justin Stagl, “Vom Dialog zum Fragebogen.” 57. A. Eitelberg, “Vergleichende Gehörsprüfung.” 58. Abraham’s survey may have also been inspired by the statistical inquiries of his former teacher, the Berlin physician and anthropologist Rudolf Virchow, who had developed a questionnaire, if rather a simple one, for his 1876 study on the color of German schoolchildren’s eyes, hair, and skin; Rudolf Virchow, “Gesammtbericht,” 282. Abraham’s doctoral thesis of 1894 lists Virchow as one of his most influential professors. Otto Abraham, “Lebenslauf,” attached to “Über den Erfolg der Künstlichen Frühgeburt” (submitted on December 21, 1894), File “Otto Abraham,” Archiv der Humboldt-Universität zu Berlin. 59. Gustav Theodor Fechner, Bericht, 17. 60. Gustav Theodor Fechner, Vorschule der Ästhetik, 1:194. 61. Wilhelm Wundt, “Ueber psychologische Methoden,” 7. 62. Carl Lorenz, “Untersuchungen.” 63. In an 1883 article on psychological methodology, Wundt discussed the questionnaires described by Charles Darwin in 1872 and by science writer Grant Allen in 1879. Charles Darwin, Expression of the Emotions, 15; Grant Allen, Colour-Sense, 205. See also Wilhelm Wundt, Logik, 494. 64. Wundt, Logik, 494. 65. Francis Galton, Inquiries, 378–79. 66. The first questionnaire used in elite studies in Germany was designed by Sebald R. Steinmetz, “Der Nachwuchs der Begabten.” See Anthony Oberschall, Empirical Social Research, 89.
The Testing of a Hundred Listeners 71 67. On “traditions of testability,” see Edward W. Constant, Origins of the Turbojet Revolution, 22–23. 68. Abraham, “Das absolute Tonbewußtsein,” 1–2, 5–6, 73. 69. Carl Stumpf, Origins of Music, 35–36. 70. Ibid., 36. 71. Ibid., 37. 72. On bioacoustics in the 1920s to 1940s, see Joeri Bruyninckx in this volume. 73. See Ziegler, Die Wachszylinder, 21, and the notes to the accompanying CD, 15. 74. Otto Abraham and Erich Moritz von Hornbostel, “Über die Bedeutung des Phonographen,” 228. 75. Ibid., 228–29. 76. Ibid., 222. 77. Erich M. von Hornbostel, “Über vergleichende akustische und musikpsychologische Untersuchungen,” 153–56. 78. Ibid., 149. 79. Ibid., 155. 80. Otto Abraham, “Tonometrische Untersuchungen,” 1. 81. A selection of the digitized recordings is available at https://acoustics.mpiwg-berlin.mpg. de/text/index-experimental-cylinders-berliner-phonogramm-archiv-ca-1907-1916. 82. In the tests using the “Song of the Germans,” the intonation was most constant for octaves, followed by the fifths, the minor sevenths, the fourths, the minor thirds, the major thirds, the major sixths, the major seconds, and finally (least constantly) the minor second. Abraham, “Tonometrische Untersuchungen,” 3. Listen to the tests at https://acoustics. mpiwg-berlin.mpg.de/node/865; https://acoustics.mpiwg-berlin.mpg.de/node/866. 83. For archival materials related to Abraham’s tests, see https://acoustics.mpiwg-berlin.mpg. de/sets/clusters/testing-hearing/testing-one-hundred-listeners-tkaczyk. 84. Sprung and Sprung, “Wir brauchen einen Mann,” 214. 85. Hugo Riemann, “Neue Beiträge.” See also Riemann, “Ideas for a Study ‘On the Imagination of Tone.’ ” 86. Carl E. Seashore, Measurement of Musical Talent; Seashore, Manual of Instructions. Further material on the Seashore tests can be found at https://acoustics.mpiwg-berlin.mpg.de/sets/ clusters/testing-hearing/testing-one-hundred-listeners-tkaczyk. 87. Seashore, Measurement of Musical Talent, 133. In Psychology of Music of 1938 (182–99), Seashore proposes to analyze instrumental sounds by a Henrici Harmonic Analyzer and reconstruct them synthetically, functionally anticipating the later RCA synthesizer as designed by his student Harry Olson. I am grateful to Martin Brody for sharing his paper “The Enabling Instrument: Milton Babbitt and the RCA Synthesizer,” presented at the workshop “Opening the Doors of the Studio,” June 24–25, 2019, Max Planck Institute for the History of Science, Berlin. 88. Seashore, Measurement of Musical Talent, 9. 89. Joseph G. Saetveit, Don Lewis, and Carl E. Seashore, Revision of the Seashore Measures, 7–9, 9–13. 90. Carl E. Seashore, “Individual and Racial Inheritance of Musical Traits,” esp. 238. I thank Alexander W. Cowan for sharing his paper “Eugenics at the Eastman School: Music Psychology and the Racialization of Musical Talents,” presented at the annual meeting of the American Musicological Society in 2017. Cowan’s paper
72 Sorting and Screening Human Hearers supplies evidence for Seashore’s involvement in several eugenics-oriented projects in the 1920s and 1930s. 91. One of Seashore’s major critics was James Mursell. Mursell, Human Values in Music Education. See Estelle R. Jorgensen, “The Seashore-Mursell Debate.” 92. On the use of the Seashore test in anthropology, industry, and the military, see Saetveit, Lewis, and Seashore, Revision of the Seashore Measures, 47–48. The last version of the tests was Joseph G. Saetveit, Don Lewis, and Carl E. Seashore, Seashore Measure of Musical Talents Manual. 93. Herbert Wing, Standardized Tests of Musical Intelligence. See Edwin Gordon, Introduction, 51–61, esp. 60. 94. The University of Arts in Graz, Austria, for example, asks applicants for the sound engineering program to undergo a 1960s version of the test: http://iem.kug.ac.at/lehre/ elektrotechnik-toningenieur/zulassung/zulassungspruefung-bachelor-elektrotechnik- toningenieur/seashore-test.html. 95. A comparable but different case is the philosopher Max Scheler, who agreed with Abraham and Hornbostel on the nonexistence of biological differences but took their interest in the culturally molded ear to argue for “spiritual” differences between European and non-European cultures. Benjamin Steege, “Between Race and Culture.” 96. Dietmar Schenk, Die Hochschule für Musik, 85–86. 97. Note in “Jahresberichte der Staatlich akademischen Hochschule für Musik Berlin, 1921– 1924,” p. 11, Bibliothek der Universität der Künste, Berlin. 98. Georg Schünemann, “Ueber Musikerziehung,” in “Jahresberichte der Staatlich akademischen Hochschule für Musik Berlin, 1925– 1927,” pp. 18– 27, Bibliothek der Universität der Künste, Berlin. The tests were subsequently published in Georg Schünemann, “Experimentelle und erkenntnistheoretische Musikerziehung.” For additional material related to Schünemann’s tests, see https://acoustics.mpiwg-berlin.mpg. de/sets/clusters/testing-hearing/testing-one-hundred-listeners-tkaczyk. 99. Schünemann, “Experimentelle und erkenntnistheoretische Musikerziehung,” 40. 100. Ibid., 41. 101. “Aufnahmebedingungen für die Staatliche akademische Hochschule für Musik in Berlin, zu Charlottenburg, Fasanenstraße 1, 1927,” Archiv der Universität der Künste, Berlin, Bestand 1, D27 and D28. Entrance exams have existed at Berlin’s Academy of Music, now part of the University of the Arts, since its foundation in 1882, but we have no details of the nature of these early tests. “Auszug aus dem Statut von 1882,” p. 7, Archiv der Universität der Künste, Berlin, Bestand 1, D26. 102. See Eberhard Preußner, “Musikalische Eignungsprüfungen,” 113. 103. Schünemann, “Experimentelle und erkenntnistheoretische Musikerziehung,” 41. 104. Georg Schünemann, “Neue Musikberufe,” 40. 105. Schenk, Die Hochschule für Musik, 265, 267. 106. “Das sprechende Herz.” For the experiments carried out by Schünemann, with the radiologists Dieter Gottheiner and Heinz Große, on modes of synchronizing x-ray photography and sound recording between 1930 and 1933, see the documents held in the Archiv der Universität der Künste, Berlin, Bestand 1b/1, 1b/7, 1b/8, and 1b/28. 107. See Viktoria Tkaczyk, “Archival Traces of Applied Research.” 108. Felix Auerbach, “Das absolute Tonbewußtsein,” 106, 109. 109. Ibid., 110–11.
The Testing of a Hundred Listeners 73 110. Otto Abraham, “Das absolute Tonbewußstsein und die Musik,” 486. 111. Ibid., 489–90. 112. A turning point was Jim Johnson [i.e., Bruno Latour], “Mixing Humans and Nonhumans Together.”
References Abraham, Otto. “Das absolute Tonbewußtsein: Psychologisch- musikalische Studie.” Sammelbände der internationalen Musikgesellschaft 3, no. 1 (1901): 1–86. Abraham, Otto. “Das absolute Tonbewußtsein und die Musik.” Sammelbände der Internationalen Musikgesellschaft 8 (1907): 486–91. Abraham, Otto. “Tonometrische Untersuchungen an einem deutschen Volkslied.” Psychologische Forschung: Zeitschrift für Psychologie und ihre Grenzwissenschaften 4 (1923): 1–22. Abraham, Otto, and Erich Moritz von Hornbostel. “Über die Bedeutung des Phonographen für vergleichende Musikwissenschaft.” Zeitschrift für Ethnologie 36, no. 2 (1904): 222–36. Allen, Grant. The Colour- Sense: Its Origin and Development. An Essay in Comparative Psychology. Boston: Houghton, Osgood, & Co., 1879. Ash, Mitchell G. Gestalt Psychology in German Culture, 1870–1967: Holism and the Quest for Objectivity. Cambridge: Cambridge University Press, 1995. Auerbach, Felix. “Das absolute Tonbewußtsein und die Musik.” Sammelbände der Internationalen Musikgesellschaft 8, no. 1 (1906): 105–12. Ballet, Gilbert. Le langage intérieur et les diverses formes de l’aphasie. 2nd, rev. ed. Paris: Félix Alcan, 1888. Benedict, Carl. “Abraham, Salomon Otto.” Neue Deutsche Biographie 1 (1953): 24. Boring, Edwin G. “The Psychology of Controversy.” Psychological Review 36 (1929): 97–121. Breidbach, Olaf. Die Materialisierung des Ichs: Zur Geschichte der Hirnforschung im 19. und 20. Jahrhundert. Frankfurt am Main: Suhrkamp, 1997. Christensen, Dieter. “Erich M. von Hornbostel, Carl Stumpf, and the Institutionalization of Comparative Musicology.” In Comparative Musicology and Anthropology of Music: Essays on the History of Ethnomusicology, edited by Bruno Nettl and Philip Vilas Bohlmann, 201–9. Chicago: University of Chicago Press, 1991. Constant, Edward W. The Origins of the Turbojet Revolution. Baltimore, MD: Johns Hopkins University Press, 1980. Darwin, Charles. The Expression of the Emotions in Man and Animals. London: John Murray, 1872. Deutsch, Diana. “Absolute Pitch.” In The Psychology of Music, 3rd ed., 141–82. London: Academic Press, 2013. Dilthey, Wilhelm. Introduction to the Human Sciences. Translated by Michael Neville. Princeton, NJ: Princeton University Press, [1883] 1991. Dilthey, Wilhelm, and Graf Paul Yorck von Wartenburg. Briefwechsel zwischen Wilhelm Dilthey und dem Grafen Paul Yorck v. Wartenburg, 1877–1897. Edited by Erich Rothacker. Halle an der Saale: Max Niemeyer, 1923. Eitelberg, A. “Vergleichende Gehörsprüfung an 100 Individuen mittels Stimmgabeln, Uhr und Flüstersprache.” Zeitschrift für Ohrenheilkunde 16 (1886): 31–52, 87–114. Ellis, Alexander J. Existing Dialectal as Compared with West Saxon Pronunciation. Part 5 of On Early English Pronunciation. London: Trübner & Co., 1889.
74 Sorting and Screening Human Hearers Ellis, Alexander J. “On the Sensitiveness of the Ear to Pitch and Change of Pitch in Music.” Proceedings of the Musical Association 3 (1876): 1–31. Fechner, Gustav Theodor. Bericht über das auf der Dresdner Holbein-Ausstellung ausgelegte Album. Mit einigen persönlichen Nebenbemerkungen. Leipzig: Breitkopf & Härtel, 1872. Fechner, Gustav Theodor. Vorschule der Ästhetik. 2 vols. Leipzig: Breitkopf & Härtel, 1876. Galton, Francis. “The History of Twins, as a Criterion of the Relative Powers of Nature and Nurture.” Journal of the Anthropological Institute of Great Britain and Ireland 5 (1876): 391–406. Galton, Francis. Inquiries into Human Faculty and Its Development. London: MacMillan, 1883. Gordon, Edwin. Introduction to Research and the Psychology of Music. Chicago: GIA, 1998. Hagner, Michael. Der Geist bei der Arbeit: Historische Untersuchungen zur Hirnforschung. Göttingen: Wallstein, 2006. Harrington, Anne. Medicine, Mind, and the Double Brain: A Study in Nineteenth-Century Thought. Princeton, NJ: Princeton University Press, 1987. Hornbostel, Erich Moritz von. “Über vergleichende akustische und musikpsychologische Untersuchungen.” Beiträge zur Akustik und Musikwissenschaft 5 (1910): 143–67. Hui, Alexandra. The Psychophysiological Ear: Musical Experiments, Experimental Sounds 1840– 1910. Cambridge, MA: MIT Press, 2013. Jackson, Myles W. Harmonious Triads: Physicists, Musicians, and Instrument Makers in Nineteenth-Century Germany. Cambridge, MA: MIT Press, 2006. Johnson, Jim [i.e., Bruno Latour]. “Mixing Humans and Nonhumans Together: The Sociology of a Door-Closer.” Social Problems 35 (1988): 298–310. Jorgensen, Estelle R. “The Seashore-Mursell Debate on the Psychology of Music Revisited.” In Advances in Social-Psychology and Music Education Research, edited by Patrice Madura Ward-Steinman, 63–76. Farnham, UK: Ashgate, 2011. Klotz, Sebastian, ed. Vom tönenden Wirbel menschlichen Tuns: Erich M. von Hornbostel als Gestaltpsychologe, Archivar und Musikwissenschaftler. Berlin: Schibri-Verlag, 1998. Kursell, Julia. “A Gray Box: The Phonograph in Laboratory Experiments and Fieldwork, 1900–1920.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 176–97. Oxford: Oxford University Press, 2012. Lorenz, Carl. “Untersuchungen über die Auffassung von Tondistanzen.” Philosophische Studien 6 (1891): 27–103. Mursell, James. Human Values in Music Education. New York: Silver Burdett, 1934. Oberschall, Anthony. Empirical Social Research in Germany, 1848–1914. Paris: Mouton, 1965. Preußner, Eberhard. “Musikalische Eignungsprüfungen.” Die Musikpflege 1, no. 6 (1930): 112–14. Preyer, William T. Über die Grenzen der Tonwahrnehmung. Jena: H. Dufft, 1876. Riemann, Hugo. “Ideas for a Study ‘On the Imagination of Tone’” (1914). Translated by Robert Wason and Elizabeth West Marvin. Journal of Music Theory 36, no. 1 (1992): 81–117. Riemann, Hugo. “Neue Beiträge zu einer Lehre von den Tonvorstellungen.” Jahrbuch der Musikbibliothek Peters 23 (1916): 1–22. Riemann, Hugo. “Die systematische Ausbildung des musikalischen Gehörs.” Der Klavier- Lehrer: Musik-paedogogische Zeitschrift 18, no. 15 (1882): 209–12. Saetveit, Joseph G., Don Lewis, and Carl E. Seashore. Revision of the Seashore Measures of Musical Talents. Iowa City: University of Iowa Press, 1940. Saetveit, Joseph G., Don Lewis, and Carl Emil Seashore. Seashore Measure of Musical Talents Manual. 2nd ed. New York: Psychological Corporation, 1960. Scheibler, Johann Heinrich. Der physikalische und musikalische Tonmesser welcher durch den Pendel, dem Auge sichtbar, die absoluten Vibrationen der Töne, der Haupt-Gattungen von Combinations-Tönen, so wie die schärfste Genauigkeit gleichschwebender und mathematischer Accorde beweist. Essen: Bädeker, 1834.
The Testing of a Hundred Listeners 75 Schenk, Dietmar. Die Hochschule für Musik zu Berlin: Preußens Konservatorium zwischen romantischem Klassizismus und neuer Musik, 1869–1932/33. Stuttgart: Franz Steiner, 2004. Scherer, Wolfgang. “Hör- Versuche: Die experimentelle Decodierung des musikalischen Hörens um 1900.” In Armaturen der Sinne: Literarische und technische Medien 1870 bis 1920, edited by Jochen Hörisch and Michael Wetzel, 107–36. Munich: Wilhelm Fink, 1990. Schünemann, Georg. “Experimentelle und erkenntnistheoretische Musikerziehung.” In Musik in Volk, Schule und Kirche: Vorträge der V. Reichsschulmusikwoche in Darmstadt (11–16. Oktober 1926), edited by Zentralinstitut für Erziehung und Unterricht, 33–42. Leipzig: Quelle & Meyer, 1927. Schünemann, Georg. “Neue Musikberufe.” Die Musikpflege 1, no. 1 (1930): 40–43. Seashore, Carl E. “Individual and Racial Inheritance of Musical Traits.” In Eugenics, Genetics and the Family, vol. 1 of Scientific Papers of the Second International Congress of Eugenics, 231–38. Baltimore: Williams & Wilkins Company, 1923. Seashore, Carl Emil. Manual of Instructions and Interpretations for Measures of Musical Talent. New York: Columbia Gramophone Company (Educational Department), 1919. Seashore, Carl Emil. The Measurement of Musical Talent. Reprinted from The Musical Quarterly, 1915. New York: G. Schirmer, 1919. Seashore, Carl Emil. Psychology of Music. New York, London: McGraw-Hill, 1938. Siegert, Bernhard. “Das Leben zählt nicht: Natur-und Geisteswissenschaften bei Dilthey aus mediengeschichtlicher Sicht.” In Medien: Dreizehn Vorträge zur Medienkultur, edited by Claus Pias, 161–82. Weimar: Verlag und Datenbank für Geisteswissenschaften, 1999. “Das sprechende Herz.” Vossische Zeitung, January 25, 1933: 3. Sprung, Helga, and Lothar Sprung. “‘Wir brauchen einen Mann, welcher heimisch ist in . . . der experimentellen Psychologie’—Carl Stumpf in seiner Berliner Zeit (1894–1936).” In Zur Geschichte der Psychologie, 2nd, rev. ed., edited by Lothar Sprung and Wolfgang Schönpflug, 201–25. Frankfurt am Main: Peter Lang, 2003. Stagl, Justin. “Vom Dialog zum Fragebogen: Miszellen zur Geschichte der Umfrage.” Kölner Zeitschrift für Soziologie und Sozialpsychologie 31, no. 3 (1979): 611–38. Steege, Benjamin. “Between Race and Culture: Hearing Japanese Music in Berlin.” History of Humanities 2, no. 2 (2017): 361–74. Steinmetz, Sebald Rudolf. “Der Nachwuchs der Begabten.” Zeitschrift für Sozialwissenschaft 7 (1904): 1–25. Stumpf, Carl. The Origins of Music. Edited and translated by David Trippett. Oxford: Oxford University Press, [1911] 2012. Stumpf, Carl. “Das psychologische Institut.” In Geschichte der Königlichen Friedrich- Wilhelms-Universität zu Berlin, edited by Max Lenz, 202–7. Halle an der Saale: Verlag der Buchhandlung des Waisenhauses, 1910. Stumpf, Carl. Die Sprachlaute: Experimentell-phonetische Untersuchungen nebst einem Anhang über Instrumentenklänge. Berlin: Springer, 1926. Stumpf, Carl. Tonpsychologie. 2 vols. Leipzig: Hirzel, 1883–1890. Stumpf, Carl. “Über Vergleichungen von Tondistanzen.” Zeitschrift für Psychologie und Physiologie der Sinnesorgane 1 (1890): 419–62. Tkaczyk, Viktoria. “Archival Traces of Applied Research: Psychotechnics and Language Planning in Interwar Germany.” In “Listening to the Archive: Sound Data in the Humanities and the Sciences,” edited by Carolyn Birdsall and Viktoria Tkaczyk. Special issue of Technology and Culture 60, no. 2, Supplement (2019), 64–95. Trippett, David. “Carl Stumpf: A Reluctant Revolutionary.” In Carl Stumpf, The Origins of Music, edited and translated by David Trippett, 17–30. Oxford: Oxford University Press, 2012. Virchow, Rudolf. “Gesammtbericht über die von der deutschen anthropologischen Gesellschaft veranlassten Erhebungen über die Farbe der Haut, der Haare und der Augen der Schulkinder in Deutschland.” Archiv für Anthropologie 16 (1886): 275–466.
76 Sorting and Screening Human Hearers Wing, Herbert. Standardized Tests of Musical Intelligence (1 sound tape reel, 1 scoring key, and 1 answer sheet). Windsor, UK: NFER Publishing Company, 1961. Wundt, Wilhelm. Methodenlehre, vol. 2 of Logik: Eine Untersuchung der Principien der Erkenntniss und der Methoden wissenschaftlicher Forschung. Stuttgart: Ferdinand Enke, 1883. Wundt, Wilhelm. “Ueber psychologische Methoden.” Philosophische Studien 1 (1883): 1–38. Ziegler, Susanne. Die Wachszylinder des Berliner Phonogramm- Archiv. Berlin: Staatliche Museen zu Berlin, 2006.
3 Murray Island Versus Aberdeenshire Contextualizing the Cross-Cultural Hearing Tests of the Cambridge Anthropological Expedition to Torres Straits, 1898–1899 Sebastian Klotz
When the Cambridge biologist Alfred Cort Haddon visited Thursday Island in the Torres Straits in 1888 to study the war dances and life of the “savages,” he could not have anticipated that he would return to the region ten years later with a full expedition. The Cambridge Anthropological Expedition to Torres Straits (CAETS), which began in April 1898, pursued the anthropological study of the inhabitants of the Torres Strait Islands, north of Australia.1 Its members included, besides Haddon, the anthropologists Charles G. Seligman and William Halse Rivers Rivers, the psychologist Charles Myers, the ethnologists William McDougall and Anthony Wilkin, and the linguist Sidney Ray. Their combined expertise ranged from medicine, psychology, physical anthropology, and biology to ethnology, linguistics, and music. The CAETS was a key endeavor in the formative years of these disciplines. It foreshadowed the rise of multifactor, large-scale ethnopsychological tests in different cultural environments following a systematic, theory-driven design, at a time when the core tenets of these diverse disciplines were being negotiated. The hearing tests discussed in this chapter thus epitomize the volatile dynamics of the modern sciences and humanities. They form a representative arena of inquiry that connects university laboratories to fieldwork in Oceania and Scotland, mass screenings in science museums to the soundproof rooms set up at World Exhibitions, and the specialist scientific communities of Britain to those of Germany and the United States. Although the research motivations and outcomes of the CAETS have been studied extensively, especially on the occasion of the expedition’s centenary,2 its hearing tests deserve more attention, as a practice that organized interactions, induced new social roles, and shaped experimental standards in very specific ways. In particular, the CAETS has remained almost unnoticed in the history of music psychology.3 My chapter shows that hearing tests—and the portable Sebastian Klotz, Murray Island Versus Aberdeenshire In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0004.
78 Sorting and Screening Human Hearers
devices they employed, such as the Galton whistle—mediated a whole series of translations: translations from the laboratory to the field; from anthropology to psychology and eventually to sociology; from discursive paradigms of genealogy and racial inequality to quantitative measurements; from ethnically monolithic to comparative, age-and gender-specific tests; from neutral scientific observers and “standard subjects” to reagents in the experiment; from apparently static, involuntary stimulus responses to adaptive learning processes; and from episodic, individual performances to systematic mass screenings, in combination with the application of psychometry and statistics. I seek to unravel the multiple discursive and methodological agendas of hearing tests during the CAETS and in follow-up surveys, focusing on the accounts in volume II, part II of the expedition’s six-volume report (published between 1901 and 1935). This material is juxtaposed with German research traditions in early ethnology, empirical psychology, and tone psychology, and with a current of U.S. psychology that adopted and modified the CAETS methodology of hearing tests. Drawing on critical epistemologies of hearing, I explore the research expectations and ideologies around hearing in Western and non-Western contexts. This perspective is committed to understanding the actual practices that researchers carried out in the field, where routines of observation, collection, and analysis functioned differently from an in-the- classroom context and often called for contingent, ad hoc decisions. Of particular interest are the presumed boundaries between different social, cultural, and ethnic groups that hearing tests attempted to reinforce. Imagined boundaries dividing the reaction capabilities of Westerners and non-Westerners actively constructed the field, informing cultural and moral geographies. As I will show, they also forged complicities between testers and test subjects and thus begin to cast light on how exactly “results” translated into “knowledge.” Assumed distinctions pose a problem when they prove to be wrong—in the case of hearing, no persuasive connection could be drawn between hearing acuity and intelligence, and the results crippled the attempt to locate “savagery” and consolidate cultural difference. These turn-of-the- century hearing tests, pursuing the seemingly simple and entirely plausible task of identifying hearing acuities and thresholds, were complicated by a variety of factors that turned out to be extremely difficult to handle on the ground.
The CAETS Hearing Tests The region chosen by Charles Myers for his tests was no virgin territory—no ideal, static island world where measurements would reveal a lived primitivism
Murray Island Versus Aberdeenshire 79
untarnished by commerce or modernity. On the contrary, by 1898 Oceania had witnessed over a century of intense commercial, artistic, and missionary encounters with a variety of international actors.4 It had served as a reference point for elaborate narratives of race as early as the eighteenth century5 and now emerged as a hub for field research by major new players in the formative years of professional zoology, ethnology (Völkerkunde), anthropology, and comparative musicology. Germany, the Netherlands, Britain, France, and the United States had all staked scientific claims in the region,6 but Britain was the only power that initiated an expedition project of such a size and duration. During the CAETS, the islanders of the Torres Straits and British Borneo became subjects of extensive physiological and psychological measurements across all the modalities of the senses.7 Although the hearing test results reported by Charles Myers in the form of measurement tables seem transparent, objective, and precise, in fact, they were implicated in complex narratives of scientific racism, developmentalism, and evolutionism.8 Born in 1873, Myers was a graduate in natural sciences from Cambridge University and a physician who ventured into psychology, a rare choice at a time when the field still had a poor reputation at Cambridge.9 The hearing tests he performed and presented were, as I will show, ill suited to reaching definite conclusions on either developmentalism or evolutionism. Instead, they exemplify a culture of measurement that represented “objectivity” at its best, yet offered problematic evidence that was resistant to easy academic acceptance—especially in the field of human hearing and tonal consciousness, which was only gradually admitting the cultural constructedness of hearing practices. Myers himself identified this problem. He carried out a follow-up survey among test subjects in Scotland in the hope of establishing a suitable standard against which he could measure the data obtained in the Torres Straits. Finding no substantial difference between the two surveys, Myers hesitated to actively refer to theories of race and heredity in his discussion of the results.10 This was no minor flaw—at stake were both the validity of the results in the domain of hearing and the very justification of the CAETS. As it turned out, the hearing tests not only tested hearing acuities among specific actors in the Torres Straits and in Aberdeenshire but also tested the convictions that informed the CAETS. The CAETS researchers were unable to mend the growing rift between reflexive anthropologies and the peculiar logic of cultures of measurement. The CAETS hearing tests, perhaps for the very first time, fully exposed a problematic relationship that continues to permeate anthropology and cross-cultural psychology today. Initial steps in ethnographic fieldwork had been undertaken by British scientists much closer to home, in the context of eugenics and psychometric
80 Sorting and Screening Human Hearers
measurements in the mid-1880s. Francis Galton opened an anthropometric laboratory at the International Health Exhibition in 1888, with some ten thousand people participating. In exchange for participation and a small fee, they received a chart with their measurements.11 At this time, Galton had a device built from his own design for measuring reaction time: a pendulum attached to a thread whose oscillations were measured.12 Later, he founded a permanent Anthropometric Laboratory at the London Science Museum.13 The range of hearing tests performed by Galton and his colleagues was wide, from simple reaction-time measurements to complex tasks including overtone recognition, the least noticeable difference in the intensity of sound at three degrees of loudness, and the accuracy of detecting sound direction.14 Large-scale testing of, if possible, every human capacity had thus already become established as an accepted practice, in Galton’s case serving to support eugenic theories and differential psychology. Eugenics implemented a kind of “biological engineering,” and hearing tests were one way—bringing together biology, psychology, and anthropology—of accessing individual capabilities to assist eugenics in its project of “racial improvement.”15 Galton measured reaction times in simple monomodal tasks and related the results to intelligence against the backdrop of his eugenic ideas, arguing that faster reaction times were proof of higher intelligence. His “Galton whistle” makes it possible to modulate a tone across a wide-frequency band into the ultrasonic realm and test differential hearing ability. Through adjustable tubes, the whistle can emit sounds that reach the hearing threshold; tube length and air pressure modulate the frequency. The whistle, crucial to a workable application of psychometrics, was initially developed for animal testing. Galton’s whistle was one of the devices Myers used during the CAETS, but Galton also consolidated another important prerequisite for Myers’s large- scale survey: statistics. In Galton’s writings, statistics emerged at the interface of anthropometry, plant genetics, kinship, biometrics, social psychology, and mathematics (curve formulae), yielding theoretically informed concepts of variation and correlation. The CAETS hearing test tables display averages and mean variations. An utterly mechanical device thus worked in concert with statistics to record the data of large portions of the population, both at home and abroad, and achieve data-based evidence that could be generalized within eugenic and psychometric discourses. The nation became an “object of scientific regulation and expertise.”16 From 1891 to 1893, the CAETS’s initiator, Haddon—a friend of Galton’s17— had himself undertaken extensive anthropometric measurements in Ireland. He was searching for remote, “primitive” communities to whom he could apply comparative anatomy measurements. Myers’s survey in Scotland,
Murray Island Versus Aberdeenshire 81
intended to complement his Torres Straits hearing tests, followed the same pattern by targeting a rural community on the British mainland that he considered to be unaffected by modern urban life. The aim of the expedition led by Haddon was to understand and document the vanishing local cultures of the British Isles by means of an ethnographic survey and write a “racial history of the United Kingdom.”18 We might regard this as the background against which the CAETS set off in pursuit of a global data basis to test the validity of evolutionary and racial narratives that would confirm the Victorian hypothesis of human beings’ natural inequality.19 Testing hearing seemed to offer an excellent means to that end. In 1890, the German philosopher and psychologist Carl Stumpf had anticipated that examining hearing would be a useful comparative strategy, since it mobilized all the auxiliary disciplines of psychology proper: self-observation, statistics, physiological facts and hypotheses, the comparison of peoples across time, biographical data, and more.20 In addition, argued Stumpf, hearing tests promised to supply comparable data across all cultures due to the universality of music. This avoided the problems of language-based interaction, which required the ethnographer to speak the language of each culture under study— an ideal impossible to achieve. British and American publications in anthropology, ethnology, and psychology show that these developments in Germany were closely observed and much discussed. Apart from the work of Stumpf, Wilhelm Wundt had attracted a large following, with numerous doctoral students who spread experimental practices across Britain and the United States. Among these were James McKeen Cattell, who later studied with Galton, and Edward B. Titchener, who translated Wundt’s seminal work into English as Principles of Physiological Psychology in 1904. Influenced by evolutionism, Wundt had also turned to the collective aspects of psychology in his nonempirical project Völkerpsychologie (“psychology of peoples”). His students expanded Völkerpsychologie into anthropological and ethnographic field research, as in the case of the CAETS, following Wundt’s protocol in preferring simple psychological judgments that would connect to physiological indicators and immediate sensations.21 But acceptance of these practices in Cambridge was slow, making it rather surprising that the CAETS members were so determined to advance experimental research. It was only in 1891 that the board of the university provided a small budget for apparatuses in psychophysics, and even in 1899 Cambridge had neither a laboratory nor a reader in psychology.22 The CAETS members may have been exposed to experimental practices during their medical training, giving them a head start over those trained in anthropology and psychology.
82 Sorting and Screening Human Hearers
British academics of the period shared the assumptions of evolutionism and scientific racism, epitomized by “systematic efforts made in various branches of natural history . . . to theorize physical differences between human groups as innate, morally and intellectually determinant, and possibly original.”23 In this framework, primitive people were thought to have greater hearing acuity but generally lesser intelligence.24 According to Herbert Spencer, the hierarchy of human faculties ascended from reflexes, instincts, and habits to memory, the imagination, and ultimately reason.25 An uneven distribution of mental energy prevented primitive people from fully developing intellectual skills—too much of their mental capacity was absorbed by the lower faculties.26 Hearing and reaction times in general were considered such lower faculties, in which non-Westerners would therefore perform better than Westerners,27 though Myers himself was unable to find persuasive evidence of this: previous observations had not unanimously proved any superiority in non-Western hearing abilities.28 Counting hearing as one of the lesser faculties collided with the notion propounded in German idealist aesthetics that the sense of hearing quintessentially represents reason and human subjectivity and has great epistemic power. This tension between the German and British rationales, and apparently between the positions of Galton and Spencer concerning hearing and intelligence, would play out in an arena parallel to hearing tests, the field of tone psychology. The founder of the Institute of Psychology at Berlin University, Carl Stumpf, and his circle explored tonal consciousness and the perception of timbre in ways far exceeding the functionality of an auditory stimulus as used in elementary hearing tests. This was a path not followed by the British, whose background lay in medicine, biology, neurology, anthropology, and psychology rather than in philosophy and phenomenology. The contrast becomes clear in Stumpf ’s founding of a phonogram archive in Berlin and his institute’s systematic research in psychoacoustics and comparative musicology.29 In the British context, the psychophysiological agenda of CAETS seems to bring together various turn-of-the-century lines of thinking—early anthropometry, psychophysics, and differential psychology, the latter comprising both field research in the non-Western world and Galton’s early anthropometric mass screenings in London.30 In the case of auditory measurements, the professionalization of otology was also relevant, providing high-precision instruments and improved understanding both of the physiological processes surrounding hearing and of possible remedies for hearing deficits (see the Introduction, Mara Mills, and Viktoria Tkaczyk in this volume). In addition, institutions such as museums and exhibitions (to be discussed later) emerged
Murray Island Versus Aberdeenshire 83
as new cultural brokers, while instrument makers and lab outfitters contributed the necessary equipment. However, the physical locale that the CAETS planned to explore had little in common with a professional psychological and anthropometric laboratory. To sum up the most important setbacks: some islands were populated mainly by foreigners; locals from Murray Island turned up for tests only irregularly and in greatly varying numbers; testing equipment failed en masse under the adverse conditions; and many Murray Islanders had impaired hearing due to extensive diving for seashells.31 In fact, often the hearing tests became otological treatments for those suffering from hearing loss. CAETS member William MacDougall lost interest in the expedition altogether.32 The classification “primitive” also proved problematic. The discourse of the primitive and “the mental and sensory capacities of the primitive people” clearly informed the CAETS from the outset,33 but as expedition members complained, the islanders were undergoing “rapid modification.”34 Subverting the coherence of the primitive as a social and anthropological concept, that process pointed both to cultural adaptability and to the individual differences that Myers wished to reveal. “Primitive” ethnic and social aggregates were far from monodimensional, he argued; they showed great variety and their own historic dynamics. This remarkable insight, contesting the contemporary discourse about primitivism, was reflected in the CAETS researchers’ refusal to draw clear-cut conclusions in support of primitivist and racial paradigms. The hearing tests administered by Charles Myers—who, apart from his formal qualifications, was also a gifted musician35—relied on an extensive array of equipment: an aural speculum, a watch, a stopwatch, a vertical tele scopic tube of adjustable length (to measure the hearing discrimination of various intensities, depending on the impact noise of a ball dropping inside the tube), Politzer’s acoumeter or Hörmesser, Runne’s clock, a Galton whistle, and tuning forks mounted on resonance chambers, played with a violin bow.36 The Hörmesser acoumeter, invented by physician and otologist Adam Politzer in the 1870s, was held between the experimenter’s fingers and could produce a controlled acoustic stimulus for testing hearing, either by air or bone conduction.37 Runne’s clock, which appears to have been similar to a stopwatch, emitted a regular number of ticks per second and could easily be halted. Examining the “delicacy of hearing” had become an accepted practice in British anthropology with the 1892 edition of the Notes and Queries on Anthropology, which featured a separate section on hearing written by John George Garson.38 Tests were to be performed either “by words in speaking” or with a watch, and followed a choreography of observer and subject, monaural and binaural tests, open and closed eyes.39 Garson’s survey addressed
84 Sorting and Screening Human Hearers
“Acuteness of sight. Delicacy of hearing. Aptitude to music.”40 Myers considerably refined these suggestions and seems to have firmly settled hearing and music within the curriculum of British anthropological fieldwork. During the CAETS, he aimed to examine auditory acuity, the upper tone limit, and the smallest tone difference.41 Due to the difficult circumstances in the natural environment, Myers developed some ingenious experimental arrangements that involved a co- researcher and extensive self-testing. It turned out that only the stopwatch and Runne’s clock were of any value for testing hearing acuity. During a night test series undertaken by Rivers at Mabuiag, the Politzer acoumeter proved useful only because environmental distractions were reduced. Varying environmental factors impacted on the reliability of the auditory acuity results, so that Myers could achieve only rough estimates.42 In the absence of accepted norms, Myers tells us, he implemented either himself or Rivers as a standard against which to measure the islanders’ relative auditory acuity. Illusory sensations (a reported tick where there was none) are included in a mean- based threshold calculation. Myers even considers the assumed self-image of islanders who may feel ill-equipped for the test due to their apparent deafness. Myers is aware that more active involvement and dedication on the part of the islanders would have yielded better results. The first data chart, with the results of hearing acuity tests at varying distances from Runne’s clock (Figure 3.1), shows that Myers generally performs better than the young Murray Islanders, but that there are cases of identical acuities, expressed in an identical distance from the stimulus. Myers also seems to have a much higher acuity than Rivers: whereas Myers can hear a clock tick from 5.00 m, Rivers needs to stand much closer to the clock, 1.75 m. Some local individuals were tested twice because, as Myers observed, hearing acuity varies from day to day.43
The Oceania Hearing Tests and the Aberdeen Campaign In the conclusion of his contribution on “Hearing,” Myers shows his reflexive turn of mind. Instead of simply presenting the data in the tables, which make up an impressive set, he asks whether his own and his colleagues’ results can pass as a European standard, and why the fairly good performance of young islanders (boys and girls) diminishes during adulthood. He reflects on his own state of health, affected by the expedition, which has compromised his sense of hearing so that it can hardly be accepted as a standard—but still
Taken from Charles Samuel Myers, “Hearing,” in Physiology and Psychology, edited by Alfred Cort Haddon (Cambridge: Cambridge University Press, 1903), 146.
Figure 3.1 Top section of the auditory acuity data chart, giving the distance from Runne’s clock for the right and left ear, respectively. Charles Samuel Myers himself (C. S. M.) and his colleague William Halse Rivers Rivers (W. H. R. R.) are “standard-observers.” The chart is a first exercise in gender-and age-specific cross-cultural experimental psychology.
86 Sorting and Screening Human Hearers
supposes that his and his colleagues’ acuity is the same in the Torres Straits as in Europe. Despite the many volatile factors involved, he concludes that “the general auditory acuity of the islanders in the Torres Straits is inferior to that of Europeans.”44 Young islanders are on a par with the expedition members, and it is the inferior results of the adults, due to damaged eardrums, that leads him to this conclusion. What the table does not show is Myers’s pondering of the circumstances and his unease in implementing himself and Rivers as a standard. In determining the capacity to hear the highest audible tone, Myers used a Galton whistle. He compares a sample from Aberdeenshire with the CAETS results (Figure 3.2). Shorter whistle lengths represent higher frequencies and thus a higher upper limit of hearing. Yet the distinctions are minimal, with just a minor advantage for the children from Aberdeenshire over those from Murray Island. Myers phrases this very cautiously: “From these data it appears that there is only a small difference.”45 Even with regard to the comparison of adults, there are no significant differences between Murray Islanders and individuals from Aberdeenshire as far as the upper limit of hearing is concerned, and especially when cases of impaired hearing on Murray Island are discounted. The concluding chart from Myers is shown in Figure 3.3. Myers set up another procedure, using tuning forks to measure the smallest perceptible difference, which I will quote in full because it allows us to observe how experimental settings were created and adapted in the pioneering stage of experimental research: After numerous trials of various methods, I found that the experiments were best conducted in the following manner. Having assured myself that the subject thoroughly understood what was required of him, I began presenting an interval so large that he could not fail to appreciate it. Next, I rapidly and roughly arrived at an interval which was too small for his correct appreciation. I worked then gradually towards the discrimination-threshold from a point at a little distance above it.46
The sample consisted of twenty-three male adults and twelve boys from Murray Island and twelve children and twenty-one adults from Aberdeenshire. Myers carefully discusses his subjects’ musical exposure, acknowledging that Murray Island children share a similar musical education with their Scottish counterparts, both being well versed in European airs.47 As to testing procedure, the Austrian-born chemist, Gestalt psychologist, and comparative and systematic musicologist Erich von Hornbostel, based at Stumpf ’s psychological institute at the time, noted that Myers combined the
Taken from Charles Samuel Myers, “Hearing,” in Physiology and Psychology, edited by Alfred Cort Haddon (Cambridge: Cambridge University Press, 1903), 152.
Figure 3.2 Comparative chart giving the average whistle lengths for the highest audible tone for individual subjects from child populations in Murray Island and Aberdeenshire, with average and mean variation for the various age groups.
88 Sorting and Screening Human Hearers
Figure 3.3 Age-specific averages for the whistle lengths of a Galton whistle to measure the highest audible tone, given for subjects from Murray Island and Aberdeenshire. Adapted from Charles Samuel Myers, “Hearing,” in Physiology and Psychology, edited by Alfred Cort Haddon (Cambridge: Cambridge University Press, 1903), 154.
“method of minimal deviation with the method of right and wrong cases,”48 using pairs of comparisons and repeated trials to observe long-term entrainment effects. Von Hornbostel closely monitored the results of the expedition and promoted comparative testing along the lines of the CAETS. With the rise of modern musical psychoacoustics around 1900, it became evident that the sense of hearing is strongly adaptive and differs in sensitivity across the hearing range, complicating a seemingly elementary task such as the one applied by Myers. The graduation of tonal perceptivity is not evenly structured, so that it matters a great deal which frequencies are selected for an experiment.49 Myers collected an impressive amount of data, with around five thousand tone judgments on Murray Island and in Aberdeenshire.50 The tables encapsulate the complexity of his experimental procedure, using variable forks, multiple experimental sessions, and the identification of perceptible difference across an individually tailored band of frequencies. His experience alerted him to the effects of adaptation and taught him to distinguish between performances in which he had told subjects that they were right or had given them no response, as well as to distinguish between groups of subjects depending on their performance and learning abilities, which are also generally noted in the tables. Myers’s classification into “musical,” “doubtfully musical,” and “unmusical” subjects is only applied to the Aberdeenshire sample.51 Perhaps he felt uncomfortable ascribing these categories to individuals from another culture whose musicality he had witnessed and documented in a separate chapter of the CAETS reports.52 The topic of the smallest perceptible tone difference is clearly the most complex in terms of the experimental arrangement, number of repeated expositions, and volume of data, as well as placing the most onerous demands on the subjects. Each sitting, made up of many dozen sequences of slightly changed tonal pairs, lasted from twenty to thirty minutes.53 Again, Myers is unable to find any
Murray Island Versus Aberdeenshire 89
relevant differences. He refrains from commenting upon the observed difference of eight vibrations per second on the tone c1 between the subjects from Aberdeen and from Murray Island, which implies that the European subjects could discriminate slightly better. He does, though, ponder the effect that this auditory disposition of Murray Islanders might have on their music, thought to privilege smaller intervals and quarter tones compared to Western music.54 The evidence counters the expectation that the islanders will have better hearing discrimination than Europeans, and the proximity of Torres Strait Islanders and Europeans is confirmed in the reaction-time chapter (again written by Myers), at least in the auditory domain.55 Working toward a new paradigm of individual difference, cultural and social factors, and adaptive behavior and learning processes,56 Myers challenges the accepted claim that substantial differences in reaction times signal differences in mental performance between Westerners and non-Westerners. Cultural difference, it turned out, could not be experimentally proven in the domain of psychophysics.57
Musical Recordings and Transcriptions During the CAETS In parallel to the CAETS testing program, Myers studied the music of the island’s inhabitants.58 The hearing tests, which took place in a highly formalized frame (psychological experiments) defined by the British researchers, are, so to speak, counterbalanced by the active production of musical culture by the Murray Islanders. But the analysis of the Murray Islanders’ musical practices gave rise to a situation in which Myers himself was exposed to a “hearing test,” for the notation of the music implied a complex procedure that Myers lays out in detail.59 For his analysis, Myers produced twenty phonographic cylinder recordings across the various genres and social and religious contexts of local music.60 He even documented himself singing a Murray Island song.61 These are probably the earliest phonographic recordings made by British researchers.62 They proved useful for Myers’s attempt to study and notate the music, a lengthy process that was aided by a metronome and a Tonmesser (“tonometer”), which consisted of sixty-five metal tongues that worked like tuning forks to enable swift and precise identification of pitches and tonal distances.63 Myers’s procedure is quite extraordinary given that systematic reflection on transcriptions would only begin in the first decade of the twentieth century.64 However, the phonographic recordings were a research tool for Myers and did not form part of the published reports.65
90 Sorting and Screening Human Hearers
Myers singles out characteristic features of song structure, intonation patterns, the minor role of rhythm, and the preferred tonal system—yet he does not contextualize the findings of his hearing tests against the background of this analysis of the islanders’ music, except for a brief remark in passing.66 As well as taking into account the specificities in cognitive and cultural concepts of hearing and of music, Myers might have connected psychological performance data to musical practices to design more appropriate hearing tests. Instead, he treated the tests as an abstract, rational, standardized task. When he recorded himself singing, listened to songs, and produced transcriptions, Myers became, so to speak, an experimental subject. I am not suggesting any cultural or methodological reciprocity, however intriguing it would be to gauge Myers’s listening expertise. Still, his description of the musical culture is striking: it shows how the “experimentalizing life” approach, epitomized by the hearing tests, is complemented by a “making culture” approach that allows the subjects of the psychological investigation to articulate their own culture and gain agency. Both operations involve elaborate discursive systems, such as statistical tables and musical notation, and various mechanical and technical apparatuses. They might seem to emblematize objective evidence, yet they are permeated by insecurities, imprecisions due to distractions, makeshift calibrations, added material (here, the test series from Scotland), and notational transcriptions that do not adequately capture the music performed. The strength—or perhaps modernity—of the CAETS reports, as far as Myers’s sections are concerned, lies in the fact that these problems are in no way hidden. They are made manifest as a part of the labor of science and of psychological, anthropological, and cultural interpretation.
The Impact of the CAETS Research Paradoxically, despite the enormous methodological challenges, the CAETS foregrounded structured, multidisciplinary ethnographic testing and large- scale psychological research. CAETS members were aware of the use of the phonograph and early ethnomusicological analysis in other fieldwork programs, such as those of the American Bureau of Ethnology, and in the early armchair ethnology of Carl Stumpf, who also used a phonograph.67 They swiftly integrated hearing tests and phonographic recordings into their research syllabus. This is remarkable considering that phonogram archives would not be established until 1900, and that phonographic research was rarely coupled with psychological tests at this early stage.68 The CAETS
Murray Island Versus Aberdeenshire 91
researchers were pioneering cross-cultural comparison and multisite ethnology by seeking complementary European data. In their view, however, this did not suffice to bridge the growing divide between reflexive anthropology and the cultures of measurement. The data charts could not possibly capture the multifarious aspects of theory building, personal interactions, Myers’s methodological hesitations, and more. The results offer seemingly plausible, yet isolated, evidence that is anything but self-explanatory. How should projections of higher or lower intelligence be brought to bear upon the hearing test data? Was the ideal experimental setup—with a modern, alert, eloquent subject faithfully reporting responses during measurements to an active, dedicated, reliable, efficient Western scientist—suited to the cultural situation under review on Murray Island? Myers was aware of these tensions and drew very cautious conclusions from his experimental results. The hearing tests fully expose the problematic relationship between theorizing and data collection that permeates the expedition as a whole.69 Given the vastness of the project and the monumental edition of the reports, it is interesting to note that the CAETS stimulated no follow-up studies with the same lavish budget and multidisciplinary research team. Myers himself, for reasons that would need to be explored separately, ignored the comparative anthropological approach in both his Text-Book on Experimental Psychology and the 1912 edition of Notes and Queries.70 This publication makes no reference to the methodologies or findings of the CAETS. On the other hand, Myers remained interested in psychoacoustic experiments, as can be gathered from the equipment of the Cambridge laboratory that opened in 1913, made possible by Myers’s own financial commitment. He had been appointed to a new lectureship in experimental psychology in 1907.71 The CAETS’s pioneering fieldwork certainly resounded in the field of music psychology and comparative musicology. Erich von Hornbostel, a dedicated methodologist who was trained in Stumpf ’s circle, immediately discussed it in a long review essay of the relevant report. He generally accepted the CAETS approach, and referred to it again in his important 1905 paper on the outlook of comparative musicology, though he advocated further tests applying more exact methods (I will return to von Hornbostel’s criticisms later).72
Toward Large-Scale Hearing Tests: The 1904 World’s Fair in St. Louis Myers’s hearing tests did act as a model for at least one survey campaign, this time in the United States, which greatly expanded the diversity of ethnic
92 Sorting and Screening Human Hearers
groups involved. The case sheds light on the longevity of personal networks, the impact of Wilhelm Wundt’s experimental research, the rise of public exhibitions as a novel site for screening hearing, and the speculative potential of interpretations drawn from the results of these tests. The 1904 St. Louis World’s Fair featured a psychometric and anthropometric unit led by Robert Sessions Woodworth, who had been chosen on the advice of James McKeen Cattell and Franz Boas. Woodworth had undertaken postdoctoral studies in England and carried out measurements on New York City immigrants, later becoming a psychologist at Columbia University.73 At the time, many U.S. psychologists had earned their doctorates at Wundt’s famous laboratory in Leipzig, and Woodworth was clearly aware of developments both in Germany and in England. He hired Frank Bruner to perform extensive testing in a soundproof room set up on the exposition grounds in St. Louis. Bruner published his data in a longer study, part of his PhD thesis, in which he considered the history of tests, their methodologies, and the conclusions to be drawn from them.74 The chief concern of Bruner’s measurements was to confirm differences in performance between the “primitive races” and Westerners. As in the case of the CAETS, experimenters were eager to establish a Western norm, this time embodied by two hundred white middle-class male adults.75 Authorized visitors were instructed how to measure one another; double-testing was introduced. The other ethnic groups appointed for the hearing tests, Indians, Filipinos, Ainu, and Pygmies,76 were far more restricted in their movements, some of them being subject to military orders. Bruner confines himself to the simple acuity and upper hearing threshold tests and aims to discount “constitutional differences.”77 For the upper limit of hearing, he spells out the need to define standards for the stimuli used— otherwise, Bruner insists, screenings cannot be compared. He describes discrepancies in the results of instrument makers and scientific investigators, reflecting on matters of gradation and their relationship to the relative sensibility of the ear.78 To test the upper range of audibility in New York, Bruner had used a Galton whistle set up in the Columbia University Psychological Laboratory, which had a soundproof room; at the St. Louis Exposition, a similar room was used, with subjects sitting in the dark to reduce distractions.79 After the sessions, Bruner regretted not having undertaken any intelligence tests to correlate his hearing test results against general intelligence.80 Reading his account, it becomes obvious that the hearing test data did not stand alone but required contextualization—yet such contextualization proved difficult, because it was unclear how far cultural adaptations and the social environment affected the assumed constitutional factors. Even within
Murray Island Versus Aberdeenshire 93
the distinct, seemingly ethnically homogeneous groups examined at the exposition, Bruner detected wide variations. This added to the problem of an accepted stimulus, which he tried to solve by designing an adjustable telephone device on one side of a mobile screen for voltage-controlled stimulus generation (ticks), with the subject sitting on other side. The subject placed one ear on a cushion, thus separated from the telephone receiver, to avoid direct contact. The other ear was covered with cotton. The controlled induction noise of the circuit served as a stimulus.81 Despite these precautions, the danger of hallucinations and uncertainty among the subjects could not be ruled out. Bruner observes a superiority on the part of the whites over the Indians, but concludes rather cautiously that this holds true only for the particular Indians he reviewed.82 His overall conclusion of white superiority is placed in the framework of dynamic psychology, underlining the connection between intellectual and sensory capacities. According to Bruner, modern man is exposed to more complex and varied stimuli than those “surrounding the life of the savage,” resulting in a wider variety of motor responses.83 In this way, Bruner integrates hearing—in the sense of responding to stimuli, not as reflexive aesthetic listening—into the complexities of modern urban life; it emerges as part and parcel of the modern condition. Motoric responses, once thought to be a lower faculty and a natural advantage of the “savage” races, are thus fundamentally reformulated. Conceptually, both the CAETS and Bruner’s experiments “discredited one central tenet of Scientific Racism—the ‘Spencer hypothesis’ of marked primitive superiority in basic functions.”84 Bruner’s worries are symptomatic of the complications that dogged hearing tests. Technically, standardized stimuli were hard to generate; methodologically, there was no agreement on whether acuity tests really isolated acuity. At the same time, it was unclear how acuity relates to the sense of tone (in the German discourse, Tonsinn), a wider cognitive capacity to recognize and actively engage with tonal systems. Sociologically and culturally, it was impossible to relate factors thought to be innate to individual subjects’ cultural and social conditions, which may have affected innate capacities.
Transforming the Contexts and Actors of Hearing Tests Bruner’s discussion of his results raises the question of the objectivity of data that seem to lend themselves to almost any conclusion. The St. Louis measurements invite a re-examination of the differing institutional contexts of early hearing tests: science museums, privately funded and university-based
94 Sorting and Screening Human Hearers
psychological laboratories, international expositions, and learned societies. The agents of cultural mediation that allowed these tests to gain popularity now encompassed exhibition and museum managers, extending the body of brokers that had included missionaries, informants, and interpreters during the CAETS. Hearing tests created specific communities: immigrants in the case of Woodworth’s measurements, the museum public in the days of Galton’s lab, men as a preferred group for tests, and complementary pairs of social aggregates such as rural versus urban, Western versus non-Western, literate versus nonliterate, standard versus nonstandard, and controlled groups versus freely visiting participants. The tests also connected geographically remote groups, as in the case of the Murray Islanders versus Aberdonians, or assembled different ethnic groups in one place, as during the World’s Fair. The tests relied on and exploited the growing mobility both of researchers and of the ethnic groups they studied, while at the same time creating new connections between these groups. In the case of the Murray Islanders, the local population was actively interested in the results of the tests and insisted on seeing the academic publication. They can be regarded as co-producers of the knowledge found in the measurement tables. Even if no long-term conclusions can be drawn from the CAETS hearing tests, they testify to the cultural labor of the tests having taken place, with all the necessary on-site arrangements, initiations into the test protocol, interruptions of the survey, and intense personal interaction between experimenters and test subjects. Despite their prominence, the hearing tests in the CAETS format failed to establish a tradition. The racial imperative, disciplines involved, and individual biographical orientations all underwent major changes after 1900, preventing the tests from feeding into a coherent framework. Further complicating their acceptance was the growing endorsement of introspective, phenomenological inquiries that would couple experimental empirical work with informed judgments from participants. This particular approach was advanced by the Stumpf school, and as references to ongoing research in Germany in Myers’s subsequent publications testify, it was followed with great interest in Britain. Myers himself admitted that introspective findings were not intended during the CAETS.85 The tendency to develop richer phenomenological concepts of auditory perception pointed to a shortcoming in the simple reaction and acuity tests, and indeed to the overall understanding of what the CAETS hearing tests were all about: if hearing is a higher cognitive function, and if it articulates very subtle individual, cultural, and racial differences, it cannot possibly be accessed through minimalist, one-dimensional testing. The heavily rationalized approach, fixated on acuities and thresholds and disguising itself as
Murray Island Versus Aberdeenshire 95
a natural task, implied an instinctive, intuitive response to the auditory stimulus, as if evaluative, emotive, memory-based, and analytical dimensions could be ignored. Research on timbre perception showed, however, that the human capacity of hearing is a highly elaborate, time-critical gift that has to be situated within broader notions of sensibility, responsiveness, selective listening, and gestalt processing. Otto Abraham addressed the complexities of “absolute tone consciousness” and potential modes of methodological exploration as early as 1901, and Stumpf explored the perception and recognition of intervals and chords with extremely short exposure among Western observers with great refinement in 1909 (see Viktoria Tkaczyk in this volume).86 This work was demanding in terms of theory building and transformation into viable experimental setups, so that any correlation of reaction-time data with constructs such as race and intelligence, or evolution and cultural genetics, must have appeared increasingly speculative. Thanks to Stumpf ’s discussion in Tonpsychologie, Myers was aware of attention spans and other individual and circumstantial factors impacting upon the measurements, but he was unable to develop a more sophisticated testing procedure on the spot during the CAETS. In fact, Myers was in any event more interested in individual differences than in racial traits, and those differences proved not to match up with attributions of racial qualities.87
Hearing Tests in German Multidisciplinary Psychology Among the methodological repercussions of the CAETS was an important innovation within early twentieth-century German psychology: multidisciplinary Sammelforschung, or “collecting research” in the sense of work that crossed disciplinary boundaries to integrate a much broader range of findings. Especially in the anthropological wing of this integrative trend, which arose out of psychology and was foreshadowed by Galton’s work, growing attention accrued to the processuality of culture, convergences of different racial and cultural traits in individuals, the influence of cultural conditions on psychological capacities, and the interplay between inner processes and material contexts.88 This would imply that intelligence was not a static capacity that could be reduced to a mean representing the average intelligence of one culture, but was rather a result of complex appropriations and exchanges with foreign cultural traits that were integrated into the culture of an individual.
96 Sorting and Screening Human Hearers
Early German anthropological and ethnological questionnaires (Fragesammlungen) disregarded ethnopsychological aspects, focusing instead on the legal and social structures of the ethnic groups under review. It was only in psychological Sammelforschung, influenced by the CAETS, that hearing tests became a standard procedure to be performed by nonspecialist travelers on the ethnic groups they encountered.89 Before the anthropologist, sociologist, and proponent of Sammelforschung Richard Thurnwald set out on a journey to Oceania in 1906, he asked Stumpf for instructions on how to perform psychological examinations on primitive peoples. This is clearly a direct effect of the CAETS, the first volumes of which had already been published at that point. Stumpf eventually passed Thurnwald’s call for theoretically informed, cross-disciplinary testing on to the newly established Institut für angewandte Psychologie und psychologische Sammelforschung (Institute of Applied Psychology and Psychological Sammelforschung), based in Potsdam.90 Thurnwald’s request, mediated by Stumpf, for input on his testing procedures triggered articulate responses from a number of renowned scholars. Their contributions were coordinated by the founders of the Institute of Applied Psychology, William Stern and Otto Lipmann. Important German psychologists and philosophers including Hermann Ebbinghaus, Oswald Külpe, and Friedrich Schumann contributed to a first set of ethnographic instructions, which also contained essays by two British CAETS members, Myers and McDougall. The perception of music was covered by Erich von Hornbostel. This first version of the Anleitungen zur psychologischen Untersuchung primitiver Menschen (Instructions for the psychological investigation of primitive people) had a typescript run of just twelve copies. In addition to musical perception, it addressed all the human senses; sensations of space, time, and number; memory capacities based on individual sense impressions and on mathematical sequences; the scope of consciousness; attention thresholds; the formation of judgments; astronomy; and popular imagination. The section on tone sensations included technical instructions on using the phonograph.91 Thurnwald used this informal draft of the testing instructions during his expedition to the South Pacific in 1906–9.92 As the manual was much in demand, the Institute of Applied Psychology decided to publish a more comprehensive edition. An illustrious commission was recruited that included Otto Abraham, Franz Boas, Rivers, Myers, McDougall, Felix von Luschan, Thurnwald himself, and Max Wertheimer.93 Surprisingly, out of a commission of thirty, only seven members managed to submit a contribution to the expanded version of the instructions that was published in 1912, with a version printed as a supplement to the institute’s
Murray Island Versus Aberdeenshire 97
journal reaching a far broader readership.94 Neither Stumpf himself nor the prolific von Hornbostel was among them. Their own research embraced not only the psychological aspect, which could be studied in the form of tests, but also the musical aspect, which made music part of the cognitive and cultural system of the ethnic groups to be studied. This broad, phenomenological, and introspective approach made them hesitant to accept physiological data as the chief clue to an ethnic group’s psychological and anthropological makeup. Questionnaires and instructions of this kind, by implying that travelers would perform the tests themselves, forced them into a new kind of encounter with their test subjects. To this extent, the tests structured the travelers’ daily routine and prefabricated their notion of the Other. Although the Anleitung manuals had to allow enough scope for actual new events and reactions, their master discourse— saturated with ethnological expertise— usually anticipated many possible outcomes of a given encounter. At the same time, the instructions gave no rationale or full background regarding either their own intentions or the future uses of the data; the remote-controlled envoy out in the field had to know only enough to perform his or her duties. Whereas Myers had carried out the tests and analysis himself, these instructions anticipate an amateur observer who was to forward the data to the specialists back home. Some ethnologists conceded to their fieldwork contributors a degree of freedom in handling the instructions. In 1903, for instance, Thurnwald suggested that travelers make brief, one-paragraph summaries on a set of questions and send them straight to the ethnographic museum in Berlin. The British counterpart of the German Anleitungen, the constantly revised and updated Notes and Queries on Anthropology, followed a similar path, abandoning the “leading questions” kind of arrangement in favor of a narrative form in the 1912 edition.95 This change was more than a stylistic shift. It indicates a new attitude to the study of other cultures. The narrative form encourages the travelers to situate themselves more openly within the ethnological master discourse, as tailored to their field activities and their actual observations on the ground. It tolerates a more explorative kind of investigation, in which travelers can make firsthand sense of their interactions with the indigenous population and of what they see, hear, and smell. This new attitude not only left more space for the traveling researcher’s imagination but also allowed indigenous populations to unfold their cultural expressions beyond the question/answer type of encounter. Toward the end of the first decade of the twentieth century, psychologists, some ethnologists, and comparative musicologists used the insights gained
98 Sorting and Screening Human Hearers
from incoming travel accounts and data sheets to create a more pluralistic account of cultural polygenism. This is not addressed in the rhetoric of the early Anleitungen itself, which is geared to monographic observation of particular populations. Indirectly, though, the genre of Anleitungen contributed to more complex considerations of a variety of political, cultural, and economic factors that shaped the individual ethnic “biography” of a foreign people. In his opening essay to the Institute of Applied Psychology’s 1912 collection Vorschläge zur psychologischen Untersuchung primitiver Menschen, Thurnwald stresses the cultural processes of which foreign populations formed a part. Criticizing the ethnographic spell cast by the museum, which privileges objects, he tries to identify the forces that bring cultural and psychological factors into convergence. Thurnwald’s emphasis on culture-specific methods coincides with Stumpf ’s statement in Anfänge der Musik (The origins of music), published in 1911, that there is no fundamental difference in sense perceptions between Western and “primitive” populations.96 Von Hornbostel makes a related point. In a 1910 essay on comparative acoustic and psychological examinations, he questions the notion that the reported impressive performance by non-Western ethnic groups,97 observed in the search for distinct racial traits, relates to their sensorium.98 In the first section of the paper, on Tonsinn, he carefully discusses the hearing tests performed by Myers during the CAETS and by Bruner at St. Louis, arguing that though experimenters believe they are examining sensory capacities, in fact, they are measuring their subjects’ ability to focus on the given task.99 Likewise, von Hornbostel doubts that there is any correlation between hearing acuity and musical talent. By delving further into acoustic memory and tone consciousness (Tonbewußtsein), he exceeds the aims of the CAETS psychological survey, which was nevertheless an important cornerstone for his empirical comparative research.100 In a draft text for the 1912 Vorschläge, not included in the final publication, von Hornbostel details a series of complex experiments on the Tonsinn of the “primitives,” addressing no fewer than 138 aspects.101 Hearing sensitivity over long distances, the manufacturing of instruments on a particular model, the response to unfamiliar recordings, modified speeds of the rotating cylinders, and the display of European-style musical intervals are all elements of von Hornbostel’s procedure. At times, they require careful arrangements of groups in the open air, some of the members blindfolded, that are reminiscent of film scripts. Von Hornbostel actually recommends researchers to set up the tests as games and offer small rewards. The draft has separate entries for “tests without apparatuses,” “tests with the phonograph,” “tests with the tonometer,” “physiological examinations of the hearing
Murray Island Versus Aberdeenshire 99
capacity,” and more. Advanced experimenters are given additional advice for more complex procedures.
Hearing Tests and the Dynamics of Modern Science Ethnopsychological research as inspired by the CAETS members, Thurnwald, and their colleagues—of which hearing tests formed a part—took shape contemporaneously with advanced photogrammetry and the definition of German industrial norms (DIN). It coincided with the emergence of a wider Prussian culture of measurement, standardization, the establishment of archives such as the Phonogramm-Archiv, and novel collecting practices as represented in the Anleitungen.102 Unlike the related cases of anthropometric physical measurements and collectable ethnographica, this research addressed a human capacity that seemed just as tangible, but in fact turned out to be too complex a construct to be brought under control. The noninvasive “experimentalizing life” format of hearing tests exposed Western assumptions, the arbitrariness of gradations, and the impossibility of defining shared stimuli and shared standards of measurement within the community of researchers. Hearing tests nonetheless enriched the anthropometric, psychological, and early comparative and culturalist agenda by promising to explore and measure the connection between a sensory modality and intelligence and by applying mobile, robust test equipment such as the Galton whistle. The question “How do people hear?” was intended to address a key human faculty, yet the signal-based approach chosen to identify thresholds was entirely nonsemantic and examined simple reactions only. The testing of more complex constructs, such as just perceptible differences, again targeted a discriminatory capacity rather than “hearing” as such, which involves many other cognitive resources and processes. Yet it is clearly here that more complex and valuable insights could have been gained. What started out as a physical examination, in which the subject was largely passive, turned out to be more demanding than anticipated. The testing process required subjects to be active, alert, attentive, efficient, and reliable in communicating test responses. In doing so, it inscribed into the senses and interactive situations of both Oceanic subjects and Western researchers a specifically Western and modern mindset that experimentalized mental functions on the basis of Western standards and introduced a performance-oriented standard that would inform the emerging field of industrial psychology, army intelligence tests, and mass health screenings. It is ironic that these very tests, and the frankness with
100 Sorting and Screening Human Hearers
which Myers communicated his ambivalent and inconclusive results, eventually contributed to the decline of scientific racism.103 The refinement of hearing tests as portrayed in this chapter, beginning with the CAETS, took place in a period when hearing was being fundamentally reassessed in the field of tone psychology and in research on tonal consciousness, informed by phenomenology and philosophy. Before these new insights could be translated into more appropriate hearing tests under lab conditions, researchers operating outside the laboratory had already ventured into new arenas such as public museums, expeditions with makeshift hearing tests, and World Fairs. The format implicated researchers and subjects alike in novel social, cultural, methodological, and technical situations that tell us more about the dynamics of these interactions and research ideologies than about their actual topic, human hearing. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/caets-crosscultural-hearing-tests-Klotz.
Notes 1. A map is given in Alfred Cort Haddon, Arts and Crafts, ix. 2. Anita Herle and Sandra Rouse, “Introduction”; Graham Richards, ‘Race’; Ben Shephard, Headhunters. 3. However, Bennett Zon acknowledges the impact of the CAETS on British ethnomusicology. Zon, “Music of Non-Western Nations.” 4. See Vanessa Agnew, “Encounter Music in Oceania.” 5. Bronwen Douglas and Chris Ballard, Foreign Bodies. 6. The Vienna Phonogrammarchiv, the Phonogramm-Archiv in Berlin, the Deutsche Marine- Expedition, and the Kaiserin Augusta River Expedition were covering the area, as were individual collectors. See Adrienne L. Kaeppler and Don Niles, “Music and Dance of New Guinea,” 479–80. A map showing the geopolitical situation of the region in 1914 can be found at https://commons.wikimedia.org/wiki/File:East_Asia_and_Oceania_1914-de.svg. 7. Herle and Rouse, “Introduction,” 4, summarizes all the volumes of the reports with their contents and year of publication. 8. Bennett Zon, Representing Non-Western Music, 159–217. 9. Myers rose to prominence when he coined the term “shell shock” during World War I. He served in the Army Medical Corps and was a protagonist of industrial psychology. 10. David Howes, “Senses,” 439. 11. J. S. Jones, “Galton Laboratory.” 12. Francis Galton, “An Instrument”; Galton, “Exhibition of Instruments.” 13. Jones, “Galton Laboratory.” 14. James McKeen Cattell, “Remarks,” 378.
Murray Island Versus Aberdeenshire 101 15. Marius Turda, Modernism and Eugenics, 2. Turda does not refer specifically to hearing tests. 16. Ibid., 6. 17. Shephard, Headhunters, 81. 18. Ibid., 47. 19. Ibid., 45. 20. Carl Stumpf, Tonpsychologie, vi. 21. Mitchell G. Ash, Gestalt Psychology, 22–25. 22. See https://www.psychol.cam.ac.uk/about-us/history. Wundt’s Leipzig laboratory opened in 1879. 23. Bronwen Douglas, “Foreign Bodies in Oceania,” 5. It is symptomatic that the two disciplines that spelled out these endeavors most persuasively, biology and anthropology, form the backbone of the CAETS project. 24. Shephard, Headhunters, 45. 25. Richards, ‘Race,’ 25. 26. See Zon, Representing Non-Western Music, 227, with a detailed discussion of the CAETS and its relevance to comparative musicology and emerging ethnomusicology; also Sven Werkmeister, Kulturen jenseits der Schrift, 169. 27. Richards, ‘Race,’ 19, 37. 28. Charles Samuel Myers, “Hearing,” 143. 29. Kurt Blaukopf traces the genealogy of the empiricist impulse in nineteenth-and early twentieth-century Austrian and German scholarship, which was relevant to the Stumpf school. Blaukopf, Pioniere. 30. Douglas and Ballard, Foreign Bodies, xiii. 31. Shephard, Headhunters, 51, 54. 32. Ibid., 61. 33. Alfred C. Haddon, Head-Hunters, viii. 34. [Haddon], Physiology and Psychology, series introduction, n.p. 35. Shephard, Headhunters, 24. 36. Based on Myers’s account in “Hearing.” 37. See Politzer, “Ueber einen einheitlichen Hörmesser”: http://www.politzersociety.org/content.php?conid=683. 38. John George Garson and Charles Hercules Read, Notes and Queries, 44–48. 39. Ibid., 45. 40. Ibid., 48. 41. Myers, “Hearing,” 141. 42. Ibid., 145. 43. Ibid., 151. 44. Ibid. 45. Ibid., 152. 46. Ibid., 158. 47. Ibid., 156. 48. Erich M. von Hornbostel, “Charles S. Myers,” 238. 49. See Horst-Peter Hesse, “Psychoakustik des musikalischen Hörens,” 213, 215. 50. Myers, “Hearing,” 165. 51. Ibid., 164. 52. Charles Samuel Myers, “Music.”
102 Sorting and Screening Human Hearers 53. Myers, “Hearing,” 159. 54. Ibid., 168. 55. Charles Samuel Myers, “Reaction-Times,” 220. 56. Werkmeister, Kulturen jenseits der Schrift, 171; Zon, Representing Non-Western Music, 12, 227. 57. Werkmeister, Kulturen jenseits der Schrift, 174. 58. Myers, “Music.” 59. Ibid., 242. 60. Meyers sent a selection of these recordings to the Berlin Phonogramm-Archiv for duplication, with extensive written documentation in the form of letters to Erich von Hornbostel. Susanne Ziegler, Die Wachszylinder, 230, 231. The recordings are in the British Library and can be accessed at http://sounds.bl.uk/world-and-traditional-music/ethnographic- wax-cylinders. Recordings made during the CAETS are also available via the National Film and Sound Archive of Australia: https:// www.nfsa.gov.au/ collection/ curated/ recordings-cambridge-anthropological-expedition-torres-straits-alfred-cort-haddon. 61. The transcription of this piece, part of the “Malu songs,” can be found in Myers, “Music,” 245. Myers’s singing sample is at http:// sounds.bl.uk/ world- and- traditional- music/ Ethnographic-wax-cylinders/025M-C0080X1096XX-0100V0. 62. Martin Clayton, “Ethnographic Wax Cylinders,” 69. 63. Myers, “Music,” 242. 64. Given that this volume was published in 1912, more than a decade after the CAETS, Myers would have had time to absorb the methodologies proposed by early comparative musicology, although he makes no reference to either Abraham or von Hornbostel in his chapter on music. Their seminal essay “Vorschläge für die Transkription exotischer Melodien” was published in 1909. In December 1905, Myers probably visited Stumpf and von Hornbostel in Berlin. Clayton, “Ethnographic Wax Cylinders,” 81. 65. Themed collections of non-Western music with extensive scholarly liner notes would be released later, when the commercial disc replaced the phonographic cylinder. 66. Myers, “Hearing,” 168. 67. Jesse Walter Fewkes advised Haddon of the usefulness of the phonograph as early as 1890. Clayton, “Ethnographic Wax Cylinders,” 69. Stumpf ’s research was “armchair” in the sense that although he had made recordings in German prisoner-of-war camps, he did not carry out fieldwork outside Germany. Carl Stumpf, “Lieder der Bellakula-Indianer.” 68. In the case of Myers, the recordings and the psychological surveys went hand in hand. For a fuller picture, Martin Clayton’s account of the phonographic activities will need to be complemented by work on the hearing tests. 69. See Herle and Rouse, “Introduction,” 19. 70. Charles Samuel Myers, Text-Book (the first edition appeared in 1911); Barbara W. Freire- Marreco and John Linton Myres, Notes and Queries. 71. See https://www.psychol.cam.ac.uk/archived-news/2013/2013-15-05-centenary; https:// www.psychol.cam.ac.uk/about-us/history. 72. Hornbostel, “Charles Myers,” 239; Hornbostel, “Die Probleme der vergleichenden Musikwissenschaft,” 96. 73. Nancy J. Parezo and Don D. Fowler, Anthropology Goes to the Fair, 311. 74. Frank G. Bruner, Hearing of Primitive Peoples, 312. 75. Parezo and Fowler, Anthropology Goes to the Fair, 213.
Murray Island Versus Aberdeenshire 103 76. The terms are Bruner’s. Apparently, Bruner tested “Indian” pupils from the same school in Chilocco that von Hornbostel had visited and worked with during his field trip to the United States. 77. Bruner, Hearing of Primitive Peoples, 2, 5. 78. Ibid., 12–14. 79. Ibid., 21, 25, 30. 80. Ibid., 40. 81. Ibid., 71. The procedure is discussed by von Hornbostel in “Über vergleichende akustische und musikpsychologische Untersuchungen,” 147. 82. Bruner, Hearing of Primitive Peoples, 87, 98. 83. Ibid., 111–12. 84. Richards, ‘Race,’ 67. 85. Myers, “Hearing,” 165. 86. Otto Abraham, “Das absolute Tonbewußtsein”; Carl Stumpf, “Über das Erkennen.” 87. Zon, Representing Non-Western Music, 177–95. 88. Richard Thurnwald, “Probleme der ethno-psychologischen Forschung.” William Stern and Otto Lipmann introduced psychological Sammelforschung as an integrative platform with the aim of linking specialist psychological applications to the overall program of psychology and to contexts outside psychology (such as pedagogy, law, medicine, and the arts). The initiative was consolidated in an institute and a journal. 89. The impact of the CAETS on Thurnwald has been noted by Marion Melk-Koch, “Auf der Suche nach der menschlichen Gesellschaft,” 58, 59. 90. The institute’s mission is outlined in the journal’s first issue. William Stern and Otto Lipmann, “Zur Einführung.” 91. This genealogy of the 1912 Vorschläge zur psychologischen Untersuchung primitiver Menschen is given by the institute directors, Stern and Lipmann, “Mitteilungen.” 92. Thurnwald, “Probleme der ethno-psychologischen Forschung,” 15. In 1906, Thurnwald reworked an ethnographic questionnaire by Steinmetz. Sebald Rudolf Steinmetz and Richard Thurnwald, Ethnographische Fragesammlung. 93. Felix von Luschan directed the Oceanic collections at Berlin’s Völkerkundemuseum, the institution that had appointed Thurnwald for the expedition. Max Wertheimer was a gifted psychologist who would discover important principles of gestalt perception only a few years later. 94. Institut für Angewandte Psychologie und Psychologische Sammelforschung, Vorschläge. A second volume was anticipated but apparently never appeared. 95. Freire-Marreco and Myres, Notes and Queries. This edition had four CAETS members on its editorial board; the same group consulted for the German suggestions on psychological tests in 1906 and 1912. 96. Carl Stumpf, Anfänge der Musik. 97. Naturvölker. Von Hornbostel calls them “sog. Primitive,” so-called primitives. “Über vergleichende akustische und musikpsychologische Untersuchungen,” 154. 98. Ibid., 146. 99. Ibid., 146–55. 100. Ibid., 149, 153. 101. On this manuscript and its context, see Lars-Christian Koch, “Images of Sound.” Von Hornbostel had included “experiments with non-European subjects” as early as 1910, in
104 Sorting and Screening Human Hearers his seminal “Über vergleichende akustische und musikpsychologische Untersuchungen,” though Myers’s CAETS activities precede this publication by more than a decade. 102. The Prussian branch of this research tendency was particularly systematic in orientation, stimulated by the state-funded research that led to the rise of German “big science.” The reality sometimes fell short of this vision, at least in the case of the Berlin Phonogramm- Archiv. Despite its changing affiliations over the years, it was always short of funding. 103. According to Richards, Haddon “moved somewhat beyond classic Scientific Racism, even while retaining some of its perspectives.” Richards, ‘Race,’ 50; in his later work, Myers moved toward cultural adaptationism, “which effectively removed race as a conceptual underlay in the anthropological methodology of its day.” Zon, Representing Non-Western Music, 13.
References Abraham, Otto. “Das absolute Tonbewußtsein: Psychologisch- musikalische Studie.” Sammelbände der Internationalen Musikgesellschaft 3, no. 1 (1901): 1–86. Abraham, Otto, and Erich Moritz von Hornbostel. “Vorschläge für die Transkription exotischer Melodien.” Sammelbände der Internationalen Musikgesellschaft 11 (1909–10): 1–25. Agnew, Vanessa. “Encounter Music in Oceania: Cross- Cultural Musical Exchange in Eighteenth-and Early Nineteenth-Century Voyage Accounts.” In The Cambridge History of World Music, edited by Philip V. Bohlman, 183–201. Cambridge: Cambridge University Press, 2013. Ash, Mitchell G. Gestalt Psychology in German Culture, 1890–1967: Holism and the Quest for Objectivity. Cambridge: Cambridge University Press, 1995. Blaukopf, Kurt. Pioniere empiristischer Musikforschung: Österreich und Böhmen als Wiege der modernen Kunstsoziologie. Vienna: Hölder-Pichler-Tempsky, 1995. Bruner, Frank G. The Hearing of Primitive Peoples: An Experimental Study of the Auditory Acuity of the Upper Limit of Hearing of Whites, Indians, Filipinos, Ainu and African Pigmies. New York: Science Press, 1908. Cattell, James McKeen. “Remarks Following ‘Mental Tests and Measurements.’” Mind 15 (1890): 373–80. Clayton, Martin. “Ethnographic Wax Cylinders at the British Library National Sound Archive: A Brief History and Description of the Collection.” British Journal of Ethnomusicology 5 (1996): 67–92. Douglas, Bronwen. “Foreign Bodies in Oceania.” In Foreign Bodies: Oceania and the Science of Race 1750–1940, edited by Bronwen Douglas and Chris Ballard, 3–30. Acton, NSW: Australian National University Press, 2008. Douglas, Bronwen, and Chris Ballard, eds. Foreign Bodies: Oceania and the Science of Race 1750–1940. Acton, NSW: Australian National University Press, 2008. Freire-Marreco, Barbara W., and John Linton Myres, eds. Notes and Queries on Anthropology. 4th ed. London: Royal Anthropological Institute, 1912. Galton, Francis. “Exhibition of Instruments (1) for Testing the Perception of Differences of Tint, and (2) for Determining Reaction-time.” Journal of the Anthropological Institute 19 (1890): 27–29. Galton, Francis. “An Instrument for Measuring Reaction Time.” Report of the British Association for the Advancement of Science 59 (1889): 784–85.
Murray Island Versus Aberdeenshire 105 Garson, John George, and Charles Hercules Read, eds. Notes and Queries on Anthropology. 2nd ed. London: Royal Anthropological Institute, 1892.Haddon, Alfred C. Head-Hunters: Black, White, and Brown. London: Methuen, 1901. [Haddon, Alfred Cort, ed.]. Arts and Crafts, vol. 4 of Reports of the Cambridge Anthropological Expedition to Torres Straits. Cambridge: Cambridge University Press, 1912. [Haddon, Alfred Cort, ed.]. Physiology and Psychology: Hearing, Smell, Taste, Cutaneous Sensations, Muscular Sense, Variations of Blood-Pressure, Reaction-Times, vol. 2 of Reports of the Cambridge Anthropological Expedition to Torres Straits. Cambridge: Cambridge University Press, 1903. Herle, Anita, and Sandra Rouse. “Introduction: Cambridge and the Torres Strait.” In Cambridge and the Torres Strait: Centenary Essays on the 1898 Anthropological Expedition, edited by Anita Herle and Sandra Rouse, 1–22. Cambridge: Cambridge University Press, 1998. Hesse, Horst-Peter. “Psychoakustik des musikalischen Hörens.” Enzyklopädie der Psychologie, edited by Thomas H. Stoffer and Rolf Oerter, D/7, 1:203–49. Göttingen: Hogrefe, 2005. Hornbostel, Erich Moritz von. “Charles S. Myers: (Sinnesphysiologischer psychologischer Teil der) Reports of the Cambridge Anthropological Expedition to Torres Straits.” Zeitschrift für Psychologie und Physiologie der Sinnesorgane 36 (1904): 237–39. Hornbostel, Erich Moritz von. “Die Probleme der vergleichenden Musikwissenschaft.” Zeitschrift der Internationalen Musikgesellschaft 7, no. 3 (1905–6): 85–97. Hornbostel, Erich Moritz von. “Über vergleichende akustische und musikpsychologische Untersuchungen.” Beiträge zur Akustik und Musikwissenschaft 5 (1910): 143–67. Howes, David. “The Senses: Polysensoriality.” In A Companion to the Anthropology of the Body and Embodiment, edited by Frances E. Mascia-Lees, 435–50. Chichester: Wiley-Blackwell, 2011. Institut für Angewandte Psychologie und Psychologische Sammelforschung Berlin, ed. Vorschläge zur psychologischen Untersuchung primitiver Menschen. Leipzig: A. J. Barth, 1912. Jones, J. S. “The Galton Laboratory, University College London.” In Sir Francis Galton, FRS: The Legacy of His Ideas, edited by Milo Keynes, 190–94. Basingstoke: Palgrave Macmillan, 1993. Kaeppler, Adrienne L., and Don Niles. “The Music and Dance of New Guinea.” In The Garland Encyclopedia of World Music 9: Australia and the Pacific Islands, edited by Adrienne L. Kaeppler und J. W. Love, 472–87. New York: Garland, 1998. Koch, Lars-Christian. “Images of Sound: Erich M. von Hornbostel and the Berlin Phonogram Archive.” In The Cambridge History of World Music, edited by Philip V. Bohlman, 475–97. Cambridge: Cambridge University Press, 2013. Melk-Koch, Marion. Auf der Suche nach der menschlichen Gesellschaft: Richard Thurnwald. Berlin: Staatliche Museen Preußischer Kulturbesitz, 1989. Myers, Charles Samuel. “Hearing.” In Physiology and Psychology, vol. 2 of Reports of the Cambridge Anthropological Expedition to Torres Straits, edited by Alfred Cort Haddon, 141– 68. Cambridge: Cambridge University Press, 1903. Myers, Charles Samuel. “Music.” In Arts and Crafts, vol. 4 of Reports of the Cambridge Anthropological Expedition to Torres Straits, edited by Alfred Cort Haddon, 238– 69. Cambridge: Cambridge University Press, 1912. Myers, Charles Samuel. “Reaction-Times.” In Physiology and Psychology, vol. 2 of Reports of the Cambridge Anthropological Expedition to Torres Straits, edited by Alfred Cort Haddon, 205–23. Cambridge: Cambridge University Press, 1903. Myers, Charles Samuel. A Text-Book of Experimental Psychology with Laboratory Exercises. 2nd ed. Cambridge: Cambridge University Press, 1922. Parezo, Nancy J., and Don D. Fowler. Anthropology Goes to the Fair: The 1904 Louisiana Purchase Exposition. Lincoln: University of Nebraska Press, 2007. Politzer, Adam. “Über einen einheitlichen Hörmesser.” Archiv für Ohrenheilkunde 12 (1877): 104–9.
106 Sorting and Screening Human Hearers Richards, Graham. ‘Race’, Racism and Psychology: Towards a Reflexive History. 2nd ed. London: Routledge, 2012. Shephard, Ben. Headhunters: The Search for a Science of the Mind. London: Bodley Head, 2014. Steinmetz, Sebald Rudolf, and Richard Thurnwald. Ethnographische Fragesammlung zur Erforschung des sozialen Lebens der Volker ausserhalb des modernen europäisch- amerikanischen Kulturkreises. Berlin: Decker, 1906. Stern, William, and Otto Lipmann. “Mitteilungen.” Zeitschrift für angewandte Psychologie und psychologische Sammelforschung 6 (1912): 116–17. Stern, William, and Otto Lipmann. “Zur Einführung.” Zeitschrift für angewandte Psychologie und psychologische Sammelforschung 1 (1908): i–iii. Stumpf, Carl. Die Anfänge der Musik. Leipzig: Barth, 1911. Stumpf, Carl. “Lieder der Bellakula-Indianer.” Vierteljahrsschrift für Musikwissenschaft 2, no. 4 (1886): 405–26. Stumpf, Carl. Tonpsychologie. Vol. 1. Leipzig: S. Hirzel, 1883. Stumpf, Carl. “Über das Erkennen von Intervallen und Akkorden bei sehr kurzer Dauer.” Beiträge zur Akustik und Musikwissenschaft 4 (1909): 1–39. Thurnwald, Richard. “Probleme der ethno-psychologischen Forschung.” In Vorschläge zur psychologischen Untersuchung primitiver Menschen, edited by Institut für Angewandte Psychologie und Psychologische Sammelforschung Berlin, 1–27. Leipzig: A. J. Barth, 1912. Turda, Marius. Modernism and Eugenics. Basingstoke: Palgrave Macmillan, 2010. Werkmeister, Sven. Kulturen jenseits der Schrift: Zur Figur des Primitiven in Ethnologie, Kulturtheorie und Literatur um 1900. Munich: Wilhelm Fink, 2010. Ziegler, Susanne. Die Wachzylinder des Berliner Phonogramm- Archivs. Berlin: Staatliche Museen zu Berlin, 2006. Zon, Bennett. “The Music of Non- Western Nations and the Evolution of British Ethnomusicology.” In The Cambridge History of World Music, edited by Philip V. Bohlman, 299–318. Cambridge: Cambridge University Press, 2013. Zon, Bennett. Representing Non-Western Music in Nineteenth-Century Britain. Rochester, NY: University of Rochester Press, 2007.
4 Hearing Perfection Emily I. Dolan
Introduction: The Violin, Central and Strange Within the canon of Western art music, the violin is a central piece of musical technology. It forms the core of the orchestra; its leading performers and the flamboyant imagery they inspire have shaped ideas and ideals of virtuosity. People have long projected themselves onto the instrument: its sonority is often described in human terms—the violin cries, it sobs, it laughs—while its iconic shape, with those elegant, shapely curves, invites anthropomorphizing.1 The relative continuity of the violin’s design over the past three hundred years makes it easy to recognize the instrument across time and to see the present in the past and the past in the present. The finest violins are celebrity objects that fetch millions of dollars in auctions; they have names and well-known provenances. And yet, for all its centrality, the violin also stands apart from many other instruments. Part of this can be attributed to the violin’s development and how radically its history differs from that of keyboards and wind instruments. The violin emerges in the early sixteenth century without a clear record. It has no single inventor. Its construction reflects and incorporates several different instrumental types, the rebec, the lira da bracchio, and the so-called Renaissance fiddle. By contrast, wind and keyboard instruments—though they have similarly complicated early histories—begin to find their modern forms in the late eighteenth and nineteenth centuries, and their technological evolution is far more legible. Today, this separation continues to be palpable within the organological community, in which scholars of the violin have, by and large, operated apart from those who study the histories of other instruments. The distinction is clear on the pages of the two flagship journals, the Galpin Society Journal and the Journal of the American Musical Instrument Society. Both cover a diverse range of topics, with essays on both mainstream and unusual instruments and studies of makers, collectors, and iconography. But although articles on violin-family instruments and makers are not absent, they form a Emily I. Dolan, Hearing Perfection In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0005.
110 Designing Instruments, Calibrating Machines
notably smaller percentage in comparison to the topics of keyboard and wind instruments.2 This division is bound up with the values that undergird organology as a discipline: the field has long emphasized origins and original makers. Scholars have measured and traced instrumental development and evolution through the careful investigation of well-preserved specimens.3 There is much value in this approach, for part of the appeal of studying musical technologies— instruments and other media—is a sense of conceptual solidity. A piece of technology often harbors the ability to serve as a kind of sonic archive of soundworlds, an access point to past listening cultures. Studying technology can counterbalance the transcendental qualities of music by binding it in a particular time and place. But violins, especially the most celebrated examples, are long-lived objects that circulate, undergoing repairs, alterations, and upgrades. Indeed, this is another important difference between violins and other instruments: the most celebrated examples of the instrumental species—those by Stradivari and Guarneri del Gesù—are nearly three hundred years old. The very things that enable a violin to be used for several centuries are the same things that erase its originary information. As a result, the violin thwarts narratives of linear technological progress; the story of the violin is one that continually folds back upon itself. This temporal looping is the jumping-off point for what follows, in which I trace one practice that has helped shape the violin’s strange relationship with history: comparative testing. Since the early nineteenth century, new violins have been compared, measured, and tested against older instruments. The results of such tests have been remarkably—even astonishingly—consistent since the early nineteenth century. And yet these outcomes have not had the impact one might expect. This is not to say that the tests are impotent: on the contrary, they have played a crucial role in defining the basic nature of the violin. In following that long genealogy, this chapter moves, like the violin, backward and forward in time, surveying a broad swath of the instrument’s sometimes tangled history.4 I have a twofold goal. First, I hope to show the ways in which testing helped a particular configuration of values to coalesce around the instrument. To test the violin is to direct meticulous and sustained attention to the instrument. This attention has helped to shape what the violin is understood to be, both as an object of scientific inquiry and as an aesthetic and social object. Second, I want to suggest ways of thinking about the relationships between technology and the history of music more generally.
Hearing Perfection 111
Perfection and Critique The violin has long attracted talk of perfection. In the nineteenth century, it was frequently held up as an example of an already perfect instrument to which other instruments could aspire. The violin did not participate in the heady discussions of technological progress that surrounded wind and keyboard instruments in the nineteenth century. For example, in the jury reports from the 1851 Great Exhibition in the Crystal Palace, the discussion of “Bow Instruments” is limited to a few sentences of brief praise for the instruments of the famed French luthier Jean-Baptiste Vuillaume; other instruments—the organ, the pianoforte, and brass and woodwind instruments—all warrant much longer discussions, several of which invoke their longer historical development.5 Lengthy histories of the violin could be found elsewhere, and they took more hagiographic forms, as in François-Joseph Fétis’s 1856 Antonio Stradivari, luthier célèbre, which explored the famed maker, the history and construction of stringed instruments going back to antiquity, and the development of the modern bow.6 In 1884, the English scholar and amateur violin maker Edward Heron-Allen, best known for his work in Persian studies, declared: “The violin, called as it justly is ‘the king of instruments,’ is perhaps the only human contrivance, which, taken as a whole, may be pronounced to be—perfect.”7 This sense of perfection continues to shape the rhetoric around the instrument. Consider this passage from the current Grove Music article on the violin: The violin is one of the most perfect instruments acoustically and has extraordinary musical versatility. In beauty and emotional appeal its tone rivals that of its model, the human voice, but at the same time the violin is capable of particular agility and brilliant figuration, making possible in one instrument the expression of moods and effects that may range, depending on the will and skill of the player, from the lyric and tender to the brilliant and dramatic. Its capacity for sustained tone is remarkable, and scarcely another instrument can produce so many nuances of expression and intensity. The violin can play all the chromatic semitones or even microtones over a four-octave range, and, to a limited extent, the playing of chords is within its powers. In short, the violin represents one of the greatest triumphs of instrument making.8
This charged description reads more like a continuation of Heron-Allen’s paean than a twenty-first-century scholarly account. There is much that begs
112 Designing Instruments, Calibrating Machines
to be unpacked here. What, for example, might it mean for an entire instrumental species to be “acoustically perfect”? One might expect such an invocation to be followed by a technical description of the resonating properties of the instrument or its powers of projection, but the description that follows in the encyclopedia article—which invokes the instrument’s range, timbre, and versatility—does not speak to acoustics. What does “perfect” actually mean? Why should this adjective adhere to an instrument that is so notoriously hard to play and frequently causes neck, back, arm, and wrist injuries? Or whose sound can be so strident and harsh in the hands of all but advanced performers? Here “perfection” stands in for a wide range of things: the instrument’s adaptability, global use, robust history, and legibility of performance—the clear correlation between gesture and sound, which creates the conditions for particular experiences of virtuosity. The Grove article is an example of the kind of celebratory writing that surrounds the violin, clinging with more tenacity than it does to the music performed using the instrument. One might, for example, call to mind all of the critical scholarship that has addressed the origins of Beethoven’s mythic status and contextualized his greatness, from that of Scott Burnham to Tia DeNora to Nicholas Mathew.9 Scholars have Beethoven Hero, but we have as yet no equivalent Stradivarius Hero, no comprehensive study examining the historical formation of the lofty rhetoric and systems of value that surround the violin.10 The dearth of critical literature about the violin’s values and status does not, however, imply that the instrument has not been studied rigorously. More than any other musical instrument, the violin has been subjected to scientific study and analysis. Some of this research has focused on obtaining a working understanding of the physics of the violin, an a posteriori theory that explains the artisanal practices of luthiers and violinists; some has sought to make interventions into makers’ practices, offering improvements to the instrument’s construction; last, there has been research on specific violins, often though not exclusively “Old Italians,” carried out for the express purpose of understanding and recreating their properties.11 Across the board, much of this research has been driven by violin makers and restorers. The Violin Society of America, for example, was founded in 1973 “for the purpose of promoting the art and science of making, repairing and preserving stringed musical instruments and their bows.”12 Since 1975, the society has sponsored twenty-two competitions for instrument makers, and its journal is one of the main organs for string-instrument research. Browsing the recent issues of the Journal of the Violin Society of America, we find articles on topics such as wolf-tone beats on the cello, x-ray studies of Cremonese instruments, and
Hearing Perfection 113
analyses of the shape of the arching of violin plates. The journal has always invited historical studies as well, but its primary goal is ultimately practical: to help makers and restorers improve their craft and violin dealers authenticate their instruments. Outside of lutherie, the violin has been taken up by physicists and acousticians since the early nineteenth century. Félix Savart investigated the nature of the vibrating body of the violin; Hermann von Helmholtz researched the patterns of motions of vibrating strings and the nature of the soundwaves produced by the bow (sawtooth waves).13 In some cases, these studies aimed to understand the violin as a sounding object and to improve its construction; in others, the violin was used as a scientific instrument to test acoustic properties of different materials. In other words, the violin moved rather seamlessly between being an object of study and a tool for study. An additional strand of research has been produced by scientists for whom the violin was a pet project, while their main areas of research lay elsewhere. Thus, the physicist Frederick Saunders, distinguished first for his work in atomic spectroscopy, carried out extensive research on the violin throughout his career. In 1949, he began collaborating with violin maker and researcher Carleen Hutchins, who was particularly interested in studying the tuning of the plates of the violin.14 Together, they produced a set of experimental violins and later, with Joseph C. Schelleng and Robert E. Fryxell, founded the Catgut Acoustical Society (CAS).15 In the early 1980s, the biochemist Joseph Nagyvary turned to the study of the varnish of Stradivari instruments (a subject of speculation since the nineteenth century), publishing extensively on the chemical makeup of Stradivari’s varnish and then, in collaboration with luthier Guang-Yue Chen, making string instruments. Today, these instruments are advertised as being “based on 40 years of research on Stradivari, Guarneris, and Amatis.”16 That this research has not sought to interrogate or deconstruct the violin’s mythic status is hardly surprising. For the maker, the violin is very much a living thing, an object that continues to present problems to be solved. For the scientist, it is precisely the instrument’s status that motivates the research and bestows value on the results. And for the dealer, these mythologies are key to the objects’ market values.
Testing Today Recently, several experiments on the violin have taken aim directly at the key assumption that has motivated so much research into the violin, namely, that Old Italians truly are the very “best” violins. Just as the notion that the violin
114 Designing Instruments, Calibrating Machines
is, as an instrumental species, “perfect,” this notion of the superiority of the Old Italians is likewise layered: it encompasses everything from beliefs in the instruments’ playability, to their quality of sound and their ability to project, to their durability (i.e., that a Stradivari instrument will continue to be a fine instrument for years to come). In other words, these are the most perfect examples of a perfect class of instruments. These experiments on Old Italians were led by Paris-based researcher Claudia Fritz, in collaboration with violin maker Joseph Curtin, computational statistician Jacques Poitevineau, assessment expert Palmer Morrel- Samuels, and musical string designer Fan-Chia Tao. Their study was motivated by the unquestioned assumption of the superiority of the instruments of Stradivari and Guarneri, despite the fact that, in the words of physicist Gabriel Weinreich, “no [objectively measurable] specification which successfully defines even coarse divisions in instrument quality is known.”17 The first experiment was carried out in 2010 in Indianapolis, when Fritz brought together twenty-one violinists of varying playing levels—from advanced amateur to professional—to test their instrument preferences. The experiment involved six instruments: three old (a Stradivari and two by Guarneri del Gesù) and three new instruments by contemporary makers. The experiment was double blind; players wore dark glasses and the instruments were perfumed to disguise their smell. The violinists were each given an hour with the six instruments and asked to choose the instrument “they would most like to take home with them.” Each player was also asked to rank the “best” and “worst” instruments according to “range of tone color, projection, playability, and response.”18 One of the newly made violins was, by a significant margin, the most preferred instrument; the Stradivari was, also by a large margin, the least favorite. Various criticisms were voiced concerning the experiment, with objections to the level of the performers, the dry acoustics of the space in which the violins were tested (the test was carried out in a hotel room), the short amount of time players had with the instruments, and the small number of violins included in the study.19 In September 2012, Fritz carried out a much more extensive experiment in Paris. The test began with twenty-four violins, which were first evaluated by a team of elite violin makers and reduced to twelve instruments, six new (but “antiqued” with respect to appearance) and six old. This time, the performers included only professional soloists. Again the experiment was double blind, though the disguising scent was not used this time. First, in a relatively dry space, the violinists were asked to imagine they had to choose an instrument to replace their own violin “for recitals and concerto performances in an upcoming tour.”20 To this end, they were asked to
Hearing Perfection 115
choose their favorite four instruments and then to rate each instrument on a scale of 1 to 10 for loudness, estimated projection, playability, tone quality, articulation and clarity, and overall preference. In the next stage, the performers tested the violins in a concert hall, where they worked with a professional accompanist and received feedback from a “designated listener” in the audience. The nature of this listener was not specified: it could be a friend, another test- taker, or Hugues Borsarello, a professional violinist who was available to be consulted as an audience member. Borsarello could also be asked to play a violin so that the player could, him-or herself, listen to the violin in the audience. Again, players were asked to rank the violins. Finally, they were asked to play a randomized series of instruments for thirty seconds and guess whether each instrument was old or new. In the end, a newly made violin again won as the most preferred instrument; again, a Stradivari was ranked last. Furthermore, in the final part of the test, players were not reliably able to tell whether they were playing a new or an old violin. There was, however, a tendency to assume a preferred instrument was an old violin. While the researchers acknowledged the limitations of the study’s representativeness (in terms of both the players and the instruments), they concluded that, “given the stature and experience of our soloists, continuing claims for the existence of playing qualities unique to Old Italian violins are strongly in need of empirical support.”21 Predictably, these studies drew widespread attention in the international popular press, with titles such as “Les Stradivarius face à la science” (Le Figaro), “A Strad? Violinists Can’t Tell” (New York Times), and “Los Stradivarius no son tan buenos” (Huffington Post Spain).22 The results of the experiments fly in the face of the extraordinary prices fetched by Old Italians, prices that have recently reached nearly $16 million.23 The notion that these instruments are overhyped and not so special after all makes for delicious headlines. Such stories of underdog success resonate with the blind tests that have been carried out with wines, such as the famous “Judgment of Paris” test of 1976, in which a California Cabernet Sauvignon from Stag’s Leap Wine Cellars took first place over French wines.24 These tests are not the only ones of late to take aim at the mystique of the Old Italian instruments. In 2010, Jean-Philippe Echard—a chemist and researcher at the Musée de musique in Paris—and his research team published the results of their chemical analysis of the finish on Stradivarius instruments. As mentioned earlier, Stradivari’s varnish has attracted much attention, with some, such as Joseph Nagyvary, believing that the secret to these instruments’ sound lies wholly in the special nature of the varnish compound. For the first time, Echard and his team were able to obtain actual samples from four of
116 Designing Instruments, Calibrating Machines
Stradivari’s violins, along with an additional sample from a viola d’amore. The secret is that there was no secret: Stradivarius used common ingredients. Traveling back in time, we quickly discover that the results of these recent tests perhaps should not be so shocking. In 1994, ocean acoustics specialist and amateur violinist Herman Medwin, in collaboration with the Tokyo String Quartet, organized various tests of old and new instruments in MIT’s Kresge Auditorium, as part of the biannual meeting of the Acoustical Society of America. New instruments were heard to be “better acoustically balanced” and “tonally richer” than the older instruments.25 Going back further, in the late 1910s, Walter Willson Cobbett (1847–1937) founded the Cobbett Competition for British makers to promote the merits of British violins. The first test was held in 1918, a second in 1923. After the top violins were chosen in the second test, the winning instrument, made by Alfred Vincent, was played behind a screen in alternation with a Stradivari and the audience members were asked to show their preference. Vincent’s instrument won.26 Dig a little deeper, and it emerges that a whole series of tests were carried out in the early twentieth century, pitting old and new violins against each other; many of them were explicitly intended to unsettle the extremely high prices commanded by Old Italians. Le Monde musical organized several battles of new and old instruments played in the dark. In the second of these, a newly made cello won out over the Davidov Stradivari. The third, held in 1912, involved violins. A new instrument by the Belgian maker Auguste Falisse took first place and a French instrument made in 1911 second place, with Old Italians taking third and fourth place.27 Around the same time, the American Guild of Violinists held a similar contest during its second annual convention in Chicago. New instruments took the first four places.28
Mythic Origins One possible approach to understanding these unexpected results is to tackle them through history and trace the origins of the instruments’ mythic status—that is, to seek out the formation of the rhetoric of perfection and the high status of Old Italians. It is not too hard to locate what might be called premythic periods or moments of uncertainty about the instrument’s design, future, and standard bearers. In the late eighteenth century, Cremonese instruments did not yet dominate discussions of the violin. The lengthy entry on the violin in the Encyclopédie makes no mention of Italian instruments. Nor was the instrument necessarily seen as “perfect.” Leopold Mozart, in his 1756 treatise on playing the violin, could complain:
Hearing Perfection 117 How comes it then that the violins are so unlike each other? How comes it that one sounds powerful and another weak? Why has this one, so to speak, a shrill tone; that one a wooden tone; this one a rough, screaming tone; that one a sad and muffled tone? . . . It is wholly due to the difference between the work of one man and that of another. They all decide the height, thickness, and so on by the eye, never attaining any fixed principles; so that that while one succeeds the other fails. This is an evil which indeed robs music of much of its beauty.29
For Mozart, the solution lay in “accurate research” into the instrument and its construction. “No one will take it amiss,” he continued, “if I say frankly that more depends on accurate research into the making of instruments than on the effort of scientists to prove why two consecutive octaves or fifths do not fall pleasingly on the ear.”30 But—and crucially—his criticism of the violin did not invoke older violins as models of lost perfection: Leopold Mozart’s focus is on the instruments being made, sold, and played upon at the time. In 1803, the violinist Johann Friedrich Schubert, writing in the Allgemeine musikalische Zeitung, bemoaned the fact that, over the previous half century, every other instrument had been subject to significant improvements and that the violin alone had been left behind. The only change, Schubert noted, had been the lengthening of the fingerboard, which in his view merely harmed the vibratory power of the instrument.31 Why, he asked, were builders not experimenting with different materials or with altering the form? Schubert put forth a few suggestions for (untested) modifications that he believed would make the tone of the violin stronger, fuller, and more beautiful. Some months later, the Bohemian violin maker Franz Anton Ernst responded, saying that everyone had ideas for improvements, but he had researched all the options, more or less, and they were all impossible.32 Yet it is also around this time when we begin to see Old Italian instruments held up as models of excellence. In a lengthy Allgemeine musikalische Zeitung article on the construction of the violin, the author (likely Johann Conrad Petiscus) complained about the inconsistent construction of contemporary instruments. He remarked that the names of Amati, Guarneri, Stradivari, Stainer, and Klotz “were in everyone’s mouths” and wished that “their instruments were also in everyone’s hands!”33 In 1806, the amateur musician and violin enthusiast Antoine Sibire published La Chélonomie, ou Le parfait Luthier. Chélonomie is a fanciful title that refers to the ancient Greek word for “turtle,” invoking ancient string instruments that were made with turtle shells. Here we find ardent praise of Old Italians: “Antonio Stradivarius. At this august and venerable name, I deeply bow to the patriarch of Luthiers.”34
118 Designing Instruments, Calibrating Machines
At the same time, interest in experimentation with the violin’s form continued. In 1817, the naval engineer François Chanot—who hailed from a family of violin makers—filed a patent for his own new and improved violin, which included the twelve-page “Mémoire sur un Essay tendant à perfectionner le violon, alto, Base, et Contre-base.” The essay’s rhetoric is striking: the violin can and should be improved.35 It could be made cheaper and made to sound better and to sound excellent even when brand new. In particular, Chanot was concerned with the length of fibers in the wood and believed that the longer and more uninterrupted the fibers, the better the instrument would resonate. He transformed the f-holes to a sleeker, less angled shape that facilitated more continuous wood, and eliminated the sharp corners for the same reason. He inverted the scroll, so that it encumbered the performer less when tuning. The guitar-shaped instrument Chanot designed is stylish and precociously modern (Figure 4.1). A few years later, Félix Savart published his Mémoire sur la construction des instruments à cordes et à archet.36 Savart first set out to explain how the
Figure 4.1 Violin, François Chanot, 1818 (E. 454). Musée de la musique, Philharmonie de Paris. Photo by Jean-Claude Billing. Used with kind permission.
Hearing Perfection 119
violin worked and to understand wood as a material, studying the relationship between the size of the instrument, the length of the strings, and the instrument’s overall power. To better understand the instrument, he subjected it to testing: using Ernst Chladni’s method of sprinkling sand on the various parts of the violin to render visible their vibratory patterns, he concluded that wood vibrates neither as homogenously as metal or glass nor, as Maupertuis had speculated, as a series of wholly independent wood fibers. Savart believed that the secret to making the ideal instrument lay in achieving the greatest possible amount of vibration, through symmetry and regularity—in other words, the secret to achieving perfection lay purely in acoustics. Accordingly, Savart created his own violin on a new model: the trapezoidal violin (Figure 4.2). He addressed the possible suspicions this violin would raise, acknowledging that “many people are rightly enamored with the merit of the Italian violins. . . . [H]owever, I believe that they will soon recognize the many faults of these instruments, and that they will eventually give preference to a simpler, less expensive construction that is more likely to give good results
Figure 4.2 Trapezoidal violin, Félix Savart, 1819 (E. 372). Musée de la musique, Philharmonie de Paris. Photo by Jean-Claude Billing. Used with kind permission.
120 Designing Instruments, Calibrating Machines
in almost every case.”37 Savart did not believe that his instrument sounded identical to “ordinary model violins.” He noted that the timbre was different; his violin was a little less brilliant, but it had great equality of sound, and it was “purer, softer, more mellow, speaking directly to the soul.”38 The slightly duller sound, he argued, was merely the lack of “noise,” because the instrument was built on rational principles. Savart did not explain this concept of noise; presumably he implied a lack of nonharmonic sounds. He did not openly acknowledge that what he labeled noise might be regarded by others as central to the identity of the violin. Both Chanot’s and Savart’s instruments were also tested and judged by the Académie des sciences and the Académie des beaux-arts.39 In 1817, Chanot’s violin was played by the violinist Alexandre-Jean Boucher in alternation with a Stradivari for a mixed committee drawn from both academies. He performed in a neighboring room, out of sight. According to the minutes detailing the test, the committee continuously mistook Chanot’s newly made instrument for the Stradivari.40 Savart’s instrument was presented to a mixed committee in 1819 (he printed the committee’s report at the end of his own Mémoire), likewise played by a professional violinist—Lefebvre—in alternation with a violin by Guarneri.41 The committee initially declared that the old violin had more brightness. But when the instruments were played out of sight, the two instruments were easily mistaken for each other and the members unanimously concluded that the “new violin could pass for an excellent violin.”42 So, from the beginning of formal tests pitting new and old violins against each other, new violins have won.43 Indeed, I have yet to find a test in which an Old Italian won out over a newly made instrument. When we step back and contemplate the violin’s two-hundred-year history, we find that the continual success of new violins in these tests does not ultimately upend either the value or desirability of the older instruments. There are several points to make about this. Both Savart’s and Chanot’s violins were met with genuine and robust excitement—one of Chanot’s instruments, for example, won the silver medal in the 1819 Exposition des produits de l’industrie française.44 Enough instruments were made following Chanot’s design that examples are easy to find in musical instrument collections today.45 Neither design, however, achieved long-term success,46 and by 1828, Fétis would describe both experimental models with some disdain.47 Chanot’s and Savart’s instruments eventually became canonized cautionary tales, appearing in histories of the violin as two of the most successful failed violin innovations. In Heron-Allen’s massive survey of violin making, Chanot and Savart are the stars of the fifth chapter, marvelously titled “The Violin, Its Vagaries and Its Variegators,” which chronicles a list of experimental instruments, including violins made
Hearing Perfection 121
from leather and clay.48 Chanot and Savart, in other words, are only two of the many experimental violin forms to appear—and disappear—during the nineteenth century. The ultimate failure of these experiments helped to reinforce the fixity of the standard violin model, proving that it could not be improved upon. But even more importantly, the practice of testing—regardless of the results—helped to cement the older instruments’ standard-bearing status.
Models and Copies So far, this seems like a straightforward story about the endurance of the ordinary violin model and the consolidation of the Old Italians’ status. But the relationship between new and old violins is actually more convoluted. Let us return to Sibire’s Chélonomie. One key element to this text is that Sibire wrote it with the help of his friend Nicholas Lupot, the great French violin maker. The book ends with ardent acclamation of Lupot as the master copier and updater of Old Italian violins: Faithful imitator of the greatest of Luthiers Lupot has recreated the precious varnish; Its harmonious tone comes from its color, And the copy is the model.49
Sibire’s volume and his praise of Stradivarius might have contributed to the status of old instruments, but it also helped support the value of Lupot’s copies. This kind of copying was also central to the success of the famed violin maker Jean-Baptiste Vuillaume.50 His instruments were hailed as affordable alternatives to the violins of Old Italian masters: musicians no longer needed to give up a year’s salary to buy their instrument.51 In 1834, Vuillaume won his second silver medal in the Exposition des produits de l’industrie française. The exhibition catalog lauds his instruments for their ability to mislead: “Indeed, his violins deceive the eye by their external forms, which identify them with the forms of those old productions that are so sought after, and the ear by the quality of sound, which is the same as that of these models.”52 Vuillaume also collected and sold old instruments. Famously, in 1855, on hearing that the great Italian instrument collector Luigi Tarisio had died, he made a whirlwind trip to Italy to purchase Tarisio’s collection.53 Old Italians served as models for Vuillaume’s copies. It is worth pausing here to consider what is meant by “copy.” Both the older instruments that Vuillaume collected and resold and the copies he made of
122 Designing Instruments, Calibrating Machines
them were modified and updated: he lengthened the fingerboard, changed the angle of the neck, and reinforced the bass bar inside the instrument to strengthen it. This in itself is not a new practice—throughout the history of music, old instruments have been constantly updated and altered to meet the needs of new musical practices. What was new was Vuillaume’s attitude toward his alterations: he shrewdly understated the changes he made to old instruments. In 1864, when Vuillaume sold to Lady Anne Blunt the 1721 Stradivari now known by her name, he signed a certificate of authenticity: “I guarantee the complete authenticity of this instrument which came into my possession with the original fingerboard and without having been opened: everything is original and I have not disturbed anything except what is necessary from the playing point of view.” This sounds as if he merely has reglued a seam, but he goes on: “I had to change the [bass] bar, and lengthen the neck to modern dimensions, but I conserved its original neck. This fine instrument is therefore absolutely complete and in an exceptionally rare state of preservation.”54 These changes were not necessary to make the instrument playable, but they were necessary to make it conform to nineteenth- century standards. Today, Stradivari and del Gesù instruments are as much products of the nineteenth century as they are of the early eighteenth century. Alterations like Vuillaume’s—along with major changes to the bow and strings, not discussed here—allowed the instrument to be loud enough and to participate in contemporary music making; they have been radically downplayed in the history of the instrument since the nineteenth century. It is this moment, one could argue, in which the modern violin, with its attendant values, was truly born. We can begin, then, to offer a sketch of the violin as an object. First, the violin has what we might describe as a fitful relationship with history, one that obscures time: old violins are subtly made new; new violins are aggressively made old. Furthermore, old and new violins are bound together in a relationship of mutual calibration, an eternal circle whereby new instruments are judged against old instruments, which are in turn updated to meet new standards of playing (for a related case, see Jonathan Sterne in this volume). Second, since the eighteenth century and up to the present day, the violin has demanded rigorous scientific analysis and testing. It has begged for explanations that appeal to physics, chemistry, dendrochronology and dendroarchaeology, and climatology.55 This forms a central part of the image of the violin, making it an object that embodies advanced science. Crucially, the many forms of testing covered here—from formal to informal tests, from Savart’s experiments to Fritz’s double-blind comparisons—all contribute to this image. But the relationship between
Hearing Perfection 123
science and aesthetics is not an easy one. Take, for example, the 1998 film The Red Violin, directed by François Girard,56 which follows the journey of a violin made by the fictional Cremonese luthier Nicolò Bussotti from its crafting in 1681 through its many players, ranging from an Austrian orphan to several generations of Romani musicians, a Paganini-like virtuoso Frederick Pope, and a Shanghai political officer, Xiang Pei, who hides the instrument during the Chinese Cultural Revolution. These story lines are linked and framed by the instrument’s auction in Montreal. Although the violin is fictional, it references two real instruments: the 1720 Stradivarius with red varnish, sometimes called the “Red Mendelssohn,” that had similarly disappeared, reappearing in 1990, and the 1743 Guarneri, known as “Il Cannone,” which was owned by Niccolò Paganini and meticulously copied by the French luthier Jean-Baptiste Vuillaume. For all the transhistorical drama of the film—death, sex, drugs, betrayal, persecution—one of the most striking and tense scenes takes place in a makeshift acoustics laboratory, in which a violin appraiser, Charles Morritz (Samuel Jackson), is working in secret with a restorer, Evan Williams (Don McKellar, who also cowrote the script), to authenticate the instrument. The room is improbably darkened, signaling the after-hours, off-the-books nature of their research. Two instruments—the red violin and a late nineteenth- century copy—are hooked up to a tangle of acoustical equipment. Williams declares the red violin to be “the single most perfect acoustic machine” he has ever seen; Morritz marvels at finding this most desired object, “the perfect marriage of science and beauty.” Each covets the violin, but when Williams declares that he would like to “take it apart. Find out how it works. Take some eigenmode readings on the individual plates,” Jackson’s character dismisses his analytical desire: “You don’t get it.” The restorer replies, “Oh yes, yes I do. Take a look at this. This is Mode 1, .06 kHz. Watch the response curve here. . . . I’ll ease it up slowly for you.” Herein lies the dramatic crux of the scene: test tones ascend the harmonic series—effectively creating a horror-film soundtrack— while the violin quivers, resonating with increasing intensity. We are made to feel, as Morritz does, that terrible violence is being committed to the instrument: “Stop!” he demands. Though Morritz praises the instrument’s “perfect marriage” of science and beauty, it is not a happy marriage. On the side of science, it is comparative testing that authenticates the violin; furthermore, both men praise the violin as perfect, and that perfection is defined in terms of its acoustical properties. Science has power: we hear about eigenmode readings and response curves (even if what those are or what they reveal remains opaque). At the same time, however, the scene stages the dismissal of science, first through Morritz’s
124 Designing Instruments, Calibrating Machines
horror at Williams’s deconstructive desires and then, more urgently, in his demand that the demonstration stop. Compare this scene with another, from a more recent film. In 2013, the violinist Daniel Hope starred in and helped produce the documentary Secrets of the Violin.57 Much attention is paid here to the high values attached to Cremonese instruments. One of the subthemes is fraud, and during the course of the film, the London-based violin maker Florian Leonhard works to produce a copy of Hope’s Guarneri violin, the “Ex-Lipinski,” made in 1737. This is framed as a challenge and an experiment to see if Leonhard can copy the particular, unique qualities of Hope’s instrument. A particularly telling scene begins in the Fiemme Valley in the Italian Alps, the source of the wood for Stradivari’s violins. Hope hikes to meet the forest ranger Marcello Mazzucchi while telling us that Stradivarius would choose his wood by knocking on trees with a hammer, at midnight, under a full moon. When he meets Mazzucchi, after warm greetings, Hope asks, “Why is this forest so special?” One waits with bated breath for the explanation. Perhaps Mazzucchi will talk about the special microclimate of the forest, or maybe he will delve into the theory that the quality of the wood was tied to the “Little Ice Age”—the theory that the sharp dip in temperatures between 1645 and 1715 helped slow growth, making the wood denser.58 Wonderfully, Mazzucchi responds, in Italian, “This is a magical forest.” He then launches into an impassioned description of the emotions invoked by the trees, the play of the light, and the forest’s “infinite silence.” But would it have mattered if Mazzucchi had offered a detailed explanation of the history and properties of the forest here? These two scenes stage precisely what we have seen across the instrument’s lengthy history: the violin has long demanded scientific explanation, but it also resists, with just as much force, the scientific conclusions that have been drawn.59 Scientific analysis does not demystify the instrument, but only contributes to its allure, adding another coat of mythological varnish. The diverse forms of testing that we have encountered here must also be understood as performative and aesthetic acts. What really matters when acousticians, chemists, and climate scientists pour attention over these instruments to understand the secrets of their construction is the lavishness of their attention, not the results.60 The violin, I argue, invites special classification, for it is a particular kind of technology. Its development from the early nineteenth century onward traces an almost paradoxical trajectory. We do not simply witness the growing valuation of old instruments within Western art music (since they are not exactly old instruments), nor merely the formation of objects that magically resist scientific explanation. Taken together, these histories reveal the formation of
Hearing Perfection 125
what I want to call mendacious technology. These instruments pretend to be old and unchanging; they pretend to be perfect, even when tests suggest otherwise; they continue to harbor secrets, even when scientific explanations have been offered. This is, importantly, not to condemn either the instrument or its makers. Rather, mendacious technology performs a productive, even essential, role within musical history, for mendaciousness also implies a capacity for storytelling. We believe the lie because it is the foundation of a musical tradition that is focused on musics of the past. But the past of music history is not truly past (or at least not yet), and the violin, with its messy conglomeration of social, aesthetic, scientific, and monetary values, is one element that continually animates other times. Our access to the musical canon of art music rests on a careful obfuscating of technological history; mendacious technologies enable us to travel through time.
Acknowledgments This chapter benefited greatly from the careful and thoughtful input from this volume’s editors and the keen insights of Daniel Callahan, Alex Csiszar, and Jonathan Sterne.
Notes 1. For a discussion of a similarly “humanized” instrument, the Turkish saz, see Eliot Bates, “Social Lives of Musical Instruments.” 2. This is partially due to the existence of other specialized journals such as The Strad or Strings. These magazines, despite containing well-researched pieces, aim for a largely popular and not a scholarly audience. 3. Note, for example, Geerten Verberkmoes’s emphasis in his 2013 article that Boussu’s instruments held in the Musical Instrument Museum in Brussels are important because they are “preserved in near original, unaltered condition.” Verberkmoes, “Benoit Joseph Boussu,” 117. 4. The transversal thinking in this chapter is inspired in part by Roger Moseley’s recent Keys to Play, tracing the long history of the keyboard interface, a subject that takes him from the ancient world to the present. In looking at this broad history, I do not mean to imply that there are no important or vital local histories to be told about the violin: there are hundreds of them. 5. Reports by the Juries, 324–35. 6. François-Joseph Fétis, Antoine Stradivari. See also texts such as Charles Goffrie, The Violin. Today, nonacademically focused books on the violin abound, such as Toby Faber, Stradivari’s Genius.
126 Designing Instruments, Calibrating Machines 7. Edward Heron-Allen, Violin-Making, 104. 8. David D. Boyden et al., “Violin.” 9. Scott G. Burnham, Beethoven Hero; Tia DeNora, Beethoven; Nicholas Mathew, Political Beethoven. 10. This is poised to change: Montreal-based violin maker Thomas Wilder, who completed an MA thesis on the idea of patina and the violin, is now writing a dissertation about the rise of fetish value of the violin in the nineteenth century, and musicologist Christina Linsenmeyer, who completed her dissertation in 2011, is working on a full-length monograph dealing with the status of Stradivari violins. Thomas Wilder, “Patina”; Wilder, “Forging of an Icon”; Christina Linsenmeyer, “Competing with Cremona.” 11. See, for example, Oliver E. Rodgers, “Tonal Tests.” 12. Violin Society of America, “About the VSA,” accessed October 21, 2019, https://www. vsaweb.org/About. 13. For more detailed histories and summaries of violin research, see Joseph Curtin and Thomas D. Rossing, “Violin”; Carleen M. Hutchins, “History of Violin Research.” 14. See, for example, Carleen M. Hutchins, “Acoustics of Violin Plates.” 15. An account of the founding of the CAS is in Hutchins, “History of Violin Research,” 1429. In 2004, the CAS merged with the Violin Society of America. 16. “About Nagyvary Violins,” accessed October 21, 2019, http://www.nagyvaryviolins.com/ about.html. 17. Claudia Fritz et al., “Player Preferences,” 760, quoting Gabriel Weinreich, “What Science Knows About Violins.” 18. Fritz et al., “Player Preferences,” 761. 19. Claudia Fritz addressed a number of the criticisms in an FAQ on her website. “Response to Criticisms of Preferences Among Old and New Violins,” accessed October 21, 2019, http:// www.lam.jussieu.fr/Membres/Fritz/HomePage/Indianapolis_FAQ.html. 20. Claudia Fritz et al., “Soloist Evaluations,” supporting information (SI), 1. The full instructions that were given to participants are included in the SI text, available at http:// w ww.pnas.org/ c ontent/ s uppl/ 2 014/ 0 4/ 0 4/ 1 323367111.DCSupplemental/ pnas.201323367SI.pdf#nameddest=STXT. 21. Fritz et al., “Soloist Evaluations.” Since this study, Fritz and her colleagues have published a further study of the powers of projection of old and new violins; again, new violins were heard to project better. Claudia Fritz et al., “Listener Evaluations.” 22. Thierry Hillériteau, “Les Stradivarius face à la science”; Pam Belluck, “A Strad?”; Miguel Ángel Criado, “Los Stradivarius no son tan buenos.” 23. The “Lady Blunt” Stradivari violin sold in June 2011 through the Tarisio auction house for $15.9 million. “ ‘Lady Blunt’ Stradivarius of 1721,” accessed October 21, 2019, https:// tarisio.com/auctions/notable-sales/lady-blunt-stradivarius-of-1721/. 24. The story of this test is entertainingly retold in George M. Taber, Judgment of Paris. A number of tests of French versus American wines have been carried out since, with similar results. 25. Malcolm Browne, “Perfect Violin.” 26. “A Violin Contest.” 27. “Salle des Agriculteurs.” 28. Earl Drake, “Display and Contest of Instruments,” 14–16. 29. Leopold Mozart, Treatise on the Fundamental Principles of Violin Playing, 14–15. 30. Ibid., 15.
Hearing Perfection 127 31. Johann Friedrich Schubert, “Über den mechanischen Bau der Violine.” 32. Franz Anton Ernst, “Noch etwas über den Bau der Geige.” 33. [Johann Conrad] P[etiscus], “Über die Violin,” col. 822. 34. L’abbé Sibire, La chélonomie, 119. 35. François Chanot, “Procédés de construction des instruments de musique à cordes et à archet,” French Patent, December 11, 1817. 36. Félix Savart, Mémoire sur la construction des instruments. 37. Ibid., 75. 38. Ibid., 75–76. 39. For a detailed account of these tests and the judges involved in them, see Linsenmeyer, “Competing with Cremona,” 101–38. 40. “Séance du samedi 26 juillet 1817.” Chanot presented a second time to the Académie des beaux-arts in 1819, bringing in an entire quartet of instruments made on his new model. 41. François-Joseph Fétis identifies the make of the instrument used in alternation with Savart’s trapezoidal violin in Fétis, “Exposition des produits de l’industrie,” 36. 42. Savart, Mémoire sur la construction des instruments, 115. 43. This is not an exhaustive discussion of instruments that were tested in this way. Linsenmeyer brings much-needed attention to Jacques-Pierre Thibout’s presentations made to the Académie des beaux-arts in 1820 and 1827. Thibout’s instruments, unlike Chanot’s (which he heavily critiqued) and Savart’s, retained the traditional form. These were also tested against Old Italians and were heard to be superior. See Linsenmeyer, “Competing with Cremona,” 144–56. 44. See Louis Costaz, Rapport du jury central, 266–67. 45. For example, MIMO (Musical Instrument Museums Online; http:// www.mimo- international.com/), which gives access to many public collections in Europe, has fifteen guitar-shaped instruments in the style of Chanot (including a cello and a viola). The Museum of Fine Arts in Boston has a guitar-shaped violin made by an American maker in the 1840s. 46. Chanot’s design has been revived recently: Luis Leguia’s company, Luis and Clark, has been making carbon fiber instruments in this shape (https://luisandclark.com/, accessed October 21, 2019). 47. See, for example, Fétis, “Exposition des produits de l’industrie.” 48. Heron-Allen, Violin-Making, 104–21. 49. “Du plus grand des Luthiers imitateur fidèle, /Lupot a recréé le vernis précieux; /C’est de son coloris le ton harmonieux, /Et la copie est le modèle.” Sibire, La chélonomie, 149. 50. According to Fétis in his report for the 1855 Exposition universelle, Vuillaume collaborated with Savart to carry out analyses on Old Italians to improve his copies. F. J. Fétis, “Exposé historique,” 698. 51. “Du perfectionnement des instrumens.” Praise for Vuillaume’s affordability was commonplace. 52. “Ses violons trompent, en effet, la vue par des formes extérieures qui les identifient avec celles de ces antiques productions si recherchées; et l’oreille par la qualité des sons qui sont les mêmes que ceux de ces modèles.” Gabriel-Victor Moléon, A. Cochaud, and A.-O. Paulin-Desormeaux, Musée industriel, 3:204. 53. For a discussion of this trip, see David Schoenbaum, The Violin, 153–54. 54. Vuillaume, quoted in Jean- Philippe Echard et al., “Documentary and Material Evidence,” 183.
128 Designing Instruments, Calibrating Machines 55. See, for example, Henri Grissino-Mayer, Paul R. Sheppard, and Malcolm K. Cleaveland, “Dendroarchaeological Re-examination”; Lloyd Burckle and Henri D. Grissino-Mayer, “Stradivari, Violins, Tree Rings.” 56. François Girard, The Red Violin. 57. Nicole Kraack, Secrets of the Violin. 58. See, for example, Burckle and Grissino-Mayer, “Stradivari, Violins, Tree Rings.” 59. See, for example, Philip Ball, “Science Can Only Tell Us So Much,” which was published in response to Claudia Fritz’s 2017 study. 60. It remains to be seen if Claudia Fritz’s experiments will have an impact on future violin making, potentially breaking the cycle of mutual calibration. Fritz has reported that some violin makers are delighted by the results of her tests because “they feel like they are liberated . . . because they feel less pressure to copy Strad.” Fritz, “Is a Stradivarius Violin Easier to Hear?”
References Ball, Philip. “Science Can Only Tell Us So Much About Stradivarius Violins.” Nature News, May 8, 2017, https://www.nature.com/news/science-can-tell-us-only-so-much-aboutstradivarius-violins-1.21954. Bates, Eliot. “The Social Lives of Musical Instruments.” Ethnomusicology 56, no. 3 (2012): 363–95. Belluck, Pam. “A Strad? Violinists Can’t Tell.” New York Times, April 7, 2014, sec. Science. Boyden, David D., Peter Walls, Peter Holman, Karel Moens, Robin Stowell, Anthony Barnett, Matt Glaser, Alyn Shipton, Peter Cooke, Alastair Dick, and Chris Goertzen. “Violin.” Grove Music Online. Oxford Music Online, January 20, 2001, http://www.oxfordmusiconline.com/ subscriber/article/grove/music/41161. Browne, Malcolm. “Perfect Violin: Does Artistry or Physics Hold Secret?” New York Times, June 14, 1994, sec. Science. Burckle, Lloyd, and Henri D. Grissino-Mayer. “Stradivari, Violins, Tree Rings, and the Maunder Minimum: A Hypothesis.” Dendrochronologia 21, no. 1 (2003): 41–45. Burnham, Scott G. Beethoven Hero. Princeton, NJ: Princeton University Press, 1995. Costaz, Louis, ed. Rapport du jury central sur les produits de l’industrie française. Paris: Imprimerie royale, 1819. Criado, Miguel Ángel. “Los Stradivarius no son tan buenos.” El Huffington Post (Spain), July 4, 2014. Curtin, Joseph, and Thomas D. Rossing. “Violin.” In The Science of String Instruments, edited by Thomas D. Rossing, 209–44. New York: Springer, 2010. DeNora, Tia. Beethoven and the Construction of Genius: Musical Politics in Vienna, 1792–1803. Berkeley: University of California Press, 2008. Drake, Earl. “The Display and Contest of Instruments.” Violinist 12, no. 3 (1912): 12–16. “Du perfectionnement des instrumens à cordes et des archets.” Revue et gazette musicale, May 1836: 168–69. Echard, Jean-Philippe, Justine Provino, Thierry Maniguet, Christine Laloue, Joël Dugot, and Stéphane Vaiedelich. “Documentary and Material Evidence of Nineteenth-Century Interventions on Musical Instruments of the Collection of the Musée de la musique in Paris.” In Conservation in the Nineteenth Century, edited by Isabelle Brajer, 181–194. London: Archetype Publications, 2013.
Hearing Perfection 129 Ernst, Franz Anton. “Noch etwas über den Bau der Geige.” Allgemeine musikalische Zeitung 7 (1804): cols. 49–56. Faber, Toby. Stradivari’s Genius: Five Violins, One Cello, and Three Centuries of Enduring Perfection. New York: Random House, 2005. Fétis, François-Joseph. Antoine Stradivari, luthier célèbre connu sous le nom de Stradivarius; précédé de recherches historiques et critiques sur l’origine et les transformations des instruments à archet et suivi d’analyses théoriques sur l’archet et sur François Tourte, auteur de ses derniers perfectionnements. Paris: Vuillaume, 1856. Fétis, F[rançois] J[oseph]. “Exposé historique de la formation et des variations de systèmes dans la fabrication des instruments de musique.” In Exposition universelle de 1855: Rapports du jury mixte international, 2: 657–708. Paris: Imprimerie impériale, 1856. Fétis, François- Joseph. “Exposition des produits de l’industrie: violons, altos et bases, perfectionnés par M. Thibout.” Revue Musicale 2 (1828): 25–36. Fritz, Claudia. “Is a Stradivarius Violin Easier to Hear? Science Says Nope.” All Things Considered, NPR, May 8, 2017. Fritz, Claudia, Joseph Curtin, Jacques Poitevineau, and Fan-Chia Tao. “Listener Evaluations of New and Old Italian Violins.” Proceedings of the National Academy of Sciences of the United States of America 114, no. 21 (2017): 5395–400. Fritz, Claudia, Joseph Curtin, Jacques Poitevineau, Hugues Borsarello, Indiana Wollman, Fan- Chia Tao, and Thierry Ghasarossian. “Soloist Evaluations of Six Old Italian and Six New Violins.” Proceedings of the National Academy of Sciences of the United States of America 111, no. 20 (2014): 7224–229. Fritz, Claudia, Joseph Curtin, Jacques Poitevineau, Palmer Morrel-Samuels, and Fan-Chia Tao. “Player Preferences Among New and Old Violins.” Proceedings of the National Academy of Sciences of the United States of America 109, no. 3 (2012): 760–63. Girard, François, dir. The Red Violin. New Line International; Channel Four films; Telefilm Canada; Rhombus Media/Mikado production, 1998. Goffrie, Charles. The Violin: A Condensed History of the Violin, Its Perfection and Its Famous Makers. Philadelphia: G. André, 1878. Grissino-Mayer, Henri, Paul R. Sheppard, and Malcolm K. Cleaveland. “A Dendroarchaeological Re-examination of the ‘Messiah’ Violin and Other Instruments Attributed to Antonio Stradivari.” Journal of Archaeological Science 31, no. 2 (2004): 167–74. Heron-Allen, Edward. Violin-Making, as It Was and Is: Being a Historical, Practical, and Theoretical Treatise on the Science and Art of Violin-Making, for the Use of Violin Makers, and Players, Amateur and Professional. 2nd ed. London: Ward, Lock, & Co., 1885. Hillériteau, Thierry. “Les Stradivarius face à la science.” Le Figaro, October 9, 2012, sec. Musique. Hutchins, Carleen M. “The Acoustics of Violin Plates.” Scientific American 245, no. 4 (October 1981): 170–86. Hutchins, Carleen M. “A History of Violin Research.” Journal of the Acoustical Society of America 73, no. 5 (1983): 1421–440. Kraack, Nicole, dir. Secrets of the Violin /Das Geschäft mit der Geige. Signed Media in co- production with ZDF/Arte, 2013. Linsenmeyer, Christina. “Competing with Cremona: Violin Making Innovation and Tradition in Paris (1802–1851).” PhD diss., Washington University, 2011. Mathew, Nicholas. Political Beethoven. Cambridge: Cambridge University Press, 2012. Moléon, Jean- Gabriel- Victor, A. Cochaud, and A.- O. Paulin- Desormeaux, eds. Musée industriel: Description complète de l’exposition des produits de l’industrie française faite en 1834. Paris: Bureau de la Société Polytechnique et du Recueil Industriel, 1838. Moseley, Roger. Keys to Play: Music as a Ludic Medium from Apollo to Nintendo. Oakland: University of California Press, 2016.
130 Designing Instruments, Calibrating Machines Mozart, Leopold. A Treatise on the Fundamental Principles of Violin Playing (1756). Translated by Editha Knocker. New York: Oxford University Press, 1985. P[etiscus], [Johann Conrad]. “Über die Violin.” Allgemeine musikalische Zeitung 10 (1808): cols. 785–96, 801–11, 817–29. Reports by the Juries on the Subjects in the Thirty Classes into Which the Exhibition Was Divided. London: William Clowes & Sons, 1852. Rodgers, Oliver E. “Tonal Tests of Prizewinning Violins at the 2004 VSA Competition.” VSA Papers 1, no. 1 (2005): 75–95. “Salle des Agriculteurs.” Le Guide musical 58, nos. 27–28 (1912): 459–60. Savart, Félix. Mémoire sur la construction des instruments à cordes et à archet. Paris: Deterville, 1819. Schoenbaum, David. The Violin: A Social History of the World’s Most Versatile Instrument. New York: W. W. Norton, 2013. Schubert, Johann Friedrich. “Über den mechanischen Bau der Violine.” Allgemeine musikalische Zeitung 5 (1803): cols. 769–77. “Séance du samedi 26 juillet 1817.” In Procès-verbaux de l’Académie des Beaux-arts, edited by Jean-Michel Leniaud and Catherine Giraudon, 2:159–62. Paris: École nationale des chartes, 2002. Sibire, L’abbé. La chélonomie ou Le parfait Luthier (1806). Brussels: Weissenbruch, 1823. Taber, George M. The Judgment of Paris: California vs. France and the Historic 1976 Paris Tasting That Revolutionized Wine. New York: Scribner, 2006. Verberkmoes, Geerten. “Benoit Joseph Boussu (1703–1773): Violin Maker and Notary.” Galpin Society Journal 66 (2013): 117–38. “A Violin Contest: Old Versus New.” Musical Times 64, no. 962 (1923): 250–52. Weinreich, Gabriel. “What Science Knows About Violins—and What It Does Not Know.” American Journal of Physics 61, no. 12 (1993): 1067–77. Wilder, Thomas. “The Forging of an Icon: The Violin in Nineteenth-Century London.” PhD diss., Cambridge University, 2016. Wilder, Thomas. “Patina and the Role of Nostalgia in the Field of Stringed Instrument Cultural Production.” Master’s thesis, McGill University, 2007.
5 Opelt’s Siren and the Technologies of Musical Hearing Alexander Rehding
With their bloodcurdling mechanical screams, sirens have long enjoyed emblematic status in the soundscape of modernity.1 Their role in testing hearing may at first seem less than obvious—unless, perhaps, we consider all kinds of liminality, including the pain threshold above 100 dB. To explain how sirens contributed to a better understanding of human hearing in the earlier part of their history, we have to strip away a few layers and return to the nineteenth century, to a time before sirens were able to turn the volume up to eleven; that is, to a time before the siren became electric and was marketed by companies such as Decot or Sterling circa 1900.2 That is the sound of the siren that we now know, but the nineteenth-century siren was a very different beast. During this earlier time, we find accounts of the siren such as the following lengthy passage, published in 1852 in Neue Zeitschrift für Musik, the central organ for progressive music, as part of a series of anonymous “Acoustical Letters” that communicated a popular version of scientific knowledge to an audience of musicians. The scenario described here recalls an extraordinary experiment with an unusual type of siren, which the letter writer attributes to the physicist Heinrich Wilhelm Dove (1803–1879): In the center of a large darkened room there is a rod, which under certain conditions is set to vibrate [in Schwingungen versetzt wird]. And every care must be taken that the vibrations are in regular succession and that their velocity can be continually increased. I step into the dark room . . . at a time when the rod makes but few vibrations. Neither ear nor eye tells me anything about the presence of this rod. But I feel it vibrating: my hand can sense its throbbing when it touches the rod. The vibrations become faster. They reach a certain number and now I perceive a noise. It is individual beats or thumps [Schläge oder Stöße] that I can discriminate with my ear: small explosions, whose rapid succession I can barely follow. The vibrations of the rod increase relentlessly. The explosions follow one after another
Alexander Rehding, Opelt’s Siren and the Technologies of Musical Hearing In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0006.
132 Designing Instruments, Calibrating Machines in more and more rapid succession, they become ever stronger. There is one moment when my ear can no longer discriminate; in my mind they merge into one; I perceive only a whirring noise—and suddenly a low bass tone strikes my ear. It is of such stunning intensity that no other sound—not my voice, not a musical instrument, not even the sound of an organ—can be heard at all. This tone continually rises in pitch. It passes through the entire medium range up to the shrillest high- pitched sound, which pierces our ear with unbearable intensity. But now everything falls back into dead silence. Still astonished by what I heard, I suddenly feel a pleasant warmth radiating from the very point where just now the sounds died away, as cozy as a fireplace. But everything around me is still plunged into darkness. Meanwhile the vibrations become faster still. A weak red light emerges; it becomes more and more intense, the rod is glowing. First red, then it turns yellow, then blue. It passes through all the colors until, after violet, everything sinks back into the night.3
The erotic undertones in this luxurious imagery are hard to miss. The narrator concludes this breathless account with a learned reference to the mythological sirens: “This is the siren of our century, which in its characteristic effect can well measure up to the one that once lured Odysseus.”4 No matter what we make here of the gender-bending image of the irresistibly alluring siren with its—or her?—throbbing rod, the passage describes a mechanism that has no counterpart in reality. Dove, the physicist who is cited as the authority behind this experiment, did in fact present a new siren model, but his mechanism is unrecognizable in the description here.5 It seems more likely that the passage is derived from a text on acoustics by physicist August Seebeck (1805–1849) that was published in the popular series Repertorium der Physik in 1849, for which Dove was a coeditor. This text occasionally draws on the vibration of straight rods to explain the physics of periodic sounds.6 But whatever the precise source, we can safely assume that no such experiment took place in reality. Once we strip the description of its voluptuous and fanciful elements, what remains is the idea that vibrations can manifest themselves in various sensory guises—running the whole gamut from individual pulses, to sound, to heat, to colors—and that the siren was able to demonstrate an underlying continuity across different perceptual dimensions. Of course, this extravagant account goes far beyond what was possible in empirical reality. The influential physiologist and physicist Hermann von Helmholtz (1821–1894) reported that he barely managed to bring the siren to high enough speed to figure out the frequency at which human pitch perception would end.7 The core of this acoustical letter, however, touches on a real phenomenon: there is
Opelt’s Siren and the Technologies of Musical Hearing 133
a point in human auditory perception, which was successfully identified with a siren, when pulsations are no longer perceived as individual events, but fuse into a continuous pitch. Conventional sirens, to be sure, did not produce the “explosions” that the throbbing rod seemed to elicit, only soft air puffs, but the principle holds. Before we explore this phenomenon further, we should pause to take note of the strange literary genre to which this text pertains. Calling it popular science may be the best we have to offer, but this term barely does it justice. These acoustical letters seem to be less interested in conveying scientific facts to a lay audience, in the manner that Helmholtz and others were eager to do,8 than in using recent scientific findings as a basis for poetic reflection, serving as inspiration for musicians who define themselves primarily as artists, not as scientists. Germany, it is true, prided itself on a broad definition of Wissenschaft (science) that included not only the natural sciences but also any kind of systematic knowledge acquisition, including artistic and literary production. But by the mid-nineteenth century acoustical knowledge and compositional practice were tied together only with the loosest of strings, and both domains were sufficiently specialized that there was no simple way to find a common language. A few years after the first publication of the acoustical letters, our popular scientist compiled his reflections into an epistolary pamphlet and, now giving up his anonymity, revealed himself as Richard Pohl (1826–1896), a figure from Richard Wagner’s inner circle.9 In light of his authorship, it is difficult not to read the passage against an imaginary soundtrack taken from the opening of Rheingold, with its famous Ur-sound, a gradual buildup of overtones of E-flat, which eventually leads into the finale of the music drama depicting the rainbow bridge that guides the gods into Valhalla.10 (This soundtrack has to remain imaginary, since Wagner didn’t actually compose the score until 1853.11) For all the physical problems with Pohl’s understanding of the siren, what he outlines is the basis of a music that was driven less by traditional compositional skills such as motivic invention or voice-leading expertise than by a thorough grasp of the physics of sound and the specificities of our perceptual apparatus—in other words, a music that works at the limits of audition, a music that probes and tests hearing itself. The question of what exactly “testing” means within the realm of music deserves a word of explanation. On a technical basis, music constitutes a subset of sounds—musical sounds are typically identified as those that are principally based on periodic waves, and these are perceived as having a definite pitch. And on a discursive basis, music constitutes a regulatory
134 Designing Instruments, Calibrating Machines
framework with carefully guarded boundaries. We don’t need to delve here into the institutions that uphold those boundaries—in the European nineteenth century, the principal forces include the etiquette and rituals of the concert hall, criticism and aesthetic principles, and the pedagogy of conservatory teaching. These technical and institutional levels join forces in that they relegate those sounds that they consider unmusical to the status of “noise”— as music’s unruly Other. Music theorists have traditionally enforced the boundaries between music and nonmusic (or noise) by invoking the “laws of music.” But these are not laws of nature. Rhetorical posturing notwithstanding, the rules and commonalities that music theory observes are liable to change, just as music and tastes change. Music, after all, is a cultural product. It is not an experimental practice, and musicians conduct experiments only in a very limited sense: on one level, musical compositions can be thought of as experimental challenges to our hearing or to our musical understanding—these terms are often regarded as interchangeable in musical discourse. As a consequence, the act of composing, the acoustical features of musical sound, and the mechanisms of the human auditory system are usually considered, if they are considered at all, to be largely coterminous. The differences between those three domains usually pass unnoticed, precisely because the boundaries of the discourse network have been regulated so as to minimize any discrepancies. But this is also why those rare interventions that explicitly invoke the broader field of experimental acoustics deserve special attention: by bringing to bear a field of knowledge that is typically excluded from consideration, musicians are forced to account for the equivocations that otherwise dominate the discourse. We are therefore likely to get furthest in our inquiry if we take a broad approach to the notion of testing, in its dual meaning as experiment/experience and attestation (see the Introduction in this volume). In this sense, Pohl’s thought experiment envisioned (or en-audition-ed) an experience that would draw out its subjects in ways comparable to the challenges Wagner presented to his audiences’ ears. And not unlike Wagner’s music, which in Nietzsche’s judgment was capable of hypnotizing its listeners, robbing its audiences of agency, Pohl was blithely uninterested in specifying the listener who experienced the siren. In invoking the “siren of our century,” Pohl alludes to the mesmerizing vocal allure of its ancient cousins: Pohl’s subject, the poetic “I” of the scientific test, remained an anonymous sensorium that did little more than to offer a receptive canvas for the stimuli emanating from the siren.12 Pohl’s science fiction imagined a Wagner-inspired acoustics of the future.
Opelt’s Siren and the Technologies of Musical Hearing 135
The Challenge of the Siren To approach a more accurate idea of the gauntlet thrown down by the siren to conventional theories of sound and the functioning of the human ear, we may dig a little deeper still, to a well-known chapter in the history of acoustics.13 In 1819, the French engineer Charles Cagniard de la Tour (1777–1859) revealed his new noisemaking device, which he called the “sirène”—an unhappy appellation, some later commentators felt.14 As a French speaker, Cagniard thought of sirens as mermaids, especially since his amphibious apparatus had the advantage of working under water as well as in air.15 The alarm function that has given the siren its place in modern consciousness, incidentally, could not have been further from the mind of its French inventor: his acoustic instrument was first and foremost a device to produce a pitch and measure its frequency exactly, something that could not be done with the monochord, the traditional acoustical experimental apparatus of choice. 16 Cagniard’s interest was particularly piqued by the novel sound production of this apparatus. He explained at the outset of his publication: “If, as physicists believe, instrumental sound is principally based on the regular succession of multiple impulses [chocs] passed on to the atmosphere by means of their vibrations, it seems only natural to think that with a mechanism put together so as to stir the air with the same speed and the same regularity, one could give rise to sound generation.”17 The sound generation mechanism of the siren, then, followed fundamentally different laws from those of conventional musical instruments, which were based on either strings or pipes. Like its ancient cousins, whom Homer depicted as pure disembodied voices, Cagniard’s modern siren was distinguished by the absence of a sounding body. Instead, Cagniard argued, it created sounds out of thin air. Figure 5.1 shows the apparatus that Cagniard constructed. A bellows is attached to the chamber at the bottom of the device (marked A and B in the images). Air is forced up and through the metal disk that sits at the top of the air chamber. The disk, shown in the two smaller diagrams on the right, has holes at regular intervals, which are drilled diagonally through the metal so as to set the disk into rotation. This can best be seen in the cross-section shown in the bottom right diagram. As the air pressure increases, the sound rises in both pitch and volume. Later siren types disconnected the air mechanism from the rotation mechanism to decouple volume from frequency modulation. What made the mechanism of the siren so controversial was the fact that it put to the test conventional theories about the nature of sound.18 Its mechanism was significantly different from traditional musical instruments, which
136 Designing Instruments, Calibrating Machines
Figure 5.1 The “sirène” developed by Charles Cagniard de la Tour. Hermann von Helmholtz, On the Sensations of Tone (1885), 12.
all produced steady-pitch sounds by means of plucking strings or blowing pipes, and which were all known to vibrate in a more or less complex sinusoidal waveform, producing their characteristic sound. No such back-and- forth oscillation patterns could be shown for the sound production of the siren: it was best understood as no more than a short “on” impulse followed by a longer “off ” period. Its sound derived from a succession of interruptions— or chocs, in Cagniard’s description. The venerable experimental acoustician Ernst Chladni (1756–1826) may have been the first to realize the revolutionary potential inherent in the siren.19 Chladni argued, along the lines of Cagniard’s initial reasoning, that the siren forces us to expand the very conception of what “tone” is, as it does not require “a standing vibration from a sounding body” but may, more broadly, be thought of as a series of “sufficiently strong and rapid impulses” that are “in some way communicated across the surrounding medium to the ear.”20 In other words, in the siren what caused the sensation of sound in the ear was not the regular alternating rarefaction and compression of the air around an assumed zero point of normal pressure induced by the vibrating resonant body of an instrument to produce a soundwave. Much more abstractly,
Opelt’s Siren and the Technologies of Musical Hearing 137
the mere effect of a repeated pattern of intermittently increased air pressure sufficed to generate sound. In marked distinction to the Leibnizian scientific axiom natura non facit saltus (nature does not make leaps), the siren suggested, as Philip von Hilgers has pointed out, the possibility that discrete mechanical events could give rise to a fused sonic phenomenon.21 A good decade after Cagniard’s presentation of the siren, an experimental physicist and a theoretical physicist—August Seebeck, whom we encountered before, and Georg Simon Ohm (1798–1854)—fought out a paradigmatic battle over the significance of this device.22 This much-discussed dispute has become something of a primal scene for sound studies and has received much critical attention in recent years; it needs to be recalled here only insofar as it is necessary to understand the mechanism of the siren.23 Seebeck took up Cagniard’s challenge and argued that the generation of sound through a pulsation of air puffs in the siren contradicted the idea that all sounds are based on sine waves. He inferred that information about pitch was given to the ear by means of a pulsation and argued that the sinusoidal form of the wave, which had traditionally been assumed to be crucial, was in fact not an essential part of sound generation. Ohm, in contrast, held with Joseph Fourier (1768–1830) that all complex waves can be analyzed into their sinusoidal components of various frequencies and amplitudes corresponding to multiples of the fundamental frequency. A violin string or an air column typically vibrates in multiple modes simultaneously to produce a more complex waveform and a rich timbre. A schooled ear, Ohm’s assumption ran, was capable up to a point of breaking down the complex waveform into its constituent frequencies. Thus, by picking out not only the fundamental frequency but also some of the upper partials of a musical sound, the ear could perceive, decompose, and analyze the multiple modes of vibrations that had produced the sound in the first place. Ohm concluded that the siren, its discrete “on” and “off ” periods forming something approaching what we would now call a square wave, was no exception to this rule—all it presented was a somewhat more complex sinusoid. Seebeck’s invitation to Ohm to visit him in Dresden and inspect the experimental setup himself, urging Ohm not to “seal his ears before the song of his siren,”24 went unheeded. After a formal, somewhat testy response to Seebeck in the same journal, Poggendorffs Annalen,25 Ohm never returned to the matter, and Seebeck died soon thereafter. In the 1840s, the dispute seemed to have been settled in Seebeck’s favor. It was up to Hermann von Helmholtz a decade later to reconcile the two sides. Helmholtz leaned on the distinction between Klang, the spectral conception of sound as composed of fundamental
138 Designing Instruments, Calibrating Machines
and upper partials, and Ton, the blended aural impression of the periodic vibration: Seebeck contends that Ohm’s definition of a tone [Ton] is too narrow, that besides the wave motion corresponding to the fundamental pitch other elements of the Fourier spectrum can reinforce the sensation of the fundamental pitch. This is quite correct if he understands by tone that which we have just called “sonority” [Klang] . . . while Ohm’s definition of the tone indeed seems to denote that which is the most basic element in the activity of the auditory nerves.26
In truly Solomonic fashion, Helmholtz relocated the battleground away from the modes of production and toward the physiology of the ear, where he found it possible to reconcile the ideas of both, and still managed to pass the crown to Ohm, who had come close to being slighted by the history of acoustics.27 Besides the protracted controversy about the physics of sound generation that it inspired, the siren ultimately came into focus for its ability to demonstrate a point that had previously been a theoretical possibility but not a practical, testable reality. A series of individual sounds gives rise to the sensation of one continuous musical tone, the pitch of which depends on the frequency of the periodic impulses. The discrete impulses from which the sound is produced merge into one overriding acoustical impression. This point was explored particularly by the French physicist Félix Savart, who built a separate mechanism based on the findings of Cagniard’s siren: the Savart wheel, a metal cog with a large number of teeth that, while rotating, strike a vibrating hard surface in regular intervals.28 (Anyone who has put a playing card between the spokes of a revolving bicycle wheel has experienced this mechanism.) The wheel had the distinct advantage over the siren of producing forceful individual pulses or chocs without the latter’s soft air puffs, which were a constant source of annoyance to researchers. Savart found that when the wheel rotated at slow velocity, the clicks of the individual teeth formed a steady rhythmic pulsation, but at a certain frequency, the rhythm became a continuous sound. Savart identified the lower threshold at 25 to 30 beats per second (bps) and the upper at 12,000 to 15,000 bps; they are now determined at ca. 20 Hz and 20,000 Hz.29 Savart’s interest in this device lay, once again, in testing hearing, or more specifically in testing the upper and lower limits of human hearing. But the ramifications of these findings for music are immense. This is where the discussion among physicists converges with Richard Pohl’s acoustical fantasy that we encountered earlier. “There is one moment,” Pohl wrote, “when my ear can no longer discriminate; in my mind, they merge into one.”30 The experiments surrounding the siren show that the two
Opelt’s Siren and the Technologies of Musical Hearing 139
parameters of rhythm and pitch are not the perceptually discrete categories that we treat them as, but that they exist on a continuum in different regions of the temporal axis. (This duality between the two parameters can be seen every day in conventional musical notation, which uses distinct systems of signification for rhythms and pitches: durations are indicated by a relatively complex system of note heads, stems, and flags, while pitches are indicated by the positions of the note head on the staff.) Of course, very large organ pipes—C1 is around 20 Hz—can produce similar impressions: we tend to hear more of a pulsation than a real pitch. But Savart’s wheel, with its ability to vary sound frequency, was particularly adept at demonstrating aurally that the boundaries between the two apparently separate dimensions were porous and could be transgressed. In the sonic knowledge that Cagniard’s invention of the siren unlocked, whether in the guise of Savart’s wheel or Pohl’s fantastical throbbing rod, the two perceptual parameters become a continuum—one can be transformed into the other.
Opelt’s Siren: Testing Hearing, Testing Music Among the musicians who were eager to exploit this scientific insight was Friedrich Wilhelm Opelt (1794–1863). Opelt may be the most important music theorist you have never heard of. He was busy in his day job as a tax collector, a profession in which he rose through the ranks in various Saxon towns, culminating with an appointment as privy counselor in the royal financial ministry in Dresden.31 Beyond his career as a government official, he had many hobbies that expanded his mathematical interests into a variety of fields of knowledge. He can best be described as a dedicated amateur and autodidact, a dabbler and tinkerer, a jack of all trades operating at the highest level, though outside the institutional framework of any specific discipline. In his spare time, he engaged in such mathematical projects as actuarial calculations for a possible state pension system for the state of Saxony or the translation of an important mechanics textbook from French into German. Today, he is perhaps best known for his fairly accurate calculations of the diameter of lunar craters, which he tackled together with his son Otto Moritz. In recognition of their accomplishments in this field, the father-son pair even had one crater named after them (48 km in diameter), located in the Mare Nubium. One of these projects in applied mathematics was Opelt’s contribution to music theory first published in 1834 in a short pamphlet called Über die Natur der Musik, which was later followed by the more extensive Allgemeine Theorie der Musik (1852).32 This was an unusual project, as the ancient connections
140 Designing Instruments, Calibrating Machines
between mathematics and music had become increasingly irrelevant to modern composition. Opelt sought to breathe new life into the symbiotic relationship by approaching it from the perspective of modern experimental science. Anxious to ensure the best possible reception of his work, ideally an international splash, Opelt placed previews in the major musical papers of his day all over Europe, including the Leipziger allgemeine musikalische Zeitung, the Neue Zeitschrift für Musik, and the English Harmonicon.33 There was little response—the influential Belgian musicologist François Joseph Fétis (1784– 1871), who included Opelt in his Biographie universelle des musiciens, concluded laconically that in all likelihood the Leipzig critic Gottfried Wilhelm Fink (1783–1846) and he himself were the only people to have read Opelt’s work.34 This sparse reception has persisted into the present. If we add the names of the Prussian physics educator Ernst Robel (ca. 1845?–1920?), the Berlin physiologist and acoustician Karl Ludolf Schaefer (1866–1932), and the director of the West German Radio recording studio in Cologne, Herbert Eimert (1897–1972), we have a fairly complete bibliography of those referencing Opelt’s musical theories. This slim reception proves nothing so much as that Opelt was taking a radically different tack from most of his musical contemporaries. Opelt took the insights offered by Cagniard’s siren and ran with them. He developed the mechanism further to propose a polyphonic siren, which served as the main tool with which to test and demonstrate his music-theoretical ideas. His starting point was the basic observation that it was possible to play two sounds at the same time on one siren, by imprinting two independent concentric circles of holes on the same disk. If the holes between the two rows were, say, in the ratio of 2:3, the resulting interval would be a fifth, and both pitches would rise at the same rate, preserving the interval between them independently of the frequency at which the disk was rotating. This only stands to reason, since one of the oldest insights into musical intervals, usually attributed to Pythagoras, is the understanding of musical intervals as ratios.35 Opelt’s siren showed empirically that it was possible to produce the same intervals not with two (or more) independent rows, but with only one joint row that had holes at the rate corresponding to the compound rhythm of the multiple pulsations.36 The siren he developed, as depicted in Figure 5.2a, was built out of heavy cardboard and offered a wide range of combinations. Here Opelt landed at least one coup in his attempt to promote his theories of sound: his sirens were produced by the important purveyor of acoustical devices, Rudolf Koenig in Paris. Not least thanks to Koenig’s outstanding distribution services, Opelt’s siren became more widely known than his theoretical oeuvre around it. Thus, when Dove introduced his new
Opelt’s Siren and the Technologies of Musical Hearing 141
siren type in 1851, he credited Opelt as the inventor of the multiphonic siren and even touted this as a well-known fact.37 The historian of science David Pantalony has drawn attention to the “visual beauty” of the siren disks in Koenig’s collection.38 And rightly so: the intricate symmetrical patterns of Opelt’s siren, seen in Figure 5.2a, make for a striking visual display. It may not be a coincidence here that Opelt came from a family of weavers, a trade that he had learned in his early years: the intricate patterns of his siren show strong affinities with the punchcards used in mechanical Jacquard looms with their complex mechanisms. These spatialized compound relations, which could be articulated slowly in the rhythmic domain or rapidly in the pitch domain, Opelt called “rhythmic systems.”39 Figure 5.2b shows a simple example, a hemiolic rhythm (two against three): the compound rhythm (dum-di-di-dum) is converted into dots spaced out accordingly across a line. This pattern becomes the basis for the circles of holes on the siren disk. There is no limit to how far this idea can be taken. Opelt himself included complex rhythmic systems, such as one based on the quadruple ratio 4:5:6:8 as shown in Figure 5.2c, which in the harmonic domain corresponds to a major triad with the upper octave. This follows Pythagorean ratios, with the ratio 4:5 corresponding to the major third (in this case: C–E), 5:6 to the minor third (here: E–G), 4:6 = 2:3 to the perfect fifth (C–G), 6:8 = 3:4 to the perfect fourth (G–C), 4:8 = 1:2 to the octave (C–C), and 5:8 to the minor sixth (E–C).40 The circular diagram of Figure 5.2c shows how these ratios, converted into distances between dots, are then applied to the siren disk. The compound rhythm marks the outermost ring, whereas the four rings marked G, T, Q, and O (for root, third, fifth, and octave—or Grundton, Terz, Quinte, and Oktave) show each individual pulsation. Continuing inward, the next ring (third from the center) maps the hemiolic ratio from Figure 5.2b onto the circular disk. The music-theoretical implications of this realization are far-reaching. One of the consequences of Cagniard’s siren—most clearly demonstrated in the tests above and below the auditory threshold with the Savart wheel—was the realization that the categorical division into separate musical parameters of pulsating rhythms and pitches was an artifact imposed by the peculiarities of our auditory apparatus.41 Opelt’s polyphonic siren took this insight as its starting point and went one important step further in that it showed, more clearly than Cagniard’s monophonic model could, that even more complex rhythms can be correlated to pitches, intervals, and chords. Opelt saw this correlation of multiple musical dimensions as a way toward a comprehensive music theory that not only embraced the dimensions of both rhythm and pitch but also extended further into aspects of harmonic progression and form.42
Figure 5.2 Opelt’s siren. (a) A multiphonic Opelt siren is turned with a crank and operated by human breath. “The Siren of Science,” Harper’s New Monthly Magazine (1872): 848. Note the numerous concentric rings that appear almost like a geometric pattern. Each ring represents a “rhythmic system” in Opelt’s sense. (b) The compound rhythms are converted into appropriately spaced dots on a line, which form the basis of the rings of holes on Opelt’s siren disk, here in a hemiolic (2:3) ratio, a simple compound rhythm. (c) Even four-voiced chords could be reproduced on one circle of Opelt’s siren. The example shows, initially along a harmonic series (spelled out in musical notation at the top), how the ratios 4, 5, 6, and 8 combine into a major triad with the upper octave. These ratios are converted into a compound rhythmic system, which forms the basis of the siren pattern. From a fold-out at the end of Opelt’s Allgemeine Theorie der Musik (1852).
Opelt’s Siren and the Technologies of Musical Hearing 143
Based on his empirical brand of neo-Pythagoreanism, he regarded periods in duple meters as corresponding to the octave, or 2:1, and triple meters as the equivalent of the interval of the octave plus a fifth, or 3:1.43 Exploring such correspondences across perceptual parameters, Opelt’s theory endeavored no less than to redefine music in all its aspects from the perspective of rhythm— based on the givens of human hearing. “Music itself,” he concluded, “from the individual sound to the most complete musical edifice, rests exclusively on the rhythm given in the motion of the soundwave.”44 No wonder that Opelt thought a better name for the siren would have been “rhythmometer”—it was a device with which to measure the foundations of music against the limits of human hearing.
Opelt’s Siren as Music-Theoretical Instrument For Opelt, the essence of musical sound resided in the Knall (“bang”), an explosive noise that corresponds to Cagniard’s choc and that we would probably now call “click.”45 Starting from this ontology of clicks, he built up his entire music theory. Later he offered a definition of musical tones as “the effect of uncountably fast vibration pulses of sounding bodies, moving uniformly like a pendulum.”46 In Opelt’s world, musical parameters are merely different manifestations of clicks along a temporal axis, operating in different dimensions or frequency ranges. What characterizes tones is that their vibrations are no longer countable; those of rhythms are within the realm of the comfortably countable; and metric units, as the temporal basis of periods and larger-scale formal parts of musical compositions, are one order of magnitude slower still. We could formalize Opelt’s musical ontology as follows: tones predominantly operate at the kHz level, rhythms cluster at the Hz level, and meter operates below the Hz level along the frequency axis. Opelt’s definition of musical tones refuses to take sides across the conceptual chasm between Ohm and Seebeck. He still uses the pendulum swing as his image, indicating a sinusoidal motion, and his invocation of the “wave motion of sound” (Klangwellenbewegung) suggests a traditional—that is, continuous—conception of the soundwave. At the same time, the individual explosions, the air puffs from his siren, are the hallmark of a discrete sound generation.47 To be sure, there is no evidence that Opelt was aware of this scientific conundrum; the 1834 pamphlet Über die Natur der Musik preceded the Ohm–Seebeck dispute, and neither of his music-theoretical publications engaged with the current scientific literature.48 It seems that the questions of the physics of sound generation that so agitated Ohm and Seebeck were relatively
144 Designing Instruments, Calibrating Machines
unimportant to Opelt. He was much more focused on the question of what happens when the sonic information is converted into musical parameters in the ear. Opelt’s conception of the physiological and psychological components follows in the traditional philosophical mold: he distinguished between intellect (Verstand) and feeling (Gefühl), based on the boundary between countability and uncountability, which corresponds to the boundary between rhythm and pitch. One is reminded of Leibniz’s famous definition of music as “a hidden arithmetic exercise of the soul, which does not know that it is counting.”49 In Opelt’s case, when vibration passes the auditory threshold at 20 Hz, “feeling” takes over from “intellect.” “Feeling does not calculate,” he explained, “but it does count—insofar as this is possible without being aware of numbers.”50 Opelt posited a recursive relationship between the various parameters: the notions of dissonance and consonance that govern the classification of intervals on the pitch level should also apply at the macro-level in the rhythmic and metric dimensions, and vice versa: “Musical chords are nothing other than rhythm appearing in a different guise, manifesting itself in one fused sonority [Klang]. The effect assigned to a specific rhythm in the realm of the countable must therefore also appear in complete analogy in the realm of sonorities.”51 This relationship between perceptually distinct realms is made apparent by the spatialization of temporal relations on which the mechanism of the siren is based. As the diagrams in Figure 5.2 show, the siren disk converts the events of simple or compound rhythms into dots along a circular line, transposing the time intervals between them into spatial distances. Another way of putting this is that Opelt’s siren converts one kind of notation into another. The conventional system of staff notation must decide whether an impulse is a rhythm or a pitch, which is quite unsuitable here. Opelt’s notation is based simply on holes punched in the cardboard disk and the distances between them. Unlike conventional notation, which must be interpreted by a musician in performance, Opelt’s notation is machine readable, a semiotic system that can be applied at various powers of 10n along various ranges of the frequency axis. This notation is a critical component of the siren’s function. When the disk rotates rapidly, the inscription takes on the role of note heads of conventional notation (indicating pitches); when it rotates somewhat slowly, the same inscription takes on the role of stems, flags, or beams; and when it rotates very slowly, it represents the equivalent of bar lines and phrase slurs.52 The principles of Opelt’s machine notation are based on holes and covered distances—or, translated into processual terms, “on” and “off ” periods. This semiotic system is especially efficient at
Opelt’s Siren and the Technologies of Musical Hearing 145
highlighting— that is, demonstrating perceptibly— the correspondences between the parts that conventional notation separates into the individual dimensions of each note or phrase. It shows with ample clarity that each of these musical dimensions operates along a different range of the frequency axis, or of its inverse, the time axis. We could describe the operativity of Opelt’s notation at different levels as a form of “time axis manipulation,” to invoke a term proposed by the media theorist Friedrich Kittler, to describe one of the essential functions of writing and other media.53 On the broadest level, time axis manipulations involve “strategies of spatialization to enable one to manipulate the order of things that progress in time.”54 Opelt’s siren employs the principle of time axis manipulation effectively to “fold” the temporal axis into the three musical dimensions of pitch, rhythm, and meter and to display the underlying recursive relation that holds the three together across their auditory differences. Opelt’s siren notation differs significantly from standard musical notation in one further important aspect: Opelt’s polyphonic siren disk serves not only as a form of sonic notation but also as the material basis of sound generation itself. Sybille Krämer points out that it is a typical feature of technological media to “not only transmit data but also—somehow—bring them forth.”55 Unlike standard musical notation, which maintains a strict material separation between sheet music on the one hand and musical instrument on the other, Opelt’s siren notation serves both as the storage medium and as the device that resonifies the data. On this basis, we can configure Opelt’s siren as a “music-theoretical instrument,” an acoustic device that produces sounds and simultaneously generates insights about music—or rather, that brings together “auditory knowledge” with “knowledge about audition.”56 Put differently, Opelt’s siren tests hearing against a background of musical sound. The epistemic forte of Opelt’s music- theoretical instrument, as we saw, is to highlight the recursive relations between different musical parameters. In this function, Opelt was poised to open a whole new chapter of musical thought, unlocked by his siren’s sonic and epistemic force. One could object that Opelt’s specific definition of sound and, consequently, his entire theory of music are so selective that they have little to do with actual music. This criticism would not be wrong, as we will see in more detail in the next section, but Opelt takes the testing function of his siren quite seriously. This becomes particularly apparent in those sounds that Opelt explicitly excludes from his theory, specifically nonperiodic sounds, which would be represented in Opelt’s notation as irregular successions of pulses— in other words: noise. This category of noise Opelt describes with disarming
146 Designing Instruments, Calibrating Machines
precision as “unmeaning” (to employ the beautiful English translation used in the announcement in The Harmonicon of 1832).57 Such sounds that have no meaning are emphatically banished from Opelt’s system. Music, we could add to Opelt’s own attempted definitions, is a closed semiotic system that is rooted in the peculiarities of the auditory system and whose hidden features become manifest thanks to his siren.
Toward a Siren Principle in Music The music-theoretical instrument into which Opelt fashioned his siren was radically innovative, as Fink and Fétis quickly realized. But it was also hopelessly overdetermined. Opelt’s extremely stringent criteria—the idea that all aperiodic rhythms must be excluded as “unmeaning,” in tandem with the concept of a recursive relationship between pitches, rhythm, and meter—rule out virtually all existing music. There is at most one composition that goes some way toward employing in its musical structures the principles Opelt described. The American experimentalist Henry Cowell (1895–1967) wrote a youthful Quartet Romantic (1915) that was based on the “demonstrable physical identity between rhythm and harmony.”58 This composition for two flutes and two violins works on two levels at the same time: the four instruments play freely atonal melodies at fixed rhythms, but the rhythms themselves contain encoded pitch information, so that the proportions between the rhythmic pulsations outline an underlying tonal structure. Not dissimilar from Opelt’s “rhythmic systems,” the rhythmic structures of this piece are based on ratios corresponding to an inaudible “conceptual” chordal progression. To give just one example, in Figure 5.3a, the first measure juxtaposes six, five, four, and two pulsations, ratios that correspond to a major triad over a root in the bass in Figure 5.3b. Over the course of the piece, these rhythms actually encrypt a fully tonal Bach-style chorale in four parts, a secret meta-composition that remains unheard by human ears, along the lines indicated in Figure 5.3c. It is true that this piece does not sound remotely like a Bach chorale. But we could imagine etching the compound rhythms onto Opelt sirens and speeding them up.59 Played at a sufficiently fast tempo, the proportions of this secret music would become audible as harmonies. Henry Cowell did not know Opelt’s work, but he was clearly thinking of the mechanism of the siren, as he made clear in his treatise New Musical Resources, which was begun around the same time as the Quartet Romantic, though it was not published until 1930. Here Cowell explored
Opelt’s Siren and the Technologies of Musical Hearing 147
Figure 5.3 Henry Cowell, Quartet Romantic, Quartet Euphometric (1915). (a) The opening measures of the quartet. The pulses of each voice combine into what Opelt would call a “rhythmic system” based on ratios that can be converted into harmonies. (b) The harmonic series shows how the 2:4:5:6 ratio links the rhythms of the opening measures of the Quartet Romantic to a major chord with the octave doubled in the bass. (c) The beginning of Cowell’s analysis of the harmonies encoded in the rhythmic systems of his Quartet Romantic.
how time is “translated, as it were, into musical tone.”60 In that context, he made explicit the relationship of this idea to the mechanism of the siren: There is a well-known acoustical instrument which produces a sound broken by silences. When the silences between the sound occur not too rapidly, the result is a rhythm. When the breaks between the sound are speeded, however, they produce a new pitch in themselves, which is regulated by the rapidity of the successive silences between the sounds.61
While the notion of “regulatory silences” that he conjured up in this description must have struck most of his readers as bewildering, Cowell went on to
148 Designing Instruments, Calibrating Machines
assure his readers that his suggestion would serve to “open up new fields of rhythmical expression in music.”62 The recursivity that guided Opelt’s thoughts in developing his Allgemeine Theorie der Musik also exerted a fascination on composers who were interested in finding immanent principles to structure their musical material a generation later. In the 1950s, the German avant-garde composer Karlheinz Stockhausen (1928–2007) independently became fascinated by this idea. In a pointedly Rousseauian gesture, he described an idea that had occurred to him during a retreat in a remote village in the Swiss Alps: I recorded individual pulses from an impulse generator and spliced them together in a particular rhythm. Then I made a tape loop of this rhythm, let’s say it is tac- tac, tac, a very simple rhythm—and then I speed it up, tarac-tac, tarac-tac, tarac-tac, and so on. After a while the rhythm becomes continuous, and when I speed it up still more, you begin to hear a low tone rising in pitch. That means this little period tarac-tac, tarac-tac, which lasted about a second, is now lasting less than one- sixteenth of a second, because a frequency of around 16 cycles per second is the lower limit of the perception of pitch. . . . The timbre of this sound is also an effect of the original rhythm being tarac-tac rather than, say, tacato-tarot, tacato-tarot, which would give a different tone colour. You don’t actually hear the rhythm any more, only a specific timbre, a spectrum, which is determined by the composition of its components.63
Composing in the age of tape recording, Stockhausen had greater chances of audibly realizing this effect in composition than Cowell had had a few decades previously. Stockhausen put the principle to use in his iconic electronic composition Kontakte, which includes a famous passage (ca. 17’19”–17’54”) in which a meandering glissando is slowed down until it gradually reveals the underlying clicks from which it was constructed—in a direct reversal of the direction of the process he described in the quotation just cited. To these compositions we could further add examples from other repertoires, such as Moby’s electronic dance music track “Thousand” (1990), which made it into the Guinness Book of Records as the fastest rhythm on record. The beat pattern begins relatively slowly, but revs up until it hits 1,000 bpm, or 16.6 Hz—until, that is, it reaches the auditory threshold. The beat track stops giving a beat and begins its perceptual transformation into a clattering pitch. Despite its record-breaking nature, connoisseurs of electronic dance music generally regard this track as gimmicky, since its dance ability at this speed leaves something to be desired.
Opelt’s Siren and the Technologies of Musical Hearing 149
Constructing a coherent narrative out of these sundry examples is difficult, which is due at least in part to the overdetermined nature of Opelt’s idea. It is virtually impossible to write music that follows Opelt’s criteria stringently over the course of a whole composition. The siren principle remains, first and foremost, an effect that exploits a quirk of the human auditory apparatus, not a compositional feature in its own right. The few pieces surveyed here constitute a smattering of ideas in different genres, not a trajectory or tradition of any kind. And accordingly, it appears that each generation of composers rediscovered the effect for themselves, often believing they were the first to do so. What the difficulty of mapping Opelt’s music theory onto musical composition shows more than anything else is that he is putting a rather unusual slant on the notion of music theory itself. This is less a music theory trying to explain composed structures than one concerned with the ways in which human audition works. As Opelt’s siren demonstrates with irresistible experiential directness, he touched on a special feature of human hearing by homing in on a divergence between acoustical data and auditory impression. In this constellation, composed music serves as little more than a set of stimuli to test this feature of our audition. Like Pohl, the anonymous letter writer we encountered initially, Opelt is indifferent to his listening subjects. Hearing is a mathematical function, the site of the seamless intersection between acoustical stimuli and perceived sounds that operate variously on body and mind. None of this should distract from the important work that Opelt performed. If Opelt is now barely a footnote in the history of music, that is largely a consequence of bad timing. Fétis realized this, it seems, when he decided to do Opelt the honor of including him in his Biographie universelle among the most important musicians from past and present. The early nineteenth century was a time of great intellectual and institutional change, especially in central Europe. First of all, it was the time when music aesthetics and the more scientific branch of acoustics parted ways. In the eighteenth century, music theory usually still maintained the trappings of musical science. Around 1800, however, Ernst Chladni in Germany and Thomas Young (1773–1829) in England founded the science of acoustics, while Christian Friedrich Michaelis (1770–1834) transferred Kantian aesthetic principles to music and E. T. A. Hoffmann (1776–1822) wrote the famous review of Beethoven’s symphonies that became the foundation charter of Romantic music aesthetics. Put briefly, as a consequence of the rise of the genius composer, music was no longer centrally concerned with sound: a composition
150 Designing Instruments, Calibrating Machines
was not primarily an acoustical statement but a spiritual one. Second, it was the time when modern musical institutions—music schools and conservatories (Paris, 1795; Prague, 1811; Vienna, 1819; Leipzig, 1843)—came into being, and these new institutions required a new pedagogy.64 They played an important part in cementing the central role of the piano as the modern music- theoretical instrument. There was simply no institutional need for Opelt’s siren; it answered a question that simply did not interest most musicians at the time. Trying to put its principles on a scientific footing, Opelt’s theory attached weights to music just as it was lifting off on its Romantic flight of fancy and threatened to drag it back to the drab ground of empirical reality. Opelt remains a footnote that does not quite belong either to music theory or to acoustics. Nonetheless, his music-theoretical instrument, the siren, has a lot to offer. It unlocks a mode of thinking about music that neither the monochord nor the modern keyboard, two of the paragons of music-theoretical instruments, could open up.65 And it brings together acoustical facts of sound with perceptual aspects of music—it probes the limits of our hearing. One way of thinking about Opelt is simply as an amateur mathematician and hobby music theorist who picked up a new technical invention, ran with it as far as he could, and based a theory of music on largely indefensible premises. But it’s perhaps not wrong to take Opelt’s music-theoretical instrument seriously. Just as the siren needed to wait for the age of electricity to become a fixture in the soundscape of modernity, so Opelt’s sirenic music theory needed to wait for the age of electronic music to unfurl its full epistemic force.66 From the perspective of the Opelt siren’s binary mechanism of sound generation, the time of the siren has come only now, in our computer age, when musical sounds are not merely proto-digital, like the sequences of on/ off impulses emanating from the rotating siren disk in the form of discrete air puffs, but when our everyday musical experience is in fact digitally generated and emerges from binary codes of zeroes and ones. We need only think back to earlier generations of computers to understand that Opelt’s siren disk is nothing but a punchcard by another name. The tinny sound of one-bit music, which emanated from the tiny loudspeakers of the first generation of home computers and has since blended into the nostalgic soundtrack of the 1980s, is based on essentially the same principles. Can Opelt, a figure firmly embedded in the nineteenth century, be seen as a prophet of electronic and digital music? Herbert Eimert, one of the rare twentieth-century commentators who took note of Opelt, certainly thought so.67 (Admittedly, Eimert was probably interested in Opelt because he was eager to lay out an alternative prehistory to the lone discovery of this principle
Opelt’s Siren and the Technologies of Musical Hearing 151
that his colleague and rival Stockhausen claimed for himself.) This verdict may seem anachronistic. But if we take the longer view, then all Opelt did was to explore the numerical aspects of sound, which nineteenth-century musicians were in danger of forgetting. (The occasional invocations of Leibniz’s mathesis universalis suggest that this tradition was never quite forgotten, just temporarily ignored.) Opelt’s siren showed that music’s numerical Doppelgänger was not just an abstract Pythagorean speculation but a material reality that was created by and in the ear. Rather than an anachronism, this assessment of Opelt becomes part of a recursive historiography, which is characterized by the “turn back to long-abandoned constellations of knowledge, which suddenly re-emerge as compatible and instructive [anschluss- und aufschlussfähig] on an operative level.”68 Given that our own era is increasingly dominated by the experience of digital music in its various forms, perhaps the time has come to take another look at the digital Aufschreibesystem that Opelt’s siren outlined in the 1830s. But this historiographic aspect is only one part of the story. Understanding Opelt against the background of his time raises a different set of questions. As we saw, we can hardly do Opelt’s ideas justice if we compare them with existing compositions of his time; music history, certainly one that is measured in a repertory of musical works, is not the most relevant domain for Opelt and his siren. As Fink realized in his ecstatic review of the then unpublished manuscript of Opelt’s music theory in 1832, the major insights of the book are pegged at a higher level, putting to the test the very “nature of music”69—or more precisely, the curious connections that emerge between the acoustical features of the sound waves and our auditory apparatus. In this regard, Opelt’s theory is not that far removed from Pohl’s acoustic- musical fantasy that we encountered initially. Both must count as attempts to convey features of acoustics to an audience of musicians in a world where the chasm between the two spheres had become almost unbridgeable. Where Pohl’s account entered into the realm of the fantastical, Opelt’s built on the ground of acoustical science. But both are ultimately appeals to the musical imagination, guided by the possibilities and constraints of our auditory perception. We will probably get furthest in our attempt to understand Opelt if we think of his work in the context of a music that could have been, or one that only existed in rudiments—that is to say, in the context of a kind of music that tests and plays with the specific features of human hearing. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/opelts-siren-rehding.
152 Designing Instruments, Calibrating Machines
Acknowledgments Thanks go to Alan Bidart, Nikita Braguinski, Daniel Chua, Alexandra Hui, Thomas Y. Levin, Mara Mills, Rena Mueller, Arne Stollberg, David Trippett, Viktoria Tkaczyk, Roland Wittje, and the anonymous reviewers for their helpful comments on the manuscript.
Notes 1. I am grateful for the opportunity to return to this topic as a focal point, having explored Opelt and his siren in passing before. See Alexander Rehding, “Of Sirens Old and New”; Rehding, “Three Music-Theory Lessons”; Rehding, “Instruments of Music Theory.” 2. Robert Weber, “Die electrische Sirene.” 3. [Richard Pohl], “Akustische Briefe,” 73–74. Here and throughout, all translations are my own unless otherwise attributed. 4. Ibid., 74. 5. Heinrich Wilhelm Dove, “Beschreibung einer Lochsirene.” 6. See August Seebeck, “Ueber das Wesen der Töne,” 5. 7. Hermann von Helmholtz, On the Sensations of Tone, 174. A further complication is the fact that sounds are based on mechanical waves, whereas light is transmitted as electromagnetic waves. 8. Popular accounts include Ludwig Riemann, Populäre Darstellung der Akustik, and Ernst Mach, Einleitung in die Helmholtzsche Musiktheorie. 9. Richard Pohl, Akustische Briefe für Musiker und Musikfreunde. 10. David Trippett has explored the connections to Wagner’s music in greater detail. Trippett, Wagner’s Melodies, 384–87. 11. In fact, Wagner specified in his autobiography Mein Leben that the idea for the Rheingold prelude came to him in a vision at La Spezia on September 5, 1853. As always, Wagner’s self-serving recollections must be taken with a grain of salt. The earliest sketches, begun on November 1, 1853, only show an approximation of the final prelude, which seems to have taken shape rather more gradually. See Warren Darcy, “Creatio ex nihilo,” 79–80. 12. Pohl presents an uncomplicated view of acoustics and audition. For him, the connection between the physics of sound and harmony is immediate and interchangeable. See his Akustische Briefe, 7. 13. For modern commentary on Cagniard de la Tour, see also Peter Pesic, Music and the Making of Modern Science, 217–30; Myles Jackson, “From Scientific Instruments to Musical Instruments”; Caroline Welsh, “Die Sirene und das Klavier.” 14. See Ernst Robel, Die Sirenen, 1:12. 15. Charles Cagniard de la Tour, “Sur la sirène,” 171. 16. Various scientists can lay claim to the mechanism of the siren. In particular, the Scottish physicist John Robison (1739–1805) presented a similar mechanism that preceded Cagniard’s siren. And Savart’s wheel fulfills a similar function within acoustical research. 17. Cagniard de la Tour, “Sur la sirène,” 167–68.
Opelt’s Siren and the Technologies of Musical Hearing 153 18. August Seebeck, “Beobachtungen,” 417; see also Seebeck, “Ueber die Sirene,” 479; and in response Georg Simon Ohm, “Noch ein paar Worte.” 19. Robel, Die Sirenen, 2:3. 20. Ernst Florens Friedrich Chladni, “Ueber Töne bloß durch schnell auf einander folgende Stöße,” 460. 21. See Philipp von Hilgers, “Sirenen.” 22. Seebeck, “Beobachtungen”; Seebeck, “Ueber die Sirene”; Seebeck, “Ueber die Definition des Tones”; Georg Simon Ohm, “Ueber die Definition des Tones”; Ohm, “Noch ein paar Worte.” 23. See Veit Erlmann, Reason and Resonance; Alexandra Hui, Psychophysical Ear; Julia Kursell, Epistemologie des Hörens; Benjamin Steege, Helmholtz and the Modern Listener. 24. R. Steven Turner, “The Ohm-Seebeck Dispute,” 22. 25. Ohm, “Noch ein paar Worte.” 26. Hermann von Helmholtz, “Ueber Combinationstöne,” 527. See also Robel, Die Sirenen, 1:32. 27. See also Steege, Helmholtz and the Modern Listener, 43–79. 28. He references Cagniard’s siren in Félix Savart, “Notes sur la sensibilité,” 341 and 346–47. 29. Félix Savart, “Note sur la limite de la perception,” 72; Savart, “Notes sur la sensibilité,” 343. 30. Pohl, Akustische Briefe, 61–62. 31. Most of our knowledge of Opelt goes back to his biographical entry, by Moritz Fürstenau, in the Allgemeine deutsche Biographie. 32. Friedrich Wilhelm Opelt, Über die Natur der Musik; Opelt, Allgemeine Theorie der Musik. 33. It is likely that he sent his work to other outlets as well, but I have not found other articles or announcements about his work in other journals. 34. François-Joseph Fétis, Biographie universelle des musiciens, 6:372. Fétis further referenced Opelt’s work in his article “De la philosophie de la musique,” 4. Fink was the author of an enthusiastic review in 1832, “Neu Aufgefundenes und systematisch Durchgeführtes im Gebiete der Tonkunst.” 35. See also Julia Kursell, “Falsche Strecken, leise Töne.” 36. I thank Wolfgang Rueckner of the Natural Science Lecture Demonstration Services at Harvard University for building an Opelt siren for me. This slightly simplified model (which left out the more complex combinations of the outer rings) offered a convincing demonstration of Opelt’s observations. A video of Opelt’s siren made by the Fondazione Scienza e Tecnica (Florence) can be viewed at https://acoustics.mpiwg-berlin.mpg.de/ node/348. 37. Dove, “Beschreibung einer Lochsirene,” 596. 38. David Pantalony, Altered Sensations, 68. 39. Opelt, Allgemeine Theorie der Musik, 25. 40. The details of Opelt’s theory are more complicated. While his initial explanations are based on Pythagorean ratios, he advocates nineteen-tone equal temperament later in the book, which reproduces these ratios not precisely, but with reasonably good accuracy. Our hearing, he argues, is flexible enough to ignore these slight differences. So as not to further complicate an argument that is already difficult enough, I will leave out the question of equal temperament here, as it has no immediate bearing on the aspects of the theory under discussion. 41. Demonstrating this threshold audibly on Opelt’s siren is not straightforward. If pressured air is used, as is common in modern demonstrations, the noise of the airstream drowns
154 Designing Instruments, Calibrating Machines any sounds produced by the slowly rotating siren. Following recommendations from nineteenth-century experimenters, either the siren can be blown at with a thin straw, in which case soft individual airpuffs can be audible, or it can be struck—à la Savart wheel— with a thin glass or metal rod. 42. Opelt, Allgemeine Theorie der Musik, 58, 65. 43. Ibid., 65. 44. Ibid., 67. 45. Ibid., 2. This dramatic terminology may leave something to be desired in the experimental reality; as seen earlier, each impulse is perhaps better described as a “breath.” An 1872 magazine article explains correctly that when “the device is connected with a powerful bellows, the first noise heard is that of a rushing wind, alternately escaping and cut off.” “Siren of Science,” 885. 46. Opelt, Allgemeine Theorie der Musik, 26. 47. The earlier treatise Über die Natur der Musik spends some time explaining the physical mechanisms of sound generation in musical instruments, vibrating strings, and open/ closed tubes, suggesting a continuity between such mechanisms in the siren and in traditional instruments rather than a rupture. 48. Among Opelt’s scientific lodestars, intriguingly, we find the biologist Karl Ernst von Baer, who would in 1862 present a groundbreaking study on the relativity of temporal perception that resonates strongly with Opelt’s ideas here: Welche Auffassung der lebenden Natur ist die richtige? There is no indication, however, from the works Opelt references—mostly von Baer’s early Vorlesungen zur Anthropologie of 1824—that this conceptual convergence was anticipated. 49. Leibniz mentions this in a letter to Christian Goldbach dated April 17, 1712. Transmission is complicated by the fact that the letter states animae (of the soul), whereas the printed version, edited by Kortholt, which was widely read, changes this to animi (of the spirit). See Ulrich Leisinger, Leibniz-Reflexe, 43. Pohl’s acoustical letter describing the siren experiment uses this quotation as an epigraph. 50. Opelt, Allgemeine Theorie der Musik, 29. 51. Ibid. 52. To be sure, this last domain of meter and phrase rhythm is the least convincing of Opelt’s categories. There is no corresponding auditory threshold that separates individual notes and metric units that could be tested by the siren. This part of Opelt’s theory is largely the product of his urge toward rigorous systematicity. 53. Friedrich Kittler, “Real Time Analysis.” 54. Sybille Krämer, “Cultural Techniques of Time Axis Manipulation,” 106. 55. Sybille Krämer, “Was haben ‘Performativität’ und ‘Medialität’ miteinander zu tun?,” 23. 56. See also Daniel Morat, Viktoria Tkaczyk, and Hansjakob Ziemer, “Einleitung,” 3. 57. “A General Theory of Music,” 170. The relevant table that includes this term (“bedeutungslos”) is not included in Über die Natur der Musik, which the article reviews. It appears in the later Allgemeine Theorie der Musik, 26–29. Fink’s review, which also includes the table, clarifies in his text that the author supplied the diagram himself. 58. Henry Cowell, Quartet Romantic, Preface. 59. One big practical hurdle remains: the rotating shape of Opelt’s siren is only capable of producing a single sound at a time, not a succession of distinct sounds as most musical compositions require.
Opelt’s Siren and the Technologies of Musical Hearing 155 60. Henry Cowell, New Musical Resources, 80. 61. Ibid., 50. 62. Ibid., 51. 63. Karlheinz Stockhausen, “Four Criteria of Electronic Music,” 91–92. 64. See Johannes Forner, “Leipziger Konservatorium und ‘Leipziger Schule.’ ” 65. See Rehding, “Three Music-Theory Lessons.” 66. This applies more generally to the technical media of the nineteenth century. Friedrich Kittler outlines a number of cases in which the mechanical apparatus was long known but only came into its own as a technical medium after electrification. Kittler, Gramophone, Film, Typewriter, 3. 67. As seen in numerous entries in Herbert Eimert and Hans Ulrich Humpert, Lexikon der elektronischen Musik. 68. Ana Ofak and Philip von Hilgers, “Einleitung,” 13. One particularly satisfying aspect here, of course, is the expansion of Opelt’s own recursive methodology, with which he restructures the parameters rhythm, pitch, and meter in his music theory. 69. Fink, “Neu Aufgefundenes und systematisch Durchgeführtes,” col. 131.
References Baer, Karl Ernst von. Vorlesungen zur Anthropologie. Königsberg: Gebrüder Bornträger, 1824. Baer, Karl Ernst von. Welche Auffassung der lebenden Natur ist die richtige? Und wie ist diese Auffassung auf die Entomologie anzuwenden? Berlin: August Hirschwald, 1862. Cagniard de la Tour, Charles. “Sur la sirène, nouvelle machine d’acoustique destinée a mesurer les vibrations de l’air qui constituent le son.” Annales de la physique et la chimie 12 (1819): 167–71. Chladni, Ernst F. F. “Ueber Töne bloß durch schnell auf einander folgende Stöße, ohne einen klingenden Körper.” Annalen der Physik und Chemie 8 (1826): 453–60. Cowell, Henry. New Musical Resources. Edited by David Nicholls. Cambridge: Cambridge University Press, 1996. Cowell, Henry. Quartet Romantic, Quartet Euphometric. New York: Edition Peters, 1976. Creese, David. The Monochord in Ancient Greek Harmonic Science. Cambridge: Cambridge University Press, 2010. Darcy, Warren. “Creatio ex nihilo: The Genesis, Structure, and Meaning of the ‘Rheingold’ Prelude.” 19th-Century Music 13, no. 2 (1989): 79–100. Dove, Heinrich Wilhelm. “Beschreibung einer Lochsirene für gleichzeitige Erregung mehrerer Töne.” Annalen der Physik und Chemie 158, no. 4 (1851): 596–98. Eimert, Herbert, and Hans Ulrich Humpert. Lexikon der elektronischen Musik. Regensburg: Gustav Bosse, 1973. Erlmann, Veit. Reason and Resonance: A History of Modern Aurality. New York: Zone Books, 2010. Fétis, François-Joseph. “De la philosophie de la musique.” Revue et Gazette musicale 7, no. 1 (1840): 2–5. Fétis, François-Joseph. Biographie universelle des musiciens. Paris: Firmin Didot, 1867. Fink, Gottfried Wilhelm. “Neu Aufgefundenes und systematisch Durchgeführtes im Gebiete der Tonkunst.” Allgemeine musikalische Zeitung 34, no. 9 (1832): col. 129–35. Forner, Johannes. “Leipziger Konservatorium und ‘Leipziger Schule.’” Die Musikforschung 50 (1997): 31–36.
156 Designing Instruments, Calibrating Machines Fürstenau, Moritz. “Opelt, Friedrich Wilhelm.” In Allgemeine Deutsche Biographie, 24: 366. Leipzig: Duncker & Humblot, 1887. “A General Theory of Music.” Harmonicon 10 (1832): 170–71. Helmholtz, Hermann von. On the Sensations of Tone as a Physiological Basis for the Theory of Music. Translated by Alexander J. Ellis. London: Longmans, Green & Co., 1885. Helmholtz, Hermann von. “Ueber Combinationstöne.” Annalen der Physik und Chemie 99 (1856): 523–29. Hilgers, Philipp von. “Sirenen: Loslösungen des Klangs vom Körper.” ZwischenRäume 6 (2003): 103–21. Hui, Alexandra. The Psychophysical Ear: Musical Experiments, Experimental Sounds, 1840– 1910. Cambridge, MA: MIT Press, 2012. Jackson, Myles. “From Scientific Instruments to Musical Instruments: The Tuning Fork, the Metronome, and the Siren.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 201–23. New York: Oxford University Press, 2012. Kittler, Friedrich. Gramophone, Film, Typewriter. Translated by Geoffrey Winthrop-Young and Michael Wutz. Stanford, CA: Stanford University Press, 1999. Kittler, Friedrich. “Real Time Analysis, Time Axis Manipulation.” Translated by Geoffrey Winthrop-Young. Cultural Politics 13, no. 1 (2017): 1–18. Krämer, Sybille. “The Cultural Techniques of Time Axis Manipulation: On Friedrich Kittler’s Conception of Media.” Theory, Culture & Society 23, nos. 7–8 (2006): 93–109. Krämer, Sybille. “Was haben ‘Performativität’ und ‘Medialität’ miteinander zu tun? Plädoyer für eine in der ‘Aisthetisierung’ gründende Konzeption des Performativen.” In Performativität und Medialität, edited by Sybille Krämer, 13–32. Munich: Fink, 2004. Kursell, Julia. Epistemologie des Hörens: Zu Hermann von Helmholtz’ physiologischer Grundlegung der Musiktheorie. Munich: Fink, 2017. Kursell, Julia. “Falsche Strecken, leise Töne: Die Laute in der Musiktheorie.” In Medien vor den Medien, edited by Ana Ofak and Friedrich Kittler, 199–221. Munich: Fink, 2007. Leisinger, Ulrich. Leibniz- Reflexe in der deutschen Musiktheorie des 18. Jahrhunderts. Würzburg: Königshausen & Neumann, 1994. Mach, Ernst. Einleitung in die Helmholtzsche Musiktheorie. Graz: Leuschner und Lubenski, 1866. Morat, Daniel, Viktoria Tkaczyk, and Hansjakob Ziemer. “Einleitung.” In Wissensgeschichte des Hörens in der Moderne, edited by Netzwerk “Hör-Wissen im Wandel,” 1–19. Berlin: De Gruyter, 2017. Ofak, Ana, and Philipp von Hilgers. “Einleitung.” In Rekursionen: Von Faltungen des Denkens, edited by Philipp von Hilgers and Ana Ofak, 7–21. Munich: Fink, 2010. Ohm, Georg Simon. “Noch ein Paar Worte über die Definition des Tones.” Annalen der Physik und Chemie 62, no. 5 (1844): 1–18. Ohm, Georg Simon. “Ueber die Definition des Tones, nebst daran geknüpfter Theorie der Sirene und ähnlicher tonbildender Vorrichtungen.” Annalen der Physik und Chemie 59, no. 8 (1843): 513–65. Opelt, Friedrich Wilhelm. Allgemeine Theorie der Musik. Leipzig: Johann Ambrosius Barth, 1852. Opelt, Friedrich Wilhelm. Über die Natur der Musik: Ein vorläufiger Auszug Aus der bereits auf Unterzeichnung angekündigten “Allgemeinen Theorie der Musik.” Leipzig: Hermann und Langbein, 1834. Pantalony, David. Altered Sensations: Rudolph Koenig’s Acoustical Workshop in Nineteenth- Century Paris. Dordrecht: Springer, 2009. Pesic, Peter. Music and the Making of Modern Science. Cambridge, MA: MIT Press, 2014. [Pohl, Richard]. “Akustische Briefe.” Neue Zeitschrift für Musik 37 (1852): 1–3, 13–15, 21–24, 35–36, 41–47, 73–76, 85–88, 185–87, 193–96, 249–51, 261–64.
Opelt’s Siren and the Technologies of Musical Hearing 157 Pohl, Richard. Akustische Briefe für Musiker und Musikfreunde. Leipzig: Bruno Hinze, 1853. Rehding, Alexander. “Instruments of Music Theory.” Music Theory Online 22, no. 4 (2016), http://mtosmt.org/issues/mto.16.22.4/mto.16.22.4.rehding.html. Rehding, Alexander. “Of Sirens Old and New.” In The Oxford Handbook of Mobile Music, edited by Sumanth Gopinath and Jason Stanyek, 2: 77–108. New York: Oxford University Press, 2014. Rehding, Alexander. “Three Music-Theory Lessons.” Journal of the Royal Musical Society 141, no. 2 (2016): 251–82. Riemann, Ludwig. Populäre Darstellung der Akustik. Braunschweig: Friedrich Vieweg und Sohn, 1896. Robel, Ernst. Die Sirenen: Ein Beitrag zur Entwickelungsgeschichte der Akustik. 4 vols. Berlin: R. Gaertner, 1891. Savart, Félix. “Note sur la limite de la perception des sons graves.” Annales de chimie et de physique 47 (1831): 69–74. Savart, Félix. “Notes sur la sensibilité de l’organe de l’ouïe.” Annales de chimie et de physique 44 (1830): 337–52. Seebeck, August. “Beobachtungen über einige Bedingungen der Entstehung von Tönen.” Annalen der Physik und Chemie 53 (1841): 417–36. Seebeck, August. “Ueber das Wesen der Töne.” In Akustik. Section 21 of Repertorium der Physik 8 (1849): 1–27. Seebeck, August. “Ueber die Definition des Tones.” Annalen der Physik und Chemie 63 (1844): 353–80. Seebeck, August. “Ueber die Sirene.” Annalen der Physik und Chemie 60 (1843): 449–81. “The Siren of Science; Or, the Mode of Numbering Sonorous Vibration.” Harper’s New Monthly Magazine 45, no. 270 (1872): 844–49. Steege, Benjamin. Helmholtz and the Modern Listener. Cambridge: Cambridge University Press, 2012. Stockhausen, Karlheinz. “Four Criteria of Electronic Music.” In Stockhausen on Music: Lectures and Interviews, edited by Robin Maconie, 88–111. London: Marion Boyars, 1989. Trippett, David. Wagner’s Melodies. Cambridge: Cambridge University Press, 2013. Turner, R. Steven. “The Ohm-Seebeck Dispute, Hermann von Helmholtz, and the Origins of Physiological Acoustics.” British Journal for the History of Science 10, no. 1 (1977): 1–24. Weber, Robert. “Die electrische Sirene.” Annalen der Physik 260, no. 4 (1885): 671–80. Welsh, Caroline. “Die Sirene und das Klavier: Vom Mythos der Sphärenharmonie zur experimentellen Sinnesphysiologie.” In Parasiten und Sirenen: Zwischenräume als Orte der materialen Wissensproduktion, edited by Bernhard J. Dotzler and Henning Schmidgen, 57– 85. Berlin: transcript-Verlag, 2008.
6 The Software Passes the Test When the User Fails It Constructing Digital Models of Analog Signal Processors Jonathan Sterne
In a prescient 1987 essay, John Mowitt noted a changing tendency in attitudes toward signal processing that accompanied the growing adoption of digital technologies for making, storing, and transmitting music. He wrote that “the fetish of noise reduction has gone hand in hand with the aggressive marketing of distortion boosters and other less obvious instrumental sources of noise.”1 Digital audio recording had been sold in terms of its promises of perfect fidelity to an original source and low signal distortion, but it turned out that these marketing and engineering points were not universally desired by users.2 Recording engineers spent the 1970s and earlier fighting against the specific noise signatures of electroacoustic devices such as compressors, tape machines, and amplifiers. They devised clever workarounds and strategies to minimize the noise and signal transformation inherent in the technologies they used.3 Now, in 1987, Mowitt noted, the drive was not just for perfect definition, but for noise and distortion as sonic effects, from the crackle of an LP record grafted onto a hip-hop recording made with a digital sampler, to simulations of tape distortion, to simulations of complex amalgams of musical instrument gear—amplifiers, compressors, echoes and delays, distortion pedals. With thirty years’ retrospection, it is clear that what Mowitt presented as an irony of digital signal processing for music was actually its historical condition. Digital signal processing could only make sense by marking itself out as distinct from analog signal processing. Once there was more than one way for everyday musicians and engineers to process and reproduce sound, aspects of the older way came into relief as aesthetic choices. If, as the martial metaphor goes, engineers of the 1960s and 1970s fought a battle against noise and distortion, we would have to ultimately characterize it as a losing battle. The result is that to a trained ear, music recorded in a particular era or a particular Jonathan Sterne, The Software Passes the Test When the User Fails It In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0007.
160 Designing Instruments, Calibrating Machines
studio environment such as Motown, Abbey Road, or Muscle Shoals had a distinctive sonic signature. That sonic signature was, of course, a combination of many things—the engineers, the equipment, the musicians, the music— and yet it is often plainly audible to listeners. For most, these aspects of sound simply fold into “the music itself,” as an aspect of timbre. And timbre, it turns out, is central to the reception and meaning of much recorded music in the twentieth and twenty-first centuries, especially popular music. Even a single note (or sometimes even less) from a well-known popular song is recognizable to listeners who know it, a feat made possible by the variability of instrumentation and production techniques across studios.4 Musicians and engineers are well aware of this phenomenon—and often (though not always) more sensitive to it than nonmusician audiences. In the context of music production and performance, they may use equipment as a shorthand for whole histories of production, performance, style, and reception, for instance naming the Roland Space Echo to reference a whole history of dub music. This is not unique to popular music or to the twentieth century, as Emily Dolan’s piece in this collection demonstrates. The cult of the Stradivarius violin is not that distant from the cult of the Stratocaster guitar. From stand-alone digital recorders to integrated computer and software systems, early commercial digital technologies presented an unexpected problem for their users because they were at first not legible within this timbral and technical history. “Early digital” now has its own retrospective sound and production style that is a result of a combination of equipment and engineering trends in the 1980s—but it would not necessarily have been apparent to listeners at the time: “It can be compared with somebody who moves into a new house. The first time he looks through the window he only sees the beautiful view. After a few days he detects a small flaw in the glass and from that moment he cannot look through the window without seeing that flaw.”5 If the “perfect sound forever” marketing campaigns of early digital audio promised musicians liberation from the sonic limits of previous generations of recording and signal processing technology, the lived reality is that many people loved their limitations, because they shaped the sound of the music.6 In the intervening decades, scholars have noted an “analog revival” in response to digital audio,7 but an equally remarkable phenomenon has occurred in the world of digital audio itself. Through an extensive labor of translation, hardware and software designers have found ways to model—that is, to emulate or, if full emulation is not possible, at least approximate—analog signal processing technologies in the digital domain. Modeling has become one of the staple offerings of the music technology industry, where one device
Constructing Digital Models of Analog Signal Processors 161
imitates some aspects of another. “Analog modeling” actually refers to a host of approaches, from attempts to model the behavior of complex circuits down to the component level, to taking actual recordings of analog signal processing devices or even physical spaces and convolving them with other digitally recorded sounds (a process I explain below), to attempts to replicate a vague “feel” or “vibe” using analog devices as inspiration, to just adding depictions of “wood panels” to the window for a software plug-in. In this chapter, I analyze the process by which engineers in the commercial music technology industry model analog signal processing in the digital domain.8 Based on participant observation as well as research into the history of the technologies I am writing about, I describe the ways in which digital models at once test the hearing of machinery—how a given technology transduces, registers, and represents sound to itself and to human auditors—and use the machinery to test the hearing of users. The chapter loosely follows the development of two different scenarios. First, I consider a model of a spring reverb device, the AKG BX20, at Universal Audio in Scotts Valley, California. I became involved with Universal Audio as part of a larger ethnographic project on signal processing technologies and the people who produce them. Since the early 2010s, I have visited dozens of companies, laboratories, and workshops. I focus on Universal Audio because they embody a particularly strong example of a set of ideas about modeling and testing, and because of the access they granted me to their modeling process. Following current ethnographic practice, I will sometimes refer to myself (and my hearing) within the narrative so that I do not create the illusion that by dint of my authorship of this chapter I am also a universal or perfect auditor.9 I then consider modeled amplifiers for electric guitars developed by Line 6 and other modeling companies. For the BX20, I examine the production of a model because I was able to participate in it; for the guitar amplifiers, I consider the reception of models, especially as they move across different user communities. Most discourse around modeling is producer discourse, which is to say musicians and engineers are primarily talking with one another when they talk about modeling, though it is useful to distinguish people who produce music from people who produce models, even if many of the people who make models also use them. Even if they are themselves musicians, audiences and fans mostly spend little time worrying about the qualities of this or that signal processing technique (if they are even aware of them at all). They will instead simply talk about the music, and maybe the sound of the music. When modeling changes the look of musicians’ equipment, this is sometimes a cause for commentary, a scenario to which we will return later.
162 Designing Instruments, Calibrating Machines
Digital Audio Models of Analog Devices: Tests and Definitions While every company, and every engineer, has to some extent a unique approach, there is a lexicon of common practice and terminology that exists above and beyond particular implementations. Like testing hearing for other purposes (see Mara Mills and Viktoria Tkaczyk in this volume), the testing of digital models is characterized by a quest for precision and for the quantification of previously unquantified dimensions of sound as it is transduced out of the sonic domain and back again. But unlike in medical or telephonic contexts, a deliberately aesthetic dimension enters into the engineering process when the goal is making music. Telecommunications has its aesthetics as well, to be sure, but the aesthetic aspect is highlighted in the engineering of sound recording and signal processing technology for music. Because “sounding good”—a recurrent trope in engineers’ talk—is a foremost concern, we gain a particularly useful insight into the politics of transduction: how cultural, historical, and economic relations are rendered in the sonic realm, and how dimensions of sound and sound technologies come to have value. To “sound good” is to invoke a history of sounds and practices, which implicitly values some aspects of those practices over others. Whether it is a test of how a signal processor receives and renders a signal or a test of how a listener responds to a digital model of an analog device, the moment of testing hearing is where these relations are negotiated, refracted, and brought to life. As part of the emergence of a digital device or piece of software, listening tests—whether strict A/B/X tests or more performative comparisons—help engineers define what is essential to a sonic technology and what is superfluous. A/B and A/B/X tests are among the most common listener testing regimes for audio technologies for music. Listeners are given two sounds, “A” and “B,” and asked whether they can tell the difference. The A/B/X test gives the listeners sounds A and B, and then a third sound, “X,” which they are asked to identify as A or B. If the user is right half the time or less, they are guessing and the two sounds are considered indistinguishable. A/B/X tests have a long history in consumer audio. It is not clear when they emerged as the standard for testing digital models of analog processors, but some of the earliest entrants into the digital modeling business, such as Native Instruments, were already doing user testing with A/B/X in the late 1990s. People who build digital models of analog processors will use this kind of testing in constructing the model, but also in final quality assurance testing before releasing a program to the general public.10
Constructing Digital Models of Analog Signal Processors 163
Listening tests matter to engineers because they design digital models to reproduce the sonic characteristics and processing behavior of other devices (which they then model in code). The digital model may or may not reproduce other aspects of the experience of using the technology. All digital models have software dimensions, and some digital models exist purely as software. A digital model of an analog device such as a guitar amplifier or reverberator might exist in the following forms: • As a plugin to enhance the functionality of other software, for example, in digital recording and mixing • As a piece of stand-alone computer software • As a smartphone app • As a stand-alone hardware device—a physical box—with digital signal processing built into it • As a physical box or device that shares the functionality of the analog device and has similar controls and interface elements • As a physical box or device that shares the functionality of the analog device but looks completely different To be sure, design matters. It is part of the user experience of the model, and hardware characteristics such as knobs and materials, or graphical interfaces, or the lack of either in a command-line software environment imply a whole set of intended meanings and user scripts. Yet it is important to note that sonically speaking, the model can be completely accurate even if it looks or feels nothing like the device it is said to model. This range of interpretations is most clear if we consider software models. Software may look entirely different from the analog device it models, with interface features that do not match the analog interface at all, or it may have a skeuomorphic appearance, mimicking the device’s look. A skeuomorph is, following Katherine Hayles, “a design feature that is no longer functional in itself but refers back to a feature that was functional at an earlier time.”11 Hayles treats skeuomorphism as a visual phenomenon, and graphic user interfaces generally do as well. But as I show in this chapter, it is both a visual and sonic phenomenon. Mowitt observed that when confronted with new, “improved” technologies, musicians and artists immediately began finding ways to get them to behave—and sound— like older technologies. To understand the relation between visual and sonic skeuomorphism, consider the interfaces of three software-based analog- modeling filters. Figure 6.1 shows a skeuomorphic interface for a digital model of the Roland SH-101 synthesizer, a picture of which is set right below it. Although the layout is not identical, the same functions are contained in
164 Designing Instruments, Calibrating Machines
Figure 6.1 Togu Audio Line’s Bassline 101 versus a Roland SH101. The “VCF” appears in roughly the same place in both the device interface and the skeuomorphic software interface. Author’s screenshot, from The Prodigy Live, http://theprodigy.info/equipment/sh-101.shtml.
the top row of controls for both; the labels and look are maintained. Thus, the layout of controls for the software version will be familiar to anyone who has used the hardware synthesizer, and the filter (the section labeled VCF) is located in the same position as on the control panel of the analog device. This is a classic example of skeuomorphism, like the “desktop” metaphor used in graphical operating systems for computers with its “file folders” and “recycling bins.” Software models may also depart from any specific analog hardware. Figure 6.2 shows a SoundToys plugin called FilterFreak (top) that is supposed to model different behaviors of analog filters. It looks like a
Constructing Digital Models of Analog Signal Processors 165
Figure 6.2 Diminishing skeuomorphism: SoundToys FilterFreak versus Ableton Auto Filter. Author’s screenshots.
skeuomorphic representation of a specific analog device, but it represents no device in particular. The screenshot shows knobs meant to represent adjustable parameters, frequency curves set inside frames meant to resemble oscilloscope screens, and on/off toggles represented as switches, and the interface even has gratuitous “wood panels” on its sides, evoking the look of 1970s audio hardware. In contrast, the Auto Filter plugin for Ableton Live (the lower part of Figure 6.2), which performs similar functions to FilterFreak, has no obvious visual similarity to any hardware, real or imagined.12 There are still circles, squares, and frequency curves, but no attempt is made to have them look like knobs, buttons, or screens. Instead, they are presented to the user in a flat, easy-to-render, Bauhaus-inspired design and paired with standard graphical user interface elements such as pulldown menus (which FilterFreak conceals behind buttons). Yet the Auto Filter plugin does also model analog hardware. The “PRD” setting in the middle indicates that its algorithm is set to emulate a resonant four-pole ladder filter similar to that on a Moog Prodigy synthesizer. Command-line-based
166 Designing Instruments, Calibrating Machines
audio programs such as Pure Data, which are controlled entirely through text commands and have no graphical user interface at all, also have the means to model the workings of the Moog. Although one can have a more or less skeuomorphic interface for digital models of an analog device, it is not necessary. But what exactly is a digital model of an analog device? Every term in that sentence raises thorny historical and epistemological questions. Recent scholarship has problematized the concepts of analog, digital, and model. For the purposes of this chapter, I will keep close to the actors’ categories, which are very much not precise analytical definitions. In the music technology world, analog refers to a retrospective definition that encompasses all audio technologies before or without a digital element. However, it should be noted that not all technologies grouped under this definition are actually analog audio technologies, and not all of them are analog in the same way. This retrospective usage of the term “analog” actually works to legitimate the digital technologies that are said to come after them, by lumping everything outside digital technology into a single, if incoherent, category. Digital is generally assumed to refer to any technology with a microprocessor, dating it to the 1970s and integrated circuits—but this too is an actors’ category that elides the ideological work of the definition, since in many cases the things other than the microprocessor (including some of the things that are at the core of what a technology does—sending sound out of a speaker, say) do not exist in the digital domain at all, and digital technologies do not necessarily need to have processors. Like the term “analog,” the term “digital” cannot be used without an attendant value judgment: critics will refer to digital sound as “cold” or “lifeless,” while other users will praise digital technologies for their miniaturization, affordability, control options, flexibility, or verisimilitude.13 As for model, historians and philosophers of science have endlessly debated what models are and do. Models may “aim to lay bare the essential principles according to which this or that domain of phenomena operate,”14 but in signal processing contexts, they tend to do so dynamically, so that relations may change in real time. That, indeed, describes how people in the digital modeling industry talk about what they are doing. What counts as essential or superficial is a social and cultural question—and thus the model does epistemological and cultural work at the same time.15 The final term in the formulation, “device,” has been the subject of much less discussion in recent scholarship. In science and technology studies, it is perhaps most famously associated with Albert Borgmann’s “device paradigm,” whereby
Constructing Digital Models of Analog Signal Processors 167 those aspects of properties of a device that provide the answer to “What is the device for?” constitute its commodity, and they remain relatively fixed. The other properties are changeable and are changed, normally on the basis of scientific insight and engineering ingenuity, to make the commodity still more available. Hence every device has functional equivalents, and equivalent devices may be physically and structurally very dissimilar from one another.16
For Borgmann, devices are fundamentally commodities, and though I might quibble as to whether this is sufficient for a general model of the device, the referents of most digital models, as well as the models themselves, are certainly commodities. Not only are they bought and sold, but they are subject to commodity fetishism, and, as Louise Meintjes has shown, a great deal of studio practice involves musicians acting on their beliefs in the magic or power contained within equipment.17 Borgmann finds that the separation of means and ends in technology raises a serious political problem, because device status is designed to occlude both the inner workings and the social workings of a technology. Borgmann’s device paradigm thus partakes of what writers in the science and technology studies tradition have called “black-boxing,” what writers in cinema studies have called the “concealment of the apparatus,” and what writers in the Marxist tradition have called “reification.”18 All of these terms have different theoretical implications and political resonances, but they all focus on defining which aspects of a technology are to be in the foreground of users’ attention and which are to be hidden from users. Thus, in the definition of both the analog device and the digital model, the circumscription of what counts as “inside” and “outside” the device is what makes the model possible. The listening test plays a crucial role in defining the analog technology retrospectively and the digital model prospectively, since the test performs the divisions of inside and outside, consecrating different aspects of the technology, its sound, and the experience of using it as essential while denigrating other aspects as superficial to the model.19
On Reverbs in Closets It is February 2012, and I am sitting with my laptop at a makeshift desk in a storage room in Universal Audio’s headquarters in Scotts Valley, California. Universal Audio is best known for its work in digital modeling and for reviving old analog equipment. Its reputation was built around the LA-2A leveling amplifier and the 1176 limiting amplifier, both of which were prized by
168 Designing Instruments, Calibrating Machines
recording engineers for the specific ways they behaved as they affected signals. When it was refounded in 1999, Universal Audio (originally founded in 1958) began manufacturing replicas of these old units, right down to the photocell in the LA-2A’s circuitry that used to modulate sound. But the company also built digital models of these and many other devices to be used as software plugins, to enhance the functionality, experience, and sound of music mixing programs such as ProTools. These general-purpose programs, called Digital Audio Workstations (DAWs), record multitrack audio and MIDI data (Musical Instrument Digital Interface—a control protocol for various kinds of devices to communicate with one another) and facilitate mixing, editing, and creative signal processing in real time. Almost every commercial music recording made today passes through a DAW at some point in its production process, and usually the DAW plays a central technical role. My host is Jonathan Abel, a mathematician and signal processing expert who contributed to Universal Audio’s original LA-2A and 1176 digital models.20 Abel is something of a historian of these devices, having spent a lot of time listening to different models and versions, visiting them in professional recording studios, and discussing their use history with recording engineers. He helped to build Universal Audio’s paradigm (along with his colleague Dave Berners and several others) and to set up a course in signal processing for audio at Stanford’s Center for Research in Computer Music and Acoustics. Abel and I have talked over his approach to digital modeling at length. He has been generous with his time, explaining the math of analog and digital signal processing to me, and I have sat in on his class. For Abel, there is always more to talk about, another layer to peel away, another pool of questions to dive into. He is fascinated by the smallest mathematical details of what these devices do to sound, and he has a deep sense of aesthetic appreciation for the gear, understanding himself as part of a lineage of signal processing researchers. He frequently spoke about the philosophical questions raised by the mathematical and operational challenges of modeling. I have hooked up my laptop to an audio interface, which is connected to an AKG BX20 reverberator (AKG is the company and BX20 is the model number). The AKG BX20 is a behemoth of a device: large, heavy, wooden, difficult to move. First released in the late 1960s, it is a classic example of a truly analog device: it uses the behavior of an ensemble of materials—electricity, coils, magnets, springs—as an analog of the behavior of sound in a room containing a microphone and a sound source. Once sounds have been converted to electrical signals, they can be sent through the BX20, where the diffusion behavior of the spring adds a sense of audible ambience to the sound,
Constructing Digital Models of Analog Signal Processors 169
as if it resounded in a chamber or hall. Controls allow the user to damp the spring, blend the spring’s “wet” reverberate signal with the “dry” signal that is not run through the spring, and modulate the volume of the sound. The BX20 has two channels, so it can work in stereo. Transduced into electricity, sound enters the BX20 through a cable plugged into another device. It is then amplified and transduced into mechanical vibrations, which are sent into a spring, converted back into electricity at the other end of the spring, sent to another amplifier and combined with the input signal at the output. This allows the user to choose how much of the processed signal to combine with the original signal. The feedback of the spring system is also transmitted along with the initial sound as modulated by the moving spring. Once out of the BX20, the combined signal can then be run into a mixing board, amplifier, and speakers and transformed back into audible sound. BX20s were popular in recording studios from the late 1960s on, adding a signature ambience to various famous recordings (today, this can be heard on albums by artists such as Norah Jones and Jack White). Along with other mechanical reverbs, they were often part of a studio’s mystique or signature sound. Lore even developed around some specific, individual devices, such as the Echoplate III (a competitor to the BX20) at Muscle Shoals studio, which supposedly benefited from the particular humidity of Sheffield, Alabama, and contributed to the sound for which the studio was known. When I encountered it, the unit had been relocated to a studio space inside a house in suburban Huntsville, Alabama. I confess that upon hearing it in April 2015, I was not immediately able to distinguish its special characteristics in contrast to other plate reverbs I have heard in person or on recordings before or since. This raises a host of issues around comparing sounds to one another. Did the device need the humidity to have a special sound? Were my own shortcomings as a listener the explanation for why I could not hear the difference? Or was the entire scenario a kind of commodity fetishism, condensing all sorts of history, work processes, musical practice, and recording practice into a single device in the suburban Huntsville closet before me? It is impossible to know precisely why that particular Echoplate III is able to retain its significance to a community of users. Back in that storage room in California in 2012—also really more of a big closet—Jonathan and I are sending signals through the BX20 and recording them. I am using Ableton Live, a popular recording and playback program that is a cross between a traditional DAW and a sampler, allowing the flexible, fast, and repeated playback and recording of samples of recorded audio. We use three files, set up as samples in my program, that Jonathan has given me: a pulse that makes a “ping” sound; a short sine sweep through the whole audible
170 Designing Instruments, Calibrating Machines
spectrum that yields a sort of springy “boing”; and a longer sine sweep that sounds like a very broadband siren firing up. Jonathan changes the settings on the AKG; I run the sound through and record it. We listen back to the recording. After a run through the BX20 on a particularly long setting, he smiles at me and admires the way the sound decays. An ear for the subtle dimensions of timbre and time is a part of the modeling process. If modeling is a producer discourse, the biggest connoisseurs of all are the people making the models. Nearly everyone I met at Universal Audio had an appreciation for the old analog gear they were modeling. Like many digital audio companies I visited—Line 6, Native Instruments, Elektron, Teenage Engineering (to name a few)—Universal Audio has a special space in its building stocked with prized examples of vintage sound equipment, a kind of shrine to the sonic and signal processing histories of recorded music in the second half of the twentieth century. Although Universal Audio specializes in the signal processing part of the recording and mixing process, its studio contains plenty of vintage keyboards, amplifiers, and musical instruments. For all this mythologizing of equipment inside the actual spaces that modeling companies inhabit, treating the equipment as a stand-in or condensation of musical and aesthetic histories, the modeling process is fundamentally a process of demystification. This is crucial: even though finished software products are black-boxed from users—you cannot know the algorithm that governs the behavior of Universal Audio’s BX20 unless you sign a nondisclosure agreement and have a “need to know”—and fascination with old audio gear depends on a layer of mystification, the work of building digital models of analog devices requires every “mystery” of an old analog device such as the spring reverb to be explained or classified.21 The engineers at Universal Audio want to know how and why the BX20 produces the sounds that it does, and the recordings we took are a path into that process, if not exactly the first step. Taken together, the sounds I sent through the BX20 tested its behavior— how it hears sound and how it plays sound back. By measuring the difference between the files we used to test the device and the recording, by sending the original signal through the device at different settings, Jonathan and the engineers at Universal Audio can construct a model of what the BX20 is doing to the sound and then build a working model able to imitate it. The file that is the result of the difference between the sounds we sent into the BX20 and the sounds we recorded out of it provides a baseline for evaluating what the unit does to sound. The difference between the two files is treated as an “impulse response” that can be applied to other signals. If the BX20’s impulse response is applied to a dry recording (say, of a voice in a dampened room or studio) through a process called convolution, it will sound
Constructing Digital Models of Analog Signal Processors 171
as if that voice went through the BX20. In digital signal processing, convolution multiplies the spectra of two audio signals to combine them, expressing their overlap: for instance, if you convolved a recording of my voice with a filter that removes all the low frequencies of my voice, the result would be a sound of my voice with the low frequencies attenuated. But convolution is time invariant; it does not change over time. Since it is based on a recording of a particular sound at a particular time, it cannot give anything other than a snapshot of the BX20’s behavior.22 Rather than representing the chaotic behavior of the spring inside the BX20, it represents one instance of the behavior of the circuitry inside the BX20, a single sonic performance. Thus, convolution begins the modeling process but does not finish it. In the months following my visit, working with his former graduate student Sean Coffin, Abel first tested the impulse responses taken from the BX20 against the same sounds being run through the BX20. They were seeking a “perceptual match,” where the listener cannot tell the difference between a convolved signal and a signal actually sent through a BX20. The listeners here are the people working on the algorithm—Abel, Coffin, friends of both, employees of Universal Audio, and others who may be brought in for listening tests.23 Like other kinds of listening tests, these test both the user and the technology at the same time. The goal of the listening test is for the technology to pass it, not the user. If the impulse response is good enough, the user will fail the test. If the user passes the test and can tell the two technologies apart, then the impulse response has failed. As Dave Berners, Universal Audio’s chief scientist, explained: To me the function of listening is to find bugs, and that’s my own opinion. We have a really great . . . person at UA, Will Shanks, who does . . . qualification that’s separate from our quality assurance team. He’ll do A/B tests; sometimes he’ll bring in other people to do subjective listening tests. But he’s really the person who makes sure that things are going the way we want them to. And he provides us a lot of feedback. He’ll tell us qualitatively what’s going on when he listens to something. But in my mind, the way it fits into the design process is, if there’s something that he can tell that’s different, it means that there’s a bug. I don’t want to have him say, well, it should be a little bit brighter, and then I’ll say, okay, I’ll just put a little EQ and make it a little brighter. It’s not an iterative thing where we try to, based on listening, converge towards something. When the design gets to that stage, it should already be converged. It should be a model of the real process and be identical. And if it isn’t, then if something sounds different, the way that it sounds different can tell us what is likely to be wrong. So it’s really useful to have the listening feedback [to tell us] what way the sound is not right, it tells us where in the algorithm to look
172 Designing Instruments, Calibrating Machines for a mistake. But nevertheless I still feel like the listening’s there to find problems more than it is to nudge things or tweak them. There’s very few cases where we do any design based on the perception. I would feel really vulnerable if that were the case.24
Trevor Pinch notes that technological tests are usually understood as performances to be witnessed by others: Will Shanks comes through in Dave Berners’s explanation as a kind of super listener, someone who stands in for future audiences. In other cases, a computer program can just as easily serve as a witness for this kind of test as a person can: the goal is simply to know whether an imagined future listener can tell the difference or not. Witnessing the test can be delegated, but there must be a human or technological witness.25 The goal here is to establish a logic of sonic equivalence between two sounds: a digital recording fed into an algorithm on one side, and a physical device on the other. According to the logic of the test, the equivalence only happens when the two sounds are indistinguishable—but in fact the testing scenario itself, with its A/B/X structure and “this or that” choice, already establishes a logic of equivalence. As Berners puts it, “it should be a model of the real process and be identical.” It does not matter that one contender is a giant box with springs inside and the other is a piece of code that takes up an infinitesimal space on a hard drive. Those differences are excluded from the test before the fact. Here, sound is the basis of commensurability between two operationally distinct technologies, and the test performs that moment of commensuration. In the test, the sound—really the hearing of the sound—is what creates the relationship between the two other elements.26 That is why, for Berners, iterative design or aesthetic judgment would make the test and the engineer “vulnerable”: they would undermine the possibility of equivalence. The devices modeled by Universal Audio, such as the 1176 or BX20, are already widely respected in the world of professional audio engineering, and are tied to famous recordings and the sounds achieved by famous musicians. Iterative listening and design has already occurred. The model simply consecrates the tangled musical and social relations that resonate inside the sound. Put simply, the modeling process decrees that “good sound” has already been decided by a history of practice.
Contested Models: Guitar Amplifiers The condition of commensurability necessary for digital models of analog devices is also one of the most contested dimensions of the whole
Constructing Digital Models of Analog Signal Processors 173
enterprise. What constitutes a satisfactory model, for whom, and under what circumstances? When can a musician, artist, or recordist substitute a digital model for the device that it models? My visit to Line 6, a company that made its name with digital models of famous guitar amplifiers, helped me to understand what can happen when different user populations have different ideas about commensurability. An analog guitar amplifier sends sound as an electrical signal through a series of processing stages, where electrical operations have an analogous relationship to sonic processes: electrical clipping leading to harmonic distortion, the roar or grind of a distorted electric guitar. As with the studio equipment discussed in the previous section, specific brands and models of guitar amplifiers are famous for the characteristics they impart to sounds. While the BX20 is targeted for a single use—mixdown of prerecorded material—the guitar amplifier model here is targeted to at least three very different user bases that crisscross recording studios, live performance, and home practice: studio engineers who are not guitarists and are seeking to change the sound of a guitar performance; guitarists who are recording or practicing at home in situations where using a guitar amplifier at full volume would be problematic; and performing guitarists who want the flexibility of having many more types of amplifiers, with significant savings in cost and space (some amplifiers are not only expensive but also larger than refrigerators). For an engineer, an amplifier is another flavor of signal processing that goes into the recording. For an electric guitarist, the amplifier is literally part of the instrument. Whereas the sound generation and amplification on an acoustic guitar are already separate— the strings and fretboard make the sound, the body amplifies it—on an electric guitar sound generation results from a considerably more complex system. Electrical pickups (usually magnets) turn the vibrations of the strings into electrical signals, which are sent down a cable to an amplifier, which processes the sound and then sends it out a set of speakers (indeed, when guitarists speak of the “amp,” they are often speaking synecdochically of the amplifier-speaker system). Without the amplifier, the guitar makes a sound, but not very much of one. With the amplifier, the guitar can become something like a controller, as single notes ring out and have impossible sustain, harmonics above and below sounded notes are synthesized and emphasized, and modulations such as built- in spring reverbs (smaller than the AKG) change the sound. Played at moderately high volumes, the guitar/ player/ amplifier/ room tetrad forms a kind of cybernetic system, a situation crystallized in the concept of feedback that is so central to both guitar playing and cybernetic theorizing. Guitar feedback, where the sound of the amplifier vibrates the strings and is detected by the pickups, which in turn send it to the amplifier for further
174 Designing Instruments, Calibrating Machines
amplification, is an instantiation of the more general cybernetic concept of feedback.27 This feedback question goes to the heart of why some guitarists reject digital models of amplifiers that otherwise would pass listening tests as indistinguishable from the analog device. Those who reject the digital models will often complain that they don’t feel the same, though this aspect of feeling is almost impossible to quantify. The difference between a software model and a physical device in the room can be important for someone playing an instrument, even if it is not audible to a third party. It is one thing for two recordings to be “perceptually equivalent”—for a listener to be unable to tell the difference between, for instance, a Marshall amplifier recorded live and a dry signal run through a piece of software that algorithmically models the behavior of the Marshall amp. For a mixing engineer who is not a guitarist, this equivalence is sufficient. For an audience member listening to the recording, it is also sufficient. But for a guitarist used to a set of physical interactions with an amplifier as part of their instrument, it may not be enough. Marcus Ryle, Line 6’s president and cofounder, explained to me: In the end, the only way people can actually decide whether we did a good job or not is to listen to the actual amps we modeled, which may not sound the same as the amps you own. But we do double-blind listening tests here with outside folks. . . . We’ve done it with everyone from artists to enthusiasts. We did it once at the LA amp show, where these are the real aficionados of boutique tube amps. And an interesting by-product that happens, I think at that show close to half the people we offered to did not want to take the test. The test was simply, here’s a guitar, there’s two amplifiers behind this sheet. Here’s your A/B foot switch, just play. Switch as long as you want, play whatever you want and just identify which is the modeling amp. And it seems that there’s people that don’t want to do that. That gets into issues far beyond technology, right? And the fact that for many people, it’s not possible for us to hear with just our ears.28
Ryle’s “far beyond technology” shows the definitional work attached to the A/B or A/B/X test. By defining the visceral, felt, and unheard dimensions of guitar playing as “far beyond technology,” Ryle is defining them as not relevant to the model. The distinct impression I got from our conversation was that the problem was one of mystification—that guitarists believe there is some magic in the devices, whereas to the engineer, they can be explained and those explanations can be operationalized in a model. There is another possible interpretation, however. It is entirely possible for the sound of a model to be indistinguishable from the sound of an analog
Constructing Digital Models of Analog Signal Processors 175
amplifier, for the behaviors around harmonic distortion or tone shaping to be identical, and for there still to be a difference. For one thing, the A/B/X test only tests the person’s hearing, not the guitar: the pickups “hear” the string (in some languages, the word for “pickup” is the same as “microphone” even though they work on different mechanical principles), but in a room they also recycle the signal from the amplifier, forming the basis of the feedback loop. Even if the sound of an amplified electric guitar can be recorded and recreated with completely different equipment, the experience may not be the same for the guitarist, and this matters precisely because modeling is so saturated with producer discourse. The embodied experience of playing at a lower volume, without a physical amplifier in the room, will be entirely different from the experience of playing at high volume. In Sensing Sound, Nina Sun Eidsheim uses the term “sonic reduction” to describe understandings of sound that treat it as a disembodied phenomenon. In the case of playing an electric guitar, sounding and hearing are multisensory phenomena for both the human being and the equipment. If the embodiment of the signal processor matters, then so too must the body of the listener or the user.29 The notion of the embodied guitarist is of course a highly gendered one— not to mention one weighted with racial and sexual overtones.30 Amplifiers also carry tremendous symbolic freight. A wall of amps for a large-scale concert once synecdochically represented the same kinds of power and mastery over nature that control over the guitar did. Their sheer size and sonic force reinforced the imagery of the individual musician as masterful and in control. For quite a long time, however, large guitar amplifiers have simply been unnecessary for most commercial music performance: they are louder than needed, they are less controllable than the installed mixer and amplifier setup found in performance venues, and, for the touring musician not rich enough to hire a fleet of roadies and ship gear across oceans, they are also expensive and difficult to move. Thus, in recent years, performing musicians in some genres have moved away from amplifiers, but in every instance I have found, their alternative is an intentional or inadvertent commentary on the symbolism of the now-absent wall of amplifiers, whether the replacement is a stack of washing machines, eye-searing LED displays, or giant artworks (Figure 6.3).31 Thus, to say that the presence or absence of an amplifier is a purely sonic question is to miss the cultural work that amplifiers do for both musicians and audiences. Line 6 deals with this problem of different needs for different user bases by making software plugins, digital modeling boxes that look nothing like amps, and various kinds of amplifiers. As a business, Line 6 takes a pragmatic approach rather than committing to a single philosophy. But for many users, the “device” meaning of amplifier goes beyond its sonic
176 Designing Instruments, Calibrating Machines
Figure 6.3 Extreme metal band Meshuggah in front of a custom stage backdrop. Photo courtesy of Carrie Rentschler.
characteristics—to gendered, embodied allusions to power and control that are intimately tied up with musical subcultures, genre identities, and the experience of making music. This is precisely the kind of collapse we often find in technologies: the device itself exists within the web of much larger and more sophisticated social relationships—commercial, financial, artistic, experiential, interpersonal. While there is much more to be said about the gender politics of guitar playing, especially given the increasing prominence of women and genderqueer guitarists in recent years, to treat the sound as the “thing” in amplification is clearly to single out only one part of the process. Whether or not there is an audible sonic difference between a Line 6 model and the analog amplifier it models, for many users there remains an irreducible cultural difference that must be negotiated one way or another.
Proof Is Not Enough If listening tests are the moment of proof for a digital model, they do not form the only basis on which digital models are built. Impulse responses are not sufficient either, because of their nature as single samples of an otherwise
Constructing Digital Models of Analog Signal Processors 177
dynamic process. This is where different schools of analog modeling diverge. Some treat the impulse response as the “truest” representation of an analog device, because it bears a relationship of cause and effect to the specific device— this is the approach of the company Acoustica Audio with its Nebula software, and is that found in the secondary market of sample libraries of impulse responses that can be loaded into digital modeling hardware or software. This approach works fine if the analog device is “linear and time invariant,” which is to say that the relationship between the input and output of the system is a linear map (given several inputs to the system with corresponding outputs, the output that corresponds to the sum of these inputs will be the sum of their corresponding outputs) and the system is time invariant (whenever you use the system, you will get the same result). But many analog systems are not linear and time invariant. The relationship between inputs and outputs of the analog system may not be a linear map and may not be time invariant—for instance, if a system behaves differently as it heats up (which is the case for tube amplifiers and for some analog synthesizers). Because of the possibility of these nonlinearities, other companies and users argue, a sampling-and- convolution-based approach is often insufficient on its own because analog devices vary their output over time and over settings. One solution is to add carefully calibrated degrees of randomness to convolution processing, which produces nonlinearity, but not in the way that an analog device would produce it. Universal Audio takes a different line, modeling the behavior of the different elements of the device to produce a digital model that both sounds like the analog device and processes like the analog device. There is a good degree of translation involved—the math for digital signal processing and the math for analog signal processing are not the same. But through years of research and experience, companies making digital models, such as Universal Audio and Line 6, build up sets of processing routines that describe different aspects of the way analog devices operate. The amplifier example shows the limits of the sonic model, and yet engineers go to great lengths to match interface and behavioral elements of a digital model to its analog referent, within the limits of the digital domain and interface. I return to my work with Universal Audio to explain how this happens. After the BX20’s impulse responses were judged to be perceptually equivalent, Abel and Coffin essentially reverse-engineered it. Using the signal flow diagram underlying the BX20, patents, schematics, and prior research, Abel and Coffin tried to create an algorithm that would reproduce the behavior of the BX20 at equivalent settings. As Abel explained to me: “We made a computational model of the impulse responses that would perceptually reproduce the measurements at the measured knob positions.”32 In other words, using a
178 Designing Instruments, Calibrating Machines
skeuomorphic interface that looked like a two-dimensional representation of the BX20’s control panel (Figure 6.4), their digital model should work just like the analog BX20. A visual representation of a knob turned to 3:00 on Abel and Coffin’s model should work just like a knob turned to 3 on the analog device. Of course, “work just like” has to be operationalized here, and the only two options for operationalizing it are listening tests and measured response curves. Again, the testing scenario defines functions for the device and its digital model. But so does prior work in the field. Abel had previously been involved in modeling another famous device, the Roland RE-201 Space Echo—a signature psychedelic effect used in everything from dub reggae to progressive rock to techno. Part of the Space Echo’s “spaceyness” comes from a spring reverb attached to its main function, a tape delay. Abel and Dave Berners had spent considerable time studying spring physics and modeling the physics of the spring inside the Space Echo, developing a theory of how it transformed sound and a mathematics to represent and operationalize that theory in Universal
Figure 6.4 Universal Audio’s skeuomorphic panel. Universal Audio website, http://www.uaudio.com/uad-plugins/reverbs/akg-bx-20-reverb.html.
Constructing Digital Models of Analog Signal Processors 179
Audio’s model.33 Thus, when confronted with the BX20, Abel already had a working model of a spring and could begin by adjusting it for the BX20’s much larger spring system. This was also true of the BX20’s other components, such as the amplifiers. Universal Audio essentially had a library of building blocks and, crucially, a set of submodels of how those different elements of technologies interact. This process of developing models of components and relations among components converts signal processing into a kind of story that can be told about a device: if it has these kinds of op-amps, and those kinds of potentiometers, then Universal Audio’s engineers expect it will behave in a certain way. To be clear, a narrative is not a universal, ontological condition of signal processing, but it may be a common social precondition for digital models and anything that involves reverse-engineering or repair.34 Engineers need a way to explain what happens inside a circuit, which has both a temporal dimension (a series of events occur) and a spatial dimension (a topology or shape). Narratives, stories of what happens when, why, and how, become the glue for keeping together understandings of the parts of a circuit in an account of its behavior as a whole. Each submodel—how a spring works, how a resistor works—contributes to engineers’ un-black-boxing of the hardware device. The narrative is the moment that identifies its completion. So when Universal Audio’s engineers assembled their operational submodels and tuned them, they constructed a working behavioral model of the BX20, which was simultaneously a story of how it worked. At that point, they aimed to get the behavioral model to produce sounds that were perceptual matches to the impulse responses of the original BX20. Thus, the tests that Abel, Berners, and Ryle all spoke of were tests of the device as well as of the story of the device. In testing, a chain of equivalence is set up: Analog BX20 at a given setting ←listener→ Impulse responses at a given setting Impulse responses at a given setting ←listener→ BX20 operational model Analog BX20 ←→ BX20 digital model
As Berners explained, this final round of testing is “quality assurance”—a kind of epistemic sealant. The goal is perceptual equivalence at the level of hearing. The software passes the test when the user fails it. But another set of equivalences is also being shored up, because both the visual and sonic dimensions of the BX20 are rendered skeuomorphically. Universal Audio’s business model is to depict the analog device as much as possible in the digital interface. The knobs on the digital model of the BX20 interface are a perfect example: they are not necessary for the software code to do
180 Designing Instruments, Calibrating Machines
its work on the transduced sound. Given that most users of Universal Audio software are working with computer mice, a slider would probably be easier to use, but because the interface looks like the BX20, it helps to reinforce the idea that the interface sounds like the BX20. Employing visual skeuomorphism to represent sonic commensurability is a long-standing practice in the modeling world, and is a longer part of the history of commercial music technologies. Leo Fender used auto-body paints for guitars in part because they were widely available and in part to associate his guitars with other kinds of commodities; Tara Rodgers has shown that wood panels on synthesizers come out of a tradition of wood-paneling other consumer goods to make them seem more “organic” and draw attention away from the social relationship in which they are embedded. The same can be said here: a skeuomorphic interface, as opposed to an interface designed around user practice in software, creates an analogy of practice for users and a logic of equivalence—one may be substituted for the other.35 This substitutionalist logic has been in operation for most of the history of commercial use of digital audio software. Even something as basic as the visual representation of an analog mixing board in ProTools, a popular digital mixing program, follows this logic.36 As Evan Brooks, ProTools’s co-inventor, explained to me, “To have a separate mixer was really just an attempt to gain— to give people some familiarity with the process. When you’re moving from analog over to a digital way of doing things we didn’t completely change their overall view, and so we felt that—people’s workflow was still kind of, at least mentally if nothing else, divided into concepts of tracking and then mixing.”37 Brooks here is referring to the mixer window in ProTools, which is meant to represent in two dimensions on the computer screen the physical controls of a hardware mixer. Visual skeuomorphism is rhetorical; it is a story about the ordering of tasks and operations in the creative process. In preserving the task sequence from analog device to digital model, the modeling industry can be said to be inherently conservative. It uses a rhetoric of democratization—more people have more access to more tools at lower cost—and this is probably true.38 In that context, the model performs a kind of canonizing process, aestheticizing a set of relations into “workflow”—record, then mix; reverberate this way, not that way. To offer “something familiar” is to begin to set the terms of how musical work is supposed to be done, and the use of models points back to tradition. This conservatism is not an inherently bad thing—tradition is an important part of music education in almost all contexts, and practice and imitation is an easy way for beginners to learn outside of formal educational contexts. At another level, the conservatism actually masks a transformation of relations; behind the avowed conservatism of modeling is industrial competition and
Constructing Digital Models of Analog Signal Processors 181
change. Brooks’s “attempt to give some familiarity” links together two technical practices that have no necessary, preordained relationship and suggests a relationship of descent and morphology in the gesture: “See, listen, it is the same as . . . .” Other software companies have gone in other directions, with other consequences. Thus, procedurally, to create a digital model of an analog audio device, engineers follow a process of defining, testing, refining, and redefining, with listening tests at every stage of the process. But socially, to create a digital model requires establishing relations of equivalence at each stage. Listening— whether done by people or machines—is a crucial part of this practice, but so is the development of a model of how the analog device “does what is does.” As it is developed, the model circumscribes essential dimensions of the analog device and brackets off nonessential elements. At different stages, listening tests establish equivalence and consecrate it. We might therefore be led to ask how much of the sound of a digital model is essential and how much is superficial—can digital models of analog devices be said to be sonic skeuomorphs as well as visual skeuomorphs? Or to put it another way, does a difference matter if you can’t hear it? It may matter in a host of ways that are deliberately set aside in the moment of testing. Perhaps the difference matters aesthetically, to the degree that aesthetics are not reducible to measurable perception. But the differences that matter may also lie further out—bodily, yes, but also technologically, culturally, politically, and economically. Who gets to signal process and under what conditions is a central question of media theory, and the very question that is left aside at the moment of the listening test. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/digital-models-sterne.
Acknowledgments In addition to audiences at the Max Planck Institute for the History of Science and the International Society for Intermedial Aesthetics, I wish to offer thanks to Carrie Rentschler, Ky Brooks, Gabriella Coleman, Emily Dolan, Mara Mills, Alix Hui, and Viktoria Tkaczyk for their comments on previous drafts, Jonathan Abel for educating me about signal processing and for the access needed to make this paper happen, and Kate Sturge, who is the most thorough editor I know.
182 Designing Instruments, Calibrating Machines
Notes 1. John Mowitt, “Sound of Music,” 194. 2. Of course, the idea of perfect fidelity was itself a fiction, but that is a topic I have covered at length elsewhere. Jonathan Sterne, Audible Past, 215–86. 3. Susan Schmidt Horning, Chasing Sound. 4. Recounted in Daniel Levitin, This Is Your Brain on Music, 151. See E. Glenn Schellenberg, Paul Iverson, and Margaret C. McKinnon, “Name That Tune.” 5. Christer Grewin and Thomas Rydén, “Subjective Assessment.” 6. Robert Poss, “Distortion Is Truth”; Albin J. Zak, Poetics of Rock; Simon Frith and Simon Zagorski-Thomas, Art of Record Production. 7. Trevor Pinch and Frank Trocco, Analog Days; Tara Rodgers, “Synthesizing Sound”; Ian Dunham, “From Kitschy to Classy.” It should be noted that this is not strictly an analog/ digital phenomenon. Musicians have long prized older equipment for its supposed characteristics. See Emily Dolan’s chapter in this volume and H. Stith Bennett, On Becoming a Rock Musician. 8. All the models considered here were designed by for-profit companies for sale to end users: musicians, audio engineers, producers, podcasters, and others. There is considerable crossover between commercial and academic contexts in music technology research (probably a bigger distinction than between nonprofit and for-profit research), but that is beyond the scope of this chapter. As I discuss later, the profit motive goes partway to explaining some aspects of analog modeling—branding and marketing most notably—but those alone are not sufficient to explain the cultural significance of the phenomenon. 9. This approach is drawn from the anthropology of sound technology. See Louise Meintjes, Sound of Africa; Stefan Helmreich, Alien Ocean; and David Novak, Japanoise, for examples of ethnographers reflexively positioning themselves with respect to their subjects and objects. 10. Author interview with Stephan Schmitt, October 27, 2011. See also Jonathan Sterne, MP3, 150–73. 11. N. Katherine Hayles, How We Became Posthuman, 17. 12. The two plugins arguably do sound different, but whether that difference is audible in a full mix is a question that would have to be answered on a case-by-case basis. 13. Georgina Born, “Computer Software”; Wendy Hui Kyong Chun, Control and Freedom; Derek Robinson, “Analog”; Jonathan Sterne, “Analog”; Benjamin Peters, “Digital.” 14. Lorraine Daston, Things That Talk, 226. 15. Joseph Klett, “Baffled by an Algorithm,” 123. 16. Albert Borgmann, Technology, 43. 17. Meintjes, Sound of Africa, 73–74. 18. Madeleine Akrich, “De- Scription of Technical Objects”; Theresa Hak Kyung Cha, Apparatus, Cinematographic Apparatus; Georg Lukács, History and Class Consciousness. 19. The recent contest around the terms other than device suggests that we need to revisit its meaning as well, especially given its ambiguity in relation to the French appareil and dispositif, two terms that have been so central to cultural theory of the past generation. See Michel Foucault, “Questions of Method”; Giorgio Agamben, What Is an Apparatus?; Karen Barad, Meeting the Universe Halfway. 20. “Original digital models” may seem like an odd locution, but it is correct here. Universal Audio has revised its models since Abel’s work on an earlier version of the software,
Constructing Digital Models of Analog Signal Processors 183 building on the paradigm he helped establish but making use of the increased processing power inside computers. 21. Strictly speaking, it would be possible to model some aspects of a device, and it would also be possible to model a device based purely on measuring inputs and outputs. But in practice, because it is engineers doing the modeling, they tend to reverse-engineer anything that they plan to model. 22. For more on convolution, see Jonathan Sterne, “Space Within Space.” 23. I discuss the concept of “expert listeners” brought in for listening tests in Sterne, MP3, 163–73. 24. Author interview with Dave Berners, May 13, 2011. All interviews are lightly edited from spoken language for readability. 25. Trevor Pinch, “ ‘Testing—One, Two, Three . . .,” 26. 26. Dylan Mulvin, “Reference Materials,” 12. 27. Norbert Wiener, Human Use of Human Beings; Steve Waksman, Instruments of Desire; Novak, Japanoise, 139–68. 28. Author interview with Marcus Ryle, July 15, 2011. 29. Nina Sun Eidsheim, Sensing Sound, 2, 120–29. As someone who has used feedback in my own music, I have experienced the interaction between a vibrating instrument and an amplifier, and the only way I can describe it is that it is like shaping electricity and sound together, as if they were clay on a rotating potter’s wheel. But it is certainly possible to achieve this with digital audio technologies, so long as there is also an amplifier. 30. Waksman, Instruments of Desire, 188; Mavis Bayton, “Women and the Electric Guitar.” 31. Other examples are Rush (see http://www.cygnus-x1.net/links/rush/mobile-productionsmonthly-09.2011.php, or search for “Rush tour washing machines”) and Sleater-Kinney (https:// w ww.inlander.com/ s pokane/ c oncert- review- s leater- k inney- g oes- big- on- tour-opening-night-at-the-fox/Content?oid=18425517, or search for “Sleater Kinney tour LED”). 32. Jonathan Abel, email to author, August 5, 2015. 33. Jonathan Abel and Dave Berners, “Digital Emulation.” 34. It is no accident that old repair manuals for analog gear are an important resource for engineers who model analog equipment. Whether to model something or to fix it, one needs an understanding of how the component parts fit together. Steven Jackson, “Rethinking Repair.” 35. Tara Rodgers, “Into the Woods.” 36. See, for example, https://www.youtube.com/watch?v=qIj9WTTkA5o, or search for “Pro Tools mixer window.” 37. Author interview with Evan Brooks, February 15, 2012. 38. Walter Benjamin, “Work of Art.”
References Abel, Jonathan, and Dave Berners. “Digital Emulation of the Roland RE-201 Space Echo.” Lecture at the Centre for Interdisciplinary Research in Music Media and Technology, McGill University, November 2, 2006. Agamben, Giorgio. What Is an Apparatus? And Other Essays. Translated by David Kishik and Stefan Pedatella. Stanford, CA: Stanford University Press, 2009.
184 Designing Instruments, Calibrating Machines Akrich, Madeleine. “The De-Scription of Technical Objects.” In Shaping Technology, Building Society: Studies in Sociotechnical Change, edited by Wiebe E. Bijker and John Law, 205–24. Cambridge, MA: MIT Press, 1992. Barad, Karen. Meeting the Universe Halfway: Quantum Physics and the Entanglement of Matter and Meaning. Durham, NC: Duke University Press, 2007. Bayton, Mavis. “Women and the Electric Guitar.” In Sexing the Groove, edited by Sheila Whitely, 37–49. New York: Routledge, 1997. Benjamin, Walter. “The Work of Art in the Age of Mechanical Reproduction.” Translated by Harry Zohn. In Illuminations, edited by Hannah Arendt, 217–51. New York: Schocken, 1968. Bennett, H. Stith. On Becoming a Rock Musician. Amherst: University of Massachusetts Press, 1980. Borgmann, Albert. Technology and the Character of Contemporary Life. Chicago: University of Chicago Press, 1984. Born, Georgina. “Computer Software as a Medium: Textuality, Orality, and Sociality in an Artificial Intelligence Research Culture.” In Rethinking Visual Anthropology, edited by Marcus Banks and Howard Morphy, 139–69. New Haven, CT: Yale University Press, 1999. Cha, Theresa Hak Kyung. Apparatus, Cinematographic Apparatus: Selected Writings. New York: Tanam Press, 1980. Chun, Wendy Hui Kyong. Control and Freedom: Power and Paranoia in the Age of Fiber Optics. Cambridge, MA: MIT Press, 2005. Daston, Lorraine. Things That Talk: Object Lessons from Art and Science. New York: Zone Books, 2004. Dunham, Ian. “From Kitschy to Classy: Reviving the Roland TR-808.” Sounding Out! The Sound Studies Blog, June 9, 2014, https://soundstudiesblog.com/2014/06/09/808/. Eidsheim, Nina Sun. Sensing Sound: Singing and Listening as Vibrational Practice. Durham, NC: Duke University Press, 2015. Foucault, Michel. “Questions of Method.” Translated by Colin Gordon. In The Foucault Effect: Studies in Governmentality, edited by Graham Burchell, Colin Gordon, and Peter Miller, 73–86. Chicago: University of Chicago Press, 1991. Frith, Simon, and Simon Zagorski-Thomas, eds. The Art of Record Production: An Introductory Reader for a New Academic Field. Burlington, VT: Ashgate, 2012. Grewin, Christer, and Thomas Rydén. “Subjective Assessment of Low Bit-Rate Audio Codecs.” Paper presented at the 10th International Conference of the Audio Engineering Society, London, September 1991. Hayles, N. Katherine. How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics. Chicago: University of Chicago Press, 1999. Helmreich, Stefan. Alien Ocean: Anthropological Voyages in Microbial Seas. Berkeley: University of California Press, 2009. Horning, Susan Schmidt. Chasing Sound: Technology, Culture, and the Art of Studio Recording from Edison to the LP. Baltimore, MD: Johns Hopkins University Press, 2013. Jackson, Steven. “Rethinking Repair.” In Media Technologies: Essays on Communication, Materiality, and Society, edited by Tarleton Gillespie, Pablo J. Boczkowski, and Kirsten A. Foot, 221–39. Cambridge, MA: MIT Press, 2014. Klett, Joseph. “Baffled by an Algorithm.” In Algorithmic Cultures: Essays on Meaning, Performance and New Technologies, edited by Robert Seyfert and Jonathan Roberge, 111–27. London: Routledge, 2016. Levitin, Daniel J. This Is Your Brain on Music: The Science of a Human Obsession. New York: Plume, 2006. Lukács, Georg. History and Class Consciousness: Studies in Marxist Dialectics. Translated by Rodney Livingstone. Cambridge, MA: MIT Press, 1971.
Constructing Digital Models of Analog Signal Processors 185 Meintjes, Louise. Sound of Africa!: Making Music Zulu in a South African Studio. Durham, NC: Duke University Press, 2003. Mowitt, John. “The Sound of Music in the Era of Its Electronic Reproducibility.” In Music and Society: The Politics of Composition, Performance and Reception, edited by Richard Leppert and Susan McClary, 173–97. New York: Cambridge University Press, 1987. Mulvin, Dylan. “Reference Materials: The People, Places and Things of Making Measurements.” PhD diss., McGill University, 2015. Novak, David. Japanoise: Music at the Edge of Circulation. Durham, NC: Duke University Press, 2013. Peters, Ben. “Digital.” In Digital Keywords: A Vocabulary of Information Society and Culture, edited by Ben Peters, 93–108. Princeton, NJ: Princeton University Press, 2016. Pinch, Trevor. “‘Testing—One, Two, Three . . . Testing!’: Toward a Sociology of Testing.” Science, Technology, & Human Values 18, no. 1 (1993): 25–41. Pinch, Trevor, and Frank Trocco. Analog Days: The Invention and Impact of the Moog Synthesizer. Cambridge, MA: Harvard University Press, 2002. Poss, Robert M. “Distortion Is Truth.” Leonardo Music Journal 8, no. 1 (1998): 45–48. Robinson, Derek. “Analog.” In Software Studies: A Lexicon, edited by Matthew Fuller, 21–31. Cambridge, MA: MIT Press, 2008. Rodgers, Tara. “Into the Woods: A Brief History of Wood Paneling on Synthesizers.” Sounding Out! The Sound Studies Blog, May 30, 2011, https://soundstudiesblog.com/2011/05/30/into- the-woods-a-brief-history-of-wood-paneling-on-synthesizers/. Rodgers, Tara. “Synthesizing Sound: Metaphor in Audio-Technical Discourse and Synthesis History.” PhD diss., McGill University, 2011. Schellenberg, E. Glenn, Paul Iverson, and Margaret C. McKinnon. “Name That Tune: Identifying Familiar Recordings from Brief Excerpts.” Psychonomic Bulletin & Review 6, no. 4 (1999): 641–46. Sterne, Jonathan. “Analog.” In Digital Keywords: A Vocabulary of Information Society and Culture, edited by Benjamin Peters, 31–44. Princeton, NJ: Princeton University Press, 2016. Sterne, Jonathan. The Audible Past: Cultural Origins of Sound Reproduction. Durham, NC: Duke University Press, 2003. Sterne, Jonathan. MP3: The Meaning of a Format. Durham, NC: Duke University Press, 2012. Sterne, Jonathan. “Space Within Space: Artificial Reverb and the Detachable Echo.” Grey Room 60 (Summer 2015): 110–31. Waksman, Steve. Instruments of Desire: The Electric Guitar and the Shaping of Musical Experience. Cambridge, MA: Harvard University Press, 1999. Wiener, Norbert. The Human Use of Human Beings: Cybernetics and Society. New York: Doubleday, 1954. Zak, Albin J. The Poetics of Rock: Cutting Tracks, Making Records. Berkeley: University of California Press, 2001.
7 To Hear as I Do The Concessions of Hearing in Taiwan’s Noise Management System Jennifer Hsieh
Testing Hearing in Taiwan While broadcasts of Beethoven’s Für Elise and Tekla Bądarzewska- Baranowska’s “A Maiden’s Prayer” resound nightly from loudspeakers perched atop roving garbage trucks, residents in Taiwan continue on with their lives, seemingly immune to the hubbub of the city. During interviews conducted in Taipei in 2014–15, residents would explain to me: “The garbage truck plays music, so it is not noise.” And if the monophonic, mechanical renditions of Western art music were not actually music to their ears, they nevertheless insisted that the songs were friendly reminders to take out the trash. Yet for the past forty years, noise has been the number one urban problem reported by residents in Taipei, ahead of other environmental issues such as noxious smells, air pollution, water contaminants, and illegal waste disposal.1 The discrepancy between documented reports of noise and the general, public perception of noise is even more intriguing given that, based on 24-hour noise monitoring data in Taiwan, the rate of decibel measurements that violate existing noise control standards is at an all-time low.2 Rather than being an absolute entity that exists out there in the physical environment, noise has been socially and historically produced as an object in need of elimination. In this chapter, I trace the converging roles of scientific testing and political reform in the production of everyday listening practices.3 I investigate how noise was created both as an object for government regulation and as an indicator of public opinion in connection with public health and environmental acoustics. And I examine how government research on noise facilitated the transition from authoritarian to postauthoritarian rule under the Chinese Kuomintang (KMT) regime in the 1970s and 1980s.4 At a time when the regime was losing its international standing as the sole legitimate representative of mainland China and Taiwan, global developments in noise abatement Jennifer Hsieh, To Hear as I Do In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0008.
190 Managing Sound, Assessing Space
and noise control engineering emerged as a benchmark opportunity for state-building from within Taiwan. More than the day-to-day management of noise, Taiwan’s noise control system enacted the ideology of scientific objectivity and rational governance at a time when doing so had political expedience both domestically and globally. By adopting noise abatement practices, the regime took steps toward reinventing itself from an authoritarian state, known for its thirty-eight years of martial law, to a modernizing and democratizing state. In contrast to previous noise regulations in Taiwan—such as the Noise Policing Act of 1959, which enhanced the authority of military police to control social behavior during public assembly—the Noise Control Act (NCA) of 1983 applied site- specific noise level tests, animal testing, community surveys, and human response tests in the development of legislation. Similar to noise regulations in countries such as Japan, the NCA determined permissible noise levels on the basis of a rubric of standardized decibel levels in a particular frequency band, time period, and geographic location. Given a standard of 67 dB, for instance, a sound measuring 65 dB would not be considered noise, whereas a sound measuring 67.8 dB would. Scholars such as Emily Thompson and Karin Bijsterveld have noted that noise-monitoring methods were used in Europe and North America as early as the 1930s, not too long after the invention of the decibel unit.5 But whereas noise abatement in the West started as an engineering problem in response to the sounds of industrialization, in Taiwan it developed in the 1970s and 1980s through state-sponsored activities that introduced the concept of noise to the general public. Furthermore, what was important for policymakers and researchers in creating a noise control policy was not only the adoption of noise abatement practices but also a public display of Taiwan’s technological advances. These methods were intended to eliminate individual bias, political favoritism, and inequity in the management of noise problems by subjecting all reports of noise to technological verification. Seen as a methodical, objective alternative to the coercive, secretive practices of postwar martial law, noise control management represented a deliberate effort by the state to reimagine itself as rational and transparent at the same time as it carved out a new domain for social control. The promise of technology to determine noise produced a double bind: it could verify a noise problem as easily as it could disqualify a noise problem. The noise management system thus became a modernizing project that standardized the hearing of noise through an epistemology of testing. Data on subjective and objective measurements of noise were tallied and analyzed to establish average acceptable noise standards throughout Taiwan. Doctors and
To Hear as I Do 191
engineers working for the Taiwan Environmental Protection Administration (TEPA) and its predecessor, the Ministry of Public Health, cross-referenced data from audiometric testing, socioacoustic surveys, international standards, and site-specific decibel measurements to arrive at a calculus of local thresholds for urban noise. These thresholds delineated the range of decibel levels that could be regarded as noise, and also had the effect of placing the hearing of noise under scrutiny.6 In other words, some perceptions of noise were legitimized while others were discounted, particularly those that could not be substantiated by additional measured effects. By co-producing standards for noise levels along with residents’ tolerance for noise, the testing and measuring of sounds went beyond testing the sound itself—they also managed the validity of people’s claims about noise. My chapter builds upon three existing bodies of literature: social and historical studies of noise and noise control; the quantification and standardization of perception; and East Asian modernities in a global context.7 In the literature on the history of noise and noise control, scholars have identified noise as a cultural artifact whose meaning and prominence changes in different historical, geographical, and technological contexts.8 My work adds to the literature with a study of noise in a place where the concept of noise does not necessarily presume unwanted sound. In fact, the Chinese concept of renao, or “hot-noisy,” treats the hubbub of the city as a desirable quality that gives city streets a lively, lived-in feel.9 In efforts to implement noise control in Taiwan, the cultural aesthetic of renao became an obstacle to convincing the public that noise was a problem; politicians and social critics lamented the apparently high tolerance of local residents to sounds that surely ought to be eliminated.10 Couched in the language of noise control was the discourse of modernity. The Beijing-born, Taiwan-based social critic He Fan, who published numerous columns about noise in the state-sponsored United Daily News from the 1960s to the mid- 1980s, bluntly stated that the reason locals were loud was because they had not yet become civilized, modern people.11 Lawmakers echoed this sentiment during legislative proceedings and acknowledged that the public’s embrace of renao hindered public awareness of the problem of noise.12 The establishment of noise as a problem went hand in hand with attempts by the state to make Taiwan—and Taiwanese subjects—modern. By describing the effects of testing in the creation of a noise control system, I show how quantitative measurements rendered certain sounds as noise and made them legible. Testing had the effect not only of recategorizing certain city sounds as noise but also of creating the epistemological framework for officials and residents to recognize noise in the first place. In this way, noise
192 Managing Sound, Assessing Space
became a scientific object. Through a process of identifying, reporting, and verifying noise, the state established a system that was purportedly free from the personal bias of state authorities and, instead, dependent on measurement. Drawing on the argument by scholars of science and the law that quantitative measures can form the basis for reaching consensus, my chapter engages with studies on scientific methods and testing as a mode of public knowledge formation.13 The use of scientific methods by the state was a significant aspect of political reform for a regime that was striving to distance itself from its authoritarian origins. As one legislator repeatedly mentioned during the debate on the NCA: “The noise control law needs to clearly state the standards, method, and implementation protocol. There needs to be a clear method, including the type of device that is used, for the public. Clear noise control standards, noise control zones, and prevention methods all need to be established as part of our duty to the public.”14 The legislator’s preoccupation with a transparent methodology suggests that, unlike in the previous era of governance, the legitimacy of the regime depended on government accountability and public support. A methodical approach to noise control narrowed the enforcement capabilities of the state by taking away the individual authority of enforcement officers, but it also enhanced the state’s overall authority to regulate noise, a practice that produced the act of hearing as a quantifiable form of behavior.15 Within the noise control system, the hearing of noise by individuals took place in relation to external references that either validated or rejected individual perceptions. For instance, researchers tested the hearing abilities of residents and compared the results with residents’ own attitudes to noise. Through statistical analysis, the biological act of hearing and the perception of noise both turned into immutable mobiles, alongside other numerical data including decibel measurements and noise level standards. Hearing became one of many parameters in the calculation of noise, and the right to adjudicate noise was diverted from human hearing to external test results.16 The geographic locale of my study speaks to the application of Western technologies in a postcolonial context. The anthropologist Michael Fischer writes that research on technology and science in Asia is much concerned with “the emotional and aesthetic facets of science and technology, the social worlds they create and in which they operate, as well as the uneven developments, localizations, and alternative trajectories of the sciences and technologies in different places.”17 Emerging almost half a century after similar regulations were put in place in the West, noise control in Taiwan calls attention to the global circulation of noise control practices and “the social worlds they create” in one sociopolitical context. Whereas noise abatement in Europe and North
To Hear as I Do 193
America belongs to the discourse of urbanization and industrialization at the turn of the twentieth century, noise control in Taiwan signified the promise of democratic reform together with a populist message of listening to the needs of the people. Taiwan was adopting a noise control system within a changing global context in which effective strategies for noise management were still under debate. In an interesting twist, the 1983 enactment of the NCA in Taiwan occurred just two years after the United States defunded its federal noise control program, delegating the management and enforcement of noise regulations to local and state authorities. Taiwan’s creation of a centralized form of noise control at a time when the established models for noise abatement in other countries were beginning to shift set the stage for an alternative model of noise management as part of an aspiration to liberal governance. The creation of noise control regulations, as well as the system’s interpellation of individual subjects within its framework of testing, took the form of an unprecedented modernity that transformed both hearing and noise into a state-level problem. In the rest of the article, I trace the major steps in research and public outreach from the 1960s to the 1980s, describing how Taiwan’s noise control system emerged from a contrapuntal process that tested environmental sounds and human hearing in relation to international standards and measurement methodologies. Medical researchers in the 1960s conducted audiometric testing to examine hearing loss among schoolchildren, but in the 1970s, as the environmental rights movement gained momentum worldwide, they applied their audiometry skills to a new project of noise control. Hearing tests and socioacoustic surveys soon provided human points of comparison to on-site decibel measurements of urban street noise. As a result, the noise control system incorporated Taiwanese subjects into a way of hearing noise that was constructed and produced by the state. What may once have been considered a private act of hearing became a public act, historically and socially situated within a modern state apparatus.
Before There Was Noise: Audiometry and Geopolitics The precursor to noise control research in Taiwan can be found in the field of otolaryngology and health education. Beginning around the 1960s, audiometric testing was carried out in the form of large-scale investigations of hearing health among schoolchildren. These studies were intended to identify potential obstacles to students’ academic performance and offer treatments,
194 Managing Sound, Assessing Space
such as hearing aids, to improve students’ basic learning skills and classroom participation. In these early years of audiometry, noise was neither a public health problem nor an environmental problem intrinsically. Only later was it produced as such. Dr. Wang Lao-teh, a local Taiwanese ear, nose, and throat doctor and later the head researcher for Taiwan’s noise control system, built his medical career by studying the causes of hearing loss among schoolchildren while working as a professor at National Taiwan Normal University’s Department of Health Education.18 Before Wang began conducting research on noise, he was primarily concerned with the biological causes of deafness. For one project, Wang used an American-made Beltone audiometer to test the hearing abilities of over 23,000 students, from middle school to university level, at different frequency levels from 500 to 4,000 Hz.19 These tests were used to identify the students with hearing loss at 20 dB or more and to determine the correlation between age, gender, and biomedical circumstances in hearing health. According to Wang’s studies, the suspected causes of hearing loss included nutritional deficiencies, the uneven rate of male and female adolescent development, and academic pressure. Wang explained: The reasons for hearing damage include the burdens of a heavy homework load that requires the use of the mind for long periods of time. The associated stress is most pronounced during the adolescent stages in life. It takes its toll on physical well-being and lowers resistance to illnesses. It is common for children who suffer from high fevers to then suffer damage to their auditory systems.20
According to Wang, hearing health was tied to a child’s physical disposition and susceptibility to illness. Wang’s studies demonstrate that, initially, hearing health was understood as a problem of internal medicine rather than one of external conditions. Additional research on hearing loss was conducted by Wang and his students at the Taipei School for the Blind and Deaf. These investigations focused on the developmental health of children as infants and in utero. In one study, researchers collected data on the conditions of each subject’s birth, the medical history of the mother, and the mother’s medicinal drug intake during pregnancy.21 The data was limited to medical and developmental accounts of human subjects, again suggesting that hearing health was a medical problem determined by a patient’s personal health history. During the 1960s, Wang published as many as twenty articles on hearing damage among schoolchildren. These reports appeared in the local journal
To Hear as I Do 195
of the Association of Ear, Nose, and Throat Doctors; in his own university’s journal Health Education; and through the Taiwan provincial government’s Health Education Commission. Wang’s publications rarely comment on the dangers of environmental noise. In fact, the term “noise” (zaoyin) is never mentioned in the titles of his articles in the 1960s, which instead focus on hearing impairment (tingli zhangai) and its medical causes. If environmental noise exposure was regarded as a peripheral factor in hearing loss in the 1960s, the 1970s saw a concerted effort by government agencies and health experts to establish noise as a public health problem that affected residents of all ages.22 The Central Daily News, the official newspaper of the KMT, cited Wang’s medical expertise to justify the call for new regulations on noise. One article summarizes a talk given by Wang at the second meeting of the Asia-Oceania Association of Otolaryngology Head and Neck Surgery, in which he stated that urban noise was the primary cause of hearing damage. The article concluded: “otolaryngologist Wang Lao- teh called for relevant authorities to reduce this bloodless killer called noise [sharen bujian xie de zaoyin].”23 According to official state news, the main cause of hearing loss was now recognized to be the various sources of noise audible throughout the city. Hearing loss was here connected to the external factors of environmental noise, a reversal of Wang’s earlier position—but the claims in the news report likely reflect the agenda of the state rather than Wang’s actual research. Just a few months before the publication of the news article, Wang had offered a multivariable analysis of hearing health in his university department’s monthly publication Health Education. There, he wrote: “Children’s hearing loss is affected by the social environment, climate, noise, and vitamin deficiency.”24 The article cites research from Japan, the United States, and Western Europe on the medical causes of hearing loss, including head injury, viral infections, and maternal rubella. In a medical context, noise exposure was just one of many possible causes of hearing damage. In official state discourse, however, urban noise was named as the leading cause of hearing and health problems. Identifying a new domain for the application of specialized knowledge, state media brought public attention to Wang’s research but altered the discourse from human development to public health. Studies of the biomedical conditions of hearing health in an educational setting, then, were eclipsed by state interest in the problem of urban noise. Government support for noise regulations gained momentum throughout the 1970s, beginning with a state-sponsored public opinion poll. In a 1975 survey by the Taiwan Public Opinion Association, a subsidiary of Gallup
196 Managing Sound, Assessing Space
International, a total of 1,500 residents in the three major cities of Taiwan— Taipei, Taichung, and Kaohsiung— were surveyed on the public health problems of their city. Conducted by trained pollsters, the survey asked residents to rank the leading urban pollutants including air pollution, water pollution, and noise. It concluded: “In the past ten years in the region of Taiwan, industrialization has rapidly grown and transportation methods have increased. Smoke from factories and cars have released contaminants into the air and water. Air pollution and water pollution, as well as the noise from these sources, affect the health of the people.”25 The creators of the survey likely expected air and water pollution to rank as the top environmental problems and relegated noise to a secondary clause. Contrary to those expectations, and to the surprise of environmental officials, residents identified noise as the leading urban problem in two of the three cities, Taipei and Kaohsiung, ahead of air and water pollution.26 The state laid claim to the results of this survey as a justification for dedicated government research on environmental noise. Huang Chyan-chyuan, a former student of Wang’s and later consultant to the TEPA, explained to me during an interview: “The government conducted a poll on citizens’ views of public health. No one expected noise to be a big issue, but it turned out to be the number one problem! After that, the government realized that they had to do something about noise.”27 Multiple government reports subsequently cited the findings of the 1975 poll as the initial grounds for Taiwan’s NCA, as it made the government aware of the gravity of the noise problem.28 According to these accounts, the government’s attention to the problem of noise directly resulted from their efforts to reach out to the public and listen to its needs. The public opinion report introduced a new way of thinking about noise control, one that focused on the people’s will and contrasted with previous forms of regulation, notably the Noise Policing Act of 1959. Created as part of the Social Maintenance Laws under the jurisdiction of the police, the Noise Policing Act had once expanded the authority of enforcement officers during the height of martial law. Outdoor sounds that drew public attention were explicitly regulated, including the amplified sounds emanating from street performances, public lectures, street vendors, and even schoolyards. With the advent of the NCA, the Noise Policing Act became a footnote to history, with modern-day reports claiming that its implementation failed because it lacked a clear method of enforcement.29 Suggesting a more historicized reason behind the abandonment of the Noise Policing Act, legislators discussing the NCA in 1982 alluded to the expansive powers of the police that needed to be limited by new legislation, explaining that the Noise Policing Act “was
To Hear as I Do 197
not very effective because there were no clear standards for noise and the authority to act was too broad.”30 As part of the state’s political reform, the NCA resonated with the image of a modern, rational state; the Noise Policing Act was better left forgotten. In contrast to the regulations that sought to control public behavior under martial law, noise regulations were promoted in the 1970s as being an attempt by the KMT regime to attend to the demands of the people and work with residents to improve quality of life. Public opinion polling was presented as an initial step toward democratic reform. In the introduction to the same issue of Public Opinion Poll Quarterly that published the poll on urban pollution, the editor Wu Wang-ji praised Taiwan’s participation in a global movement championing the voices of the public. Wu explained that Taiwan was following in the footsteps of developed countries in Europe and North America and lauded Taiwan’s progressive facilitation of an open and democratic environment, as “gathering the thoughts of the public through polling is one way to achieve democracy.”31 The 1975 poll offered tangible evidence that urban noise was recognized as a problem by residents, but government support for the noise control system also derived from national security concerns. Several events during the 1970s required Taiwan to re-evaluate its relationship with the international community and reassess its national agenda. In 1971, Taiwan, also known as the Republic of China (ROC), lost its membership in the United Nations and was replaced by the People’s Republic of China (PRC) as the sole representative of China. Richard Nixon’s 1972 visit to Beijing, followed by Jimmy Carter’s decision to switch formal diplomatic relations from the ROC to the PRC, further threatened political stability in Taiwan.32 The transition to a democratic state was a priority for the KMT regime as it tried to legitimize its continued rule over Taiwanese subjects. The death of the military dictator and ROC president Chiang Kai-shek in 1975 and the subsequent political rise of his son Chiang Ching-kuo allowed for a gradual transition from authoritarian rule to liberal democracy. As political scientist Thomas Gold put it, Chiang Ching-kuo initiated “enlightened policies aimed at uniting the populace” and sought to transform his own image as the feared former director of the secret police to a “warm and concerned man of the people.”33 The 1975 public opinion poll on urban pollution, published only eight months after the elder Chiang’s death, helped to usher in a new era of KMT governance.34 Given Taiwan’s growing diplomatic isolation, the development of a noise control apparatus became a strategy to guard against political uncertainty, both domestically and internationally.35
198 Managing Sound, Assessing Space
Noise Research in Taiwan: A Variable Agenda Following the formal announcement of a government-backed noise control research commission under the Ministry of Public Health, Wang and his protégés initiated a range of scientific and public relations projects on noise. Starting in 1976, the commission outlined a long-term plan to counsel and educate local citizens on the problem of noise.36 Known as the Preliminary Noise Control Project, the plan outlined a multiyear program to investigate urban noise problems and develop materials for public outreach. One concrete goal was to create new noise regulations and noise control standards specific to Taiwan, starting with Taipei. Reflecting the national and political stakes through which noise control emerged in Taiwan, the representatives present at the planning meeting spanned local and national agencies including the Ministry of the Interior’s National Police Agency, the Taipei Bureau of Education, the Taipei Police and Transportation Department, and the Ministry of Public Health’s Department of Environmental Protection.37 Joining the government officials was Huang Chyan-chyuan, who attended the meeting as the sole academic representative on behalf of Wang Lao- teh and the National Taiwan Normal University’s Health Education Department. During the meeting, the commission approved two major objectives for noise control: educational outreach and field measurements in select locations throughout Taipei. One of the most publicly visible outcomes of the Preliminary Noise Control Project was the installation of a street-side noise monitoring tower (Figure 7.1), a high-profile attempt by the government to calibrate public awareness of noise with internationally recognized decibel standards for environmental noise levels. Towering over a busy street near Taipei Main Station in 1978, an illuminated monitor resembling an athletics scoreboard displayed real-time decibel levels to pedestrians and motorists. The displays were color-coded green (below 66 dB), yellow (66 to 85 dB), and red (above 85 dB), to inform passersby whenever they were being exposed to dangerous or unacceptable levels of noise.38 The educational component of the noise tower was evident in news articles that explained to the public both the meaning of the illuminated sign and the problem of noise. In one such piece, the reporter addresses the public in a playful—yet paternalistic—tone: “Boys and girls, take a look at this photograph. Do you know what this is?” After explaining how the noise tower worked and the color-coded indicators, the author describes the social and physical problems of noise:
To Hear as I Do 199
Figure 7.1 Undated photograph (ca. 1978) of the noise-monitoring tower at Taipei Main Station. Courtesy of Ring-In Trading Development Co., Ltd., Taipei, Taiwan.
The dangers of noise are not the same as the dangers of a flood, a fire, a car wreck, or other things that are visible. Usually those who suffer from noise will find that their hearing comes and goes. They find that they cannot concentrate fully when reading books. They become nervous, annoyed, and generate feelings of insecurity. They might be more prone to losing their temper and are unable to rest peacefully. They may even suffer from high blood pressure. All of these effects will reduce an individual’s well-being.39
200 Managing Sound, Assessing Space
Critics of the noise monitoring tower questioned its effectiveness but nonetheless accepted that the numerical values communicated useful information about harmful noise.40 By visually disseminating decibels across a busy public space, the tower extracted the personal and idiosyncratic experience of hearing noise into a shared and standardizable experience. The Preliminary Noise Control Project of 1976 continued into the 1980s with a series of empirical tests. Through a comparison of collected data, including international research reports, animal testing, site-specific decibel measurements, and socioacoustic surveys, government research attempted to identify standards for noise that would be appropriate in a local, Taiwanese context.41 This process involved transposing information into quantitative data that could be compared, maneuvered, and molded into different configurations. For instance, researchers would compare a recommended set of noise level standards with site-specific noise levels in Taiwan to determine whether the recommended noise standards were feasible. Then, researchers made fine-tuned adjustments to arrive at a locally suitable noise standard that could also reflect the perceptual expectations of residents. Testing led to more testing; efforts to verify and confirm test results in one modality required testing in other modalities. Lacking a local cultural threshold for urban noise, researchers initially compared site- specific noise measurements to Japanese noise standards (Figure 7.2). In 1980, noise measurements taken over a period of 24 hours were collected using a Rion NA-20 analog sound pressure meter and Rion LR-04 printer in specified locations around Taipei. The tests found that Taipei residents were exposed to high levels of noise both day and night, and that they were living in environments ranging from 60 to 76 dB(A).42 Follow-up
Figure 7.2 Comparison of Taipei noise measurements with Japanese noise control standards in dB(A). Translated by the author from Huang Chyan-chyuan, “Problem of Urban Noise in Taiwan,” 16.
To Hear as I Do 201
research continued to compare the results of field measurements to Japanese noise standards. The results provided researchers with concrete evidence that decibel levels in Taipei far surpassed noise standards in Japan, by as much as 20 dB(A). Ultimately, the results supported the necessity of additional research into noise prevention and abatement in Taiwan.43 In contrast to Wang’s earlier studies on schoolchildren, which had examined the range of biomedical causes of hearing loss, much of the research sponsored by the government in the 1980s examined the effects of noise exposure for the purposes of creating nationwide noise level standards. In 1983, a study was conducted on laboratory test rats to investigate the health effects of noise exposure at different decibel levels. Citing the lack of adequate international research on the topic, the study aimed to collect information on suitable decibel levels for human beings. Using a Rion SF-05 noise generator, Wang and his students separated laboratory rats into six different cages for three months and exposed the cages to different noise levels, ranging in increments of ten from 40 dB(A) to 90 dB(A). After monitoring the growth rate, fertility rate, and outward behavior of the rats before, during, and after the three-month period, Wang and his students concluded that the rats exposed to 70 dB(A) or more exhibited signs of aggression, had a lower growth rate, and produced premature offspring. The report recommended that humans should not be exposed to decibel levels over 70 dB(A) and that this finding should be taken into consideration when creating nationwide noise control standards.44 Starting in the mid-1980s, a number of additional tests involving human subjects supplemented the data from the controlled, laboratory environment of animal testing. In a 1985 study, An Investigation on the Nuisance of Noise Among Chinese, socioacoustic surveys were conducted on urban residents in 1,000 households in four different suburbs of Taipei to investigate human responses to noise.45 The survey examined the extent to which noise was perceived as a problem by urban residents, to determine whether the perception of noise had any correlation to actual decibel measurements at a location. The study confirmed the laboratory test findings that had identified 70 dB(A) as the threshold for undesirable noise. Data included self-assessments by research subjects and their perception of the sounds of traffic, as well as the decibel levels measured in different neighborhoods over a 24-hour period. The report included a literature review of mainly Japanese and English-language sources that detailed the bodily effects, mental effects, and social and behavioral impact of noise. The survey also followed many of the same conventions as those published by British researchers F. J. Langdon and I. D. Griffiths in the Journal of Sound and Vibration in the 1960s and 1970s.46 Biographical information such as age, gender, educational background, profession, and socioeconomic
202 Managing Sound, Assessing Space
status was correlated with individual perceptions of noise based on temperament, physical and mental discomfort, and behavior. Through a combination of laboratory and on-site testing, local knowledge about noise was converted into quantitative data to be assessed and evaluated alongside existing domestic and international research on noise and acoustics. By analyzing international research data, applying it to the Taiwanese context, and conducting original research, Taiwanese research stayed in dialog with global developments in noise abatement. Wang’s studies departed from international prototypes by including questions on citizens’ attitudes to noise regulation. For instance, the 1985 survey asked whether or not citizens supported the noise regulations and whether or not they believed the regulations were effective.47 Conducted in the years before and after the official end of martial law in 1987, these studies are a reminder that noise control legislation was carried out alongside the state’s transition from authoritarian to postauthoritarian rule. Alluding to the political significance of the socioacoustic study, the 1985 report reiterated that although international research had already measured various human responses to different noise levels, it was important for a modern society not to overlook the people’s opinions. In other words, it was not enough simply to create a noise standard; the standard needed to be tested and evaluated in relation to public expectations of noise levels as well as to on-site noise measurements of specific locations. Noise was not only an object capable of being measured, recorded, and documented but also a product of individual perception. Underlining the dialectical process between the state and the public in the creation of both democratic governance and noise management, Wang’s 1985 report stated: “People’s attitudes to noise will inform state efforts in noise control.”48 As Sebastian Klotz argues in his contribution to this volume, tests involve an underlying reciprocity between test subject and test giver. The tests conducted by state-sponsored researchers were, on the one hand, an examination of subjects’ noise tolerance levels; on the other hand, they assessed public support for government efforts to create a new domain of governance. Also in the 1980s, the TEPA officially announced a set of noise standards that were provisionally borrowed from existing standards in Japan. When Taiwanese researchers conducted 24-hour noise measurements in 160 different locations in four urban planning zones in Taipei, they found that certain zones exceeded these noise standards by as much as 90 percent. To create a set of policies that reflected existing noise conditions, some areas of Taipei were recommended for inclusion in a separate noise control zone with more relaxed standards.49 In a report in the 1990s, city planners similarly adjusted
To Hear as I Do 203
the noise control standards by dividing areas of Taipei into different segments to accommodate differences in existing noise levels. By adjusting and shifting the noise standards in a patchwork-like fashion, urban planners made noise control zones into malleable, moveable parts that could be overlaid in different ways across the city. In some zones, the noise level standards increased by 5 to 15 dB(A) to create standards that were more feasible given the actual noise conditions of a specific area.50 As the TEPA worked to develop noise control regulations for all of Taiwan, Wang remained committed to promoting hearing health among schoolchildren. He incorporated the problem of noise into his research on children’s hearing loss while maintaining his own interest in childhood development. With support from the Ministry of Science and Technology in Taiwan, Wang published a book of materials to instruct children about paying attention to everyday sounds and stressing the importance of caring for the ears.51 For Wang, hearing was a basic necessity for learning, and he attributed students’ poor performance in schools to problems of hearing that could be treated with hearing aids or better seating arrangements. Wang’s sensitivity to childhood development stands out in his efforts to ensure the hearing health of children. In addition to avoiding loud sounds, Wang counseled, children could prevent hearing loss by following guidelines including eating a balanced diet, not placing objects into the ear canal, and not leaving water in the ear. In Wang’s research, hearing health remained multifaceted, with noise exposure just one of many risk factors. Wang’s broader concern with students’ abilities to learn in the classroom was also addressed by the TEPA, albeit with a specific focus on the object of noise. Using research similar to that on noise control standards, the TEPA took on-site measurements of noise levels in classrooms, tested the effects of noise on students’ academic performance, and carried out socioacoustic surveys on students’ perceptions of noise. In one project, officials administered verbal and arithmetic quizzes to test children’s academic performance at noise levels of 60 decibels and 75 decibels.52 They compared the results to students’ self-evaluations of their performance and concluded that students’ negative perceptions of noise did not lead to a negative effect in academic performance. In another report, researchers gauged the effects of aircraft noise in schools under flight paths by surveying students and conducting audiometric testing.53 The report determined that although students had a negative perception of the effects of aircraft noise, their hearing abilities did not differ from those of students who did not live under flight paths. Much like the other tests conducted by the TEPA, tests on the effects of noise on schoolchildren also tested the
204 Managing Sound, Assessing Space
reliability of individual claims about noise. The state’s effort to listen to the needs of the people was therefore conditioned upon the production of a state-wide, standardized set of criteria for noise. By subjecting students to behavioral tests, audiometric testing, and survey questionnaires, the TEPA produced noise as an object predicated upon human test subjects while actually circumventing human responses.
Conditional Hearing: Technoscience in Taiwan In this chapter, I have discussed how the emergence of noise as a problem in Taiwan was constructed through the combined trajectories of audiometric testing of schoolchildren, Cold War geopolitics, the transition to postauthoritarian rule, the application of international research on noise control, and the state’s adoption of a technoscientific method for policymaking. Testing hearing took on a number of different meanings. It enabled a scientific investigation of noise control regulations, and it aimed to bridge the subjective opinions of noise with quantifiable measurement of noise levels. Noise became the ideal test object because it simultaneously represented a quantifiable measurement, a subjective experience, and a direct connection between the state and the voices of the people. As an object that required verification across a number of domains, noise engaged the state in multiple forms of testing, putting the state’s regulatory authority to the test as much as individual citizens. In the case of Taiwan, testing hearing helped to mediate the transformation from authoritarian rule to liberal democracy. Within a framework of testing, noise and hearing noise were co-produced. Not regarded as a problem until the 1970s, noise was brought into being as a regulatory object by the state’s response to public opinion polling, which resulted in its prioritization of noise control research. The act of hearing noise was further solidified through government tests that surveyed residents in Taiwan about their perceptions of noise, and through outreach efforts that sought to instill a sensitivity to the dangers of certain decibel levels. Furthermore, as an object that was already being regulated in other countries around the world, noise signified the efforts of the Taiwanese state to tackle problems of modernity through the adaptation of technological instruments and standardization. In this scenario, testing was significant not only as a means of generating data but also as a means of establishing ways of knowing between test objects,
To Hear as I Do 205
test subjects, and test givers. The testing of people, animals, and the acoustic environment had a performative aspect that demonstrated the state’s efforts to connect with citizens and to justify its legitimacy to both national and international audiences. Through testing, the state created a platform to make its own decision-making process transparent to scrutiny by others, and it also identified a new aural engagement between citizens and the state. Noise abatement techniques transformed cultural understandings of noise and conditioned Taiwanese subjects’ hearing of noise through decibel measurements and statistical analysis. Hearing was not a practice belonging to either the state or its subjects; it was triangulated through the test mechanisms of technological devices and objective standards. Taiwan’s noise management system remains in use today, and it continues to play a central role in the way Taiwanese citizens handle noise problems. At the same time, the rerouting of hearing by the state has had the unanticipated effect of changing social relationships around hearing and noise. Rather than remaining a subjective experience, hearing noise has become the object of investigation under government noise standards and quantifiable decibel measurements. One instance from my fieldwork stands out and shows how individual responses to noise in Taiwan have been transformed by the government apparatus of noise control. During a visit to a noise complainant’s house, I met a woman in her forties who had endured months of sleepless nights. By the time we met, her voice was quivering, and her eyes had trouble focusing. Adjacent to the home she shared with her elderly mother was a 24- hour self-service laundromat. The intermittent starting and stopping of the laundry machines would awaken the woman in the middle of the night and prevent her from falling back asleep. The woman told me that when she first mentioned the problem to the proprietor, his response was dismissive: “If you have a noise problem, go talk to the authorities and make a report!” The proprietor was challenging the woman to take the problem to the authorities and refused to work with her directly on the issue. She explained: “I was just bothered by the noise, so I went to talk to him about it. Who knew that I had to talk to a government office to look into this problem?” When officials arrived to take decibel measurements of the laundry machines, they found that the machines did not violate the noise level standard. Caught between government regulations, decades of testing hearing, and a social system that recognized the state as the adjudicator for noise, the woman and her mother found themselves in a bureaucratic and technological struggle to have others hear as they do.
206 Managing Sound, Assessing Space
Notes 1. Mingyi xiehui [Public Opinion Association], “A Survey on How to Prevent Public Diseases.” All translations are my own unless otherwise attributed. 2. Executive Yuan Environmental Protection Administration, “Effectiveness of control measures,” accessed May 2, 2020, https://www.epa.gov.tw/eng/EAE30B4C436A1731. 3. There are social critics who also discuss the problem of urban noise from the perspective of modernity and civility, beginning in colonial Taiwan and continuing into the postwar era following the arrival of the Kuomintang regime from China. I choose not to focus on these developments in this chapter because, although they may have shaped the way in which noise was broadly understood in Taiwan at the time, they do not directly relate to my focus on scientific testing. 4. The Kuomintang is the official political party of the Chinese Nationalists, which lost the Chinese Civil War to the People’s Republic of China. Following defeat, they fled to Taiwan with the intent to eventually reclaim China. 5. Karin Bijsterveld, Mechanical Sound; Emily Thompson, Soundscape of Modernity. For more on the decibel, see Mara Mills, “Dezibel.” 6. Classification influenced the social act of hearing by turning the quality of hearing into a quantity. Geoffrey Bowker and Susan Leigh Star, Sorting Things Out; Michel Callon and John Law, “On Qualculation.” 7. Ruth Rogaski, Hygienic Modernity; Gail Hershatter, Dangerous Pleasures; Lydia Liu, Clash of Empires; C. J. W.-L. Wee, “Imagining the Fractured East Asian Modern”; Jordan Sand, “Tropical Furniture”; Prasenjit Duara, Sovereignty and Authenticity; Peter Duus, Abacus and the Sword. 8. Hillel Schwartz, Making Noise; Lilian Radovac, “Muting Dissent”; Roland Wittje, “Concepts and Significance”; Leonardo Cardoso, Sound-Politics; Steven Harris, “ ‘I Know All the Secrets of My Neighbors’ ”; Eun-sung Kim, “Sound and the Korean Public”; John M. Picker, “Soundproof Study.” 9. On renao, see David Ackerman and Kristen Walker, “Consumption of Renao”; Ching-wen Hsu, “ ‘Making Streets’ ”; Julie Chu, “Attraction of Numbers”; DJ W. Hatfield, “Heat and Noise.” 10. Referencing the cultural meaning of renao, Zhuang Jing-yuan testified in the 1983 committee review of the Noise Control Act: “Of course, we are a nation of people who love renao; as a result our noise control standards are more relaxed than other countries” (“Xianshi guoren shi hen shou renao de minzhu, yinci women de guanzhi biaozhun bi waiguo gao”). Legislative Yuan, “Committee on the Interior and Committee on Transportation and Communication Second Review of ‘Noise Control Act,’ ” January 6, 1983, 122–23. 11. He Fan, “There Is Too Much Freedom!” 12. Legislative Yuan, “Committee on the Interior and Committee on Transportation and Communication First Review of ‘Noise Control Act,’ ” December 16, 1982, 150. 13. Theodore M. Porter, Trust in Numbers, 90–91; Callon and Law, “On Qualculation.” 14. Legislative Yuan, “Committee on the Interior and Committee on Transportation and Communication First Review of ‘Noise Control Act,’ ” December 16, 1982, 150. 15. For more on systems of accountability that transform qualities into quantities, see Callon and Law, “On Qualculation.”
To Hear as I Do 207 16. On the use of quantification as a government tool to provide the appearance of transparency and legitimize policy decisions, see Sheila Jasanoff, “Transparency in Public Science”; Ian Hacking, Taming of Chance; Marilyn Strathern, “Tyranny of Transparency.” 17. Michael M. J. Fischer, “Anthropological STS in Asia,” 187. 18. Along with his students and mentees, the public health expert Huang Chyan-chyuan and the physicist Kuo Hung-liang, Wang is considered one of the first local Taiwanese experts to utilize a scientific approach to noise control. All three experts were educated under the Japanese colonial system and spent their formative years as students in Japan. 19. Wang Lao-teh and Wu Man-guei, “Taipei College and High School Students’ Hearing Examination Report,” 16. 20. Ibid., 18. 21. Zhu Yu-ming, “Causes of Deafness Among Students.” 22. In the 1970s, the primary focus of Wang and his students was the impact of noise on the general population in urban environments. Before that, however, Wang’s research team conducted research on the effects of noise on factory workers. See Wang Lao-teh, “Effects and Prevention of Factory Noise.” The effects of factory noise on workers received more attention in the 1990s and 2000s, as awareness of labor rights grew. Shu Shiwei, “How Much Do You Know About Noise and Vibration?”; “Zaoyin yanzhong weihai renti jiankang”; Tian Minghui, “Noise Pollution in our Environment.” 23. “Zaoyin: Sharen bujian xie.” References to the invisible dangers of noise were common at this time. Metaphors of noise as a “bloodless killer” and a “shapeless killer” emphasized the unobtrusive and unexpected ways in which noise could cause bodily and psychological harm. 24. Wang Lao-teh, “Educational Problems of High Frequency Hearing Impaired.” 25. Mingyi xiehui, “Survey on How to Prevent Public Diseases,” 19. 26. Ninety-one percent of the 1,500 respondents asserted that they suffered from public health problems. Noise was the main problem reported by 41.46 percent. It was the highest- ranking public problem in the poll, followed by air pollution at 40.60 percent and water pollution at 14.67 percent. Ibid., 18–19. 27. Huang Chyan-chyuan, interview with the author, June 25, 2015. 28. See Xingzheng yuan huanjing baohu shu, Noise Control Chronology. 29. Ibid., 12. 30. Legislative Yuan, “Committee on the Interior and Committee on Transportation and Communication First Review of ‘Noise Control Act,’ ” December 16, 1982, 143. 31. Wu Wang-ji, “New Tendency of the World.” 32. Hung-mao Tien, writing in 1989, portrayed Taiwan since the 1970s as follows: “On Taiwan growing diplomatic isolation has undermined the foundations of the national government’s legitimacy. Those who hold key positions in the government and the national legislatures derive their authority from their claim to represent all the provinces on the Chinese mainland. Since the international community no longer accepts that claim, the rationale for maintaining the ROC’s present state structure has lost a major source of its authority and has become subject to growing popular criticism as home. Under the circumstances, the KMT regime has found it increasingly difficult to justify its present political structure. Hence the ROC’s international status is not merely a foreign relations issue; it also affects the rationale on which the ROC’s political institutions and power are maintained on Taiwan.” Tien, Great Transition, 216.
208 Managing Sound, Assessing Space 33. Thomas B. Gold, State and Society, ix. 34. As Brian Hall notes: “Taiwan’s modernization and democratization processes are unique because they took place relatively rapidly and against the backdrop of the Cold War, a military dictatorship, growing international isolation, and pressure from Communist- controlled mainland China.” Hall, “Modernization,” 140. 35. See Tien, Great Transition; Muto Ichiyo, “Cold War and Post–Cold War Dynamics”; June Teufel Dreyer, “Evolution of Language Policies.” 36. This combination was made materially possible by Wang’s preexisting connections as an otolaryngologist: the Japan-based distributor of audiometers also produced decibel meters, and it was the Rion NA-09, an analog decibel meter, that was used by early government researchers. 37. Xingzheng yuan weisheng shu [Taiwan Ministry of Health], “Zaoyin guanzhi shiyan jihua” [Noise control preliminary trial], March 9, 1976. Archives of the Ministry of Foreign Affairs/Secretariat/Executive Council Meeting [Waijiao bu dang an /Mishu chu / Xingzheng yuan huiyi], Meetings and Proceedings of the Ministries of the Executive Yuan Sessions 314–318 [Xingzheng yuan di sanyisi zhi sanyiba ci gebu hui fu shouzhang huiyi yishi richeng ji yishi lu], Academia Historica, Taipei. 38. An Qing, “Noise! Danger! Do Not Disturb!” 39. Ibid. 40. Chen Rulen, “Noise Annoys, So What?” 41. Initial results of the preliminary project were published in a series of nine articles in a special issue of the health education journal Jiankang jiaoyu (“Zaoyin”). Other studies funded by the project include Wang Jing-mao et al., Taiwan Noise Control Implementation Plan; Wang Lao-teh et al., Effects of Noise on the Development of Animals. Additional publications introduced noise to a general audience, such as Huang Chyan-chyuan, “Taiwan Urban Noise Conditions”; Zhuang Jing-yuan, “Noise: The Annoying Symphony”; Wang Lao-teh, “Talking About the Noise Problem”; Huang Chyan-chyuan, “Problem of Urban Noise in Taiwan.” 42. Wang Jing-mao et al., Taiwan Noise Control Implementation Plan, 67. The abbreviation dB(A) stands for the a-weighted decibel calculation that is designed to approximate human hearing. 43. Huang Chyan-chyuan, “Problem of Urban Noise in Taiwan.” 44. Wang Lao-teh et al., Effects of Noise on the Development of Animals. 45. Wang Lao-teh et al., Investigation on the Nuisance of Noise among Chinese. 46. F. J. Langdon, “Noise Nuisance”; I. D. Griffiths and F. J. Langdon, “Subjective Response to Road Traffic Noise.” Questions such as “Does the noise bother you to the extent that you want to move or quit your job?” assess the extent to which noise is a disturbance in one’s daily life, while the question “During what settings do you feel bothered by noise?”— followed by the choices “During illness; in a conversation; during sleep; during a mood swing; when busy with work”—assesses the influence of personal circumstances in the perception of noise. 47. Wang Lao-teh et al., Investigation on the Nuisance of Noise among Chinese, 75. 48. Ibid., 4. 49. Huang Chyan-chyuan et al., Research and Analysis on Environmental Noise Monitoring, 175. 50. Huang Rong-cun et al., “A Study on the Adaptability of Noise Control Zoning.”
To Hear as I Do 209 51. Wang Lao-teh and Yang Hai-cheng, Earcare. 52. Huang Chyan-chyuan et al., Study of the Effects of Noise. 53. Zhuang Jing-yuan et al., Influences of Aircraft Noise on School-Age Children.
References Ackerman, David, and Kristen Walker. “Consumption of Renao at a Taiwan Night Market.” International Journal of Culture, Tourism and Hospitality Research 6, no. 3 (2012): 209–22. An Qing. “Zaoyin! Weixian! Qing buyao chao!” [Noise! Danger! Do not disturb!]. Min shen bao [The people’s news], April 5, 1978: 8. Bijsterveld, Karin. Mechanical Sound: Technology, Culture, and Public Problems of Noise in the Twentieth Century. Cambridge, MA: MIT Press, 2008. Bowker, Geoffrey C., and Susan Leigh Star. Sorting Things Out: Classification and Its Consequences. Cambridge, MA: MIT Press, 1999. Callon, Michel, and John Law. “On Qualculation, Agency, and Otherness.” Environment and Planning D: Society and Space 23, no. 5 (2005): 717–33. Cardoso, Leonardo. Sound-Politics in São Paulo. Oxford: Oxford University Press, 2019. Chen, Rulen. “Zhaoyin rao ren, you zenyang?” [Noise annoys, so what?]. Zhongguo Luntan [China forum], April 25, 1978: 5. Chu, Julie. “The Attraction of Numbers: Accounting for Ritual Expenditures in Fuzhou, China.” Anthropological Theory 10, nos. 1–2 (2010): 132–42. Dreyer, June Teufel. “The Evolution of Language Policies and National Identity in Taiwan.” In Fighting Words: Language Policy and Ethnic Relations in Asia, edited by Michael E. Brown and Sumit Ganguly, 385–410. Cambridge, MA: MIT Press, 2003. Duara, Prasenjit. Sovereignty and Authenticity: Manchukuo and the East Asian Modern. Lanham, MD: Rowman & Littlefield, 2003. Duus, Peter. The Abacus and the Sword: The Japanese Penetration of Korea, 1895–1910. Berkeley: University of California Press, 1995. Effectiveness of control measures. Executive Yuan Environmental Protection Administration, https://www.epa.gov.tw/eng/EAE30B4C436A1731. Last accessed May 2, 2020. Fischer, Michael M. J. “Anthropological STS in Asia.” Annual Review of Anthropology 45 (2016): 181–98. Gold, Thomas B. State and Society in the Taiwan Miracle. Armonk, NY: M. E. Sharpe, 1986. Griffiths, I. D., and F. J. Langdon. “Subjective Response to Road Traffic Noise.” Journal of Sound and Vibration 8, no. 1 (1968): 16–32. Hacking, Ian. The Taming of Chance. Cambridge: Cambridge University Press, 1990. Hall, Brian. “Modernization and the Social Construction of National Identity: The Case of Taiwanese Identity.” Berkeley Journal of Sociology 47 (2003): 135–69. Harris, Steven E. “‘I Know All the Secrets of My Neighbors’: The Quest for Privacy in the Era of the Separate Apartment.” In Borders of Socialism: Private Spheres of Soviet Russia, edited by Lewis H. Siegelbaum, 171–90. New York: Palgrave Macmillan, 2006. Hatfield, DJ W. “Heat and Noise.” In Taiwanese Pilgrimage to China: Ritual, Complicity, Community, 23–46. New York: Palgrave Macmillan, 2010. He, Fan. “Ziyou tai duole!” [There is too much freedom!]. Lianhebao [United daily news], October 6, 1955: 6. Hershatter, Gail. Dangerous Pleasures: Prostitution and Modernity in Twentieth- Century Shanghai. Berkeley: University of California Press, 1997.
210 Managing Sound, Assessing Space Hsu, Ching-wen. “‘Making Streets’: Planned Space and Unplanned Business in New Kujiang, Taiwan.” City and Society 22, no. 2 (2010): 286–308. Huang, Chyan-chyuan. “Taiwan diqu dushi zaoyin xiankuang ru guanzhi duice” [Taiwan urban noise conditions and control countermeasures]. You shi [Young lion] 371 (1983): 26–30. Huang, Chyan-chyuan. “Taiwan diqu dushi zaoyin wenti yuqi zhanwan” [The problem of urban noise in Taiwan and its future prospects]. Jiankang jiaoyu [Health education] 52 (1983): 16–17. Huang, Chyan-chyuan, Deng Jiaji, Ye Guolian, Lu Shichang, and Dong Zhenyin. Huanjing zaoyin zhen ce yu fenxi yanjiu [Research and analysis on environmental noise monitoring]. Taipei: Taipei Department of Environmental Protection and National Taiwan Normal University Department of Health Education, 1988. Huang, Chyan-chyuan, Shen Shi-hong, and Wu Cong-nen. Zaoyin dui xuetong xuexi de jiben nengli yingxiang zhi shiyan yanjiu [Study of the effects of noise on students’ basic learning abilities]. Taipei: Executive Yuan Ministry of Public Health and National Taiwan Normal University Department of Health Education, 1987. Huang, Rong-cun, Kuo Hung-liang, Sakurai Masako, Huang Chyan-chyuan, Wu Cong-neng, Lai Huide, Yang Xiantong, Chen Changyu, and Chen Yuru. “Taibei xian zaoyin guanzhi fenqu zhi shiqie xing yanjiou” [A study on the adaptability of noise control zoning in Taipei County]. Taipei: Taipei County Government Research Commission and National Taiwan University Department of Psychology, 1993. Ichiyo, Muto. “The Cold War and Post–Cold War Dynamics of Taiwan and East Asia in People’s Security Perspective.” Inter-Asia Cultural Studies 3, no. 1 (2002): 25–37. Jasanoff, Sheila. “Transparency in Public Science: Purposes, Reasons, Limits.” Law and Contemporary Problems 69, no. 109 (2006): 21–45. Kim, Eun-sung. “Sound and the Korean Public: Sonic Citizenship in the Governance of Apartment Floor Noise Conflicts.” Science as Culture 25, no. 4 (2016): 538–59. Langdon, F. J. “Noise Nuisance Caused by Road Traffic in Residential Areas: Part I.” Journal of Sound and Vibration 49, no. 2 (1976): 241–56. Legislative Yuan. “Lifa yuan neizheng jiaotong liang weiyuanhui shencha, ‘Zaoyin guanzhifa caoan’ di yi ci liandu huiyi jilu” [Committee on the Interior and Committee on Transportation and Communication first review of “Noise Control Act”]. Lifa yuan gongbao [Legislative proceedings] 72, no. 23 (December 16, 1982): 143–51. Legislative Yuan. “Lifa yuan neizheng jiaotong liang weiyuanhui shencha, ‘Zaoyin guanzhifa caoan’ di er ci liandu huiyi jilu [Committee on the Interior and Committee on Transportation and Communication second review of “Noise Control Act”]. Lifa yuan gongbao [Legislative proceedings] 72, no. 29 (January 6, 1983): 119–27. Liu, Lydia. The Clash of Empires: The Invention of China in Modern World Making. Cambridge, MA: Harvard University Press, 2004. Mills, Mara. “Dezibel.” In Handbuch Sound. Geschichte—Begriffe—Ansätze, edited by Daniel Morat and Hansjakob Ziemer, 52–56. Stuttgart: J. B. Metzler, 2018. Mingyi xiehui [Public Opinion Association]. “Ruhe fangzhi gonghai wenti mingyi ceyan” [A survey on how to prevent public diseases]. Mingyi zhazhi jikan [Public opinion poll quarterly] 9 (1975): 18–19. Picker, John M. “The Soundproof Study: Victorian Professionals, Work Space, and Urban Noise.” Victorian Studies 42, no. 3 (2000): 427–53. Porter, Theodore M. Trust in Numbers. Princeton, NJ: Princeton University Press, 1996. Radovac, Lilian. “Muting Dissent: New York City’s Sound Device Ordinance and the Liberalization of the Public Sphere.” Radical History Review 121 (2015): 32–50. Rogaski, Ruth. Hygienic Modernity. Berkeley: University of California Press, 2004. Sand, Jordan. “Tropical Furniture and Bodily Comportment in Colonial Asia.” Positions: East Asia Cultures Critique 21, no. 1 (2013): 95–132.
To Hear as I Do 211 Schwartz, Hillel. Making Noise: From Babel to the Big Bang and Beyond. New York: Zone Books, 2011. Shu, Shiwei. “Ning dui zaoyin han zhendong liaojie duo shao” [How much do you know about noise and vibration?]. Taiwan gonglian hui [Taiwan Federation of Trade Unions] 6 (December 31, 1991): 43. Strathern, Marilyn. “The Tyranny of Transparency.” British Educational Research Journal 26, no. 3 (2000): 309–21. Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the Culture of Listening in America, 1900–1933. Cambridge, MA: MIT Press, 2004. Tian, Minghui. “Women huanjingzhong de zaoyin wuran” [Noise pollution in our environment]. Xinshi laogong [New labor] 5 (August 1, 1988): 9. Tien, Hung-mao. The Great Transition: Political and Social Change in the Republic of China. Stanford, CA: Hoover Institution Press, 1989. Wang, Jing-mao, Li Shu-pei, Wang Lao-teh, Zhuang Jing-yuan, Huang Chyan-chyuan, Zhang Bei-li, Jiang Da-xiong, and Shen Qing-fu. Taiwan diqu zaoyin guanzhi shishi zhunbei jihua [Taiwan noise control implementation plan]. Taipei: Executive Yuan Ministry of Public Health and National Taiwan Normal University Department of Health Education, 1980. Wang, Lao-teh. “Gaodu tingli zhangai xuetong zhi jiaoyu wenti” [Educational problems of high frequency hearing impaired]. Jiankang jiaoyu [Health education] 25 (1970): 23. Wang, Lao-teh. “Gongchang zaoyin zhi yingxiang yu fangzhi” [The effects and prevention of factory noise]. Zhongxue gongyi jiaoyu yuekan [Secondary vocational education monthly], July 1979: 10–14. Wang, Lao-teh. “Qian tan zaoyin wenti” [Talking about the noise problem]. Jiankang jiaoyu [Health education] 52 (1983): 11–13. Wang, Lao-teh, and Wu Man-guei. “Taibei shi daxue xueshen, gaozhong shen de gingli diaocha baogao” [Taipei college and high school students’ hearing examination report]. Zhong deng jiaoyu [Secondary education], August 1964: 16–18. Wang, Lao- teh, and Yang Hai- cheng. Er de baojian [Earcare]. Taipei: Youth Literary Company, 1978. Wang, Lao-teh, Zhuang Jing-yuan, Huang Chyan-chyuan, Ye Guolian, and Chen Qiurong. Guoren duei zaoyin yanfan chendu zhi diaocha yanjiou [An investigation on the nuisance of noise among Chinese]. Taipei: Executive Yuan Ministry of Public Health and National Taiwan Normal University Department of Health Education, 1985. Wang, Lao-teh, Zhuang Jing-yuan, Huang Chyan-chyuan, Shen Qingfu, Ye Guolian, and Li Li-zhen. Zaoyin duei dongwu shenzhang fau yingxiang zhi yanjiou [The effects of noise on the development of animals]. Taipei: Executive Yuan Ministry of Public Health and National Taiwan Normal University Department of Health Education, 1983. Wee, C. J. W.-L. “Imagining the Fractured East Asian Modern: Commonality and Difference in Mass-Cultural Production.” Criticism 54, no. 2 (2012): 197–225. Wittje, Roland. “Concepts and Significance of Noise in Acoustics: Before and after the Great War.” Perspectives on Science 24, no. 1 (2016): 7–28. Wu, Wang-ji. “Shijie minyi de xin quxiang” [The new tendency of the world: Public opinion]. Minyi zhazhi rukan [Public opinion poll quarterly] 9 (1975): inside front cover. Xingzheng yuan huanjing baohu shu [Executive Yuan Environmental Protection Administration (TEPA)]. Zaoyin guanzhi jishi [Noise control chronology]. Taipei: Executive Yuan Environmental Protection Administration, 2012. https://www.epa.gov.tw/File/ F7629DB6D7F5BF25. “Zaoyin” [Noise]. Special issue, Jiankang jiaoyu [Health education] 39 (1977). “Zaoyin: Sharen bujian xie” [Noise: The bloodless killer]. Zhongyang ribao [Central daily news], November 16, 1971: 3.
212 Managing Sound, Assessing Space “Zaoyin yanzhong weihai renti jiankang” [The serious effects of noise on human health]. Taiwan gonglian hui [Taiwan Federation of Trade Unions] 4 (August 1, 1991): 14. Zhu, Yu-ming. “Taibei shili manglong xuexiao longshen yuanyin zhi yanjiou” [Causes of deafness among students at the Taipei School for the Blind and Deaf]. Jiankang jiaoyu [Health education] 22 (1968): 9–11. Zhuang, Jing-yuan. “Naoren de yuetuan: Zaoyin” [Noise: The annoying symphony]. In Fei we kongzhong de shashou [Flying killer in the air], 113–33. Taipei: Baike Wenhua, 1983. Zhuang, Jing-yuan, Shen Shi-hong, Wu Cong-neng, and Lin Hui-fang. Dayuan diqu fei hang zaoyin dui fujin xuetong shizhi yingxiang zhi yanjiu pinggu [Influences of aircraft noise on school-age children around the airport at Ta-yan County]. Taipei: Executive Yuan Ministry of Public Health, 1986.
8 Testing Spatial Hearing and the Development of Kunstkopf Technology, 1957–1981 Stefan Krebs
In the 1930s, research groups at Bell Telephone Laboratories in Murray Hill, New Jersey, and the Philips Research Laboratory in Eindhoven, the Netherlands, independently constructed artificial head microphones. Both groups mounted two microphones on the cheeks of a mannequin’s head, roughly where the ears would have been. Capturing sounds with this type of microphone and reproducing them with stereo head receivers promised to allow a “true auditory perspective,” defined by Bell engineers J. C. Steinberg and W. B. Snow in 1934 as sound reproduction “which preserves the spatial relationship of the original sounds.”1 During the winter of 1931–32, Snow and another Bell engineer, Karl Hammer, had developed the necessary full binaural system.2 It consisted of an artificial head (nicknamed Oscar), a set of amplifiers and equalizers, and thirty-two pairs of head receivers that “ideally” reproduced “in a distant listener’s ears . . . exact copies of the sound vibrations that would exist in his ears if he were listening directly.”3 Hammer and Snow used the system for the wire transmission of symphonic music in true auditory perspective. In Eindhoven, researchers Kornelis de Boer, Roelof Vermeulen, and Arend van Urk were initially interested in the study of perceptual questions of human spatial hearing and the development of a binaural hearing aid. First, they mounted two microphones on opposite sides of a sphere 22 cm in diameter, an apparatus they called the Kunstkopf, or “artificial head.”4 The Kunstkopf signals were reproduced through head receivers. The idea for this stationary binaural hearing aid came from the observation that a person could follow a conversation between more than two interlocutors better when listening with two ears.5 For comparison, they also built a mannequin with microphones mounted on the cheeks, but listening tests did not show perceptual differences between the sphere and the mannequin.6 Stefan Krebs, Testing Spatial Hearing and the Development of Kunstkopf Technology, 1957–1981 In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0009.
214 Managing Sound, Assessing Space
Both groups were working in a context of applied research, interested less in understanding human spatial hearing as such than in exploiting particular phenomena of human spatial hearing to construct new stereophonic sound systems or advanced hearing aids. However, they soon realized that listening tests with Kunstkopf systems also raised fundamental questions about human spatial hearing.7 Thus, Hammer and Snow noticed, most listeners reported that when they were not in the same room as Oscar, all sounds came from the rear. Even when a speaker walked around the Kunstkopf in a circle, almost all listeners had the impression that the speaker had made two semicircles behind it. Hammer and Snow called this the “localization phenomenon”—and it was one they could not explain. Unfortunately, they did not have the resources to continue investigating full binaural technology, because for commercial reasons Bell Labs’ research interest soon shifted to three-channel stereophony.8 The Eindhoven researchers, in contrast, turned to the study of human spatial hearing and used the Kunstkopf to investigate directional hearing in a follow-up project. They were especially interested in front/back discrimination, where no interaural time or level differences occur. One potential explanation for this was the filtering effect of the outer ears, causing a coloration of sounds coming from the rear. After some listening tests, de Boer and van Urk—like the Bell researchers—noticed that most test subjects located all sounds captured with the Kunstkopf behind their heads.9 The Dutch study shows how the understanding of spatial hearing and the development of Kunstkopf microphones became inextricably linked, and how scientific interests and industrial applications were entangled in the object of the Kunstkopf. Nevertheless, just as in the case of Bell, commercial considerations soon pushed research in Eindhoven away from full binaural systems and toward loudspeaker reproduction, especially for sound film.10 It was not until the late 1960s that three German research groups independently started to build new artificial heads, instigating a decade and a half of intensive research on binaural technology and human spatial hearing.11 They had different research agendas and experimental setups. For researchers at the Heinrich Hertz Institute Berlin and the Institute of Technical Acoustics at the Technical University Berlin, the Kunstkopf was a technical means to investigate room acoustics; scientists at the Third Physical Institute at the University of Göttingen used a Kunstkopf to study human spatial hearing; and engineers at the Institute of Electrical Communication Engineering at the Rhineland Westphalian Technical University (RWTH) Aachen deployed their system to research binaural hearing aids. Artificial heads captured sounds for listening tests but were also used for electroacoustic measurements. The German researchers conducted localization experiments with music, speech, noises,
The Development of Kunstkopf Technology 215
and pure tones; they measured ear signals and impulse responses, computed autocorrelations and cross-correlations of ear signals and head-related transfer functions. They tested the feasibility and reliability of the Kunstkopf itself, tested spatial hearing with the help of the Kunstkopf, and compared different listening tests with and without the Kunstkopf. Initially, testing with and of the Kunstkopf neatly combined scientific with industrial agendas. From the mid-1970s, the interest of the three German research groups slightly shifted toward the commercial exploitation of Kunstkopf technology, as building a Kunstkopf with pleasing spatial sound characteristics became more important than understanding particular hearing phenomena. However, the epistemic nature of testing made the Kunstkopf a kind of generator that not only facilitated the design of numerous test setups but also raised surprising questions about human spatial hearing and binaural technology.12 This chapter describes how the three German research teams conceptualized and investigated the epistemic and technical aspects of Kunstkopf technology. I show that artificial heads were deployed to render listening tests more reliable, but also that the Kunstkopf itself was tested to construct an equivalence relationship between binaural technology and human spatial hearing.13 Only the latter type of test enabled the researchers to argue that artificial heads could stand in for human listeners. Listening tests and electroacoustic measurements constituted a complex process of argumentation and persuasion in which the Kunstkopf played different roles in the experimental system: as an epistemic object, a technical instrument, a black box replacing human test subjects, and a commercial sound reproduction system. The different artificial heads were also embedded in different local experimental cultures that shaped the choice of tools, materials, and research methods. Finally, I show that Kunstkopf technology contributed to the notion of human spatial hearing as a mathematically describable transfer function—an epistemic reconceptualization that shifted Kunstkopf research from listening to measuring, from perceptual to physical and computational questions. The chapter’s first three sections address research activities in Berlin, Göttingen, and Aachen between 1957 and 1976; the last section then looks at the state of Kunstkopf research around 1980.
Rendering Listening Tests More Reliable, 1967–1973 Around 1967, Georg Plenge, Ralf Kürer, and Henning Wilkens, engineers at the Heinrich Hertz Institute Berlin and the Institute of Technical
216 Managing Sound, Assessing Space
Acoustics at the Technical University Berlin, were looking for new methods to evaluate the acoustic quality of concert halls.14 The project, funded by the German Research Foundation, aimed to identify correlations between electroacoustic measurements and listeners’ subjective ratings.15 Listening tests posed the problem that only one listener at a time could follow a symphonic performance at a particular seat in the house, making test series with several test subjects and for different seats very time-consuming; furthermore, the listener’s “mood” and the orchestral performance could differ from test to test. To overcome these uncertainties and make listening tests more manageable and reliable, Plenge, Kürer, and Wilkens decided to create facsimile recordings of an acoustic event at several places in a concert hall, which could then be played back in a controlled laboratory environment to any number of participants. They applied Jens Blauert’s distinction between “auditory event,” the conscious auditory perception, and “acoustic event,” the corresponding sound event,16 and were concerned to elicit an auditory event that replicated the original acoustic event. They first experimented with four-channel recordings but soon realized that only Kunstkopf recordings could transmit all the spatial characteristics needed for the qualitative evaluation of room acoustics.17 As no artificial head microphones were available on the market, Plenge, Kürer, and Wilkens had to build their own (Figure 8.1). They knew that exactly replicating the outer ears, including pinnae, ear canals, and eardrums, would be crucial to achieving good head-related signals.18 Having no experience in constructing an artificial head themselves, they asked a make-up artist and a plastic surgeon to produce a plaster copy of a human head and make rubber copies of the outer ears of Plenge and Wilkens. For the recording of symphonic music, they chose two high-grade mono studio microphones (Neumann KM83) to be mounted inside the head; smaller probe microphones were dismissed because of their limited dynamic range and lower signal-to- noise ratio. A special coupler was needed to connect 21-mm microphones to 6-mm ear canals, and was also intended to reproduce the eardrum impedance. For the reproduction of Kunstkopf recordings, the researchers used stereo headphones to make the system fully binaural.19 Before the group began to use the Kunstkopf system as a research tool, Wilkens was tasked with conducting a “prospective” test—a test, that is, “to find if the design is feasible” for spatial listening tests, taking place “before the technology is formally released into use.”20 In particular, the Kunstkopf ’s directional characteristic and frequency response had to be compared with that of a human head. The directional characteristic of a human head was usually measured with probe microphones inside the ear canal, close to the eardrum,
The Development of Kunstkopf Technology 217
Figure 8.1 The first Berlin Kunstkopf, mounted on a transport case, ca. 1969. Courtesy of Ralf Kürer.
but Wilkens rejected this method as too complex, uncomfortable, and tiring for the subjects. Instead, he chose a listening test in which subjects had to compare the audibility of differences in the loudness of sound sources from two different directions. This kind of test had previously proven very precise. After correcting the artificial head’s frequency response, Wilkens conducted two experiments on directional hearing, first that of natural hearing, then that of the artificial head system. The two test subjects were Wilkens himself and his supervisor, Plenge. The results were very encouraging: listening through the binaural system permitted good directional localization, and the accuracy of directional judgments was even higher than that in the competing system in Göttingen.21 Plenge, Kürer, and Wilkens concluded that their Kunstkopf system enabled directional hearing in the median plane and natural perception of sound sources’ reverberation, loudness, and distance.22 The successful listening tests having established a robust equivalence relationship between natural and Kunstkopf-mediated listening, the Berlin researchers could now return to their original subject: the quality of
218 Managing Sound, Assessing Space
concert hall acoustics. In his PhD project, Kürer investigated the Berlin Philharmonic auditorium, looking for correlations between impulse response measurements and subjective judgments of acoustic quality.23 He measured impulse responses, then calculated center time for different places in the auditorium. The principal correlation between center time and subjective ratings of acoustic quality, however, was investigated in a Technical University master’s project that used Kunstkopf recordings for the listening tests. There, the artificial head was a research tool, as originally envisioned, whereas Kürer used the Kunstkopf as if it represented a human test person—he integrated it into his impulse response measurements and found significant differences between its interaural signals. More generally, Kürer’s study focused on measurements rather than listening tests.24 The next investigation compared quality judgments for different seats in six concert halls. To record the same orchestral performance from different positions, more artificial heads were required. A second Kunstkopf was built, in roughly the same way as the first one.25 Given that their own listening tests had demonstrated the astonishing sound quality of the Kunstkopf recordings, the researchers asked the Berlin-based microphone company Neumann if it would be interested in commercializing an improved version. In 1971, Neumann created several preproduction models, which were then made available for the concert hall study.26 Plenge, Wilkens, and two assistants were also allowed to accompany the Berlin Philharmonic orchestra with Herbert von Karajan on a concert tour through West Germany. The orchestra agreed to perform three short pieces just before the audience entered, and back in Berlin, Wilkens used three of eight parallel binaural recordings from each concert venue for listening tests.27 All test subjects could now listen to the same performance, from the same seat in the same concert hall, under the same test conditions in the laboratory, allowing more reliable judgments than individual in situ listening tests could offer. To compare listeners’ judgments, Wilkens developed a questionnaire with nineteen contrastive pairs and asked forty test subjects to rate the acoustic quality of sixty different recordings. He concluded that the acoustic quality of particular places in a concert hall could be determined by three factors: perception of “the strength and extension of the sound source,” perception of “the clarity of the total sound,” and judgment of “the total sound in relation to tone color.” Wilkens argued that the first factor was a property of the hall, whereas the second was influenced by acoustic differences between places in the hall.28 The Kunstkopf system not only rendered listening tests more manageable but also raised new questions about human spatial hearing, notably the “in- head localization” (IHL) of sound. Until then, this phenomenon had been
The Development of Kunstkopf Technology 219
associated with sound reproduced using electroacoustic devices, but there were different and contradictory theories regarding loudspeaker reproduction and headphones. In fact, listening tests with the Berlin Kunstkopf system did not show any IHL, and thus cast doubt on the older theories. This finding prompted Georg Plenge to study IHL for his professorial dissertation, aiming to formulate a single comprehensive theory of IHL based on an extensive analysis of existing literature on IHL and some additional listening tests. The Berlin system used open headphones for sound reproduction so that listeners could also hear other sound sources, such as an additional loudspeaker, and Plenge’s experimental setup exploited this effect to study loudspeaker and headphone reproduction simultaneously. His tests with twenty listeners showed “no fundamental difference between the sounds emanating from headphones and those emanating from loudspeakers.”29 Using Atkinson and Shiffrin’s modal model of memory,30 Plenge formulated two propositions to explain IHL: first, listeners need time for adaptation if they perceive two identical ear signals; second, IHL occurs when the listeners’ short-term store does not contain sufficient information about sound field (such as room size), reverberation time, and sound source and location (such as distance).31 Although Plenge had focused on mediated listening, his findings also contributed to understanding human spatial hearing, because he confirmed the interaction of auditory spatial perception and memory. Taken as a whole, the Berlin studies showed that artificial heads were useful in rendering listening tests of room acoustics more controllable and consistent, because Kunstkopf recordings could bring different rooms—in this case concert halls—into the laboratory.32 Before they could start their main study, Plenge, Kürer, and Wilkens first had to test the Kunstkopf itself to substantiate their claim that auditory perception of binaural sound reproduction was commensurable with listening to the original acoustic event. In addition, constructing and testing an artificial head generated new scientific questions and commercial perspectives: the researchers found they could use the Kunstkopf system to investigate certain aspects of human spatial hearing and realized they had constructed a new device for music recordings that enhanced spatial perception of the original sound performance. Cooperation with Neumann facilitated the construction of more and improved artificial heads. And unlike the rather improvised first two heads, the preproduction models had similar specifications, enabling better cross-comparisons. The commercial model, named KU80, was introduced in 1972 and acquired by several research institutes and public broadcasters.33 The presentation of the first binaural radio drama, Demolition, at the International Radio Exhibition in Berlin in 1973 was a huge success, but the commercial exploitation of
220 Managing Sound, Assessing Space
Kunstkopf technology did not really take off in the following years. As I have shown elsewhere, the idea of facsimile or “real” high-fidelity recordings of a musical performance did not suit the new musical culture of multitrack, multimicrophone productions and the related recording practices that had been emerging since the late 1950s.34 The Kunstkopf “only” offered the exact reportage of a musical performance—editing and mixing Kunstkopf signals was very difficult because the spatial characteristic was so easily lost— and contemporary sound engineers wanted to perfect the recording of live performances rather than simply documenting them.35 Only in radio drama and documentaries were some fascinating Kunstkopf experiments carried out between 1973 and the end of the decade.36
Constructing the Average Human Listener, 1968–1978 In Göttingen, Peter Damaske and his doctoral student Bernhard Wagener were interested in synthetic sound fields and investigated interaural differences as a way to better understand and exploit human spatial hearing. They were particularly concerned with localization in the median plane, where no interaural time or level differences occur. Previously, researchers had measured human ear signals for the empirical analysis of the human head’s complex filtering effect, but in 1968, Damaske and Wagener proposed an artificial head system to substitute for tedious in-ear measurements. Like the Berlin group, they started to build their own Kunstkopf from scratch. A concept study was planned, aiming to demonstrate that directional observations of unmediated and mediated listening were alike.37 In his chapter in this volume, Jonathan Sterne describes similar listening tests that help to construct an equivalence relationship between analog and digital sound devices—the technology passes the test if listeners fail to perceive differences, and fails the test if listeners are able to tell the test settings apart. Damaske had to improvise. He acquired the plaster head of a shop- window mannequin from a nearby department store; it was hollow and had a wall thickness of 20 mm. Two 8-mm holes were drilled into the ears, and curved probe tubes were inserted and coupled with Sennheiser MD 321 probe microphones. Test measurements showed that the directional characteristics of the artificial head largely matched existing findings on that of the human head. Only at 8 kHz was the artificial head about 10 dB too sensitive. Damaske and Wagener assumed that the mannequin’s unnatural ear shape was causing this deviance, so they gradually adapted the artificial head’s
The Development of Kunstkopf Technology 221
directional characteristic by carving out the solid pinnae with a rotating rasp. They designed a prospective test with three listening experiments to assess the functioning of the Kunstkopf system, all carried out in the Göttingen institute’s anechoic chamber. First, they asked seven male listeners to localize a human speaker reading text from eleven different positions in the median plane. During preliminary tests, Damaske and Wagener had noticed that localization through the binaural system was more difficult, and they asked only two, experienced listeners to take part in the second experiment. This resembled the first one, but the artificial head replaced the human listener, while the two remaining listeners were located in another room and listened through the Kunstkopf using Telefunken T50 headphones. The last test had a similar setup but examined localization in the horizontal plane. Evaluation of the test results confirmed the preliminary findings that the directional hearing capability of the artificial head was less accurate than unmediated hearing. However, Damaske and Wagener concluded that the Kunstkopf system could, in principle, be used as a feasible sound reproduction device for subjective localization tests.38 In a follow-up study, Damaske analyzed frequency spectra and cross-and autocorrelations of ear signals to identify directional cues, following Joseph Licklider’s triplex theory of auditory perception.39 To justify his experimental setting, Damaske asserted a stronger similarity relationship than before. He claimed not only a sufficient similarity between unmediated and Kunstkopf- mediated listening but also that a Kunstkopf actually represented an “average human head,” so that microphone signals could substitute for human ear signals.40 First, Damaske coupled the artificial head with a Hewlett Packard loudness analyzer to study the loudness density of ear (or rather microphone) signals. As a test signal, he used broadband noise from different directions in the median plane. In a second test, he used a Princeton Applied Research Model 10 correlator to calculate cross-and autocorrelations of ear signals. Finding that correlation functions for real and “phantom” sources in the median plane differed only slightly,41 Damaske hypothesized that human spatial hearing locates phantom sound sources similarly to real sound sources. Finally, he carried out two series of listening tests, asking ten listeners to locate real and phantom sources in the median plane. The results validated his hypothesis.42 Damaske’s study demonstrated the important role Kunstkopf systems could play in the investigation of human spatial hearing, especially in combination with other new tools such as analog and digital computers. However, the strong claim that the Kunstkopf represented an average human head demanded further substantiation, and Volker Mellert was assigned the task of
222 Managing Sound, Assessing Space
improving the Kunstkopf system to this end. He first measured seventeen test subjects’ thresholds of hearing to identify correlated structures among all test subjects, then adjusted the artificial head’s frequency responses to the average threshold of hearing. New pinnae were cast in plastic, and a new diaphragm was built to replicate the human eardrum impedance. The pinnae were copies of the outer ears of a subject whose pinnae “matched the average geometrical size of the ears of 30 persons.”43 Preliminary listening tests with the new system were considered successful.44 The improved Kunstkopf was also used for several binaural recordings, including recordings of a jazz concert and of a 120-voice choir. Damaske, Wagener, and Mellert judged these “very impressive and lively,” even though their artificial head had not been built for music recordings— the probe microphones had an unfavorable signal- to- noise ratio compared to studio microphones, and the initial aim was to enhance understanding of human spatial hearing rather than produce high-fidelity recordings. The improved Kunstkopf, with its exact copies of a test person’s outer ears, was indeed more human-like than the first Göttingen head, and it was also more average than the Berlin model (which arbitrarily used casts of the outer ears of one of the researchers), even if that average was still the average of a rather small sample with a strong gender bias: all the test subjects were men. After the sudden death of Erwin Meyer, director of the Third Physical Institute, Manfred Schroeder of Bell Labs was appointed his successor. With Karl Friedrich Siebrasse and Dieter Gottlob, Schroeder built a second Kunstkopf based on Mellert’s findings and used it to compare acoustic quality in twenty-two European concert halls. Unlike Plenge and Wilkens, who had recorded a live musical performance, Schroeder, Siebrasse, and Gottlob played a reverberation-free magnetic tape recording of the fourth movement of Mozart’s Jupiter Symphony by the English Chamber Orchestra from the stage and recorded it with the artificial head, which, without further scrutiny, they considered a reliable research tool.45 Binaural reproduction was realized using a new loudspeaker system of Damaske and Mellert’s that “enabled rapid switching between different halls.”46 The facsimile character of this loudspeaker system, which unfortunately only worked in an anechoic environment, was described as “acoustic photography.”47 Instead of asking for a detailed judgment of different qualities, as Wilkens had done, Schroeder, Gottlob, and Siebrasse designed a simple preference test and asked twelve listeners which of two halls they preferred. The preference scores for all subjects and halls were accumulated in a matrix showing how many times a hall was preferred by each listener. The researchers tried to find correlations with objective parameters of the halls, such as the volume of the auditorium or
The Development of Kunstkopf Technology 223
time delay between direct sound and first reflection. The acoustic parameters were obtained from impulse response measurements using the artificial head.48 Meanwhile, Mellert continued his work on improving binaural technology, aiming to build a “subjectively relevant standard microphone” for the evaluation of sound fields.49 In 1974, he was appointed professor of applied physics at the new University of Oldenburg, where, with Reinhard Weber, he started to construct a new Kunstkopf. To make it a standard microphone, the artificial head was still based on the notion of an average human head. Mellert and Weber used a large dataset of about four thousand measurements, aggregated from existing literature, to calculate mean values of head size, ear size, and ear shape, but they used the actual ears of a test person whose pinnae were closest to the calculated “average” ears.50 The molds were produced on a computer numerical control machine, which is why the pinnae had a rather abstract cascading structure.51 The Oldenburg approach of using thousands of anthropometric measurements to determine the artificial head’s size and shape had already been applied by the microphone company Knowles Electronics in 1972. There, Mahlon D. Burkhard and his colleagues had used about five thousand measurements from male and female members of the U.S. Air Force to build KEMAR (Knowles’ Electronics Manikin for Acoustic Research). This head-and-torso simulator’s first main application was to measure hearing aids in situ; it did not play an important role in the study of human spatial hearing in the 1970s.52 The Oldenburg head originally used Schoeps studio microphones, and was later also equipped with standard microphones from Brüel & Kjær. Mellert and Weber deployed the new head for research on subjective noise evaluation.53 At least two Oldenburg heads were sold to Delta Acoustic Studio, a music recording studio in Wilster, northern Germany. Delta’s founder, Manfred Schunke, experimented with binaural recordings and, in 1978, convinced Lou Reed to produce his next three albums in Kunstkopf stereo. Delta tried to commercialize the Oldenburg head as the “Delta Head,” though with little success.54 As we have seen, the concept of the “average human listener” became a leitmotif of Kunstkopf research in Göttingen and Oldenburg. The notion of the artificial head as the perfect representation of an average human head permitted the projection that researchers could use a Kunstkopf instead of human test subjects for all kinds of measurements. It is striking that the first Göttingen head, a shop-window mannequin with no proper outer ears, was far from representing a human head. Mellert deployed listening tests and impulse response measurements of human ear signals to construct an “average
224 Managing Sound, Assessing Space
listener” and gradually adjust his Kunstkopf to the average characteristics of a few listeners. The Oldenburg head was ultimately built according to a huge set of measurements, yet the actual pinnae were still copied from a single test subject. Whereas Schroeder, Siebrasse, and Gottlob, like the Berlin researchers before them, used the Kunstkopf as a research tool to render listening tests more practicable, Damaske actually replaced human test subjects with a Kunstkopf system. He used the microphone signals as if they were ear signals, to calculate monaural autocorrelations and interaural cross-correlations. Damaske’s study also shows that Kunstkopf research on human spatial hearing depended on other newly available research instruments as well, such as correlators and loudness analyzers.
Replacing Artificial Heads with Human Heads, 1957–1976 Researchers at the Institute of Electrical Communication Engineering, RWTH Aachen, built their first artificial head as early as 1957. This wooden head was equipped with high-grade studio microphones (Neumann KM53a) mounted on the cheeks. It was constructed for Paul Scherer’s master’s project, which investigated different stereophonic systems: intensity, time of arrival, and mixed stereophony. Scherer used the Kunstkopf for localization experiments in the institute’s new anechoic chamber.55 Scherer’s project and Kunstkopf model can be situated within the strand of applied research on artificial head stereophony at Bell and Philips, where the main interest was not in understanding human spatial hearing, but in exploiting particular phenomena of spatial hearing for the construction of sound reproduction technologies.56 In a later master’s project, the Kunstkopf was equipped with outer ears, significantly improving localization accuracy.57 In 1961, Klaus Wendt, supervisor of the Kunstkopf projects, left RWTH Aachen and Kunstkopf research was discontinued. In 1964, the head of the Aachen institute, Volker Aschoff, published Über das räumliche Hören (On spatial hearing), outlining this future research topic for his acoustics researchers.58 The same year, Jens Blauert joined the institute as head of the small acoustics research team. Using insights from systems theory, he specified the methodological constraints of studying human spatial hearing in communication engineering. In listening tests, the input signal can be described as the physical quantity of an acoustic event, but the output signal is always the individual perception of a human listener. Blauert introduced the term “auditory event” to describe that subjective perception,
The Development of Kunstkopf Technology 225
emphasizing the need for communication engineers to deploy psychophysical methods for analyzing human hearing.59 Blauert’s PhD project investigated sound localization in the median plane. In line with his programmatic reminder of the psychoacoustic nature of human hearing, Blauert combined physical measurements with listening tests. In a first experiment, he asked twenty male listeners to localize sounds with their heads tied to a stand. The loudspeaker setup ensured that stimuli at both ears were identical, and the listeners had to describe the direction (front, rear, or above) from which they perceived the sound sensation. This revealed that in some frequency bands, the judgments front, rear, and above accumulated independently of the direction of incidence. Blauert had discovered “directional bands”—correlations between particular frequency ranges and directional localizations in the median plane.60 Blauert then used probe microphones at the entrance of the ear canal of several test subjects to measure sound pressure levels for identical stimuli from front, rear, and above. Finally, an experiment was conducted to verify the hypothesis of timbre difference. This time, Blauert used a probe microphone at the entrance to one ear canal of ten observers to make recordings of pink noise coming from a loudspeaker in the front, and a second recording with a loudspeaker in the rear. When the sounds were played back through loudspeakers, all ten observers reported “their sound sensation to be in front, when that signal was reproduced that was originally recorded during sound exposure from the front,” and likewise for signals from the rear.61 Blauert’s experimental dispositif differed from that of the researchers in Berlin and Göttingen. He used readily available technology, small standard microphones with tiny probe tubes, and designed a device to fix a human test subject so that the probe could be safely inserted into the ear (Figure 8.2); the groups in Berlin and Göttingen had rejected this experimental apparatus as too time-consuming and uncomfortable. During his third experiment, Blauert used test subjects as a kind of human Kunstkopf as he recorded inside their ears—a method that was later adopted by his colleagues, as we will see. In 1974, Blauert was appointed professor of communication acoustics at the Ruhr University Bochum, where he continued to work on Kunstkopf technology,62 and published Räumliches Hören (Spatial hearing), a comprehensive survey of work on theories of human spatial hearing. An enlarged and revised English edition appeared in 1983, and two years later, he published a German postscript describing recent findings and new trends. Artificial head technology was barely mentioned in the first publication but was prominently discussed in the postscript.63 Still in Aachen, Blauert and his colleagues Peter Laws and Hans-Joachim Platte refined the probe microphone method for measurements of impulse
226 Managing Sound, Assessing Space
Figure 8.2 Measuring ear signals with a probe microphone, ca. 1972. Courtesy of Peter Laws.
responses. Around 1970 or 1971, a new measuring device was built using a Brüel & Kjær probe kit, allowing relatively easy and reliable measurements about 4 mm inside the ear canal. Laws used this to investigate distance hearing and the IHL phenomenon. The experimental system again combined probe microphone measurements with listening tests. Twenty test subjects were asked to indicate the perceived distance of an acoustic event, their ear signals being measured simultaneously. Different sounds (white noise, music, and speech) were reproduced by a loudspeaker at a distance first of 3 m, then of 0.25 m, and finally through headphones. The results showed that perceived distances were identical when sound pressure signals at eardrum level were identical, even if the sound event came from two different reproduction systems.64 The impulse measuring technique required only a very short detection time and thus lent itself to serial testing. Its disadvantage was that the Fourier transformation between test signal and microphone input had to be computed; accordingly, it was not common in electroacoustics until digital computers and analog-to-digital converters became widely available. Like the Göttingen Kunstkopf, the Aachen measuring device was part of a larger experimental apparatus that rapidly expanded when new digital tools emerged.65 Although limited computing power still made calculating Fourier
The Development of Kunstkopf Technology 227
transformations a very onerous task, Blauert, Laws, and Platte described their experimental setup in a 1974 article as a robust technique to “easily” obtain “outer ear transfer functions,” also known as “head-related transfer functions,” or HRTFs. In the subsequent decades, with the increasing performance and decreasing cost of digital computers, HRTFs became standard representations of the filtering effects of head, pinnae, and torso. This reconceptualization of spatial hearing as mathematical function also implied a shift from listening tests to measurements. Starting in 1974, Laws and Platte built on the improved probe microphone device to construct a binaural apparatus for the exact replication of ear signals. Initially, this was used for measuring ear signals of hearing-impaired people.66 The new microphone array, an assemblage of standard microphone technology and custom-built mount, was presented at the 1976 meeting of the German Acoustical Society. It was a kind of duplication of the probe microphone device, and Laws and Platte used it to verify the hypothesis that similar sound pressure signals at eardrum level stimulated similar auditory events. Problems with the commercial version of the Berlin Kunstkopf, Neumann’s KU80—especially the lack of frontal localization—had raised doubts as to whether facsimile sound recordings were possible at all. Laws and Platte’s experiment contested the Berlin and Göttingen rhetoric of Kunstkopf technology as an equivalent replication of human spatial hearing. Their apparatus avoided the problems of replicating the exact geometry of the human head and outer ears, because an “average” test subject could wear the device on his or her head and become a kind of natural Kunstkopf—the Originalkopf, or “original head.”67 Preliminary listening tests with a number of “inexperienced” listeners revealed that the new apparatus enabled unambiguous and distinct frontal localizations.68 Laws and Platte emphasized inexperience because other Kunstkopf systems had been tested only by experienced listeners or even the researchers themselves. Wilkens and Plenge, for example, had been the only test subjects to construct the similarity relationship between the Berlin head and unmediated spatial hearing. This was all the more problematic because they were using a head equipped with copies of Wilkens’s pinnae—he literally heard with his own ears. A series of three RWTH experiments aimed to test the binaural quality of the “original head” device. First, eight test subjects, seated in the middle of a twelve-loudspeaker array in a darkened anechoic chamber, were asked to judge directions of sixty short examples of human speech reproduced in random order through one of the speakers. Second, a test subject wearing the new device was seated in the same setting and the sixty samples were tape- recorded; the eight subjects from test one then listened to these recordings
228 Managing Sound, Assessing Space
with headphones and again had to judge sound directions. The third test resembled the second, but instead of the human Originalkopf, the Neumann Kunstkopf was used for the recordings. It transpired that listening through the original-head system allowed almost the same localization accuracy as unmediated listening, demonstrating that facsimile recordings were indeed possible.69 Laws and Platte had constructed a new rhetorical relationship between physical copies of ear signals and identical auditory events; they also reified the more general assumption that Kunstkopf hearing and human spatial hearing were equivalent. In 1977, the institute was awarded a German Research Foundation grant to continue using the Originalkopf device, with a new, applied focus on binaural hearing aids.70 The same year, the original head was also used in the production of a radio feature on the relationship between binaural technology and spatial hearing and, in 1978, of the binaural radio drama Dignis Akritis oder Der zwiefach gezeugte Held. The human microphone was not, however, appropriate for practical use outside the laboratory.71 Instead, Laws and Platte started to construct an artificial head. Head and pinnae were cast from a person whose individual monaural transfer functions resembled the “average” transfer functions of many test subjects—this average person happening, coincidentally, to be one of the researchers.72 The first RWTH Kunstkopf model was made of plaster, with Menzolit ears.73 Around 1982, Klaus Genuit developed an improved version for a research project funded by car manufacturer Mercedes Benz, and commercialized it four years later as the “HEAD Measurement System I,” or “Aachen Head.”74 Genuit’s aim was to provide a binaural standard microphone for acoustic measurements, in particular for industrial research and development. Characteristics of an improved Kunstkopf model, designed around 1989 with simplified head and ear geometry, were later adopted in the International Telecommunication Union (ITU) test standards. For example, ITU-T recommendation P.58, first introduced in 1993, standardized the size and characteristics of head and torso simulators for electroacoustic measurements of telephone sets, headsets, and hands-free telecommunication devices.75 The acoustics group in Aachen thus deployed a different experimental dispositif to study human spatial hearing. In his path-breaking PhD dissertation, Blauert combined subjective localization tests with impulse response measurements and identified the crucial role played by certain directional bands for sound localization in the median plane. He used a small-probe microphone for in-ear measurements, which became a standard technique in the Aachen laboratory. Blauert, Laws, and Platte revised the probe microphone device and integrated analog-digital converters and digital computers into
The Development of Kunstkopf Technology 229
their experimental apparatus. Now they were able to calculate head-related transfer functions that mathematically represented the filtering function of head and outer ear. Distinguishing between acoustic events and auditory events, Blauert had emphasized the importance of psychoacoustic methods in the study of human hearing, but his colleagues Laws and Platte returned to a solely physical model of binaural sound reproduction. Their applied research background made them more interested in building a functioning technology, a working binaural hearing aid, than in understanding the mechanism behind spatial hearing. In their Originalkopf experiments, they aimed for exact replications of ear signals to elicit identical auditory perceptions, a technique that achieved very good practical results but did not enhance knowledge of human spatial hearing. When the Originalkopf was used to test the similarity relationship between Kunstkopf and spatial hearing, the results revealed deficiencies of existing artificial head systems but confirmed the basic equivalence of Kunstkopf and human ear signals.
The State of Kunstkopf Research, 1978–1981 Around 1980, Kunstkopf research in Germany came to a temporary stop as the initial groups in Berlin, Göttingen, and Aachen shifted their scientific focus. By this time, the research had reached a mature enough state for Klaus Genuit to develop an artificial head that later became a standard microphone for measuring telecommunication devices. In March 1978, researchers from several German and Danish institutes met at a workshop before the annual meeting of the German Acoustical Society to discuss the current state of Kunstkopf research. There, Jens Blauert stated that “head-related stereophony using artificial heads and similar recording devices has become the standard method in scientific research, deployed when a facsimile electroacoustic reproduction is required.”76 In Blauert’s view, artificial heads were now robust research tools that could easily fit into different experimental setups; in some cases they could even substitute for human test subjects, obviating the need for uncomfortable in-ear probe microphone measurements.77 Binaural technology had become an attractive research tool for disciplines from physics and electrical communication engineering to otology and musicology. Neumann, for example, sold early models of the KU80 to the German Association of Hearing Aid Acousticians and the Cologne Institute of Musicology.78 Two other reasons for the flourishing of Kunstkopf research during the 1970s may be mentioned. First, the wider availability of digital computers in research laboratories facilitated the use of Kunstkopf measurements to
230 Managing Sound, Assessing Space
calculate interaural cross-correlations or HRTFs. Advances in computer technology thus played a crucial role in making Kunstkopf microphones into useful probe instruments.79 Second, the Kunstkopf itself turned out to be a flexible technical instrument as well as an interesting epistemic object. Large parts of the experimental apparatus could be used either for studies of room acoustics, with the Kunstkopf acting as a recording device, or for localization experiments in the laboratory aiming to understand binaural sound reproduction. Problems that occurred during binaural listening tests, such as the inversion of directions or elevation of auditory events, generated new research questions about aspects of human spatial hearing, and further experiments then led to explanations that helped to improve binaural technology.80 However, the circulation of knowledge was hampered by a lack of standardization. All the work invested in building and fostering the similarity relationship between artificial heads and spatial hearing did not compensate for the lack of a binaural standard.81 The divergent microphones, head and pinnae models, and headphones prevented direct comparisons of results. The three research teams in Berlin, Göttingen, and Aachen alone had, between 1957 and 1978, built eight different artificial heads and even more artificial pinnae. They used different microphone models—from Neumann, Schoeps, and Brüel & Kjær—and, more importantly, microphone types (studio and probe microphones). Furthermore, much of the auxiliary equipment was custom built, from mounts to electrical equalizers. Neumann’s KU80 model was available from 1972, but researchers using rival artificial heads did not accept the KU80 as a standard Kunstkopf, although some institutes immediately acquired one. Researchers organized exchanges and comparisons of different Kunstkopf systems. The Institute for Broadcasting Technology (IRT) in Munich, for example, made comparative measurements of Neumann’s KU80 and Damaske’s artificial head, and of Neumann’s KU81 and the Aachen head, while the German public broadcaster WDR in Cologne invited scientists and engineers to a comparative listening test of Neumann’s KU81 and the Aachen head.82 But despite these forays, in 1978 the lack of standards for artificial heads and headphones was still a major obstacle for future research.83 Another problem was the diversity of test scenarios. Distances between sound source, Kunstkopf, and test subject differed significantly; sounds were reproduced using loudspeakers or headphones. Researchers utilized a wide range of test signals: for objective measurements they used noises and pure tones; for localization tests they used noises, pure tones, and live or recorded human speech; and for judgments of room acoustics they reproduced binaural recordings of live performances or recorded music. Altogether, test setups were deeply rooted in divergent experimental cultures and local laboratory
The Development of Kunstkopf Technology 231
contexts. As a result, listening tests and measurements often turned out to be incommensurable.84 Moreover, many experiments investigated isolated subphenomena of spatial hearing. In 1981, Günther Theile, who headed the department of sound recording and reproduction at the IRT in Munich, fiercely criticized the state of Kunstkopf research and theory building: “What is the point of a model of directional hearing for room-related stereophony if it is only valid for the free sound field, the horizontal plane, not for broadband loudspeaker signals and not for distance hearing?” Theile emphasized that direction and distance are inseparable spatial coordinates of one and the same auditory event: “A functional model of directional hearing is not a model of auditory perception; it does not necessarily describe a specific function of the human ear in the case of spatial hearing.” Instead, Theile advocated a holistic approach and emphasized the crucial role of auditory memory. Based on recent findings in neurophysiology and perceptual research, he formulated a new theory of sound localization, the “association model,” which consisted of two consecutive decoding processes: spatial information is analyzed and then, in a second, higher stage, the gestalt information of an acoustic event is decoded, both processes being interdependent and based on associative pattern recognition. In Theile’s view, “there is no location of an auditory event without an associated auditory gestalt.”85 His association model of human spatial hearing questioned not only the use of artificial test signals such as pure tones but also theory construction in communication engineering more generally. Theile argued that understanding human spatial hearing was a fundamentally interdisciplinary endeavor and that without greater understanding of the human brain, all communication models of sound localization would remain inadequate.86
Conclusion Between 1967 and the early 1980s, artificial heads were popular research tools and epistemic objects in scientific disciplines including physics, technical acoustics, and electrical communication engineering— particularly in Germany.87 Kunstkopf systems were deployed in different research areas, whether room acoustics, human spatial hearing, binaural hearing aids, or full binaural music reproduction. For the study of room acoustics, Kunstkopf technology promised to bring the concert hall into the laboratory. Plenge, Wilkens, and Kürer also realized that full binaural technology had the potential to fulfill the old high-fidelity quest of bringing the concert hall into the
232 Managing Sound, Assessing Space
living room, or, as they put it, of transferring the domestic listener into the concert hall. However, the commercialization of binaural recordings failed— the notion of facsimile sound reproduction had not kept up with changing musical preferences. Initially, no adequate artificial heads were on hand, so researchers had to build their own. Grounded in the tinkering culture of electrical engineering and applied physics, the different Kunstkopf models were co-created through the tools, materials, experimental cultures, and local networks available in each setting.88 All three groups took it for granted that artificial heads were replicas of male human heads—indeed, in a classic case of engineers modeling the world from their own reality, in two instances the artificial heads were even equipped with the researchers’ own pinnae. As a consequence, head-related recordings were optimized for male listeners rather than for generally smaller female heads and ears. There were other limitations to the Kunstkopf experiments. Financial and organizational constraints, as well as a lack of experience in psychoacoustic testing, restricted experimental designs to small samples of human listeners. The problematic choice of test persons—male students and colleagues— further reified the gendered approach to studying binaural technology and spatial hearing.89 Additionally, the researchers’ understanding of human spatial hearing was highly circumscribed by their experimental setups, since spatial hearing tests scrutinized fragmented and isolated subphenomena rather than real spatial hearing, as Günther Theile pointed out. All three groups argued that artificial heads represented an average listener—investing extensive work to calculate mean head-related transfer functions, for example—but the underlying measurements were still taken from very few test subjects.90 The assumed equivalence relationship between binaural technology and human spatial hearing thus rested on a small and biased data corpus yet was used as a rhetorical device to justify the substitution of human listeners with artificial head microphones. Nevertheless, the basic assumption enabled a flexible alignment of artificial head and human ear measurements, and thus facilitated experimentation by replacing tedious in-ear measurements with easily obtained Kunstkopf signals. The different test setups and techniques helped researchers to oscillate between more applied and more basic questions, between constructing binaural technology and understanding spatial hearing. Sometimes, testing was even the point of departure for an epistemic shift, such as when the prospective test of the Berlin head triggered questions about IHL. From the mid- 1970s, however, the technical improvement of head-related stereophony became more important than the investigation of human spatial hearing as
The Development of Kunstkopf Technology 233
all three groups moved toward more applied research.91 Problems with existing artificial heads, the difficulty of setting up large-scale psychoacoustic experiments, a general preference for the “physicalist method,” and commercial interests all contributed to the shift from figuring out spatial hearing to exactly replicating ear signals, as the latter technique promised the construction of a working binaural system.92 The increased interest in the physicalist method also tied in with more general advances in computer technology that paved the way for a new way of thinking about spatial hearing: head- related transfer functions—the mathematical description of the filtering effect of head and ears—now represented human spatial hearing and replaced psychoacoustic explanations. Within this new epistemic framework, testing focused on in-ear and Kunstkopf measurements instead of psychoacoustic listening tests.93 The Kunstkopf proved to be a very flexible object, and the three research groups deployed artificial heads in different research areas. In Berlin, Plenge, Kürer, and Wilkens initially studied room acoustics but later investigated the Kunstkopf as music recording technology. In Göttingen, Damaske and Mellert were at first interested in human spatial hearing, but later Siebrasse and Gottlob turned to room acoustics; after his move to Oldenburg, Mellert used the Kunstkopf for noise evaluations. In Aachen, Blauert scrutinized specific phenomena of human spatial hearing; Laws and Platte were initially interested in developing binaural hearing aids and later in more generic binaural sound reproduction technology. These differing research interests were reflected in different testing cultures: researchers in Berlin tried to improve the reliability of listening tests; testing in Göttingen aimed to construct the average human listener in order to replace him or her with an average artificial head that could be used for measurements instead of listening tests; in Aachen, Blauert combined in-ear measurements and listening tests to account for the psychoacoustic nature of his research, whereas Laws and Platte replaced the Kunstkopf with an “original head,” aiming for exact replications of ear signals. The common ground between the research groups was the Kunstkopf as an epistemic object and a test device. Testing the Kunstkopf, and testing spatial hearing with the Kunstkopf, served as a bridge between understanding human spatial hearing and constructing a new recording device—a bridge that in turn facilitated the co-development of Kunstkopf technology in science and industry. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/testing-spatial-hearing-krebs.
234 Managing Sound, Assessing Space
Acknowledgments This study was supported by the Luxembourg National Research Fund (FNR) and co-funded by the Marie Curie Actions of the European Commission (FP7-COFUND), project code 9151779. I am grateful to the editors and the participants of the Testing Hearing workshops for helpful comments on earlier versions of this paper, and would like to thank Kate Sturge for the careful language editing of the manuscript.
Notes 1. Steinberg and Snow distinguished “true auditory perspective” from “good auditory perspective.” True auditory perspective required a full binaural system consisting of a dummy- head microphone, two-channel transmission, and stereo headphones; good auditory perspective could be achieved with all kinds of multichannel sound reproduction systems, including three-channel loudspeaker systems. J. C. Steinberg and W. B. Snow, “Physical Factors.” 2. Karl E. Hammer and W. B. Snow, “Memorandum for File, MM-3950,” November 2, 1932, reproduced in Acoustical Society of America, Study of Speech. 3. Steinberg and Snow, “Physical Factors,” 245. Stephan Paul credits F. A. Firestone (University of Michigan) with building the very first mannequin for sound reproduction, between 1928 and 1930. Paul, “Binaural Recording Technology,” 769. 4. The German term was later also used in English. Kornelis de Boer and Roelof Vermeulen, “Eine Anlage für einen Schwerhörigen.” I will use the synonymous terms “artificial head” or “Kunstkopf,” and “binaural” or “head-related stereophony.” The former describe the technical object, the latter the sound reproduction system. 5. De Boer and Vermeulen aimed to exploit the “cocktail-party effect,” the ability of spatial hearing to focus auditory attention on a particular stimulus and filter out other stimuli. 6. De Boer and Vermeulen, “Eine Anlage für einen Schwerhörigen.” See also Kornelis de Boer, “Plastische Klangwiedergabe.” 7. General studies of binaural hearing date back to the late eighteenth century and intensified from the second half of the nineteenth century. For overviews, see Peter Laws, “Untersuchungen zum Entfernungshören”; Georg Plenge, “Über das Problem der intracranialen Ortung.” 8. This shift was driven by sound film. See Harvey Fletcher, “Symposium on Wire Transmission”; Robert E. McGinn, “Stokowski.” Oscar received much attention at the Chicago World’s Fair, but artificial head research was nevertheless discontinued at Bell. Cheryl Ganz, The 1933 Chicago World’s Fair, 78–79. 9. Kornelis de Boer and Arend van Urk, “Einige Einzelheiten beim Richtungshören.” 10. Roelof Vermeulen, “Vergleich zwischen wiedergegebener und echter Musik.” It is important to distinguish full binaural systems (i.e., the use of a Kunstkopf on the recording side and stereo headphones on the reproduction side) from binaural sound reproduction, often a synonym for all kinds of stereophonic sound reproduction. When McGinn describes Fletcher and Stokowski’s interest in binaural sound reproduction, for example, he does not
The Development of Kunstkopf Technology 235 refer to the dummy-head experiments that took place at the same time and also involved Stokowski and the Philadelphia Symphonic Orchestra. McGinn, “Stokowski.” 11. Artificial heads were also used and investigated by groups in Hamburg, Munich (both in West Germany), and Dresden (in East Germany). G. Krumbacher, “Über die Leistungsfähigkeit”; W. Kuhl and R. Plantz, “Kopfbezogene Stereophonie”; Werner Schirmer, “Die Richtcharakteristik des Ohres.” I focus on the three groups because their artificial heads became technical or epistemic objects for other researchers. See, for example, M. Kleiner, “Problems in the Design and Use”; Torben Poulsen, “Hörvergleich unterschiedlicher Kunstkopfsysteme.” 12. On technical instruments as “generators of surprise,” see Hans-Jörg Rheinberger, Toward a History of Epistemic Things, 31. 13. Trevor Pinch, “ ‘Testing—One, Two, Three . . . Testing!,’ ” 29. 14. Georg Plenge names the year 1967 in his professorial thesis of 1973, “Über das Problem der intracranialen Ortung.” However, the earliest activity I could identify was Wilkens’s master’s project in April 1969. Plenge’s statement indicates the conflict in priorities between Berlin and Göttingen. Henning Wilkens, “Wann wurde welcher Kunstkopf verwendet?,” undated memo, Deutsches Museum Munich, Collection Kunstkopf, Folder Wilkens 1. See also author interview with Volker Mellert, March 9, 2016, Oldenburg. 15. For the objective tests, they used impulse response measurements and the analysis of impulse response echograms. Georg Plenge, “Über die Hörbarkeit”; Ralf Kürer, “Untersuchungen zur Auswertung.” 16. Georg Plenge, Ralf Krüger [i.e., Kürer], and Henning Wilkens, “Über die Reproduktion von Hörbildern”; Jens Blauert, “Die Beschreibung von Hörversuchen.” 17. Author interview with Henning Wilkens, November 7, 2013, Munich; author interview with Ralf Kürer, September 24, 2015, Berlin. 18. At the Sixth International Congress on Acoustics (August 1968), Jens Blauert noted that the directional judgment of listeners was highly dependent on the frequency of the test signal, and thus emphasized the importance of the filtering effect of the outer ears. Jens Blauert, “Ein Beitrag zur Theorie”; Blauert, “Untersuchungen zum Richtungshören.” 19. Henning Wilkens, “Kopfbezügliche Stereophonie”; “Betr. Künstlicher Kopf,” memo dated November 10, 1970, Neumann company archives, Folder KU80 Ohrkurven; Draft patent application, May 21, 1969, Deutsches Museum Munich, Collection Kunstkopf, Folder Wilkens 1; Henning Wilkens, “Erinnerungen aus dem beruflichen Leben,” December 6, 2006, private collection Henning Wilkens; interview Wilkens; interview Kürer. 20. Pinch, “ ‘Testing—One, Two, Three . . . Testing!,’ ” 27. 21. A summary of the study was published as Wilkens, “Kopfbezügliche Stereophonie.” On the Göttingen Kunstkopf, see later in this chapter. 22. Plenge, Kürer, and Wilkens, “Über die Reproduktion von Hörbildern.” 23. For comparison, Kürer took measurements in other rooms, including two lecture rooms at the Technical University Berlin. 24. Kürer, “Untersuchungen zur Auswertung”; interview Kürer. 25. Both are in the collection of the Deutsches Museum in Munich. 26. The preproduction models were made of plaster with new pinnae copied from Wilkens’s and Plenge’s ears. Neumann provided materials to improve the coupler between ear canal and microphone, and a special kinked version of the KM83 condenser microphones. “Betr. Künstlichen Kopf,” undated notes, Neumann company archives, Folder KU80 Ohrkurven; “Betr. Künstlichen Kopf,” November 10, 1970, ibid.; interview Wilkens; interview Kürer.
236 Managing Sound, Assessing Space 27. Plenge and Wilkens used four artificial heads and the orchestra played twice so that they could record from eight different places. They decided to use recordings from three spots in each auditorium. 28. All quotations from Henning Wilkens, “Mehrdimensionale Beschreibung,” 103 (here and throughout, all translations are my own unless otherwise attributed). See also Henning Wilkens and B. Kotterba, “Vergleich der Beurteilung.” 29. Georg Plenge, “Über das Problem der Im-Kopf-Lokalisation,” 241. 30. R. C. Atkinson and R. M. Shiffrin, “Human Memory.” 31. Plenge, “Über das Problem der intracranialen Ortung.” 32. The problem of how listeners could objectively describe their subjective experience was not solved; Kunstkopf recordings could only improve listening conditions. 33. “Lieferung von Künstlichen Köpfen,” undated memo, Neumann company archives, Folder KU80 Ohrkurven. On Neumann’s KU80 and its use in radio drama and music productions, see Stefan Krebs, “Failure of Binaural Stereo.” 34. Krebs, “Failure of Binaural Stereo.” On the new recording culture, Susan Schmidt Horning, Chasing Sound. 35. See Nikolaus Schampaul, “Die Übertragung des natürlichen Klangbildes”; Hans-Peter Reinecke, “Das Ideal des naturgetreuen Klangbildes.” 36. Krebs, “Failure of Binaural Stereo.” German public service radio stations produced and broadcast at least seventy-six binaural radio dramas and twenty-five binaural documentaries between 1973 and 1980; they also made about 135 binaural music recordings, very few of which were actually broadcast. See the author’s research blog at https:// binauralrecording.wordpress.com. 37. Peter Damaske and Bernhard Wagener, “Richtungshörversuche”; interview Mellert. 38. They also presented two propositions to improve dummy-head technology and argued that for better results, listeners needed more time to adapt to the system. Damaske and Wagener, “Richtungshörversuche.” 39. Peter Damaske, “Richtungsabhängigkeit.” Licklider’s duplex (1951) and triplex (1956, including binaural perception) theories of pitch and frequency perception were based on cross-and autocorrelations of ear signals; he assumed that the nervous system could compute these functions (delay, multiplication, and averaging) in real time. Joseph Licklider, “Duplex Theory of Pitch Perception”; Licklider, “Auditory Frequency Analysis.” 40. Damaske and Wagener, “Richtungshörversuche”; Peter Damaske and Volker Mellert, “Ein Verfahren.” 41. With a stereophonic loudspeaker setup, sounds can be made to appear between the two loudspeakers, seemingly floating in the air where there is actually no speaker. These sounds are called phantom sources. 42. Damaske, “Richtungsabhängigkeit.” 43. Volker Mellert, “Construction of a Dummy Head.” 44. Peter Damaske, “Head-Related Two-Channel Stereophony.” 45. The decision to use a recording instead of live music was a practical one: Siebrasse and Gottlob had no opportunity to accompany an orchestra on tour. 46. Manfred Schroeder, Dieter Gottlob, and Karl Friedrich Siebrasse, “Comparative Study,” 1195. Damaske and Mellert had developed an alternative reproduction system with two loudspeakers using a 90-degree filter to compensate for the acoustical crosstalk of conventional stereo reproduction. Damaske and Mellert, “Ein Verfahren.”
The Development of Kunstkopf Technology 237 47. Press release, May 22, 1974, private collection Volker Mellert. Plenge, Krüger [Kürer], and Wilkens too had used a visual metaphor—Hörbild (“auditory picture”)—to describe the facsimile character of their Kunstkopf. Plenge, Krüger, and Wilkens, “Über die Reproduktion von Hörbildern.” 48. Schroeder, Gottlob, and Siebrasse, “Comparative Study.” 49. Volker Mellert, “Verbesserte Schallfeldabbildung,” 433. 50. Mellert and Weber assigned one of their students to do a literature survey of existing ear measurements. However, the pinnae were not modeled from existing ear measurements, as Martha Brech claims, but copied from a test subject’s ears. Brech, Der hörbare Raum, 199. 51. Mellert, “Verbesserte Schallfeldabbildung”; Reinhard Weber and Volker Mellert, “Ein Kunstkopf ”; interview Mellert. 52. Paul, “Binaural Recording Technology,” 775. 53. For example, Volker Mellert, “Vergleichende Beurteilung.” 54. Delta advertisement, no date, private collection Volker Mellert; Mario Scheuermann, “Rock-Star mit Kunstkopf: Lou Reed live,” clipping dated August 22, 1978, ibid. 55. Paul Scherer, “Untersuchung”; Scherer, “Über die Ortungsmöglichkeit.” Scherer used a Neumann SM2 microphone, the world’s first stereo condenser microphone, introduced only that year. The Aachen institute’s access to this high-end device was probably facilitated by contacts with Gerhart Boré, a microphone developer at Neumann. Boré had studied electrical engineering in Aachen and worked at the Institute of Electrical Communication Engineering from 1950 to 1956. 56. Bell and Philips were the main points of reference for the Aachen master’s projects of the 1950s and early 1960s. 57. Robert van Kordelaar, “Messungen am künstlichen Kopf.” Unfortunately, the available sources do not describe the construction of this artificial head in detail. 58. Volker Aschoff, Über das räumliche Hören. 59. Jens Blauert, “Zur Methode”; Blauert, “Die Beschreibung von Hörversuchen.” Blauert and Laws studied literature on statistical, psychophysiological, and psychometric methods in 1966 and 1967. Jens Blauert and Peter Laws, “Physiolog. Akustik,” lecture notes, December 12, 1966, private collection Peter Laws; Peter Laws, “Einfache Meß-und Skalierungsmethoden für psychophysiologische Größen,” lecture notes, January 11, 1967, ibid. 60. Jens Blauert, “Sound Localization,” 209. 61. Blauert, “Sound Localization,” 211; see also Blauert, “Untersuchungen zum Richtungshören.” 62. Members of Blauert’s department were involved in improving the Neumann KU80 dummy head, for example developing the outer ears of the new model KU81, released in 1981. Jens Blauert, Herbert Hudde, and Jürgen Schröter, “Kopfbezogene Mikrophonanordnung,” report, May 1980, Deutsches Museum Munich, Collection Kunstkopf, Folder Theile 1. 63. Jens Blauert, Räumliches Hören; Blauert, Spatial Hearing; Blauert, Räumliches Hören: Nachschrift. He published another postscript in 1997. 64. Peter Laws, “Entfernungshören”; Laws, “Untersuchungen zum Entfernungshören.” 65. Jens Blauert, Peter Laws, and Hans-Joachim Platte, “Impulsverfahren,” 35. 66. This can be inferred from two newspaper reports of the time and a 1974 master’s thesis. However, during interviews, Laws and Platte did not remember that the original head was only a by-product of this kind of research. Wolfgang Kempkens, “Aachens Kunstkopf kann auch sprechen,” Aachener Volkszeitung, April 17, 1975, clipping in private collection Hans-Joachim Platte; Gerd Pasch, “Jetzt gibt es den ‘Originalkopf,’ ” Aachener Nachrichten,
238 Managing Sound, Assessing Space March 4, 1978, clipping ibid.; Rolf Thome, “Räumliches Hören”; author interview with Hans-Joachim Platte, March 10, 2016, Nörten-Hardenberg; author interview with Peter Laws, April 7, 2016, Aachen. 67. The apparatus also included a microphone equalizer and a headphone equalizer that compensated for linear distortions of microphone and headphone. Hans-Joachim Platte, Peter Laws, and Harald vom Hövel, “Anordnung zur genauen Reproduktion,” 361. 68. Ibid. 69. Headphones had to be worn in a narrowly defined position for best results. Peter Laws and Hans-Joachim Platte, “Spezielle Experimente.” 70. Hans Dieter Lüke, 50 Jahre, 82. 71. Jens Clasen, “Flach atmen und am besten überhaupt nicht schlucken,” Frankfurter Rundschau, April 7, 1978, clipping in WDR Historical Archives, Folder 13630; Hans- Joachim Platte, “Stand der Kunstkopfstereofonie,” manuscript dated July 25, 1978, ibid. 72. A photo series in Hans-Joachim Platte’s private collection documents the casting of Peter Laws’s head. 73. Peter Laws and Hans-Joachim Platte, “Ein spezielles Konzept,” 28–29. 74. “Kunstkopf-Technologie von HEAD acoustics,” HEADlines, September 1995, private collection Peter Laws. In 1986, Klaus Genuit founded the Aachen-based company HEAD Acoustics, which became an internationally renowned innovator in head-related stereophony. Author interview with Klaus Genuit, July 23, 2013, Aachen. 75. Genuit to Neumann, March 10, 1982, Neumann company archives, Folder KU81 IRT. ITU rules P.57 (artificial ears) and P.58 (head and torso simulator) are based on the characteristics of the HMS II model. Klaus Genuit, “Kunstkopf-Meßsysteme,” 66. 76. Jens Blauert, “Vorbemerkung,” 195. 77. Blauert, Räumliches Hören: Nachschrift, 88. 78. “Lieferung von Künstlichen Köpfen,” undated list, Neumann company archives, Folder KU80 Ohrkurven. Little is known about the use of these artificial heads, but it seems safe to assume that they became boundary objects easing cooperation between scientific disciplines. Dummy-head systems could be integrated into experimental systems yet were robust enough to foster cooperation and the exchange of technical objects. On boundary objects, see Susan Leigh Star, “This Is Not a Boundary Object.” 79. Dieter Gottlob, “Anwendung der kopfbezogenen Stereofonie.” 80. Ibid., 214. See also Hans-Joachim Platte, “Probleme bei Messung”; Peter Laws, “Messung und Nachbildung”; Herbert Hudde, “Methoden zur Bestimmung.” 81. The ITU-T recommendations P.57 and P.58 offered no remedy, as they did not define standards for binaural sound reproduction. 82. See the following items in the Deutsches Museum Munich, Collection Kunstkopf: Measuring sheet, March 12, 1981, Folder Theile 3; Measuring sheet, March 23, 1984, Folder Theile 4; “Verlauf und Ergebnis des Vergleichs verschiedener Kunstköpfe beim WDR,” memo, May 26, 1982, ibid.; Plenge to various recipients, May 27, 1982, ibid. See also Peter Laws, Jens Blauert, and Hans-Joachim Platte, “Anmerkungen”; Poulsen, “Hörvergleich.” 83. On standardization, see Blauert, “Vorbemerkung”; Gottlob, “Anwendung der kopfbezogenen Stereofonie.” 84. In 1964, Volker Aschoff, director of the Aachen Institute of Electrical Communication Engineering, had emphasized the decisive effect of different test signals on spatial auditory perception. Aschoff, Über das räumliche Hören, 13.
The Development of Kunstkopf Technology 239 85. Günther Theile, “Zur Theorie,” 156, 158, 163. 86. Discussion of Theile’s association model is still ongoing. Author interview with Günther Theile, December 17, 2015, Munich; Email Theile to Blauert, July 19, 2007, private collection Günther Theile. 87. Researchers in East Germany also built and investigated Kunstkopf technology in the 1960s and 1970s, but the exchange of tools and ideas was hampered by the Cold War. See Schirmer, “Die Richtcharakteristik des Ohres.” 88. Blauert saw this bricolage culture as one of the main drivers of Kunstkopf research in the 1970s. Blauert, “Vorbemerkung.” 89. Janina Fels, professor of medical acoustics at the Institute of Technical Acoustics, RWTH Aachen, sees the usually small number of test subjects as one of the greatest weaknesses of Kunstkopf research past and present. Unrecorded author interview with Janina Fels and Gottfried Behler, June 1, 2016, Aachen. 90. Blauert’s team in Bochum followed this approach to identify the “typical listener,” then used his ears as a model for the new pinnae in Neumann’s KU81 model. Jens Blauert, Herbert Hudde, and Jürgen Schröter, “Kopfbezogene Mikrophonanordnung,” report, May 1980, Deutsches Museum Munich, Collection Kunstkopf, Folder Theile 1. 91. All filed patents on binaural recording and reproduction. Examples are Ralf Kürer, Georg Plenge, and Henning Wilkens, Verfahren zur räumlichen Wiedergabe von Schallsignalen und Anordnung zu seiner Durchführung, German Patent 20 23 377, filed May 9, 1970, and issued July 19, 1979; Volker Mellert, Verfahren zur richtungsgetreuen breitbandigen Abbildung von Schallfeldern, German Patent Application 25 39 390, filed September 2, 1975; Klaus Genuit, Ein breitbandiger rauscharmer Kunstkopf mit hoher Dynamik und der Eigenschaft der originalgetreuen Übertragung von Hörereignissen, German Patent Application 31 46 706, filed November 25, 1981. 92. The term “physicalist method” is Blauert’s, “Vorbemerkung.” 93. Little work was done on the inner ear or neuronal foundations of spatial hearing in the period; an exception is Volker Mellert, Karl Friedrich Siebrasse, and S. Mehrgardt, “Determination of the Transfer Function.”
References Acoustical Society of America, ed. Study of Speech and Hearing at Bell Telephone Laboratories. CD-ROM compiled by Christine M. Rankovic and Jont B. Allen, 2000. Aschoff, Volker. Über das räumliche Hören. Cologne: Westdeutscher Verlag, 1964. Atkinson, R. C., and R. M. Shiffrin. “Human Memory: A Proposed System and Its Control Processes.” In The Psychology of Learning and Motivation, edited by K. W. Spence and J. T. Spence, 2:89–195. New York: Academic Press, 1968. Blauert, Jens. “Ein Beitrag zur Theorie des Vorwaerts-Rueckwaerts-Eindrucks beim Hoeren.” In Reports of the 6th International Congress on Acoustics, 21–28 August 1968, Tokyo, edited by Y. Kohasi, A45–A48. Tokyo: Acoustical Society of Japan, 1968. Blauert, Jens. “Die Beschreibung von Hörversuchen anhand eines einfachen, systemtheoretischen Modells.” Kybernetik 5, no. 2 (1969): 45–49. Blauert, Jens. Räumliches Hören. Stuttgart: S. Hirzel, 1974. Blauert, Jens. Räumliches Hören: Nachschrift. Neue Ergebnisse und Trends seit 1972. Stuttgart: S. Hirzel, 1985.
240 Managing Sound, Assessing Space Blauert, Jens. “Sound Localization in the Median Plane.” Acustica 22, no. 4 (1969–70): 205–13. Blauert, Jens. Spatial Hearing: The Psychophysics of Human Sound Localization. Cambridge, MA: MIT Press, 1983. Blauert, Jens. “Untersuchungen zum Richtungshören in der Medianebene bei fixiertem Kopf.” PhD diss., RWTH Aachen, 1969. Blauert, Jens. “Vorbemerkung.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 195–96. Blauert, Jens. “Zur Methode der Nachrichtentechnik bei der Erforschung und Beschreibung der menschlichen Wahrnehmung.” Psychobiologie: Zeitschrift der Psychobiologischen Gesellschaft 14 (1966): 49–55. Blauert, Jens, Peter Laws, and Hans-Joachim Platte. “Impulsverfahren zur Messung der Außen ohrübertragungsfunktionen.” Acustica 31, no. 1 (1974): 35–41. Brech, Martha. Der hörbare Raum. Bielefeld: transcript-Verlag, 2015. Damaske, Peter. “Head-Related Two-Channel Stereophony with Loudspeaker Reproduction.” Journal of the Acoustical Society of America 50, no. 4 (1971): 1109–115. Damaske, Peter. “Richtungsabhängigkeit von Spektrum und Korrelationsfunktionen der an den Ohren empfangenen Signale.” Acustica 22, no. 4 (1969–70): 191–204. Damaske, Peter, and Volker Mellert. “Ein Verfahren zur richtungstreuen Schallabbildung des oberen Halbraums über zwei Lautsprecher.” Acustica 22, no. 3 (1969–70): 153–62. Damaske, Peter, and Bernhard Wagener. “Richtungshörversuche über einen nachgebildeten Kopf.” Acustica 21, no. 1 (1969): 30–35. De Boer, Kornelis. “Plastische Klangwiedergabe.” Philips’ Technische Rundschau 5, no. 4 (1940): 108–15. De Boer, Kornelis, and Arend van Urk. “Einige Einzelheiten beim Richtungshören.” Philips’ Technische Rundschau 6, no. 12 (1941): 363–68. De Boer, Kornelis, and Roeland Vermeulen. “Eine Anlage für einen Schwerhörigen.” Philips’ Technische Rundschau 4, no. 11 (1939): 329–32. Fletcher, Harvey. “Symposium on Wire Transmission of Symphonic Music and Its Reproduction in Auditory Perspective.” Bell System Technical Journal 13, no. 4 (April 1934): 239–44. Ganz, Cheryl. The 1933 Chicago World’s Fair: A Century of Progress. Urbana: University of Illinois Press, 2008. Genuit, Klaus. “Kunstkopf-Meßsysteme und die Signalverarbeitung im menschlichen Gehör.” In 50 Jahre Institut für Elektrische Nachrichtentechnik, 1950–2000, edited by Hans Dieter Lüke, 65–67. Geilenkirchen: Gatzen, 2000. Gottlob, Dieter. “Anwendung der kopfbezogenen Stereofonie in der akustischen Forschung.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 214–16. Hudde, Herbert. “Methoden zur Bestimmung der menschlichen Trommelfellimpedanz unter Berücksichtigung der Querschnittsfunktion des Ohrkanals.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 206–8. Kleiner, M. “Problems in the Design and Use of ‘Dummy- Heads.’” Acustica 41, no. 3 (1978): 183–93. Krebs, Stefan. “The Failure of Binaural Stereo: German Sound Engineers and the Introduction of Dummy Head Microphones.” ICON: Journal of the International Committee for the History of Technology 23 (2017): 113–43. Krumbacher, G. “Über die Leistungsfähigkeit der kopfbezüglichen Stereofonie.” Acustica 21, no. 5 (1969): 288–93. Kuhl, W., and R. Plantz. “Kopfbezogene Stereophonie und andere Arten der Schallübertragung im Vergleich mit dem natürlichen Hören.” Rundfunktechnische Mitteilungen 19, no. 3 (1975): 120–32. Kürer, Ralf. “Untersuchungen zur Auswertung von Impulsmessungen in der Raumakustik.” PhD diss., Technical University Berlin, 1972.
The Development of Kunstkopf Technology 241 Laws, Peter. “Entfernungshören und das Problem der Im- Kopf- Lokalisiertheit von Hörereignissen.” Acustica 29, no. 5 (1973): 243–59. Laws, Peter. “Messung und Nachbildung von Trommelfellimpedanzen.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 201–6. Laws, Peter. “Untersuchungen zum Entfernungshören und zum Problem der Im- Kopf- Lokalisiertheit von Hörereignissen.” PhD diss., RWTH Aachen, 1972. Laws, Peter, Jens Blauert, and Hans- Joachim Platte. “Anmerkungen zur stereophonen kopfbezogenen Übertragungstechnik.” Acustica 36, no. 1 (1976–77): 45–47. Laws, Peter, and Hans- Joachim Platte. “Ein spezielles Konzept zur Realisierung eines Kunstkopfes für die kopfbezogene stereofone Aufnahmetechnik.” Rundfunktechnische Mitteilungen 22, no. 1 (1978): 28–31. Laws, Peter, and Hans- Joachim Platte. “Spezielle Experimente zur kopfbezogenen Stereophonie.” In Fortschritte der Akustik. Plenarvorträge und Kurzreferate der 4. Tagung der Deutschen Arbeitsgemeinschaft für Akustik, 365–68. Weinheim: Physik Verlag, 1975. Licklider, Joseph Carl Robnett. “Auditory Frequency Analysis.” In Information Theory, edited by Colin Cherry, 253–68. London: Butterworth Scientific, 1956. Licklider, Joseph Carl Robnett. “A Duplex Theory of Pitch Perception.” Experientia 7, no. 4 (1951): 128–34. Lüke, Hans Dieter, ed. 50 Jahre Institut für Elektrische Nachrichtentechnik, 1950– 2000. Geilenkirchen: Gatzen, 2000. McGinn, Robert E. “Stokowski and the Bell Telephone Laboratories: Collaboration in the Development of High-Fidelity Sound Reproduction.” Technology and Culture 24, no. 1 (1983): 38–75. Mellert, Volker. “Construction of a Dummy Head After New Measurements of Thresholds of Hearing.” Journal of the Acoustical Society of America 51, no. 4B (1972): 1359–361. Mellert, Volker. “Verbesserte Schallfeldabbildung mit einem neuen Kunstkopf.” In Fortschritte der Akustik: Plenarvorträge und Kurzreferate der 4. Tagung der Deutschen Arbeitsgemeinschaft für Akustik, 433–36. Weinheim: Physik Verlag, 1975. Mellert, Volker. “Vergleichende Beurteilung von Verkehrsgeräuschen—erste Ergebnisse und Korrelation mit physikalischen Parametern.” In Fortschritte der Akustik: Plenarvorträge und Kurzreferate der 5. Tagung der Deutschen Arbeitsgemeinschaft für Akustik, 289–92. Düsseldorf: VDI-Verlag, 1976. Mellert, Volker, Karl Friedrich Siebrasse, and S. Mehrgardt. “Determination of the Transfer Function of the External Ear by an Impulse Response Measurement.” Journal of the Acoustical Society of America 56, no. 6 (1974): 1913–915. Paul, Stephan. “Binaural Recording Technology: A Historical Review and Possible Future Developments.” Acta Acustica united with Acustica 95, no. 5 (2009): 767–88. Pinch, Trevor. “‘Testing—One, Two, Three . . . Testing!’: Toward a Sociology of Testing.” Science, Technology, & Human Values 18, no. 1 (1993): 25–41. Platte, Hans-Joachim. “Probleme bei Messung und Nachbildung von Aussenohrübertragungsf unktionen.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 198–201. Platte, Hans- Joachim, Peter Laws, and Harald vom Hövel. “Anordnung zur genauen Reproduktion von Ohrsignalen.” In Fortschritte der Akustik: Plenarvorträge und Kurzreferate der 4. Tagung der Deutschen Arbeitsgemeinschaft für Akustik, 361–63. Weinheim: Physik Verlag, 1975. Plenge, Georg. “Über das Problem der Im-Kopf-Lokalisation.” Acustica 26, no. 5 (1972): 241–52. Plenge, Georg. “Über das Problem der intracranialen Ortung von Schallquellen bei der akustischen Wahrnehmung des Menschen.” Professorial diss., Technical University Berlin, 1973.
242 Managing Sound, Assessing Space Plenge, Georg. “Über die Hörbarkeit kleiner Änderungen der Impulsantwort eines Raumes.” Acustica 25, no. 5 (1971): 315–25. Plenge, Georg, Ralf Krüger [Kürer], and Henning Wilkens. “Über die Reproduktion von Hörbildern mit Hilfe eines künstlichen Kopfes.” In 8. Tonmeistertagung, edited by Hans- Georg Daehn, 80–83. Hamburg: VDT, 1969. Poulsen, Torben. “Hörvergleich unterschiedlicher Kunstkopfsysteme.” Rundfunktechnische Mitteilungen 22, no. 4 (1978): 211–14. Reinecke, Hans- Peter. “Das Ideal des naturgetreuen Klangbildes— ein psychologisches Problem.” In 8. Tonmeistertagung, edited by Hans- Georg Daehn, 85– 88. Hamburg: VDT, 1969. Rheinberger, Hans-Jörg. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford, CA: Stanford University Press, 1997. Schampaul, Nikolaus. “Die Übertragung des natürlichen Klangbildes.” In 8. Tonmeistertagung, edited by Hans-Georg Daehn, 45–50. Hamburg: VDT, 1969. Scherer, Paul. “Über die Ortungsmöglichkeit verschiedener stereophonischer Aufnahmeverfahren.” Nachrichtentechnische Fachberichte 15 (1959): 36–42. Scherer, Paul. “Untersuchung verschiedener Aufnahme- Anordnungen.” Master’s thesis, RWTH Aachen, 1957. Schirmer, Werner. “Die Richtcharakteristik des Ohres.” Hochfrequenztechnik und Elektroakustik 72 (1963): 39–48. Schmidt Horning, Susan. Chasing Sound: Technology, Culture and the Art of Studio Recording from Edison to the LP. Baltimore, MD: Johns Hopkins University Press, 2013. Schroeder, Manfred, Dieter Gottlob, and Karl Friedrich Siebrasse. “Comparative Study of European Concert Halls: Correlation of Subjective Preference with Geometric and Acoustic Parameters.” Journal of the Acoustical Society of America 56, no. 4 (1974): 1195–201. Star, Susan Leigh. “This Is Not a Boundary Object: Reflections on the Origins of a Concept.” Science, Technology, & Human Values 35, no. 5 (2010): 601–17. Steinberg, J. C., and W. B. Snow. “Physical Factors.” Bell System Technical Journal 13, no. 2 (1934): 245–58. Theile, Günther. “Zur Theorie der optimalen Wiedergabe von stereofonen Signalen über Lautsprecher und Kopfhörer.” Rundfunktechnische Mitteilungen 25, no. 4 (1981): 155–70. Thome, Rolf. “Räumliches Hören bei ein-und zweiohriger Hörgeräteversorgung.” Master’s thesis, RWTH Aachen, 1974. Van Kordelaar, Robert. “Messungen am künstlichen Kopf.” Master’s thesis, RWTH Aachen, 1962. Vermeulen, Roelof. “Vergleich zwischen wiedergegebener und echter Musik.” Philips’ Technische Rundschau 17, no. 6 (1955): 191–98. Weber, Reinhard, and Volker Mellert. “Ein Kunstkopf mit ‘ebenem’ Frequenzgang.” In Fortschritte der Akustik: Plenarvorträge und Kurzreferate der 6. Tagung der Deutschen Arbeitsgemeinschaft für Akustik, 645–48. Berlin: VDE-Verlag, 1978. Wilkens, Henning. “Kopfbezügliche Stereophonie— ein Hilfsmittel für Vergleich und Beurteilung verschiedener Raumeindrücke.” Acustica 26, no. 4 (1972): 213–21. Wilkens, Henning. “Mehrdimensionale Beschreibung subjektiver Beurteilungen der Akustik von Konzertsälen.” PhD diss., Technical University Berlin, 1975. Wilkens, Henning, and B. Kotterba. “Vergleich der Beurteilung verschiedener raumakustischer Situationen bei Anregung eines Raumes mit einem Orchester oder Lautsprechern.” Acustica 40, no. 5 (1978): 291–97.
9 Absorption, Transmission, Reflection Testing Materials in the Laboratory Roland Wittje
Around 1929, physics professor Johan Peter Holtsmark established the first acoustic laboratory at Norges Tekniske Høgskole (Norwegian Institute of Technology, NTH) in Trondheim. Having previously worked on the Stark effect and electron scattering, both matters related to quantum physics, Holtsmark was a newcomer to acoustics research.1 Yet within a few years he had become Norway’s foremost authority in the field of architectural acoustics and a consultant for the Norwegian Broadcasting Corporation. The standardized testing of construction materials and designs in the laboratory was central to his research and consultancy activities. Why and how did Holtsmark enter acoustics research and consulting, and what does his trajectory tell us about acoustic testing of construction materials in the interwar period? The standardized acoustic testing of construction materials and methods, and the laboratory setups for carrying out such tests, had several genealogies. Up to World War I, musical listening, musical literacy, and musical practice were the essential skills for acoustic measurement. During and after the war, however, musical listening became increasingly irrelevant to measurement— acousticians now needed skills in electrical engineering and radio to design and use electrical measurement and recording technology. Instead of the inherently subjective human ear, it was now the apparently objective electric system that defined acoustic measurement.2 Sound abatement initiatives, the rise of electroacoustic technologies, and a transformed understanding of architectural acoustics (especially through the work of Harvard physicist Wallace Sabine in the early twentieth century) all played major roles in the development of a large variety of construction materials with specific acoustic properties.3 Electroacoustic media, especially radio broadcasting and systems of sound motion pictures and amplification, required new materials to be deployed in the acoustic design of spaces for recording and playback. Standardized electroacoustic measurement facilitated standardized testing of materials by eliminating the unreliable human listener and making Roland Wittje, Absorption, Transmission, Reflection In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0010.
244 Managing Sound, Assessing Space
measurements reproducible. As well as measurement technology, standardized acoustic laboratories were required, built as acoustically “neutral” spaces where the acoustic properties of materials and designs could be isolated. A whole range of acoustic laboratory designs for different types of measurements emerged in the 1920s and 1930s, including reverberation chambers, laboratories with test fields for wall and floor construction, and anechoic chambers. Like many other scientific and engineering fields in the period, acoustics made the transition from tabletop experimentation to standardized laboratory setups and electrical measurement systems based on radio tube amplifiers.4 The establishment of materials testing laboratories in the German Empire in the 1870s and 1880s, which served as a model for Norway, was part of a program to base engineering practice and education on scientific methods and principles and to reshape the engineering disciplines as technical sciences.5 Most such laboratories in Germany were created as departments of, or in close collaboration with, the institutes of technology, Technische Hochschulen, as in the case of the Prüfungsstation für Baumaterialien (Testing Station for Construction Materials) in Berlin, which was a special department of the Technische Hochschule Berlin- Charlottenburg.6 Construction materials made up a large, even dominant part of all the materials tested, and laboratory testing was essential to the increasing standardization of both materials and methods in civil engineering—for the laboratories not only carried out standardized tests and issued certificates but also trained students in testing methods. The history of these laboratories for materials testing should be seen in the context of the emergence of other testing and standardization institutions, such as the Physikalisch-Technische Reichsanstalt (Imperial Physical Technical Institute), which opened its doors in Berlin in 1887 and served as a model for other national bureaus of standards, and the Chemisch Technische Versuchsanstalt (Chemical- Technical Experimental Testing 7 Station). Although standardized testing and certification was meant to ensure the quality of the materials and thus the safety of buildings, as a rule its logic was not so much to improve the overall quality of construction as to optimize the cost of construction while minimizing the volume of construction material needed.8 This was particularly true in interwar Europe, when a severe housing shortage led to public programs to build affordable housing for the lower middle and working classes. As the Swiss architectural historian and critic Sigfried Giedion noted in 1929, prewar housing was actually constructed too solidly. The craft tradition, building for eternity, produced luxurious and spacious living for the rich, but it made humane and healthy housing
Absorption, Transmission, Reflection 245
unaffordable for the less affluent sections of society.9 Giedion and others, such as the German architect and city planner Ernst May, argued that only industrialization and mass production could supply adequate urban housing for the poor.10 With the expansion of construction activity, as well as the industrialization of construction and the depletion of resources such as wood and natural stone, resource-saving methods and new construction materials came to the fore. Whereas earlier construction materials came ready-grown from nature, new synthetic materials—steel, glass, cement—were assembled, as Giedion put it, “from elements in the laboratory.”11 Despite the rhetoric of science-based industrialization, however, construction materials, construction methods, and architecture were still overwhelming local; they depended on the local availability and quality of materials, local practices, local industries, and local climates. At the same time, testing laboratories were set up to ensure standards and comparability for construction materials and methods that went beyond local practice and craftsmanship. The standard tests of construction materials included mechanical tests of bending, density, tensile strength, and torsion, along with physical and chemical tests of durability, weather resistance, and fire resistance.12 Although acoustic tapping had long been used in determining the properties of materials and products,13 testing the acoustic properties of construction materials was initially not part of the routines carried out in the testing laboratories. The growing din in modern cities had given rise to anti-noise societies in European and North American cities from around 1900,14 but most acousticians were mainly interested in musical sounds and did not address problems of noise and noise abatement before World War I. The wartime deployment of acoustics to develop sound location of artillery, aircraft, and submarines fundamentally changed the kinds of sounds that scientists studied and how they studied them. Acousticians’ attention, then, was drawn to noise and noise abatement by the roar of the battlefields, not by the calls of hygienists or the noisy streets of Berlin or London.15 Norway remained officially neutral during World War I and did not participate in the development of military sound locating technologies that occurred in Germany, France, Britain, the United States, and other combatant nations. In the interwar period, however, Norway experienced advances in electroacoustic media, the economic crisis, and the industrialization of construction and construction materials in a similar way to other European countries. Holtsmark established the acoustic laboratory at the NTH—and his own reputation as a Norwegian authority in architectural acoustics—at a time when the field of acoustics was undergoing rapid growth and transformation. Electrical engineers with a background in
246 Managing Sound, Assessing Space
amateur radio built and used the electroacoustic devices that became crucial to acoustics testing and consultancy. Their practices of instrument building and use connected acoustics to the research in nuclear physics and electron scattering that Holtsmark was concurrently pursuing. In addition, the rapid advance of sound motion pictures and public radio broadcasting created a market for acoustic knowledge, and thus for acoustic consulting work. Acoustic consulting encompassed not only testing acoustics in the laboratory but also fieldwork, and the Trondheim acousticians moved their measurement instruments outside the laboratory, carrying them all over Norway. The combination of media technologies, acoustic consultancy, and standardized testing created an essential relationship between the local, national, and transnational production and consumption of acoustic knowledge and practices.
An Acoustic Laboratory for the Norwegian Institute of Technology Although the systematic acoustic laboratory testing of construction materials did not become widespread until the 1920s, the acoustic testing of materials has a longer history in architectural acoustics. It was physicists and electrical engineers who pioneered standardized acoustic testing, whereas the tests determining other mechanical parameters of construction materials were usually directed by civil or mechanical engineers. In 1902, Charles Norton, electrical engineer and physics professor at the Massachusetts Institute of Technology, carried out acoustic tests on “Quilt,” a new type of padded building paper, introduced by Samuel Cabot in 1892, that could absorb sound as well as offering insulating and fire-retardant qualities.16 Norton measured heat transfer at the same time and developed processes to manufacture fireproof asbestos materials. His investigation of sound insulation was not designed to keep out city noise, but to insulate dormitories in the New England Conservatory of Music against musical sounds. In Germany, too, there was a close connection between the testing of sound insulation of construction materials or techniques and the question of heat insulation. When Oscar Knoblauch opened the first Laboratory for Technical Physics at the Technische Hochschule Munich in 1902, he initially concentrated on the thermodynamics of steam, but soon moved the laboratory’s research agenda to the heat and sound insulation of buildings.17 Between 1907 and 1911, Richard Berger experimented on sound propagation through walls in Knoblauch’s laboratory for his doctoral dissertation.18
Absorption, Transmission, Reflection 247
Wallace Sabine was interested in the sound reflection of construction materials and techniques, as opposed to Norton’s and Berger’s concerns with sound insulation. Sabine used sound- absorbing materials in his investigations of reverberation time in lecture theaters and concert halls, and made recommendations for construction. In 1911, the tile manufacturer Raphael Gustavino approached Sabine with a proposal to collaborate on producing sound-absorbing tiles that would allow churches to look Gothic but with much less reverberation than a Gothic church.19 Sabine’s task of advising architects on the use of materials to achieve particular acoustic effects was complex, since the acoustics could not easily be separated from other aspects such as aesthetics, practical use, and cost or from other physical and chemical properties such as strength and fire resistance. In the emerging acoustic testing laboratory, however, the acoustic properties of construction materials and techniques could be reduced to three parameters: the absorption, transmission, and reflection of sound. As sound measurement in the laboratory became more reductionist, it also added new parameters. Sabine carried out his initial reverberation experiments only at a single frequency, 512 cycles per second,20 but scientists soon pointed out that the acoustic behavior of a room depended on the frequency of the sound.21 Absorption, transmission, and reflection of sound had to be determined either by using complex sounds or by measuring over a greater range of frequencies. By the 1930s, there were a large number of acoustic materials on the market that advertised noise-absorbing properties.22 These new materials and construction techniques created a demand for acoustic testing and standardized methods to determine absorption coefficients and sound transmission through walls and floors,23 and acoustic laboratories were set up at the German Technische Hochschulen to carry out such tests. The Institut für Schall-und Wärmeforschung (Institute for Sound and Heat Research) at the Technische Hochschule Stuttgart, for example, was founded in 1929.24 Its director was the technical physicist Hermann Reiher, who had studied sound and heat insulation with Knoblauch in Munich.25 Like many of his acoustician peers, Reiher had a background in electrical engineering and built his own electroacoustic measurement apparatus. His Institute for Sound and Heat Research was founded as an independent division of the Materialprüfungsanstalt (Materials Testing Laboratory) of the Technische Hochschule Stuttgart and was closely affiliated with the local construction industry. Whereas the Institute for Sound and Heat Research mainly investigated the propagation of noise in residential and office buildings, other facilities placed new media technologies, especially sound motion pictures and radio broadcasting, high on their acoustic testing agenda. This was the case for
248 Managing Sound, Assessing Space
Johan Holtsmark and the acoustic laboratory at NTH. Like many others set up around 1930 in North America and Europe, the NTH acoustic laboratory aimed to facilitate new electroacoustic media technologies and the increasing industrialization of construction. Recollecting the origins of the Trondheim acoustics laboratory, Holtsmark wrote that he had followed Sabine’s pioneering research in architectural acoustics, but had decided to engage in acoustics research himself only when sound motion pictures came at the end of the 1920s. This coincided with the development of radio and studios and so forth, and it was a very hectic time with intensive work from many sides to get sound motion pictures working in the shortest time possible and to get suitable broadcasting studios for radio. . . . What we got a sense of in particular was, first, the problems with loudspeakers in the cinemas and, second, the adaptation of studios and movie theaters with suitable room acoustics. These were altogether interesting problems and the whole thing was so new that we were forced to do a great deal ourselves to sort them out.26
In 1929, Holtsmark published an article in the Norwegian engineering magazine Teknisk Ukeblad informing readers about the new acoustic laboratory at the physics department of NTH. Under the heading “Acoustic Problems,” he sketched out what he saw as the main tasks for technical acoustics and the agenda for the new laboratory: constructing devices to generate or receive sound; assessing the acoustic properties of lecture halls, concert halls, and other premises through calculations; and sound insulation.27 The first three papers on acoustics that Holtsmark and his assistants published were all on sound insulation.28 But what made Holtsmark, a physicist who had worked in the fields of atomic physics and quantum mechanics, turn to technical acoustics? The timing of the new laboratory’s establishment was by no means random—it coincided with a period when international research in technical acoustics and electroacoustics was becoming organized. The Acoustical Society of America was founded in December 1928, and in 1931, when Holtsmark joined, it already had 632 members. The Heinrich Hertz Institute for Oscillation Research in Berlin, first conceived in 1927, opened its doors in 1930. And in Norway, demand for the acoustical planning and manipulation of architectural spaces was increasing. The first sound motion picture to arrive in Norway—The Singing Fool, directed by Lloyd Bacon—was shown in the summer of 1929 at the Eldorado cinema in Oslo, which was the first Norwegian cinema to install sound equipment.29 Radio broadcasting began to mature in Norway at the same time. With a steadily growing number of listeners, radio evolved from a
Absorption, Transmission, Reflection 249
specialist activity of radio enthusiasts to a mass medium. Increasing demands were made on both the content of radio programs and their technical quality. In short, research and consultancy in technical acoustics offered great opportunities for engineers around 1929. The most important ingredient for entering into acoustics research in Trondheim, however, was the experience of electroacoustic instrumentation that Holtsmark and his assistants had gathered in previous years through amateur radio. As in other research fields at NTH’s physics department at the time, most of the scientific instrumentation was made in the department’s own workshop. Until 1939, all of Holtsmark’s assistants involved in acoustics research, consultancy, and instrument making were electrical engineers with prior connections to NTH’s Academic Radio Club. Indeed, from 1925 to 1929, the Radio Club had its premises at the physics department, where it set up a provisional broadcasting station in the spring of 1926. Holtsmark himself published in the Norwegian amateur radio journal Norsk radio.30 The NTH team’s experience in amateur radio and in supplying the Student Society with electroacoustic devices was a crucial element of the transition to standardized acoustic testing, providing the skills to tackle the radically different instrumentation and measurement practices of the new acoustics research. Before World War I, most scientific instruments in such research were closely related to musical instruments, and the acoustician’s essential skills were musical ones. In the 1920s and 1930s, musical listening was no longer required. Instead, acousticians needed the skills of a radio ham to make their own electrical devices and to control and manipulate sound and sound measurement.31 It was through these amateur activities that the Trondheim acousticians learned to build and use electroacoustic instrumentation such as microphones, loudspeakers, amplifying circuits, and high-frequency filters. Holtsmark’s most important assistants in the first years of acoustics research were Reno Berg and Vebjørn Tandberg. Berg, one of the students who had operated the Radio Club’s broadcasting station in 1926, performed most of the acoustic measurements throughout the 1930s. He stayed at NTH for the rest of his career. Holtsmark’s publications, correspondence, and memoirs indicate that his own contributions to acoustics research were mainly theoretical, conceptual, and organizational, whereas Berg carried out the practical measurements and experiments. Tandberg, the Radio Club’s chair in 1929, was the main maker of electroacoustic devices until 1933, when he left Holtsmark and NTH to start his own radio factory in Oslo—Tandberg Radio became arguably Norway’s most famous manufacturer of radio and other audio equipment.32 Tandberg later stated that, de facto, his radio factory started in the basement of the Trondheim physics department.33 He constructed
250 Managing Sound, Assessing Space
loudspeakers, condenser microphones, and the Trondheim acousticians’ first specific precision measurement instrument: an automatic reverberation and sound intensity measurement apparatus (Figure 9.1), a copy of which was built for the Norwegian Broadcasting Corporation (Norsk Rikskringkasting, NRK).34 Research and consultancy in acoustics were largely carried out by Holtsmark and Berg, while a large number of telecommunication engineers were assigned to instrument building for shorter periods. Oddvar Johannesen and Eilif Bjørnstad, for example, built a registering decibel meter for NRK between 1935 and 1937. The biographies of Tandberg and other electrical engineers demonstrate the intimate connection, if not co-production, of electroacoustic devices and skills between the acoustic laboratories, the radio manufacturing industry, and NRK.35 The great majority of the research papers by Berg and Holtsmark were written in German and published in the proceedings of the Royal Norwegian Society of Science and Letters;36 only one was published internationally in the German journal Elektrische Nachrichten-Technik, while some were written in Norwegian and published in other Norwegian professional and trade journals. This language preference made most articles intelligible for the non- Scandinavian world. As an international language of science and engineering, German thus prevailed despite NTH’s research and consultancy in acoustics being mainly directed at Norwegian society and industry. But although Holtsmark could send out preprints to researchers working in the same field, publication in Norwegian-based journals seriously limited the circulation of the papers. Articles in the proceedings of the Royal Norwegian Society of Science and Letters were not reviewed in review journals such as the German Physikalische Berichte.37 Published in an international language of science and engineering but in a local journal with limited international distribution, Holtsmark’s acoustics research and consultancy were situated between a local and a transnational knowledge economy. The Norwegian context and character of Holtsmark’s consulting activities coexisted with his and his collaborators’ participation in the international acoustics community. Holtsmark was a Norwegian delegate to international congresses that negotiated acoustical units, limits for noise, and other matters regarding acoustics.38 With Holtsmark, Berg, and the NTH electrical engineers Erik Julsrud and Fredrik Møller all being members of the Acoustical Society of America, Norway had the strongest representation of the Scandinavian countries there. From 1933 onward, Holtsmark reviewed articles from the Journal of the Acoustical Society of America for Physikalische Berichte. He maintained contact with the German acoustics community, particularly with Erwin Meyer, one of Germany’s foremost authorities in the
Absorption, Transmission, Reflection 251
field, at the Heinrich Hertz Institute in Berlin. Holtsmark visited the institute in February 1931, shortly after its opening, to study its methods of acoustic investigation.39 In summer 1935, Berg went on a lengthy study tour of England and Germany with Wilhelm Ramm and Sverre Westin. Ramm and Westin were assistants of Holtsmark’s, working on a Van de Graaff particle accelerator and on electron scattering. Sending assistants working in different fields on a joint study tour typified Holtsmark’s program of organizing various research agendas simultaneously.40 For the acoustical planning of the new home of NRK broadcasting at Marienlyst in Oslo, Holtsmark studied broadcasting houses in other countries including Germany and Britain. Whereas Holtsmark’s research on nuclear physics and electron scattering was part of an international research endeavor, with few ties to Norwegian industry or society, his work in acoustics was both locally and nationally situated, and he also participated in international networks of research and consultancy that were essential to the national and local establishment of radio broadcasting and sound motion film as large technological systems of mass media.41 As was the case with other research activities, there was little funding available in NTH’s central budget for acoustics research assistants or research instrumentation. Holtsmark’s main assistant, Reno Berg, was appointed in 1932 as one of the department’s teaching assistants and took on the teaching of technical acoustics to electrical engineering students in 1934. The costs for all other assistants, additional instrument makers in the workshop, and scientific instrumentation had to be met from other sources. Holtsmark established a mixed financing system, based on funding from various research foundations and income from consultancy clients. The funding from research foundations, among them the Norges tekniske høiskoles fond and the A/S Norsk Varekrigsfond, released Holtsmark and Berg from the pressure of depending entirely on consulting assignments—a problematic source, since consultancy clients were usually interested in quick solutions to specific technical problems, not in funding a long-term strategic research agenda that would have placed the acoustics laboratory’s research activities on a steady financial basis and enabled substantial research projects for Holtsmark’s assistants. In Holtsmark’s 1940 strategic plan for the establishment of research scholarships at NTH’s physics department, he specified three research assistants as his desired workforce for the acoustic laboratory.42 That was an ambitious target, but by this point Holtsmark had found a relatively convenient solution for his funding problem: in 1939, after at least six years of continuous consultancy for NRK, his largest client, Holtsmark managed to negotiate
252 Managing Sound, Assessing Space
an annual basic fee from NRK, allowing him to operate more independently.43 His correspondence with NRK also reveals that Holtsmark billed not only equipment, running costs, and his assistants’ consultancy fees but also a fee for his own work. NRK considered Holtsmark’s own consulting fees rather high given his professorial salary at NTH, which was, like NRK, a public institution. Holtsmark defended his fees as a legitimate compensation for the time spent on the task. It can be assumed, though this is not documented, that Holtsmark also received income from other consulting assignments—paid consultancy services were common among NTH professors from the engineering disciplines.44
Architectural Acoustics and the Consulting Scientist One of the defining features of Holtsmark’s approach to architectural acoustics was the combination of fieldwork—as architectural acoustics investigated the propagation of sound in architectural spaces outside the laboratory—and lab work, or the laboratory testing of the acoustic properties of construction materials and construction designs. The two activities were not necessarily sharply divided. The acoustic properties of rooms and buildings depended on their shape and size, but also on the acoustic properties of the construction materials and techniques used, the most common sound manipulations being the adjustment of reverberation time, the removal of sharp echoes, and sound insulation. Nonetheless, the distinction between architectural acoustics as a field practice and the testing of materials and construction designs as a laboratory practice is a crucial one due to the very different research and consulting approaches that they entailed. Architectural acoustics as fieldwork implies going out into the built environment, with its complex technical and social relationships, as opposed to the artificial environment of the testing laboratory. The investigations and consulting activities of Holtsmark and his assistants were related either to existing buildings and their modification or to the planning of new buildings. Cinemas were among the existing structures whose acoustic properties had to be modified. With the arrival of sound motion pictures, many movie theaters that had previously been used for silent pictures were acoustically inadequate for the presentation of the upgraded medium. Long reverberation times, sharp echoes, and inappropriate timbres left the electrically amplified sound of the movies unpleasant and unintelligible. Holtsmark was involved in the upgrading of many cinemas, but his largest acoustics consulting assignment in the 1930s concerned the new NRK broadcasting house at Marienlyst. His
Absorption, Transmission, Reflection 253
acoustic consulting for Marienlyst started in 1933 and construction began in 1938; it was completed only in 1950. In both the planning of new buildings and the modification of existing ones, the consultants’ findings would be presented as a report including suggestions on how to achieve the desired acoustic characteristics. These assignments did not necessarily include acoustic measurements, but in the cases where they did, the acoustic laboratory had to move out into the field. The building or room in question itself became the laboratory, with consequences both for the scientific instruments used and for research practices. The acoustic laboratory was moved out of NTH’s physics laboratory and transported literally all over Norway.45 Measurement instruments had to be made compact, portable, and robust so as to withstand traveling over long distances; instruments had to be easy to assemble and set up on site. The reverberation measurement apparatus built by Vebjørn Tandberg around 1931 and contained in a suitcase is the embodiment of an instrument built for the traveling scientist (Figure 9.1). It was
Figure 9.1 Reverberation measurement apparatus built by Vebjørn Tandberg around 1931. The apparatus shown here was kept at NTH’s physics department. Courtesy of the Department of Electronic Systems, Norwegian University of Science and Technology. Photo: Roland Wittje.
254 Managing Sound, Assessing Space
compact and could be carried by one person. The dimensions of the main apparatus in the black casing fitted the suitcase, so that the apparatus would not have to be removed for use. A compartment on the right holds the recording microphone, which can be placed at some distance from the suitcase and comes with a cable to be plugged into the main apparatus. A comparable apparatus in the laboratory would measure variable differences of sound intensity, but outside the laboratory, the noise level was so high that the apparatus worked at a fixed sound intensity.46 The portable and simplified design of the instrument did not mean that just anybody could perform meaningful measurements without training. The laboratory’s stabilization in the field and the extraction of reproducible results were far from trivial matters, and the apparatus still required special skills. Inside the lid, we see the circuit diagram that provided a map to the apparatus for the electrically literate. There is also a manual for careful calibration. Measurements of the same room by different scientists or engineers could yield very different results, and in fact the problems of calibrating the reverberation apparatus eventually meant that the copy built for NRK did not achieve the anticipated success.47 Consulting in architectural acoustics was an inherently interdisciplinary task. It involved the ability to cooperate with architects, civil engineers, and, most importantly, the building’s users. A seemingly never-ending list of concerns had to be considered: multiple uses, aesthetic and architectural factors, construction statics, fire regulations, construction budgets, aesthetic fashions, state regulations, and more. Acoustic planning and manipulation alone were not the main concern—a building’s acoustics were only one of many peripheral issues attending the planner’s main question, the practical use of the venue. Holtsmark’s reports and suggestions in architectural acoustics were clearly shaped by this complex context and indicate his ability to situate acoustical concerns within the requirements of the building’s proposed uses. The testing of construction materials and construction designs regarding their acoustic properties was radically different from the work of the acoustician in the field. Materials testing was conducted in the laboratory, and under controlled conditions. In the laboratory, one variable of a well- defined material or construction design could be singled out and investigated systematically. The acoustic laboratory for materials testing at the NTH physics department was a permanent setup. Stable conditions for investigation enabled different materials to be compared under the same conditions. A very important premise for experimental physics research, reproducibility, could thus be established—the problems of reproducibility for architectural
Absorption, Transmission, Reflection 255
acoustics in the field having become evident when architects tried and failed to reproduce the acoustic characteristics of outstanding buildings such as the Gewandhaus concert hall in Leipzig.48 The reports written on materials testing in the laboratory were thus very different from Holtsmark’s consulting reports for architectural acoustics in the field. The lab-based reports could be standardized, largely ignoring the multitechnical and social context of construction planning. It was only in the laboratory that the testing of acoustic variables such as reflection, absorption, and transmission was able to become the main or even sole concern of investigation. The distinction between laboratory research and fieldwork is important, then, but it could not always be strictly maintained for the research practices of Holtsmark and Berg. Instead, we see a complexity of interactions. In several cases, the Trondheim acousticians were granted permission to use construction sites for a more extensive series of tests, often not relevant to the consulting task. The half-finished construction sites were thereby transformed into laboratories. This relocation of the laboratory from the physics building to the construction site occurred mainly for practical reasons, since no comparable laboratory was available at NTH at the time. The construction site could sometimes prove a poor laboratory, however. In 1934, Berg and Holtsmark turned the construction site of the Handelsstandens Hus in Trondheim into a laboratory for comparing airborne and impact sound, but although they obtained useful measurements for airborne sound, conditions in their provisional laboratory were unsuitable for the planned measurements of impact sound.49 From August 1931 to August 1934, Berg and Holtsmark measured reverberation in thirty-seven public assembly rooms in Norwegian cities.50 The main instrument in these investigations was Tandberg’s suitcase—the portable automatic reverberation measurement apparatus. Twenty-five of the buildings studied were used as cinemas, thirteen as concert halls, eleven as theaters, and six as lecture halls. Several were multipurpose premises, requiring compromises to be made to obtain satisfactory reverberation characteristics for all types of arrangements. Whether the acoustic properties were satisfactory also depended on the patrons’ listening habits. Evaluations of concert halls, for example, could vary between musicians and audience,51 whereas lecture halls mainly required intelligibility of speech, and opinions as to their acoustic virtues were less likely to diverge.52 Furniture had a strong, usually positive, influence on the timbre of reverberation and was discussed and investigated in great detail.53 Apart from reverberation, sound propagation through buildings (particularly through cement ceilings and different types of sound insulation) was the main topic of the investigations and suggestions by the Trondheim acousticians.
256 Managing Sound, Assessing Space
One might ask whether scientific consultancy and electroacoustic measurement actually improved the acoustic quality of such buildings. Holtsmark’s enthusiasm for novel electroacoustic devices becomes obvious in a 1934 interview published in Oslo Illustrerte. There, he praises the microphones and loudspeakers built by his assistants in the NTH workshop. When it comes to the success of acoustical engineering as opposed to craftsmanship in construction, though, Holtsmark appears less thrilled. In the interview and elsewhere, Holtsmark claims superiority for the acoustics of many buildings constructed before the age of scientific acoustic manipulation, particularly the old Gewandhaus in Leipzig and the Freemasons’ Lodge in Oslo. In the interview, Holtsmark does not blame this on cost optimization in construction during the interwar years, but supposes that the former architects may have achieved better acoustics through their access to long experience, as opposed to the relatively short traditions of scientifically based architectural acoustics. For the construction of the large symphony studio in Oslo’s new broadcasting house, Holtsmark recommended that NRK should copy elements of “the good old halls,” such as the gallery, padded chairs, and pillars.54
Acoustic Testing in the Laboratory Craft had certainly come a long way in the history of architectural acoustics, but its knowledge could not easily be translated into the language of modern acoustics and industrial construction. For the transition from craft-based to industrial construction, the testing and standardization of construction materials was essential. The testing of all kinds of fabricated materials had held a central position in the organization of the Norwegian Institute of Technology right from its establishment in 1910. The materials testing laboratory, Materialprøvingsanstalten, was set up as an independent unit at NTH. Like most of NTH, Materialprøvingsanstalten followed a German model; in fact, its first director, the professor of mechanics Heinz Egerer, was German himself. The laboratory served a double function: like other units of NTH, it was a teaching institution to train engineers in the techniques of testing materials, and it also acted as a public testing institution for private industry and the public sector. In 1918, the Ministry of Church and Education approved the legal status of the laboratory as a public institution for testing materials. According to the regulations of 1918, only tests relating to mechanical properties of materials were carried out in the laboratory’s own facilities, which were well equipped for this task. Other tests, for example those on the chemical or electrical
Absorption, Transmission, Reflection 257
properties of materials, were delegated to the NTH department with the appropriate expertise and facilities. Accordingly, physics-based tests were to be performed at NTH’s physics department. Initially, there was very little cooperation between the physics department and Materialprøvingsanstalten. Of a total of 5,337 tests of different materials carried out between 1912 and 1927, only five were performed by the physics department, compared to 1,959 by the chemistry department, 178 by the electrical engineering department, and 100 by the geology department. The mechanical laboratory of Materialprøvingsanstalten itself carried out 3,068 tests.55 The testing of acoustical materials marked the physics department’s entrance into the business of testing materials. Many publications and archival sources bear witness to the extensive test series in the acoustical laboratory of the physics department. The tests investigated different construction materials and different construction designs for walls and floors, often testing construction designs and materials in combination. Little of this business seems to have been administered through Materialprøvingsanstalten. On average, just one test a year is mentioned in the laboratory’s annual reports during the 1930s. The avoidance of the Materialprøvingsanstalten’s regime by the physics department is perhaps not surprising: the laboratory’s testing procedures were rather formalized in character, resulting in a standardized test certificate. Because Materialprøvingsanstalten was officially approved as a public testing institution, this document had a legal status and could be used as documentation, for example, by a materials manufacturer. But most of the acoustical tests carried out by Holtsmark and Berg were on a different basis, usually commissioned by the user and not the producer of the materials. The test series also tended to be comparative, contrasting and optimizing different materials and construction designs in particular combinations.56 In addition, Holtsmark and Berg’s investigations were less standardized than the tests of the materials testing laboratory; Holtsmark wanted to conduct more open-ended scientific research and not primarily run a testing facility for the construction industry. In his consultancy work, he usually tried to persuade his client to let him investigate the problems on a scientific basis. Holtsmark wanted the investigations to be more prolonged and theoretically oriented.57 He argued that problems should be investigated in their profundity rather than treated superficially to serve temporary needs—in the long run, the client would profit from such scientific studies on the foundations of the technology under investigation.58 Of course, Holtsmark had to be sure to study the physical principles of the acoustics of construction materials and designs to stay within the boundaries of his own discipline: in the physics laboratory, the problems had to be tackled at their physical origin. The place of
258 Managing Sound, Assessing Space
architectural acoustics and testing materials within the discipline of physics could easily be contested, as is shown by the case of Hermann Reiher at the Institute for Sound and Heat Research in Stuttgart. When Reiher asked to be appointed professor of technical physics at the Technische Hochschule Stuttgart, the professor of physics, Erich Regener, refused on the grounds that Reiher’s investigations were merely practical and that Reiher lacked interest in studying the physical foundations of the phenomena.59 Holtsmark would most likely have agreed with Regener’s line of argument. In his view, the fact that acoustic knowledge and practices were mainly for local consumption did not mean they did not need to be scientifically grounded—quite the contrary.60 During the 1930s, three rooms in NTH’s physics department were at last converted to the acoustic testing laboratory,61 investigating two different types of constructions: walls and floors. Measurements of wall constructions in the sound laboratory had started as early as 1931.62 The constructions were tested either in test fields arranged in an opening between two rooms, to measure sound transmission, or within one lab room, to measure reflection and reverberation. All recordings were by microphone. The measurements were conducted over a range of audible frequencies using standardized gramophone test records.63 The instrument collection of the NTH physics department today includes an electroacoustic tone generator built there in the 1930s, suggesting that it was also used, with a loudspeaker, as a sound source in measurements. The testing of floor constructions required a test field between different stories of the building. Measurements of sound insulation or sound reflection qualities of wall constructions were always based on airborne sound, generated by means of test records or tone generator, whereas the sound insulation qualities of floors against airborne sound were of secondary interest. The researchers mainly investigated the floor’s insulation properties against impact sounds created by people walking. To produce standardized impact sounds, the Trondheim acousticians built a tapping machine that dropped small hammers onto the floor (Figure 9.2). For impact sound, sound intensity was not measured over single frequencies, as for wall insulations, but summarized across all frequencies. To account for the frequency sensibility of the ear, an ear filter had to be applied. Figure 9.3 shows the laboratory arrangement with the tapping machine and the measuring apparatus with microphone, amplifier, ear filter, and sound intensity meter. In 1931, the testing of the insulation of different floor constructions against impact sounds started at the initiative of the Oslo Municipal Sound Insulation Committee and with the support of twenty-seven interested businesses. The
Figure 9.2 Tapping machine for generating impact sound on the flooring to be tested. Reno Berg, Finn Berner, Edvard Harboe, and Johan Holtsmark, “Lydisolasjon mot bankelyd i gulvkonstruksjoner,” Byggekunst 16 (1934): 134.
BANKE.APPARAT.
MIKROFON.
FORSTERKER
φ REFILT
Figure 9.3 Setup of the laboratory for measuring impact sound on floors. Reno Berg, Finn Berner, Edvard Harboe, and Johan Holtsmark, “Lydisolasjon mot bankelyd i gulvkonstruksjoner,” Byggekunst 16 (1934): 134.
260 Managing Sound, Assessing Space
committee was one of many municipal and professional bodies instituted in Europe and North America in the late 1920s and early 1930s to deal with urban noise. As noise was increasingly perceived as a problem in modern cities, the committees became involved in noise abatement, noise measurement, standards, and regulation.64 Beginning in summer 1932, Reno Berg tested different floor assemblies. The first laboratory was set up not at NTH in Trondheim, but in the unfinished building of the Norwegian School of Veterinary Science in Oslo, probably because the test field in the Trondheim physics department, requiring a square hole of 2 by 2 m in the cement ceiling, had not yet been created. Instead, Berg took the laboratory instruments to the construction site. At least seventy-seven floor constructions were tested in the laboratory in the veterinary building and were discussed in a 1934 publication.65 The test field at NTH in Trondheim was finished by 1935.66 For laboratory measurements, the acoustic properties of the laboratory walls, ceiling, and floor had to be considered. In 1937, Holtsmark and Berg started to investigate the relationship between the insulation of wooden floors for airborne sound and for impact sound. For these measurements, the airborne sound insulation of the laboratory ceiling around the test field proved insufficient, and a sound- insulating wooden box had to be built over the test field.67 These laboratory setups show the extent to which sound measurement had become electrified. Whereas earlier generations of acousticians had relied on their listening skills and on mechanical devices, Holtsmark and his assistants used electric devices for sound production, detection, and analysis. Sound was reproduced from a test record by means of an amplifier and a loudspeaker, or even generated purely electrically by a tone generator. Microphones took on the task of listening, with filters emulating the sensitivity of the human ear. In the reverberation apparatus, time was also measured electrically through the discharge of a condenser. But electroacoustics went beyond instrumentation to become a new language for thinking and talking about sound.68 In several publications on the sound insulation of wall constructions, Berg and Holtsmark made use of equivalent circuit diagrams to represent the wall as a mechanical oscillating system. Such analogies were used in papers detailing both theoretical and experimental investigations,69 and equivalent circuit diagrams frequently served to represent acoustical vibrations in scientific papers.70 In 1930, Holtsmark added a course on acoustics, taught by his assistants, to the electrical engineering curriculum. The course was rich in analogies between electrical and acoustical oscillations, where the acoustic variables of force, speed, displacement, mass, and elasticity were translated into the electric variables of tension, current, charge, self-induction, and
Absorption, Transmission, Reflection 261
capacity.71 The use of equivalent circuit diagrams and other electroacoustic analogies in publications and in the NTH acoustics lecture script shows the presence of an electro-analog language and understanding of acoustics not only in research instrumentation but also in its representation and communication in teaching and research at NTH and elsewhere.72
Conclusion The NTH laboratory was one of many acoustic laboratories founded in Europe and North America in the interwar period. As new media technologies, new acoustic knowledge, and new practices of measuring sound spread rapidly across the world, they created local demand for implementation and expertise, for acoustic research, testing, and consultancy. In this context, Holtsmark managed to establish himself as the scientific expert on architectural acoustics and acoustic testing in Norway, and setting up the acoustic laboratory was central to his claim to expertise. His work in acoustics was situated simultaneously in a local, a national, and an international knowledge economy. Holtsmark’s acoustic consulting for NRK was part of a national framework to establish radio broadcasting in Norway. Acoustic testing of construction materials and construction designs served a local and regional construction industry but also facilitated the spread of sound motion film in Norway by adjusting existing motion picture theaters to the new medium. At the same time, Holtsmark was an active member of an international acoustic community and a Norwegian delegate in negotiations on the international acoustic standards and units that were essential to sound motion film and radio as technological systems. Holtsmark’s engagement with acoustics and the testing lab thus offers insights into a specifically local and Norwegian story of acoustics testing and consulting, but one that is embedded in the transnational transformation of acoustics research and the global spread of sound movies, radio broadcasting, and the standardized testing of construction materials. Holtsmark was a newcomer to acoustics when he set up the Trondheim laboratory in 1929. By the mid-1930s, however, he and his assistants had built up extensive expertise in constructing electroacoustic instrumentation, relying heavily on young and motivated electrical engineers with a background in amateur radio. The Trondheim acoustic laboratory exemplifies how acoustics research practices had changed in the interwar period. Earlier acoustic instruments were part of the instrument cabinet, stored away in the collection, then taken out and assembled on a table for experimentation—but the interwar acoustic laboratories were permanent setups, requiring specific
262 Managing Sound, Assessing Space
laboratory architecture and spreading over several rooms. Their instrumentation was modular and electrical. Radio engineering skills were also needed to run and interpret the tests and experiments, which explains the dominance of electrical rather than civil or mechanical engineers. Acoustic testing laboratories included rooms for measuring sound transmission through walls and floors, reverberant rooms, and ultimately anechoic chambers. Acoustics was not the only research field that underwent these transformations into specific laboratory architectures and modular instrumentation in the period: the transformation of radioactivity research into nuclear science and the particle accelerator laboratory is another prominent example, illustrated by the Trondheim Van de Graaff accelerator laboratory, but it was the acoustic laboratories that pioneered the use of radio technology and electric amplification before these entered and dominated other domains, including particle acceleration. Acoustic testing of materials in the laboratory contrasted with architectural acoustics in the field. Acoustic laboratories allowed scientists to retreat from complex and messy spaces such as lecture and concert halls, cinemas, recording studios, and private living spaces. Only in the laboratory could scientists fully isolate the acoustic properties of particular materials and designs, reducing them to absorption, transmission, and reflection, and for these to be tested in a standardized way, the laboratory had to be an acoustically neutral space. What it meant to be acoustically neutral, however, depended on the test to be carried out—the acoustically neutral laboratory was an idealization. The example of the acoustic laboratory of NTH shows how such spaces were created in practice and how they operated in the contexts of acoustic research, testing, and consulting. In Norway and elsewhere, construction materials were increasingly mass produced. The resulting standardization demanded increasing testing of construction materials, which constituted the vast majority of materials tested at NTH’s materials testing laboratory. The manufacturing of materials and the application of construction design and practices lost much of their local specificity as modern economic and industrialist logic, with its categories of production and consumption, began to dominate construction planning and execution. Ironically, as Holtsmark himself pointed out, there was a contradiction between economic optimization and progress: within one generation, construction materials had begun to be produced on the basis of modern engineering, but their quality, including their acoustic properties, had declined considerably.73 Holtsmark was not alone in criticizing the new acoustics. Hope Bagenal and Alex Wood worried in Planning for Good Acoustics in 1931: “Serious acoustic complaints often arise in the modern home. Cheap
Absorption, Transmission, Reflection 263
rapid domestic building has now superseded the slow massive kind. But mass is the first and most important defence against noise . . . and cannot be compensated for adequately by thin patent materials, however well advertised.” Bagenal and Wood concluded that “reinforced concrete structure,” in many ways the quintessence of modern architecture, “should not be used for residential or domestic buildings.”74 If the logic of standardized testing was not to improve the overall quality of construction but to optimize the cost of construction while minimizing the volume of construction material needed, this seems to have placed it directly at odds with the good acoustics we might have expected it to accomplish.75
Notes 1. At the same time as he was setting up the acoustic laboratories, Holtsmark continued to work on electron scattering and in 1933 started to build a Van de Graaff generator, which became the first particle accelerator in the Nordic countries. See Roland Wittje, “Acoustics”; Wittje, “Proton Accelerator.” 2. See Roland Wittje, Age of Electroacoustics, 200–201. 3. Emily Thompson, Soundscape of Modernity. 4. Paolo Brenni, “Physics Instruments.” 5. Karl-Heinz Manegold, Universität; Robert Fox and Anna Guagnini, Education, Technology, and Industrial Performance. 6. Adolf Martens and Max Guth, Das Königliche Materialprüfungsamt, 4; Walter Rusek, 100 Jahre Materialprüfung. 7. When the Physikalisch-Technische Reichsanstalt was established as a body independent of the university system, its founders worked hard to demarcate its mission and mandate from the existing testing laboratories at the universities and Technische Hochschulen, and to negotiate the influence of the engineering disciplines on the new institute. Cahan, Institute for an Empire, 53–56. 8. Jochen Stark and Bernd Wicht, Geschichte der Baustoffe, 14. 9. Sigfried Giedion, Befreites Wohnen, 6; see also Friederike Mehlau-Wiebking, Richard Döcker, 127–29. 10. Ernst May, “Das soziale Moment.” 11. Giedion, Befreites Wohnen, 11. 12. Otto Wawriziniok, Handbuch. 13. Richard Berger, Schalltechnik, 69. 14. Thompson, Soundscape of Modernity, 120–28, Karin Bijsterveld, Mechanical Sound. 15. Wittje, Age of Electroacoustics, 110–11, 204–205. 16. Thompson, Soundscape of Modernity, 173–75. 17. Oscar Knoblauch, Geschichte des Laboratoriums. 18. Richard Berger, Über die Schalldurchlässigkeit. 19. Thompson, Soundscape of Modernity, 175–90. 20. Ibid., 35–36.
264 Managing Sound, Assessing Space 21. See, for example, Richard Heger, “Zur Theorie.” 22. Thompson, Soundscape of Modernity, 190–207. As Thompson points out, acoustically designed construction materials were pioneered in the United States, but by the 1930s they were also widely available in Europe. 23. See, for example, Paul E. Sabine, Acoustics and Architecture, 90–97, 232–81. 24. Wittje, Age of Electroacoustics, 163–69. 25. The notions of “technical physics” and “technical acoustics” became common in Germany and other European countries after World War I to denote the kind of physics or acoustics related to industrial practice. Ibid., 116. 26. Johan Holtsmark, “Noen erindringer fra akustikkens barndom i Norge.” Manuscript for a lecture given at Trondheim on May 14, 1968, for the opening of the Department of Acoustics at NTH, 2, archive of the Acoustics group at the Department of Electronic Systems, Norwegian University of Science and Technology (NTNU), Trondheim; my translation. 27. Johan Holtsmark, “Akustiske problemer,” 281. 28. Johan Holtsmark, “Zur Definition”; Holtsmark, “Zur Theorie”; Holtsmark, “Die Schalldurchlässigkeit.” 29. Wittje, “Acoustics,” 104. 30. Ibid., 91–96. 31. Wittje, Age of Electroacoustics, 200. 32. Helmer Dahl and Arnljot Strømme Svendsen, Vebjørn Tandberg. 33. Vebjørn Tandberg, writing in Otto Delphin Amundsen, Vi fra NTH, 24; Wittje, “Acoustics,” 134–38. 34. In the following, I use the abbreviation NRK for Norsk Rikskringkasting (the Norwegian Broadcasting Corporation), though this abbreviation was not yet in common use during the 1930s. See Hans Fredrik Dahl, Hallo-Hallo, 14. 35. The concurrent production of measurement apparatus for both the NTH acoustic laboratories and the NRK was part of Holtsmark’s strategy. In the case of the decibel meter, Holtsmark argued that although the apparatus could be bought from abroad, building it at the NTH workshop would both be cheaper and develop local instrument-making skills. Wittje, “Acoustics,” 131. 36. The choice of German rather than English as the international language did not reflect a proximity on Holtsmark and Berg’s part to German rather than American research practices. Most of their references were to papers by U.S. researchers, especially from Bell Laboratories, with papers by German researchers coming second. Nevertheless, German was still the preferred international academic language for Holtsmark and other Norwegian scientists and engineers who had studied in Germany. 37. Nevertheless, Holtsmark’s research on architectural acoustics was noticed to some extent in Germany. Josef Engl referred to Holtsmark’s sound insulation measurements in his 1939 Raum-und Bauakustik, but not to papers from the Royal Norwegian Society. The only publication listed by Engl was Holtsmark and Tandberg’s 1933 paper on the reverberation apparatus: Johan Holtsmark and Vebjørn Tandberg, “Ein transportabler Apparat.” Engl, Raum-und Bauakustik, 291, 358. 38. Johan Holtsmark, “Akustikk.” 39. Holtsmark to President of NTH, November 24, 1931, Archive NTH-Fysisk Institutt, Eb:1, Statsarkivet i Trondheim.
Absorption, Transmission, Reflection 265 40. Reno Berg probably spent some time at Meyer’s acoustics department in summer 1935. See Holtsmark’s request to Meyer, June 4, 1935, Archive NTH-Fysisk Institutt, Eb:1, Statsarkivet i Trondheim. I do not have documentation confirming Berg’s stay. 41. Wittje, Age of Electroacoustics, 7–8. 42. Holtsmark to Olaf Devik, November 24, 1940, Archives of the Norwegian Broadcasting Corporation, NRK Oslo. 43. For 1939–40, Holtsmark received a contribution of NOK 15,000 for the acoustics laboratories for equipment, salaries, and other expenses. Telegrafstyret to Norsk Rikskringkasting, “Stötte til akustikk-laboratoriet ved Norges Tekniske Höiskole,” April 28, 1939, Archives of the Norwegian Broadcasting Corporation, NRK Oslo. The more complicated relationship between Holtsmark and the NRK is discussed in Wittje, “Acoustics,” 126–33. 44. In 1939 and 1940, Holtsmark had billed NOK 2,000 in fees for himself. See F. Kjelstrup, “Ad brev av 3/1-41 fra professor Holtsmark om akustiske forsøk,” January 9, 1941, and Holtsmark to Norsk Rikskringkasting, “Akustiske forsøk,” January 17, 1941, both Archives of the Norwegian Broadcasting Corporation, NRK Oslo. Consulting work by NTH professors for Norwegian industry is discussed in Tore Jørgen Hanisch and Even Lange, Vitenskap for industrien, and Ketil Gjølme Andersen and Gunnar Yttri, Et forsøk verdt. Industrial consulting by NTH engineering professors seems to have been largely uncontroversial and publicly supported. Contacts made through consulting demonstrated a professor’s proximity to Norwegian industry and ability to solve practical problems. At times, however, the neutrality of professors with strong industrial contacts was questioned. 45. I draw here on Bruno Latour, Pasteurization of France, esp. 75, where Latour talks about moving “the laboratory to the place where the phenomena to be retranslated are found.” 46. Holtsmark and Tandberg, “Ein transportabler Apparat,” 389. 47. Holtsmark, “Noen erindringer fra akustikkens barndom i Norge,” 7. 48. Holtsmark remembered from his visit to Leipzig in 1913 that although the new Leipziger Gewandhaus of 1884 exhibited good acoustic qualities, it could not rival the old Gewandhaus. Efforts in America to reproduce the old Leipziger Gewandhaus on a larger scale and with other materials clearly failed to match its acoustic qualities. Holtsmark, “Noen erindringer fra akustikkens barndom i Norge,” 1. 49. The Handelsstandens Hus was an office building in functionalist style built between 1932 and 1934 by the city’s commercial association, Trondhjems Handelsstands Forening. See Reno Berg and Johan Holtsmark, “Målinger av lydisolasjon.” 50. Reno Berg and Johan Holtsmark, “Akustiske målinger.” 51. Changes in hearing habits, and the consequent change in valuation of particular locations’ acoustic properties, are discussed in Thompson, Soundscape of Modernity, particularly 330–31. 52. Engl classified premises according to whether they prioritized intelligibility of speech (such as lecture halls, assembly halls, and stages for speech) or the performance of instrumental and vocal music (such as concert halls, operas, and churches). Engl, Raum- und Bauakustik, chs. 11 and 12. The sound movie theater is treated separately; ibid., 345. 53. Reno Berg and Johan Holtsmark, “Målinger av efterklang.” 54. “Forskning og fremskritt,” an interview with Johan Holtsmark; also Holtsmark, Einige akustische Probleme, esp. 20–24; Holtsmark, “Noen erindringer fra akustikkens barndom i Norge,” 2, particularly for its mention of the Leipziger Gewandhaus; and “NeFAS’ memorandum,” July 30, 1937, Archives of the Norwegian Broadcasting Corporation, NRK Oslo.
266 Managing Sound, Assessing Space 55. See “Prøvingsanstalten gjennem 15 år (1912–1927),” reproduced in Norges Tekniske Høiskole—beretning, 73. 56. See, for example, Reno Berg et al., “Lydisolasjon mot bankelyd.” Examining seventy-seven different combinations of materials, the tests avoided highlighting certain materials and specifying particular companies. Holtsmark also repeatedly discouraged companies from having their construction materials tested acoustically, arguing that such tests made little sense because it was not merely the material itself but its use in construction that determined the acoustical effect. See, for example, Holtsmark to Materialprøvingsanstalten, November 2, 1939, Arkiv NTH- Materialprøvingsanstalten, NA: 75, Statsarkivet i Trondheim. 57. See, for example, Holtsmark to Telegrafstyret, “Søknad om bidrag,” April 5, 1939, Archives of the Norwegian Broadcasting Corporation, NRK Oslo. In this funding application, Holtsmark contrasted practical trials to serve immediate needs with long-term and more theoretically oriented tasks, arguing that NTH should favor the latter. 58. In the interwar period, Holtsmark was not alone in advocating fundamental studies in applied and industry-related research. See Friedrich Wilhelm Hagemeyer, “Die Entstehung,” 85, on fundamental research at the Bell Telephone Laboratories, and Carl Ramsauer, Physik, esp. 126, on the AEG research institute. 59. Wittje, Age of Electroacoustics, 167–68. 60. Holtsmark presented this line of argument in a 1935 article urging NTH and Norway to become involved in fundamental nuclear physics research. Johan Holtsmark, “Kunstig radioaktivitet,” 30; see Wittje, “Acoustics,” 183. 61. Asbjørn Krokstad, “Akustisk Laboratorium ved NTH,” five-page manuscript, ca. 1968, archive of the Acoustics group at the Department of Electronic Systems, Norwegian University of Science and Technology (NTNU), Trondheim. 62. Reno Berg and Johan Holtsmark, “Die Schallabsorption.” 63. See, for example, Reno Berg and Johan Holtsmark, “Die Ausbreitung”; Berg and Holtsmark, “Målinger av efterklang”; Berg and Holtsmark, “Målinger av lydisolasjon.” 64. Bijsterveld, Mechanical Sound, 110–24. 65. Berg et al., “Lydisolasjon mot bankelyd” (also published in Teknisk Ukeblad 81, no. 33, 1934). This article, written with Finn Berner, a professor of architecture at NTH, and the civil engineer Edvard Harboe, was Berg and Holtsmark’s only joint publication with architects and civil engineers. 66. The test field is mentioned in a letter from Holtsmark to Materialprøvingsanstalten, September 25, 1935, Arkiv NTH- Materialprøvingsanstalten, NA: 61, Statsarkivet i Trondheim. 67. Reno Berg and Johan Holtsmark, “Die Schalldämmung von Holzdecken”; Berg and Holtsmark, “Die Schalldämmung von Holzdecken II.” 68. William Henry Eccles, “New Acoustics”; Wittje, Age of Electroacoustics. 69. For example, Berg and Holtsmark, “Die Schallabsorption”; Holtsmark, “Zur Theorie.” 70. In “Die Schallisolation von Doppelwänden,” for example, Berg and Holtsmark simply copied the equivalent circuit diagram from someone else’s publication. To be sure, not all acousticians of the 1930s used electrical analogies in acoustics representation. Engl, for example, limited the use of circuit diagrams in his Raum-und Bauakustik of 1939 to the description of electrical measurement instruments. 71. Wittje, “Acoustics,” 101; and Bjørn Trumpy, Akustikk, 13–14.
Absorption, Transmission, Reflection 267 72. See Wittje, Age of Electroacoustics, and Trumpy, Akustikk, for a published version of the NTH lecture script. 73. “Forskning og fremskritt.” 74. Hope Bagenal and Alex Wood, Planning for Good Acoustics, 200, 201. Original emphasis. 75. Stark and Wicht, Geschichte der Baustoffe, 14.
References Amundsen, Otto Delphin, ed. Vi fra NTH, de neste ti kull, 1920– 1929. Oslo: Dreyers Forlag, 1950. Andersen, Ketil Gjølme, and Gunnar Yttri. Et forsøk verdt: forskning og utvikling i Norsk hydro gjennom 90 år. Oslo: Universitetsforlaget, 1997. Bagenal, Hope, and Alex Wood. Planning for Good Acoustics. London: Methuen, 1931. Berg, Reno, Finn Berner, Edvard Harboe, and Johan Holtsmark. “Lydisolasjon mot bankelyd i gulvkonstruksjoner.” Byggekunst 16 (1934): 132–40. Berg, Reno, and Johan Holtsmark. “Akustiske målinger i en del forsamlingslokaler i Norge.” Det Kongelige norske videnskabers selskabs Skrifter 32 (1935): 1–22. Berg, Reno, and Johan Holtsmark. “Die Ausbreitung des Luftschalls in Gebäuden II.” Det Kongelige norske videnskabers selskab forhandlinger 7, no. 14 (1934): 43–46. Berg, Reno, and Johan Holtsmark. “Målinger av efterklang og lydabsorpsjon utført i Handelsstandens Hus, Trondheim.” Det Kongelige norske videnskabers selskab forhandlinger 7, no. 44 (1934): 157–60. Berg, Reno, and Johan Holtsmark. “Målinger av lydisolasjon mot luftlyd i en del gulv-og veggkonstruksjoner utført i Handelsstandens Hus, Trondheim.” Det Kongelige norske videnskabers selskab forhandlinger 7, no. 43 (1934): 153–56. Berg, Reno, and Johan Holtsmark. “Die Schallabsorption einiger Wände und Decken.” Det Kongelige norske videnskabers selskab forhandlinger 4, no. 36 (1931): 127–30. Berg, Reno, and Johan Holtsmark. “Die Schalldämmung von Holzdecken.” Det Kongelige norske videnskabers selskab forhandlinger 10, no. 46 (1937): 173–76. Berg, Reno, and Johan Holtsmark. “Die Schalldämmung von Holzdecken II.” Det Kongelige norske videnskabers selskab forhandlinger 12, no. 41 (1939): 149–52. Berg, Reno, and Johan Holtsmark. “Die Schallisolation von Doppelwänden I. Holzwände.” Det Kongelige norske videnskabers selskab forhandlinger 8, no. 23 (1935): 75–78. Berger, Richard. Die Schalltechnik. Braunschweig: Vieweg, 1926. Berger, Richard. Über die Schalldurchlässigkeit. Munich: Oldenbourg, 1911. Bijsterveld, Karin. Mechanical Sound: Technology, Culture, and Public Problems of Noise in the Twentieth Century. Cambridge, MA: MIT Press, 2008. Brenni, Paolo. “Physics Instruments in the Twentieth Century.” In Science in the Twentieth Century, edited by John Krige and Dominique Pestre, 741–58. Amsterdam: Harwood Academic, 1997. Cahan, David. 1989. An Institute for an Empire: The Physikalisch-Technische Reichsanstalt, 1871–1918. Cambridge: Cambridge University Press. Dahl, Hans Fredrik. Hallo-Hallo!: Kringkastingen i Norge 1920–1940. Oslo: Cappelen, 1975. Dahl, Helmer, and Arnljot Strømme Svendsen. Vebjørn Tandberg: triumf og tragedie. Bergen-Sandviken: Fagbokforlaget, 1995. Eccles, William Henry. “The New Acoustics.” Proceedings of the Physical Society 41 (1929): 231–39. Engl, Josef. Raum-und Bauakustik— ein Leitfaden für Architekten und Ingenieure. Leipzig: Akademische Verlagsgesellschaft, 1939.
268 Managing Sound, Assessing Space “Forskning og fremskritt—Høittaler kan bygges helt fullkomne.” Interview with Professor Holtsmark. Oslo Illustrerte, no. 5 (1934): 12. Fox, Robert, and Anna Guagnini, eds. Education, Technology and Industrial Performance in Europe, 1850–1939. Cambridge: Cambridge University Press, 1993. Giedion, Sigfried. Befreites Wohnen. Zurich: Orell Füssli, 1929. Hagemeyer, Friedrich Wilhelm. “Die Entstehung von Informationskonzepten in der Nachrichtentechnik: Eine Fallstudie zur Theoriebildung in der Technik in Industrie-und Kriegsforschung.” PhD diss., Freie Universität Berlin, 1979. Hanisch, Tore Jørgen, and Even Lange. Vitenskap for industrien: NTH–En høyskole i utvikling gjennom 75 år. Oslo: Universitetsforlaget, 1989. Heger, Richard. “Zur Theorie und Praxis der Raumakustik.” Zeitschrift für Architektur und Ingenieurwesen 16/57, no. 4 (1930): 309–22. Holtsmark, Johan. “Akustikk.” Hallo-Hallo 12, no. 39 (1937): 3–4. Holtsmark, Johan. “Akustiske problemer.” Teknisk Ukeblad 47, no. 27 (1929): 279–81. Holtsmark, Johan. Einige akustische Probleme im Rundfunk. Oslo: Norsk Rikskringkasting, 1936. Holtsmark, Johan. “Kunstig radioaktivitet.” Teknisk Ukeblad 82, no. 4 (1935): 28–30. Holtsmark, Johan. “Die Schalldurchlässigkeit geteilter Wände.” Det Kongelige norske videnskabers selskab forhandlinger 3, no. 16 (1930): 62–65. Holtsmark, Johan. “Zur Definition der Schalldurchlässigkeit von Wänden.” Det Kongelige norske videnskabers selskab forhandlinger 3, no. 14 (1930): 55–57. Holtsmark, Johan. “Zur Theorie der Schalldurchlässigkeit einer homogenen Wand.” Det Kongelige norske videnskabers selskab forhandlinger 3, no. 15 (1930): 58–61. Holtsmark, Johan, and Vebjørn Tandberg. “Ein transportabler Apparat zur Messung von Nachhalldauer und Schallintensität.” Elektrische Nachrichten-Technik 10, no. 10 (1933): 389–92. Knoblauch, Oscar. Die Geschichte des Laboratoriums für technische Physik der Technischen Hochschule München 1902–1934. Munich: Technische Hochschule, 1941–42. Latour, Bruno. The Pasteurization of France. Translated by Alan Sheridan and John Law. Cambridge, MA: Harvard University Press, 1988. Manegold, Karl-Heinz. Universität, Technische Hochschule, Industrie. Berlin: Duncker & Humblot, 1970. Martens, Adolf, and Max Guth. Das Königliche Materialprüfungsamt der Technischen Hochschule Berlin. Berlin: Springer, 1904. May, Ernst. “Das soziale Moment in der neuen Baukunst.” Das neue Frankfurt 2, no. 5 (1928): 77–83. Mehlau-Wiebking, Friederike. Richard Döcker: Ein Architekt im Aufbruch zur Moderne. Braunschweig: Vieweg, 1989. Norges Tekniske Høiskole, beretning for året 1925– 1926. Trondheim: Norges Tekniske Høgskole, 1927. Ramsauer, Carl. Physik, Technik, Pädagogik: Erfahrungen und Erinnerungen. Karlsruhe: Braun, 1949. Rusek, Walter. 100 Jahre Materialprüfung in Berlin: Ein Beitrag zur Technikgeschichte. Berlin: Bundesanstalt für Materialprüfung (BAM), 1971. Sabine, Paul E. Acoustics and Architecture. New York: McGraw-Hill, 1932. Stark, Jochen, and Bernd Wicht. Geschichte der Baustoffe. Wiesbaden: Bauverlag, 1998. Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the Culture of Listening in America, 1900–1933. Cambridge, MA: MIT Press, 2002. Trumpy, Bjørn. Akustikk—utvalgte forelesninger for elektroavdelingens 4. årskurs (linje for svakstrøm), 1. del. Trondheim: Tapirs forlag, 1930. Wawriziniok, Otto. Handbuch des Materialprüfungswesens. Berlin: Springer, 1908.
Absorption, Transmission, Reflection 269 Wittje, Roland. “Acoustics, Atom Smashing and Amateur Radio: Physics and Instrumentation at the Norwegian Institute of Technology in the Interwar Period.” PhD diss., Norges teknisk- naturvitenskapelige universitet, 2003. Wittje, Roland. The Age of Electroacoustics: Transforming Science and Sound. Cambridge, MA: MIT Press, 2016. Wittje, Roland. “A Proton Accelerator in Trondheim in the 1930s.” Historical Studies in the Physical and Biological Sciences 35, no. 1 (2004): 115–52.
10 Of Silent Sirens and Pied Pipers Auditory Thresholds and High-Frequency Technologies of Animal Control Joeri Bruyninckx
Audiogenic Seizures One spring night in 1939, while Jane and Clifford Morgan were cleaning glassware for one of their experiments, some of the rats in the laboratory began convulsing and finally lapsed into a coma. As advanced graduate students in experimental psychology at Rochester University, the Morgans were well familiar with what they identified as a seizure. That very spring, they had heard the psychologist Norman R. F. Maier explain at a Rochester colloquium how he had induced similar seizures in rats by driving them off a jumping stand with a sudden blast of compressed air. Only a year earlier, the American Association for the Advancement of Science (AAAS) had awarded Maier a prestigious prize for a paper in which he proposed that the seizures were actually neurological shortcuts, a result of the forceful internal conflict that he provoked in the rats by presenting them with an insoluble choice between two evils.1 But unlike Maier, the Morgans had not intentionally provoked their rats, let alone drawn them into a complex choice discrimination. Instead, they related the animals’ abnormal behavior to the only documented constant in the Maier experiments: the high-frequency hiss of the compressed air with which they had been drying their glass tubes. Within a few weeks, they rushed a preliminary article to the Journal of Comparative Psychology proposing, contra Maier, that the seizures were in fact audiogenic.2 The disagreement with Maier was pursued in a long exchange of papers, but after nearly a decade, the controversy resolved, at least publicly, in favor of the Morgans. The experimental reports seemed to validate a longstanding popular and medical concern with the role that acoustic stimulation played in various physiological and neurological pathologies.3 These reports convincingly showed that high frequencies could affect or even harm experimental animals Joeri Bruyninckx, Of Silent Sirens and Pied Pipers In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0011.
274 World as Testbed
but also, especially from the mid-1940s, gave credibility to suspicions of comparable effects on humans. In the wake of those fears, a series of academic, government-sponsored, and industry-sponsored inquiries were launched on the physical and affective powers of ultrasound—the range of waveforms above 20,000 Hz, and thus above human hearing—and finally on ways to instrumentalize them as a viable technology of sonic control. In this chapter, I take these investigations as a path into the history of testing practices, arguing that the presumed potency of high-frequency sounds hinged on a newly unfolding understanding of the relationship between human and nonhuman hearing. Following attempts by audiologists and psychologists to model human hearing on the hearing of experimental animals, a string of hearing tests now rendered nonhuman and human hearing comparable, yet still strangely incommensurable. As a standardized measurement practice, hearing tests were instrumental in making perceptual differences traceable and quantifiable; plotting these differences on a frequency spectrum, the hearing test made seemingly incongruous or even incompatible acoustic realities comparable. Among other things, that comparison showed the hearing thresholds of experimental nonhuman animals such as mice or rats to be markedly different from those of humans, often exceeding them well into the ultrasonic domain. I will show that these testing practices, and specifically the distinction between animal and human hearing thresholds that they brought into being, fed into a new promise of sonic control of bodies in the decades immediately following World War II. This promise became especially concrete with the emergence of a group of ultrasonic pest repellent technologies in the late 1950s. Hearing tests, and the technologies built on their results, made nonhuman hearing exploitable both physically and commercially, thus revealing a “politics of frequency.”4 Coined by Steve Goodman, this term seeks to capture the political potential that is exercised through sound, even in segments of the frequency spectrum inaudible to most humans. Goodman draws attention to the affective powers of the low-frequency vibrations he calls “unsound”; others have identified the politics of acoustic frequency as a means to dismantle or establish privileges for particular social groups.5 As Goodman and Mitchell Akiyama show, such means harbor a potential for sounds to be weaponized, for instance as high-frequency acoustic deterrents. Designed to interfere with hearing in a particular frequency range, they stratify subjects into those who are sonically affected and those who are not. Existing work has examined acoustic deterrents that target humans, but the direct precursors of such deterrents, pest control technologies, can reveal how the politics of frequency came to be inscribed in technology in the first place.6
Of Silent Sirens and Pied Pipers 275
The hearing test serves as a double point of entry into that politics. Its history suggests a possible genealogy for the idea of wielding frequency as an instrument of power, and the attempt to weaponize sound can be traced back to military and commercial experiments on the harmful effects of ultrasound, which drew inspiration from the Morgans’ demonstrations. But this very genealogy prompts us to consider the limits of a tactics of control that is exclusively concerned with frequency (see Jennifer Hsieh in this volume). The hearing test normalized frequency as a metric suited to comparing animal and human hearing, but in doing so, it pushed other parameters to the periphery. Although frequency was generally adopted to measure animal hearing, its users also soon discovered its shortcomings as a technique for sonically controlling nonhuman bodies. From this point of view, the history of ultrasound’s weaponization counters ontological claims about the affective potential of sounds, isolated from the particular auditory and experimental cultures in which they emerged.7 Taking the Morgans’ seizure experiments as a starting point, this chapter tracks the notion of nonhuman animals’ ability to hear and be affected by high-frequency sound as it came to animate scientific, military, and commercial research. I first examine how nonhuman animals’ hearing of sounds inaudible to humans came to interest Morgan and his colleagues at Rochester and Harvard University in the late 1930s and reconstruct the techniques they used to make nonhuman and human hearing measurable and comparable in the laboratory. I follow the appropriation of Morgan’s notion of audiogenic seizures immediately after World War II to explore ultrasound’s allegedly powerful effects and harness them commercially. Moving to the 1960s and 1970s, I illustrate the advent of one such consumer application, the ultrasonic pest repellent, which drew its legitimacy from popular reports on animal hearing and on ultrasound’s extraordinary effects. I then describe a backlash against high-frequency and high-intensity sound as a means of deterrence and the emergence of a wholly different approach, premised on a bioacoustical understanding of animal communication. Finally, I seek to explain the continued presence of high-frequency technologies of animal control despite their dubious effectiveness.
Testing the Limits of Animal Hearing The Maier controversy did much to raise Clifford Morgan’s professional visibility.8 In 1939, he moved to Harvard to work with the behaviorist Karl Lashley, focusing on food intake in white rats. As a result of his work on audiogenic
276 World as Testbed
seizures at Rochester, Morgan also quickly became part of a Harvard circle of physiologists and psychologists of hearing. This, and particularly Morgan’s acquaintance with audiologist Hallowell Davis and psychophysicist Stanley Stevens, shaped the intellectual and material context within which he and various collaborators began to determine the acoustic variables involved in inducing the seizures. Morgan had observed that the seizures were more likely when the rats were exposed to frequencies of around 20 kilocycles per second (or 20 kHz) and that the animals seemed to respond to even higher frequencies, inaudible to human hearers.9 Without reliable data available in the literature, in 1940 Morgan and his collaborator James Gould set out to test their rat subjects’ hearing thresholds in more detail, drawing on a conditioning method that had been refined in the 1930s at the University of Rochester, Morgan’s previous affiliation.10 While Morgan was a graduate student at Rochester, the experimental psychologist Elmer Culler was appointed head of the department. Culler, an expert in problems of hearing and learning, brought with him the Animal Hearing Laboratory that he had set up in Illinois with support from the American Otological Society. Its sound-isolated laboratories, surgery, and living quarters for experimental animals enabled Culler and his students to study and manipulate the auditory systems of nonhuman animals. Such work, generally applied to mammals such as rats, cats, dogs, or chimpanzees, ultimately aimed to enhance understanding of the mechanisms of human hearing. The Animal Hearing Laboratory was not the first to study nonhuman hearing. There is a long history of anecdotal observations on the mechanisms of hearing by animals other than humans, typically through histological and anatomical dissection. New electrophysiological methods and their application to studies of auditory nerve responses in the 1920s had alerted Culler to the topic. But although the group continued to use both these methods, it was best known for its motor conditioning techniques, which Culler had encountered and refined during a sabbatical year at the Cornell psychology department.11 By conditioning nonhuman subjects to respond to standard tones at particular frequencies, the group managed to transpose existing techniques for audiometric testing of humans onto animal subjects. Audiometry had already become a standard technique in the early 1920s, when American Telephone & Telegraph Co.’s investment in telephonic systems led its subsidiaries Western Electric Co. and Bell Telephone Laboratories to develop principles of audiometry into a standardized method of testing hearing for basic research in speech and hearing.12 Western Electric started to work on an instrument for measuring hearing sensitivity (and the loss
Of Silent Sirens and Pied Pipers 277
thereof) immediately after World War I and began distributing its first commercial electronic audiometer in 1922.13 Previous methods had relied on tuning forks, handclaps, or pistol shots to test auditory responsiveness, but the audiometer provided the greater degree of control and amplification necessary for measuring a subject’s sensitivity to the smallest perceptible differences in frequency.14 The audiometric test consisted of a series of pure tones, finely graded in frequency and intensity, played to test subjects who were asked to signal their awareness of them. This enabled the tester to plot a threshold of auditory perception in a new graphic form: the audiogram.15 By 1930, otologists had adopted this technique as a standard to test and compare large numbers of subjects, making it possible to specify minimum and maximum auditory thresholds for the average human listener.16 In the 1930s, experimental psychologists, too, embraced the basic principle of audiometry, if not the electric audiometer itself, in the attempt to establish a baseline measurement for their animal subjects’ hearing.17 Before-and-after measurements allowed them to assess the effects of their experimental lesions or destruction of the auditory system. To use audiometry on nonhuman animals, conditioning was vital as a way of making them into suitably capable listeners. Fixed in cages in a soundproof room, the animals were conditioned for a period of three to six months (Figure 10.1). They generally learned to associate an audible stimulus (here a pure tone, rather than a Pavlovian bell) with a mild electric shock or the delivery of food, prompting an appropriate involuntary behavioral response such as a respiratory movement or motor reaction that could be measured. The success of the conditioning relied on a precise and balanced management of affect; hence, techniques had to be continually adjusted for the temperament of each species or individual. The experiment thus not only tested the animals’ hearing sensitivity but also put them to the test as experimental subjects. Laboratory descriptions are rife with qualifications on this issue: cats, for instance, were found to be somewhat erratic in their performance, while dogs, whose hearing profile was found to be most similar to humans’, were more reliable. And whereas some individuals quickly learned to assume an optimal listening technique—tilting their head, holding their breath—others seemed overactive and tended to respond to any stimulus.18 Even “good” animals could become “nervous under the strain of repeated, intent listening for tones which are often just barely discriminable from silence,” with many reacting instead to their “own body-noises.”19 Other subjects simply refused to cooperate under test conditions. A small percentage were unable to overcome the initial fear of electric shock and died of starvation; others learned to outsmart the experimenters by responding to
278 World as Testbed
Figure 10.1 The animals’ soundproof compartment in the conditioning laboratory. A dog would be placed directly upon the table and held in place by a harness. S. Dworkin, J. Katzman, G. A. Hutchinson, and J. R. McCabe, “Hearing Acuity of Animals as Measured by Conditioning Methods,” Journal of Experimental Psychology 26, no. 3 (1940): 284.
the experimenters’ fumbling with the control dial instead of to the acoustic stimulus.20 Successful conditioning was key to mapping these animals’ hearing thresholds onto an audiogram that allowed them to be compared—with themselves (after anesthetized modification), with their conspecifics, or with the average human listener. In fact, audiograms of nonhuman hearing typically staged a cross-species comparison with the median human threshold. Such analogies underline animal hearing’s role as a model for understanding human hearing and as a testing ground for ameliorating hearing deficiencies. “By making comparative tests on our trained animals and on ourselves,” Culler assured his readers, “we have proved that a test-animal is fully as reliable a witness as a good human patient. Almost every problem that arises in the otological clinic with human beings can now be tested with experimental controls in the animal laboratory.”21 Even when the results were not immediately applicable to problems of human hearing, the human threshold added a sense of scale, illustrating the exact frequency bands in which the subject’s ability to pick up more or less loud tones departed from that of an average human listener. The animals tested tended to show comparable sensitivity in the range of human hearing, but greater sensitivity above that range. Perhaps surprisingly, the nonhuman
Of Silent Sirens and Pied Pipers 279
audiogram usually shared the scale of the human audiogram; its metric was often the same as that used by audiologists and telephone engineers. Given that the functional limitation of the standard clinical audiometer was set at a frequency of 8.192 kc/s, it is no wonder that Morgan searched in vain for data on the rat’s hearing at frequencies much higher than 8 kc/s—but even when audio oscillators were used to extend the testable frequency range, the metric of human hearing was maintained. In cases where high upper hearing thresholds were measured, most graphs summarizing hearing responses cut off at 16 kc/s, the generally accepted upper limit of human hearing at the time (Figure 10.2). The functional limitations of human hearing and the mechanical reproduction of sound thus significantly shaped testing practice. At Harvard, Morgan and Gould used the same combination of behavioral conditioning and audiometry to test the hearing sensitivity of the rat. With a beat-frequency oscillator, they collected measurements on absolute hearing thresholds for eight rats and nine human subjects between 1 and 14 kc/s, to compare their auditory sensitivity. Despite a focus on frequencies within the human hearing range, Morgan and Gould benefited from their association with work on ultrasonic frequencies by Harvard audiologists and physicists. Having borrowed a “crystal loud-speaker” from the physics laboratory of George W. Pierce (who had himself experimented with ways to translate ultrasonic frequencies into audible sound),22 they found that the rats showed remarkable auditory sensitivity even at 40 kc/s. Although the apparatus’s limitations prevented the researchers from determining a precise upper limit of
50 40 30
0
100
MAN DOG CAT CYCLES 200
14M
10
DECIBLES
20
400
M
2M
3M
5M
10M
16M
Figure 10.2 Auditory thresholds of cats and dogs compared with those of human subjects. S. Dworkin, J. Katzman, G. A. Hutchinson, and J. R. McCabe, “Hearing Acuity of Animals as Measured by Conditioning Methods,” Journal of Experimental Psychology 26, no. 3 (1940): 294.
280 World as Testbed
the rat’s hearing, this finding suggested that its optimum hearing probably lay between 15 and 60 kc/s.23 This clearly surpassed the auditory sensitivity of cats and dogs (the Animal Hearing Laboratory’s favorite test animals), whose upper limits had been determined as 50 and 38 kc/s respectively. That rats heard high frequencies better and thus more loudly than humans, the authors argued, was one probable cause of the audiogenic seizures.24 Morgan and Gould had tapped into their Harvard colleagues’ growing awareness of the acoustic territories that still lay uncharted. By 1940, one of Hallowell Davis’s graduate audiology students and a frequent collaborator of Morgan’s, Robert Galambos, using a different experimental technique that directly measured cochlear potentials in the bat ear, found that these exceeded his measurement instruments’ limit of 98 kc/s. With zoologist Donald Griffin and aided by George W. Pierce, Galambos also demonstrated that bats emitted ultrasonic squeaks that helped them in echolocation.25 Galambos’s findings led Gould and Morgan to speculate that with better sound-producing instruments, they might show the rat’s hearing spectrum to extend up to and beyond this mark as well. To be sure, these researchers were not the first to report on animals’ sensitivity to high frequencies. But technical advances in cochlear microphonic measurements, and in the instruments now available for generating and recording ultrasonic frequencies, made the extent of that sensitivity quantifiable with unprecedented exactitude.26 Such technologies, and the research by Morgan’s Harvard colleagues in audiology and psychology, intersected with military interests—for instance, in the harmful effects of noise on soldiers.27 The findings aroused further interest in animals’ extraordinary hearing abilities not only for their own sake but also for a host of military and commercial concerns during and after World War II (see Camprubí and Hui in this volume).
Silent Siren: Dangerous Sounds After World War II, air force personnel became increasingly worried about potential harm from the ultrasonic frequencies produced by the new turbojets introduced during the war.28 Although largely based on speculation, these concerns seemed validated by some clinical studies that attributed “ultrasonic sickness”—a set of indefinite symptoms such as nausea, fatigue, headache, and disturbances to neuromuscular coordination, along with psychological trauma and “feelings of impending doom”—to exposure to high-frequency noises.29 Audiologist Hallowell Davis, appointed by the Aeronautical Board as
Of Silent Sirens and Pied Pipers 281
chair of a new Ultrasonics Panel to address these worries, argued that public fear had been unnecessarily stoked by “weird stories” and the “ill-advised publication of unconfirmed, uncontrolled, and unanalyzed ‘observations’ at the anecdotal level.”30 To dispel these rumors, the U.S. military funded several research projects aiming to systematically investigate any effects of high-powered and high-frequency ultrasound on human and nonhuman mammals.31 One such project was conducted by a small group of physicists at Penn State University. The group was well situated to carry out this work. As part of their wartime research for the U.S. Army Signal Corps, exploring the use of “inaudible” high-frequency sounds for signaling in the field, laboratory director Harold Schilling and his collaborators had built the world’s most powerful ultrasonic siren. Ultrasonic generators such as the Galton whistle had been used in acoustic experiments since the nineteenth century (see Sebastian Klotz in this volume). But the “silent siren” produced frequencies in the ultrasonic range with an unprecedented level of energy—so much so that its sonic beams left uncomfortable sensations on the experimenters’ exposed skin and could set a cotton wad on fire in seconds. Further experimentation also revealed that the silent siren could burn mosquitoes in seconds and cook a mouse in its fur in under a minute. Reporting their discovery to the press, Schilling and his team made ultrasound out to be a multipurpose domestic and industrial technology, fit for laundering clothes, sterilizing food, killing germs, clearing fog, and keeping the home free of vermin. It was the pest control application that proved especially disconcerting to the public.32 Given its lethal effects on small animals, might this powerful phenomenon, though inaudible to humans, nevertheless have lasting (and possibly damaging) effects on organs, tissues, or the nervous system?33 Already, the press was reporting that earplugs had failed to protect experimenters against dizziness and loss of balance; elsewhere, the first claims for compensation were being made by laborers working around high-power sources of ultrasound such as jet engines.34 Just a few months after the siren story broke, the U.S. Air Force Wright Field Station in Ohio commissioned the Penn State research. By chance, the Penn State investigators were joined by the zoologist Hubert Frings. Initially, he advised the acousticians on measuring physiological effects in mice subjected to the siren, but he quickly assumed a more prominent role. Frings had previously worked on insect infestations, and his appointment at Penn State was partly sponsored by the local West Pennsylvania Pest Control Association. In exchange, he was expected to keep pest control professionals updated about progress in the field. Frings quickly realized that
282 World as Testbed
association with the project could help him gain access to funding and equipment as well as keeping his industrial sponsors happy.35 He managed to turn his assignment into a series of research contracts by exploring the siren’s potential applications for pest control. Although ultrasound was challenging to apply, as Frings noted in the trade journal Pest Control in 1948, its effects were far from mysterious. Referring to concerns over undetermined ultrasonic sickness, he concluded that ultrasound’s powers were special but hardly unpredictable or “spooky.”36 It proved more difficult than expected to wield those powers to practical effect. Contrary to popular predictions, the research team soon discovered that it would be extremely costly and impractical to scale up the siren to produce enough energy to destroy pests in the numbers they had hoped. However, Frings and his wife Mable (herself the holder of a bachelor’s degree in zoology and a prolific producer of bibliographic studies) picked up on the audiogenic seizures that Morgan and his various collaborators had reported in the literature. With the sponsors’ endorsement, they shifted tactics, setting up a research program that aimed to induce audiogenic seizures in their experimental white mice.37 Hubert and Mable Frings believed that if sonic stress could induce destructive tendencies in small animals, this could help answer the U.S. Air Force’s questions concerning the trauma reputedly caused by their jet engines. At the same time, Frings hoped it would satisfy his industrial sponsors’ search for effective repellents that were safe for humans.38 In 1950, the Fringses set up a laboratory with an inexpensive anechoic chamber and, helped by around fifteen undergraduate assistants, carried out basic experiments to determine the neurochemical and genetic background that made mice susceptible to seizures (Figure 10.3). The Fringses’ results were promising and met with an eager reception from both of their sponsors, the military and the pest control association, who encouraged them to continue exploring the application of sound and ultrasound as a repellent for other species. Pest control professionals had recently been experimenting with prerecorded noises resembling gunshots to scare off birds in storage hangars and around city architecture. Realizing that such noises would not only temporarily disturb the birds but also bother their human neighbors, the Fringses proposed to investigate the use of ultrasonic “shots” of pure tones instead. These would be inaudible to humans but potentially effective in scaring off pests. The U.S. Air Force offered to fund the Fringses to scale up their acoustic-repellent research into the hearing ranges of particular pests, such as the starling. Morgan’s notion of audiogenic seizure thus provided the foundation for an extensive investigation whose many
Of Silent Sirens and Pied Pipers 283
Figure 10.3 “Dr. Hubert Frings and Mable Frings, 1952,” University Photographics records (9825). Pennsylvania State University Archives, Eberly Family Special Collections Library, Penn State University Libraries. Used with permission from the Eberly Family Special Collections Library, Penn State University Libraries.
iterations would occupy Hubert and Mable Frings and their collaborators for the better part of the 1950s. In seeking to match specific data on animal hearing ranges with the repellent potential of intense high-frequency sounds, this research program integrated the laboratory animal hearing test into its attempts to develop effective sonic control. As the next section shows, the Fringses were not alone in developing such tactics into an application with commercial promise.
Ultrasonic Pied Piper: Bandwidth Control At the end of the 1950s, the first commercial ultrasonic pest repellents, using pulse tones in the high-and ultrasonic frequency spectrum, entered the consumer market. Over the next few decades, devices were sold as effective techniques for keeping granaries, gardens, and industrial facilities free of insects, rodents, and even larger mammals—or as a way of disciplining pets by conditioning the animal to abstain from barking or by restricting its ranging area. Tones in the ultrasonic frequency spectrum were said to
284 World as Testbed
be uncomfortable or unpleasant to a whole range of nonhuman animals. Although developers in some cases also claimed sonic powers that could be incapacitating or even deadly,39 more typically the instruments were promoted as a nonlethal way of repelling pests. Ultrasonic automatic whistles were advertised for preventing animal collisions with vehicles, and acoustic scarecrows were promoted as a more animal-friendly—or at least more ecological—means of control than mechanical traps or chemical repellents. One of the first instruments was marketed as the Ultrasonic Pied Piper, a playful reference to the folktale figure who lured away a rodent plague from the city of Hamelin using a magical pipe. Inventors and manufacturers of ultrasonic scarecrows did not embed their prototypes in as rigorous an experimental and testing regime as the one set up by the Fringses. Patent applications show that their claims to efficacy were generally founded on a popular understanding of recent findings in psychoacoustics and comparative psychology, which they mobilized to powerful effect. Two discursive sources were key to their authority. First, developers cited reports in the popular press describing the “discovery” that high-intensity ultrasonic waves had incapacitated insects and small rodents. Patenting and commercial advertising materials of the first designs uncritically referenced a single 1947 press report, “Army Engineers Kill Mice with New Sonic Wave,” on the Penn State silent siren.40 Second, they cited pests’ extreme aural sensitivity to high-frequency sound, usually without explicit reference to the literature. This suggests that such findings—established as fact in the scientific literature of the 1930s and 1940s—had nested in popular awareness. In time, these separate claims became not only intertwined but also conflated. Thus, Lowell Moe, who developed the ultrasonic system commercialized as the Ultrasonic Pied Piper, explained that the upper limit of the normal audio hearing spectrum for humans is approximately 15,000 cycles per second. Sound waves having frequencies beyond this range cannot be heard by humans, nor do these sound waves have any effect on the nervous system of humans. However, sound waves having ultrasonic frequencies, that is, frequencies beyond approximately 15,000 cycles per second, can be heard by and have a definite adverse effect upon small animals. . . . An ultrasonic sound wave of approximately 20,000 cycles per second not only can be heard by an average sized rat, but it also has an extremely irritating effect upon the nervous system of the rat.41
Equating the ability to hear ultrasound with the ability to be affected by it, Moe distinguished human users from nonhuman pest subjects. In doing so,
Of Silent Sirens and Pied Pipers 285
he implicitly dismissed lingering concerns over ultrasonic sickness while also extending the reach of existing experimental reports, according to which only a fraction of nonhuman animals had effectively been shown to be susceptible to audiogenic trauma. Taking advantage of the difference in hearing thresholds for different species, ultrasonic repellents broadcast an “irritating” signal just above the threshold of human hearing but well within the acute hearing spectrum of most pests, thus targeting those species’ hearing without jamming human acoustic awareness. In fact, some high-frequency sonic repellents even offered the opportunity to target a specific frequency range, which affected mosquitoes or rodents at higher ultrasonic frequencies that would remain inaudible to the family pet.42 As students of human–animal relations have pointed out, geography plays an important role in defining animals as food, pet, or pest, as useful or sentient.43 In that light, ultrasonic repellent technology may be seen as a way to further invigorate such culturally defined relations. It did so, at least conceptually, by repositioning animals relative to humans on a spectrum of auditory sensitivity—so that, in many ways, the dog seemed a more human-like listener than the rat. But it also promised to convert this perceptual difference into a spatial relation, suggesting a topographical boundary that located animals either inside (for pets) or outside (for pests) particular zones of human activity. Developers of acoustic repellents thus not only drew—however loosely— on a specific factual body of testing knowledge but also came to share some of the logic of the hearing tests I have described. As Mara Mills has argued, psychoacoustic research in the early twentieth century was driven by telephone engineers’ attempts to make telephone communications more economical.44 They did so partly by limiting the available bandwidth for telephone users to those frequencies that, according to psychoacoustic research, were essential for human speech to be intelligible to a listener over the phone. Jonathan Sterne notes that this allowed communication engineers to repurpose the bandwidth not required by these “surplus” frequencies to fit multiple conversations on a single line.45 In a strikingly similar way, inventors repurposed a body of psychoacoustic findings on the perceptual limits of human and nonhuman hearing to exploit a section of the acoustic spectrum not accessible to human listeners, using it for a commercial application of sonic deterrence.46 Relatively straightforward and inexpensive to produce, high-frequency pest repellent technologies quickly found their way to the consumer market.47 The devices, fitted in sturdy metal casings, were initially aimed at farming professionals, suburban pet owners and house owners, campers,
286 World as Testbed
and outdoor hobbyists. In subsequent decades, they were transformed into portable pens or stylish mahogany boxes, allowing them to be inconspicuously integrated into the home, the restaurant, or the corporate office.48 They were sold by small companies with names such as Sonic Technology, Supersonic Industries, or Rat- X. Initially marketed through regional newspapers among the small ads for do-it-yourself, cleaning, and gardening, from the 1980s onward they achieved a higher profile through favorable publicity in pet, agricultural, and popular scientific magazines as well as in trade and scientific journals such as the Journal of the Acoustical Society of America. Because it was inaudible to humans, the potency of commercial ultrasound did not always become apparent by itself. Homefree, the manufacturer of an ultrasonic pest control device, asked the audience of its radio advertisement to listen to four seconds of silence: “You can’t hear it work. You can only see the results.”49 Paper advertisements were complemented by front-of-shop demonstrations and instructive analogies: one salesman noted that the device emitted a sound that “to the rat’s ears” would be comparable in loudness to the audible sound of a revving motorcycle engine 50 feet from a human ear. Another began his demonstration with a screeching blast of audible sound that had prospective customers cover up their ears while he explained that its high-frequency (inaudible to them, but equally potent) equivalent would simply “zap” the pests away.50 Just as earlier hearing experiments had projected an analogy between animal and human hearing, these demonstrations made ultrasound’s potency to the animal testable through the human ear. They illustrate the gap that developers of ultrasonic pest repellents successfully straddled, at least for a while, by demonstrating the inaudible effectiveness of sonic energy while asserting its safety to human listeners. Ultrasonic repellents tapped into a space-age fascination with sonic force fields and nerve-crushing sound waves.51 A UNESCO Courier issue dedicated to noise control, for instance, reported on “the dangers of sounds we cannot hear,” juxtaposing articles on the benefits of ultrasonic repellents with a feature on the potential uses of low-frequency sound generators as acoustic weapons and a report on medical concerns regarding the physical effects of sound.52 Mobilizing popular awareness of ultrasound and its potential impacts, developers on the one hand managed to validate the counterintuitive claim that sounds well outside the periphery of human hearing could still exert observable, powerful effects. On the other hand, they presented ultrasound as a benign technology whose sonic radiation was not all-permeating but could be carefully curtailed and restricted to particular frequency bands.53
Of Silent Sirens and Pied Pipers 287
Bird-E-Vict: From Tone to Signal In spite of the subtle reframing of ultrasound as a benign but effective technology of control, animals again largely failed to cooperate. Endorsements for the devices continued to be published in trade journals and newspapers, but a growing chorus of professionals in pest control began to voice concerns that the manufacturers’ claims were grossly overstated. In the 1980s, finding themselves unable to achieve consistent results with ultrasound, pest professionals installed a series of protocols to test the devices’ efficacy in laboratories, test enclosures, and field trials. The results suggested a more nuanced picture than the jubilant anecdotal accounts: ultrasonic frequencies did seem to have real effects on some species of rodents, but usually only temporarily, until the animals became habituated to the sound. Many other pests remained unaffected. Alerted by a series of formal complaints, the Federal Trade Commission filed suits against some of the companies selling ultrasonic pest control devices for making false advertising claims.54 The manufacturer of the Homefree ultrasonic pest repellent device, for instance, was ordered to compensate its users and cease representing any other ultrasonic pest control product as an effective alternative to conventional products for eliminating various pests, unless it could offer reliable evidence to substantiate its claims. The order specified that such evidence must take the form of “tests, experiments, analyses, research studies, or other evaluations,” which would be adequate only if conducted “in an objective manner by persons qualified to do so, using procedures generally accepted in the relevant professions or sciences to yield accurate, reliable, and reproducible results.”55 To be sure, the issue to be proven was that the ultrasonic device had an effect of any kind on animals, not that it was safe for humans. Aside from these legal worries, professionals in the pest control industry identified several problems with high-frequency sound generators. Despite its powerful aura, ultrasound turned out to be impractical for covering large areas, because the higher the frequency of a wave, the higher its directionality and the rapidity of its decay. Pests were, moreover, reported to have quickly become immune, habituated, or even physically deaf to the aggressive frequencies. One of the key problems identified was the random nature of the signal with which the animals were targeted. In keeping with psychoacoustic testing practices, experimental psychologists had used pure tones to test animal hearing, and many developers of commercial high-frequency devices relied on that feature. But whereas pure tones had served a practical purpose for testing—allowing experimenters to eliminate uncertainty regarding the tone to which a subject was responding—the developers of high-frequency
288 World as Testbed
devices used the pure tone simply as a signal that was convenient to produce. Several repellent developers responded to the animals’ indifference by introducing new combinations and variations of frequencies intended to avoid habituation. However, it soon transpired that high tones alone were often no more effective than scarecrows in repelling pests. The acoustic noise that had been introduced to interfere with rodent activity became white noise, whose ever-present signal the animals quickly learned to ignore as a communicatory gesture.56 Among the first to observe that the characteristics of ultrasonic sound generators, quite apart from their cost, prohibited them from being scaled up were the Penn State zoologists Hubert and Mable Frings.57 Some time in 1954, they changed tactics, substituting an ethology of communication for the psychoacoustics of hearing, and began to investigate the use of meaningful sound sequences as opposed to randomly chosen high tones. The Fringses stumbled upon the importance of communicative signals while examining the effect of high-frequency sound on roosting starlings and pigeons in an air force hangar and various city parks. To test the birds’ sensitivity to ultrasonic frequencies, the research assistants had set out to catch wild test subjects. It soon turned out that the individuals they managed to seize would scare away the rest of the roosts with their shrieking protests. In a last-ditch attempt at success, the Fringses and their collaborators recorded this “distress call” on an early-model tape recorder used for dictating and mounted it on a panel truck to broadcast around buildings and parks that were affected by large roosts. This interest in bird behavior moved their investigations into the terrain of the newly emerging interdiscipline of bioacoustics.58 Influenced by new technologies of recording (particularly the gramophone and the magnetic tape recorder) and visualization (such as the sound spectrograph) in the early 1950s, pioneering bioacousticians had turned to the physical characteristics of vocalization and their communicative functions. Employing the magnetic tape recorder to broadcast prerecorded calls back at the animals, the Fringses’ work also invoked a tradition of playback experiments in the field.59 That tradition was almost as old as mechanical sound recording itself, but magnetic tape recorders granted investigators an unprecedented flexibility to set up such experiments in the field: they were portable enough to be carried around in the field, and tape itself also allowed the parameters of a playback sound to be manipulated in significant ways. By cutting and pasting prerecorded vocalizations in any desired order, investigators could study how particular acoustic characteristics influenced their communicatory function. In the 1950s and 1960s, a time when recording techniques were yielding more detailed understanding of the acoustics and behavioral functions of particular
Of Silent Sirens and Pied Pipers 289
animal calls, the Fringses deployed this ethological research to control pests.60 Their first experiments with the playback truck were such a success that when their research was published, an enterprising Brooklyn-based manufacturer of automatic answering machines and message repeaters, Mohawk Business Machines Company, proposed to integrate the recordings and instructions into its own continuous-loop cartridge technology to create a commercial bird repelling system, which was eventually marketed as Bird-E-Vict.61 Whereas Bird-E-Vict applied a simple playback mechanism for prerecorded distress signals, also audible to human listeners, the technologies that followed promised a more sophisticated approach to pest control using the analysis of animal communication. One notable instance was a synthetic animal-sound generator marketed as AV-Alarm and Transonic by the California company Santa Rita Technology in the 1980s. Like the Bird- E-Vict, the AV-Alarm emerged as a commercial spin-off of military research conducted on the initiative of the Wright-Patterson Air Force base. In the early 1960s, the base’s new bionics division engaged communication engineer John L. Stewart to work on an electronic analog model of the human ear and part of its nervous system. This served research on various kinds of speech processing and its application to hearing aids and the disruption of enemy communications. In the mid-1960s, Stewart began to extend his concept to other models of animal hearing, such as that of bats, birds, and insects. He hypothesized that his theory of auditory signal processing by the ear would allow him to reverse-engineer particular sounds to stimulate annoyance, distress, or anxiety. Stewart distinguished sounds that simply jammed human and animal communications, which were often inefficient because the nonhuman ear adapted to them, from the more effective “commanding sounds” that are difficult to adapt to, such as birds’ alarm and distress cries.62 The AV-Alarm capitalized on this distinction, synthesizing natural bird calls by varying their acoustic contents and thereby behavioral meanings. It falls outside the scope of this chapter to examine Stewart’s research in more detail. But what the AV-Alarm and Bird-E-Vict illustrate is that a new type of acoustic technology wielded an altogether different type of signal to deter animal species considered pests.
Thresholds of the Imagination Our understanding of what we know we cannot hear is defined in part by what we know we can hear. Average hearing is marked at one end by deafness and hearing loss and at the other by prodigious degrees of auditory sensitivity.
290 World as Testbed
To an average human listener, a phenomenology of extraordinary auditory sensitivity is simply inaccessible except through imagination, projection, and extrapolation. It was therefore only by deploying electrophysiological measurements of cochlear responses, anechoic chambers, audiograms, playback devices, and ultrasonic generators that audiologists, psychologists, and bioacousticians were able to open up a contact zone where the acoustic realities of pest and pets could be mapped onto the more familiar acoustic reality of an average human listener. By materializing those incommensurable realities as partly overlapping domains of auditory perception, an assemblage of experimental technologies enabled users to imagine new conditions of interaction with the animal world. High-frequency and ultrasonic technologies of control constituted one such condition of interaction. These technologies were made possible by the gap between thresholds of human and nonhuman hearing that materialized in the tests, but they may also have been conditioned by the material settings in which the thresholds were determined. Acoustic technologies of animal control shared with the nonhuman animal hearing test an implicit assumption, namely that it is possible to control the body’s physiological reaction to sound in a basic fight, flight, or startle response. In effect, pest repellents took the acoustic techniques that had been used to manage and condition rats’ behavioral responses to sound and escalated them into a technological network of behavioral control. Although the Ultrasonic Pied Piper and the Bird-E-Vict targeted different frequency bandwidths and used different acoustic signals as deterrents, both sought to pack sonic excitation into a mechanism of control. This promise of control is more easily understood if we conceive of the upper threshold of hearing as a threshold of the imagination. The developers of ultrasonic pest repellents managed to frame their customers’ very inability to hear ultrasound as proof of its sonic power, coupling animals’ ability to hear ultrasound with the assumption that they were susceptible to its effects. Within this logic, the imperceptibility of sound in the ultrasonic spectrum was no longer cause for concern, but rather an illustration of its effectiveness in distinguishing between human and nonhuman, pet and pest. A cunning salesperson could present piercing high-frequency noise to prospective customers in the showroom, then reassure them that their physical discomfort would become the pest’s as the dial was turned from the audible to the ultrasonic signal. In a fascinating inversion, the concern with noise as damaging to body and mind—which had prompted such insistent noise abatement campaigns since the early twentieth century—here became precisely a resource for the technology’s success.63 But if the limited human capacity for imagining extreme auditory sensitivity affords a possible explanation for the
Of Silent Sirens and Pied Pipers 291
technology’s continued application, it may also help us understand why high- frequency generators often failed as technologies of control in practice. This becomes evident in a juxtaposition of the failure of the Pied Piper and the success of Bird-E-Vict. Though both operationalized a logic of sonic excitation, they targeted animal hearing using remarkably different acoustic effects. Developers of ultrasonic repellents such as the Pied Piper typically framed its powers in measures of frequency and intensity that they adopted from the hearing test. In fact, the developers drew so explicitly on the hearing test as a way of validating their technologies that, to weaponize the allegedly harmful effects of intense ultrasound, they deployed the same default pure high-frequency tone that was used in the hearing test. In contrast, by targeting animal populations with what it claimed were behaviorally meaningful alarm signals, the Bird-E-Vict problematized that reliance on the hearing test as a failure to understand the ways animals listen—ways that a standardized hearing test was never designed to measure. A comparison between the Pied Piper and the Bird-E-Vict suggests that, however loud and intense, a noise the human listener reads as containing a communicatory gesture (“go away!”) may fail to act as a signal for the nonhuman listener. To the animal listener, the noise of a pure tone may be just that: noise. It may be only fitting, then, that the use of high-frequency sound ultimately came to be regarded as a technology more successful in dispersing humans than in controlling its original target, pests. The much-discussed Mosquito, for instance, is marketed as an ultrasonic youth deterrent technology that discourages lingering in public spaces. Nor, perhaps, is it surprising that the application of a singular piercing tone to deter a teenage population sensitive enough to its high frequency has been challenged by a different approach— conventional speaker systems that blast out classical music.64 Replacing a physical source of interference with a cultural one may, with some imagination, be regarded as the equivalent of broadcasting the behaviorally meaningful alarm call of the crow. Regardless of their effectiveness, such technologies deserve a place in the history of technology. The historical trajectory of ultrasonic applications is a reminder of the many technological systems that continue to saturate our landscape, imperceptible to some but not to others. This trajectory also helps us understand a continued fascination with the weaponization of ultrasound that is both seemingly recurrent and often built upon dubious foundations.65 In spite of scientific skepticism, ultrasound continues to be imagined as an effective repellent, from the Brazilian radio station that broadcast a high- frequency 15-kHz tone under its music to allow its listeners to enjoy radio unbothered by mosquitoes, to the accusations by the U.S. government in
292 World as Testbed
2018 that embassy staff in Cuba and China had been targeted by acoustic attacks.66 That is not to deny that various acoustic weapons, across the entire frequency spectrum, have been and continue to be described in the military literature and technology patents for aggressive purposes. Indeed, state and military actors have actively experimented with harnessing their use for crowd control and urban combat, in addition to uses in monitoring subaquatic movements and environments (see Camprubí and Hui in this volume).67 Rather, part of the attraction of these technologies may precisely be that, regardless of their efficacy above human hearing thresholds, they effectively exploit thresholds of the imagination. Above all, these technologies call attention to the subtle ways in which the expectations and norms of particular auditory, experimental, and commercial cultures become confounded in a technology. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/auditory-thresholds-animal-control-Bruyninckx.
Acknowledgments I am grateful to the organizers and participants of the authors’ workshop in preparation of this volume for many stimulating discussions, especially to Alexandra Hui, Mara Mills, and Viktoria Tkaczyk, and to Karin Bijsterveld, Lino Camprubí, Stefan Krebs, and Jonathan Sterne for their suggestions and feedback. Dominique Robert offered helpful comments on an earlier version of the paper. Fabian Voigtschild traced sources related to ultrasonic animal repellents.
Notes 1. Donald A. Dewsbury, “On Publishing Controversy”; Eliot Stellar and Gardner Lindzey, “Clifford T. Morgan.” 2. Clifford T. Morgan and Jane D. Morgan, “Auditory Induction.” The phenomenon had already been observed in 1914, in a rat responding to jingling keys. See Henry H. Donaldson, Rat. 3. Much of the historical scholarship on this topic has focused on music and noise. For musical effects on the body, see James Kennaway, Music and the Nerves; Kennaway, Bad Vibrations; Shelley Trower, Senses of Vibration. In the twentieth century, noise abatement campaigns drew on a medicalization of noise critique that framed urban noise as responsible for the nervous overstimulation of city dwellers. James G. Mansell, Age of Noise.
Of Silent Sirens and Pied Pipers 293 4. Steve Goodman, Sonic Warfare. Closest to my use of this term is Mitchell Akiyama’s, whose analysis of Mosquito identifies ultrasonic frequency as a strategy for privileging access to public space for some age groups at the expense of others. Akiyama, “Silent Alarm.” 5. Michelle Friedner and Stefan Helmreich identify low frequencies inaudible to both hearing and Deaf communities as a promising contact zone between the two, and between the scholarly fields organized around their embodied experiences, sound studies and deaf studies. Friedner and Helmreich, “Sound Studies Meets Deaf Studies.” 6. I follow up here on Goodman’s (Sonic Warfare, 22, 184) discussion of Mosquito as a rodent-repellent ultrasound technology redeployed in UK public space. Though this genealogy is not explicitly acknowledged in the origin stories projected by the Mosquito inventor Howard Stapleton and the company Compound Security Systems, the system instrumentalizes insights with a much longer history. 7. I refer here to Brian Kane’s critique of an “ontological turn” in sound studies, which concerns itself with the material-affective force that sound and vibration exert on the body (Kane identifies Goodman’s work as a prominent example). Kane, “Sound Studies Without Auditory Culture.” 8. Clifford Morgan’s star rose quickly, and he became assistant professor of biology at Johns Hopkins University in 1943. Maier, meanwhile, retreated into relative obscurity. See James H. Capshew, Psychologists on the March. The controversy strained personal relations between the two. Although the dispute publicly played out in Morgan’s favor, Donald Dewsbury shows that Morgan may actually have conceded to Maier in private. Dewsbury, “On Publishing Controversy.” 9. Before the hertz was adopted, frequencies were stated in cycles per second (c/s). One hertz equals one cycle per second. 10. James Gould and Clifford T. Morgan, “Auditory Sensitivity in the Rat”; Gould and Morgan, “Hearing in the Rat.” 11. Wilfred John Brogden, “Elmer Augustine Kurtz Culler.” 12. See Mara Mills in this volume. For a history of audiometry, see Mills, Deafening. 13. Audiometers had been in existence since at least 1880, but the Western Electric 1A Audiometer, using newly developed radio and telephonic technologies, significantly improved upon earlier versions. The audiometer’s development accompanied a desire to objectify the experience of living subjects. See Mills, Deafening; Emily Thompson, Soundscape of Modernity, 146–48; Jonathan Sterne, MP3, 36–40. 14. Thompson, Soundscape of Modernity, 146–47. 15. Mills, Deafening, 129–31. 16. See also Margaret Floy Washburn, Animal Mind. 17. Audiometers were employed at Robert M. Yerkes’s Laboratory of Comparative Psychobiology and Neurophysiology at Yale University, at the Laboratory of Physiology at McGill University, and at Harvard. Most experimenters, however, made do with a home- built set of oscillators and amplifiers. 18. Fred A. Mettler et al., “Acoustic Value,” 479. 19. Elmer Culler et al., “Measurements of Acuity,” 225. 20. In a benchmark compendium, bioacoustician William Tavolga relates how a particular squirrelfish, nicknamed “Clever Hans,” was trained to respond to an acoustic stimulus by jumping from one part of a tank to another. Researchers were pleased to work with this fish as it learned quickly and demonstrated a remarkable sensitivity to the stimulus. But
294 World as Testbed when its hearing thresholds seemed to sink lower each day, they became suspicious. After testing the fish without any sound, they discovered that Hans was watching the researchers and responding not to sound, but to the motion of the scientists as they reached for the test button. Tavolga, “Retrospect and Prospect,” 574. 21. Elmer A. Culler, “Research.” 22. Frederick A. Saunders and Frederick V. Hunt, George Washington Pierce. 23. Gould and Morgan, “Auditory Sensitivity in the Rat,” 327. 24. Ibid. 25. Robert Galambos, “Cochlear Potentials”; Donald R. Griffin and Robert Galambos, “Sensory Basis of Obstacle Avoidance.” The relationship between Griffin and Galambos is detailed in Charles G. Gross, “Donald R. Griffin.” Morgan collaborated with Galambos on several experiments related to hearing tests. In one series, they used a bullhorn to test the effect of lower-frequency but high-intensity sounds, showing intensity to be an important parameter. 26. This was Galambos’s own assessment. Galambos, “Robert Galambos.” 27. Hallowell Davis et al., “Temporary Deafness.” 28. In the late 1920s, waterborne ultrasound had been found to cause harm to small fish, plants, and amphibians. See Robert W. Wood and Alfred L. Loomis, “Physical and Biological Effects.” During World War II, German scientists found that the sound fields around their early turbojet engines contained a level of energy in the ultrasonic frequency range that, they postulated, would cause considerable harm to the human ear if it were to become audible. For an overview, see Henning E. von Gierke et al., “Noise Field of a Turbo-Jet Engine.” 29. Crit Pharris, “Ultrasonic Sickness.” 30. Hallowell Davis, “Biological and Psychological Effects.” 31. After World War II, the Wright-Patterson Field Station hosted the Air Material Command, which became a major postwar research and development organization. It focused on improving weapon systems but also carried out secondary projects. 32. See, for instance, “Danger of Unheard Noise”; W. W. Morris, “Silent Sounds Are Hot”; “Sound Kills Mouse.” 33. This was an open question for the Penn State researchers. Hubert Frings, Scientific Scobberlotching, 65. 34. “Reveal Sound Waves Can Kill Mice”; Karl F. Graff, “History of Ultrasonics,” 82–83. 35. Frings, Scientific Scobberlotching, 64–65. 36. Hubert Frings, “Pest Control with Sound Waves.” 37. Hubert Frings, Mable Frings, and Alan Kivert, “Behavior Patterns”; Hubert Frings and Mable Frings, “Acoustical Determinants.” 38. Frings, Scientific Scobberlotching, 68–69. Frings had just learned firsthand about the toxic effects of chemical repellents such as DDT and chlordane, which had temporarily affected his vision in one eye. Ibid., 76–77. 39. George F. Quittner, Pest Control. 40. “Army Engineers Kill Mice.” The same article was quoted as the single foundation for claims about the lethal effects of ultrasound on pests, for instance in patent applications by Lowell A. Moe (Ultrasonic Systems), Norman R. Evans (Device for Chasing Pests), and Benson Carlin (Nuisance Control Technique). 41. Moe, Ultrasonic Systems. 42. For instance, Evans, Device for Chasing Pests.
Of Silent Sirens and Pied Pipers 295 43. Chris Philo and Chris Wilbert, Animal Spaces. 44. Discussing the application of hearing tests in telephone engineering, Mills uses the term “ergonomopolitics” to describe the forms of power exercised through an ergonomic optimization of human users’ relationship with objects. Mills, “Deafening,” 130. See also Mara Mills, “Dead Room.” 45. Sterne, MP3, 45–48. 46. Some of the first developers had experience in both the military and in electrical engineering. During the war, Lowell Moe worked at the Wright Field Aircraft Radio Laboratory in the field of microwave radar before founding the American Engineering Company and becoming director of Electronics Research, where he developed the Ultrasonic Pied Piper. Another pioneer, Benson Carlin, was a radio engineer for the Signal Corps and specialized in ultrasonic applications. After the war, he turned his expertise to the development of a commercial system at Sperry Instruments for detecting flaws in metal products, which made use of ultrasound, and patented an instrument for ultrasonic “nuisance control” in 1960. 47. Popular Electronics even published circuit schemes of an ultrasonic generator for do-it- yourself enthusiasts to experiment with. Lyman Greenlee, “Electronic Pest Control.” 48. As architectural historian Mark Wigley has argued, by the 1960s the suburban American domestic exterior had become increasingly electrified, with electrical systems operating and controlling fences, sprinkler systems, and video cameras. Ultrasonic repellents tapped into a similar desire to control the domestic sphere at a distance and to erect subtle yet effective borders. Wigley, “Electric Lawn.” 49. Federal Trade Commission, Federal Trade Commission Decisions, 76. 50. “Rats Meet ‘Pied Piper.’ ” 51. In one illustration of this appeal, readers of the comic series The Flash were introduced in 1959 to a new villainous character named “Pied Piper,” a deaf musical prodigy who uses sonic technology to hypnotize people and cause vibrational damage in his opponents. A short story by science fiction author J. G. Ballard from 1963 described the possibility of inaudible but neuro-affective high-frequency music. Ballard, “Sound-Sweep.” 52. “Danger of Sounds,” 28. 53. At least one patent claimed its device could emit acoustic frequencies that resonated with frequencies in the brain, suiting it for use in riot control and animal dispersal. Edward G. Longinette and Charles W. Porter, Riot Control Devices. In the 1970s, the popular press reported on alleged uses of ultrasound in crowd dispersal. R. Rodwell, “ ‘Squawk Box’ Technology,” 667. These form part of a long history of attempts to use sound as a military weapon. See Daria Vaisman, “Acoustics of War”; Juliette Volcler, Extremely Loud; Goodman, Sonic Warfare. 54. For example, Complaint in the Matter of Sentronic Controls Corporation, et al. Consent Order in Regard to Alleged Violation of Sec. 5 of the Federal Trade Commission Act, 1985. Similar complaints were made against other producers and marketers of ultrasonic pest control devices. Saga International, Inc., for instance, was ordered in 1986 to refund the full price of the product to consumers. Federal Trade Commission, Federal Trade Commission Decisions, 62–85. 55. Ibid., 81. 56. This continual inversion of noise is the topic of Michel Serres’s philosophical treatise The Parasite, which uses a fable about an interrupted meal between two rats to theorize how signal and interference unsettle each other.
296 World as Testbed 57. Other researchers in Canada and Britain had come to a similar conclusion, based on field experiments. They found that although birds tended to respond to the high-frequency tones of an air-raid siren broadcast in the field, its effect did not last enough to make it an economical proposition. “Birds Undaunted by Sound”; G. Thiessen et al., “Acoustic Irritation Threshold.” 58. Hubert Frings later served as the secretary of the International Committee on Biological Acoustics, whose inaugural meeting was held in 1956 at Penn State. 59. Gregory Radick, Simian Tongue. 60. See Joeri Bruyninckx, Listening in the Field, ch. 4. 61. The air force insisted that Penn State apply for a patent on this method, so that military interests could be protected and the college could benefit from it to further research. See Frings, Scientific Scobberlotching, 119. One student collaborator on the project later developed the concept into a business called Bio-Sonic Control Co. 62. John L. Stewart, “Experiments with Sounds.” 63. Karin Bijsterveld, Mechanical Sound; Mansell, Age of Noise. 64. Anne Midgette, “Blasting Mozart.” 65. Carl Zimmer, “ ‘Sonic Attack.’ ” 66. William Kremer, “Ultrasound Mosquito Repellents”; “Mosquito Fumigation.” Ironically, according to recent reports, a mosquito repellent pesticide may have been responsible for the physical complaints and illnesses experienced by U.S. and Canadian diplomats, earlier ascribed to the effects of a “secret ultrasonic weapon.” 67. Examples are the “Acoustic Squawk Box” (1973) and the HyperSonic Sound(r) technology developed by the Long Range Acoustic Device Corporation. For an overview of (largely failed) attempts at developing such acoustic weapons, see Neil Davison, ‘Non-Lethal’ Weapons, ch. 7. On scientific skepticism regarding their effectiveness, see Jürgen Altmann, “Acoustic Weapons.”
References Akiyama, Mitchell. “Silent Alarm: The Mosquito Youth Deterrent and the Politics of Frequency.” Canadian Journal of Communication 35, no. 3 (2010): 455–71. Altmann, Jürgen. “Acoustic Weapons: A Prospective Assessment.” Science & Global Security 9 (2001): 165–234. “Army Engineers Kill Mice with New Sonic Wave.” Washington Times-Herald, November 30, 1947. Ballard, J. G. “The Sound-Sweep.” Science Fantasy 13, no. 39 (1960): 2–39. Bijsterveld, Karin. Mechanical Sound: Technology, Culture, and Public Problems of Noise in the Twentieth Century. Cambridge, MA: MIT Press, 2008. “Birds Undaunted by Sound.” Science News-Letter 73, no. 18 (1958): 279. Brogden, Wilfred John. “Elmer Augustine Kurtz Culler: 1889–1961.” American Journal of Psychology 75, no. 1 (1962): 155–60. Bruyninckx, Joeri. Listening in the Field: Recording and the Science of Birdsong. Cambridge, MA: MIT Press, 2018. Capshew, James H. Psychologists on the March: Science, Practice, and Professional Identity in America, 1929–1969. Cambridge: Cambridge University Press, 1999.
Of Silent Sirens and Pied Pipers 297 Carlin, Benson. Nuisance Control Technique and Apparatus Therefor. U.S. Patent 2,922,999A, filed January 17, 1956, and issued January 26, 1960. Culler, Elmer A. “Research on the Central Auditory Mechanisms: Some Recent Methods and Results.” Ohio Journal of Science 41 (1941): 117–22. Culler, Elmer, Glen Finch, Edward Girden, and Wilfred John Brogden. “Measurements of Acuity by the Conditioned-Response Technique.” Journal of General Psychology 12 (1935): 223–27. “The Danger of Sounds We Cannot Hear.” UNESCO Courier, July 1967: 26–31. “Danger of Unheard Noise.” Science News-Letter 53, no. 18 (1948): 274. Davis, Hallowell. “Biological and Psychological Effects of Ultrasonics.” Journal of the Acoustical Society of America 20, no. 5 (1948): 589. Davis, Hallowell, Clifford T. Morgan, Joe E. Hawkins Jr., Robert Galambos, and F. W. Smith. “Temporary Deafness Following Exposure to Loud Tones and Noise.” Acta Oto- Laryngologica. Supplementum 88 (1950): 1–56. Davison, Neil. ‘Non-Lethal’ Weapons. Basingstoke, UK: Palgrave Macmillan, 2009. Dewsbury, Donald A. “On Publishing Controversy: Norman R. F. Maier and the Genesis of Seizures.” American Psychologist 48, no. 8 (1993): 869–77. Donaldson, Henry H. The Rat: Reference Tables and Data for the Albino Rat and the Norway Rat. Philadelphia: Wistar Institute of Anatomy and Biology, 1915. Dworkin, S., J. Katzman, G. A. Hutchinson, and J. R. McCabe. “Hearing Acuity of Animals as Measured by Conditioning Methods.” Journal of Experimental Psychology 26, no. 3 (1940): 281–98. Evans, Norman R. Device for Chasing Pests. U.S. Patent 3,058,103, filed September 10, 1958, and issued October 9, 1962. Federal Trade Commission. Federal Trade Commission Decisions: Findings, Opinions and Orders. Vol. 108, July–December. Washington, DC: U.S. Government Printing Office, 1986. Friedner, Michelle, and Stefan Helmreich. “Sound Studies Meets Deaf Studies.” Senses and Society 7, no. 1 (2012): 72–86. Frings, Hubert. “Pest Control with Sound Waves: Ultrasonics as a Possibility in the Future of Rodent and Insect Control.” Pest Control 16, no. 4 (1948): 9. Frings, Hubert. The Scientific Scobberlotching of Hubert and Mable Frings. N.p.: BTcurlewPress, 2015. Frings, Hubert, and Mable Frings. “Acoustical Determinants of Audiogenic Seizures in Laboratory Mice.” Journal of the Acoustical Society of America 24, no. 2 (1952): 163–69. Frings, Hubert, Mable Frings, and Alan Kivert. “Behavior Patterns of the Laboratory Mouse Under Auditory Stress.” Journal of Mammalogy 32, no. 1 (1951): 60–76. Galambos, Robert. “Cochlear Potentials Elicited from Bats by Supersonic Sounds.” Journal of the Acoustical Society of America 14, no. 41 (1942): 41–49. Galambos, Robert. “Robert Galambos.” In The History of Neuroscience in Autobiography, vol. 1, edited by Larry R. Squire, 178–220. Washington, DC: Society for Neuroscience, 1996. Gierke, Henning E. von, Horace O. Parrack, William J. Gannon, and R. G. Hansen. “The Noise Field of a Turbo-Jet Engine.” Journal of the Acoustical Society of America 24, no. 2 (1952): 169. Goodman, Steve. Sonic Warfare: Sound, Affect, and the Ecology of Fear. Cambridge, MA: MIT Press, 2012. Gould, James, and Clifford T. Morgan. “Auditory Sensitivity in the Rat.” Journal of Comparative Psychology 34, no. 3 (1942): 321–29. Gould, James, and Clifford T. Morgan. “Hearing in the Rat at High Frequencies.” Science 94, no. 2433 (1942): 168. Graff, Karl F. “A History of Ultrasonics.” In Physical Acoustics: Principles and Methods, vol. 15, edited by Warren P. Mason and R. N. Thurston, 1–97. New York: Academic Press, 1981.
298 World as Testbed Greenlee, Lyman. “Electronic Pest Control: Ultrasonics Forces Rodents and Insects to Depart for Quieter Surroundings.” Popular Electronics, July 1, 1972: 47–50. Griffin, Donald R., and Robert Galambos. “The Sensory Basis of Obstacle Avoidance by Flying Bats.” Journal of Experimental Zoology 86, no. 3 (1941): 481–506. Gross, Charles G. “Donald R. Griffin, 1915– 2003.” In Biographical Memoirs, vol. 86. Washington, DC: National Academies Press, 2005. Kane, Brian. “Sound Studies Without Auditory Culture: A Critique of the Ontological Turn.” Sound Studies 1, no. 1 (2015): 2–21. Kennaway, James. Bad Vibrations: The History of the Idea of Music as a Cause of Disease. London: Routledge, 2016. Kennaway, James, ed. Music and the Nerves, 1700–1900. Basingstoke: Palgrave Macmillan, 2014. Kremer, William. “Ultrasound Mosquito Repellents: Zapping the Myth.” BBC News, December 11, 2012. Longinette, Edward G., and Charles W. Porter. Riot Control Devices Employing a Modulated Stimulus Frequency. U.S. Patent 3,557,899, filed January 10, 1967, and issued January 26, 1971. Mansell, James G. The Age of Noise in Britain: Hearing Modernity. Urbana: University of Illinois Press, 2017. Mettler, Fred A., Glen Finch, Edward Girden, and Elmer Culler. “Acoustic Value of the Several Components of the Auditory Pathway.” Brain 57, no. 4 (1934): 475–83. Midgette, Anne. “Blasting Mozart to Drive Criminals Away.” Washington Post, January 20, 2012. Mills, Mara. “The Dead Room: Deafness and Communication Engineering.” PhD diss., Harvard University, 2008. Mills, Mara. “Deafening: Noise and the Engineering of Communication in the Telephone System.” Grey Room 43 (2011): 118–43. Moe, Lowell A. Ultrasonic Systems. U.S. Patent 3,157,153A, filed on October 22, 1962, and issued on November 17, 1964. Morgan, Clifford T., and Jane D. Morgan. “Auditory Induction of an Abnormal Pattern of Behavior in Rats.” Journal of Comparative Psychology 27, no. 3 (1939): 505–8. Morris, W. W. “Silent Sounds Are Hot.” Popular Mechanics, February 1948: 135–38. “Mosquito Fumigation May Have Caused Mystery ‘Havana Syndrome,’ Study Says.” The Guardian, September 19, 2019. Pharris, Crit. “Ultrasonic Sickness.” American Industrial Hygiene Association Quarterly 9, no. 3 (1948): 57–62. Philo, Chris, and Chris Wilbert. Animal Spaces, Beastly Places: New Geographies of Human- Animal Relations. London: Routledge, 2005. Quittner, George F. Pest Control. U.S. Patent 3,138,138, filed November 8, 1961, and issued June 23, 1964. Radick, Gregory. The Simian Tongue: The Long Debate about Animal Language. Chicago: University of Chicago Press, 2007. “Rats Meet ‘Pied Piper’: Ultrasonics.” New York Times, March 26, 1972: 72. “Reveal Sound Waves Can Kill Mice, Flies.” Chicago Tribune, November 30, 1947: 23. Rodwell, R. “‘Squawk Box’ Technology.” New Scientist 59, no. 864 (1973): 667–68. Saunders, Frederick A., and Frederick V. Hunt. George Washington Pierce, 1872–1956: A Biographical Memoir. Washington, DC: National Academy of Sciences, 1956. Serres, Michel. The Parasite. Translated by Lawrence R. Schehr. Minneapolis: University of Minnesota Press, 2007. “Sound Kills Mouse.” Science News-Letter 52, no. 18 (1947): 247. Stellar, Eliot, and Gardner Lindzey. “Clifford T. Morgan: 1915–1976.” American Journal of Psychology 91, no. 2 (1978): 343–48. Sterne, Jonathan. MP3: The Meaning of a Format. Durham, NC: Duke University Press, 2012.
Of Silent Sirens and Pied Pipers 299 Stewart, John L. “Experiments with Sounds in Repelling Mammals.” In Proceedings of the 6th Vertebrate Pest Conference, edited by Warren V. Johnson and Rex E. Marsh, 222–26. Davis, CA: California Vertebrate Pest Committee, 1974. Tavolga, William N. “Retrospect and Prospect—Listening through a Wet Filter.” In Hearing and Sound Communication in Fishes, edited by Arthur N. Popper and Richard R. Fay, 573–88. New York: Springer, 1981. Thiessen, G., E. A. G. Shaw, R. D. Harris, J. B. Gollop, and H. R. Webster. “Acoustic Irritation Threshold of Peking Ducks and Other Domestic and Wild Fowl.” Journal of the Acoustical Society of America 29, no. 12 (1957): 1301–306. Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the Culture of Listening in America. Cambridge, MA: MIT Press, 2004. Trower, Shelley. Senses of Vibration: A History of the Pleasure and Pain of Sound. New York: Continuum, 2012. Vaisman, Daria. “The Acoustics of War.” Cabinet Magazine 5 (Winter 2001– 2), http:// cabinetmagazine.org/issues/5/acousticsofwar.php. Volcler, Juliette. Extremely Loud: Sound as a Weapon. New York: New Press, 2013. Washburn, Margaret Floy. The Animal Mind: A Text- Book of Comparative Psychology. New York: MacMillan, 1923. Wigley, Mark. “The Electric Lawn.” In The American Lawn, edited by Georges Teyssot, 154–95. New York: Princeton Architectural Press, 1999. Wood, Robert W., and Alfred L. Loomis. “The Physical and Biological Effects of High- Frequency Sound- Waves of Great Intensity.” The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science 4, no. 22 (1927): 417–36. Zimmer, Carl. “A ‘Sonic Attack’ on Diplomats in Cuba?” New York Times, October 5, 2017.
11 Testing the Underwater Ear Hearing, Standardizing, and Classifying Marine Sounds from World War I to the Cold War Lino Camprubí and Alexandra Hui
For humans, listening in the ocean is a profoundly technological experience. Prior to World War I, only a few fishing and whaling cultures listened for their quarry with naked ears—Malaysian fish listeners, for example, or whalers listening through the hulls of their ships—and the oceans were traditionally understood to be silent. The difficulties for the unmediated human ear to make sense of underwater sound rendered the ocean a mysterious and sonically ambiguous space. Indeed, Jacques Cousteau titled his 1953 autobiographical book—and later, film—The Silent World. And yet Cousteau and oceanographers like him knew that the oceans were actually quite noisy. As we discuss in this chapter, hydrophones (underwater microphones) provided a technological means of hearing ocean sounds that would put an end to this sonic ambiguity. They did so (this is our first thesis) together with the development of the discipline of marine biology, which itself grew out of a call to both order and classify the newly audible ocean sounds. During the two world wars and then the Cold War, submarine warfare motivated military interest in underwater sounds of any origin. Marine biologists contributed to the effort by helping identify the myriad sounds that could hinder detection of an enemy vessel by sonar operators. Shared technologies and goals meant shared forms of listening. Sonar technicians and marine biologists heard the ocean the same way. Then, as they began to listen to different sounds and listen for different sounds, their ears began to diverge, as did their definitions of signal and noise. Less than twenty years after Cousteau’s autobiography, the marine biologists Roger and Katy Payne released their LP Songs of the Humpback Whale (1970), an album that turned listening below the surface from a military exercise into an aesthetic experience and informed the environmentalist agenda for years to come.1 We attempt here to chart that divergence and explain how underwater listening speciated over the course of the twentieth century. We do so by tracing Lino Camprubí and Alexandra Hui, Testing the Underwater Ear In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0012.
302 World as Testbed
the various meanings and roles of testing in this story. By looking at testing, our paper places technological listening at the center of the transformation of the oceans from a world of silence into a sonically rich space. Trained human ears, machine listening, animal sounds, transducers, and various visual representations of acoustic phenomena retooled the ocean epistemically and ontologically. Testing brought these many entities into contact. As an actors’ category, testing appears in four separate but intertwined contexts in this chapter: testing listening equipment, testing marine animals’ sounds, testing hearing of humans and nonhumans, and testing the accuracy of the graphic representations of sound. Each of these instances captured more than one meaning of “testing,” but what was common to all of them was that testing worked as a mediator between different sonic realms. The testing of hydrophones facilitated the identification of certain marine species as sonic creatures; testing animals in the laboratory enabled new interpretations of the signals that hydrophones were receiving; recordings of submarine engines and cetaceans were employed to test recruits for submarine warfare; humans and animals were tested together to assess their respective hearing ranges, probing the limits of the definition of the hearing sense; and the experienced ears of marine biologists put spectrograms (and the interpretation of them) to the test.2 Once the connections between these sonic realms were established through testing, these realities came to function as proxies of one another, enabling the calibration, standardization, and mobility of sonic technologies and phenomena across the world’s oceans.3 By rendering alien realities commensurable (and this is our second thesis), testing performed the dual epistemological and ontological work of making the underwater world knowable by sonifying it. Sonification, in turn, facilitated further testing. Underwater eavesdropping tested the human limits of sound, expanding audibility from the low frequencies of waves and whales to the high frequencies of propellers and shrimp. While we address the international development of technologies and ears for underwater listening from World War I onward in a general way, we focus on U.S. Navy personnel and marine biologists during World War II and the Cold War. We do not wish to imply that this was exclusively an American story, exclusively a military story, or only an ocean story. Scientists have long grappled with listening to birds and other terrestrial creatures, usually without the military looking over their shoulders.4 But we have chosen this episode in particular because it was unique in scale, rapidity, and technological sophistication. Hydrophones and submarines were available to only a handful of navies around the world, and the U.S. Office of Naval Research (ONR) had the budget and workforce to push its listening agenda. By teasing out and
Testing the Underwater Ear 303
contextualizing the various tests of the underwater ear mobilized in antisubmarine warfare, we bear witness to the rapid sonification of the ocean.
From a Silent World into a Cacophony of Sounds Underwater sound propagation had been a subject of scientific interest since at least Leonardo da Vinci, but it was not till the early nineteenth century that several experiments determined the speed of sound underwater—the birth of underwater acoustics is conventionally attributed to the research performed by Jean-Daniel Colladon and Charles-François Sturm in 1826.5 As larger, faster, mechanized ships populated the seas, listening to engine sounds by submerging the receiving end of an ear horn became common practice among both scientists and amateurs interested in identifying and measuring underwater sounds. This curiosity dovetailed with that of physicists and engineers investigating wireless communication and safe navigation. In an age of electromagnetism and empire, converting energy from one form into another was key for governments and companies desiring to connect and control far- flung places.6 Researchers began to experiment with transducing submerged mechanical waves into sounds that humans could hear without getting wet. By 1889, carbon microphones had been adapted to underwater conditions through the attachment of vibratory metal diaphragms—American engineer Elisha Gray dubbed them “hydrophones.”7 These transducers were either tuned to particular frequencies to receive specific signals or opened to all audible underwater sounds8 In World War I, German U-boats transformed the ocean depths into battlefields, motivating all navies involved to ask their physicists and engineers to improve on these early underwater sound detection and identification technologies and practices.9 In 1917, French physicist Paul Langevin produced an effective active hydrophone that emitted pulses at high frequencies through piezoelectric quartz crystals and, in turn, received their echoes.10 Factoring time with the velocity of sound propagation, sonar (as it would soon be known) enabled submarine seekers to calculate the distance separating the receivers from the reflecting object. During the interwar period, the accuracy of both “passive” and “active” hydrophones greatly increased, aided by innovations in telecommunications technologies, particularly the telephone. By the 1940s, hydrophones could focus on certain directions through binaural recording, filter out or highlight certain frequencies, amplify ranges of interest, and even detect mechanical waves at frequencies lower than 20 Hz and transform them into signals audible to humans.11
304 World as Testbed
The effectiveness of these technologies depended on trained and tested human ears. Already in World War I, Emily M. Smith and Frederic C. Bartlett of the Cambridge Psychological Laboratory attempted to “devise and apply a series of tests for the selection of candidates for the Anti-Submarine Service.”12 In 1918, their main challenge was to test the testing instruments they had created. One produced continuous electronic sounds imitating a submarine, for instance, and another measured physiological and psychological responses to these sounds under varying conditions: dark and light, day and night, binaural and monoaural, and differing durations and intensities of stimuli. Smith and Bartlett often conducted experiments on one another—they eventually married, but unfortunately only he would go on to become a leading figure in British psychology.13 Among the variables they considered was “the [subject’s] order of testing,” which enabled them to test the effects that the experimenters’ fatigue and other subjective factors could have on test results.14 Smith and Bartlett also tested different classifications of sounds, proposing terms that depended on analogies with color, weight, timbre, simplicity, activity, or presence and the distinction between “heard or unheard” and “subjective or objective” sounds. Interest in standardizing tests and training declined in the following decades, due both to the absence of a submarine threat during peacetime and to extensive research into a variety of methods for automating sound reception, including oscillographs that graphically recorded signals.15 Sonar, which became the standard tool for measuring the ocean depths (an activity known as “sounding” long before acoustics was involved) and mapping the ocean floor, depended on acoustics but not on human hearing.16 Interpreting the echoes of sonar against moving objects, however, remained the task of human sonar operators, or “ping men.” By the late 1930s, in the context of rearmament and escalating international tensions, selection and training had come back to the foreground. In particular, tests were developed to determine which trainees had the aptitude to distinguish the sonar echo from the background noise of the ocean—that is, a relatively pure tone from what was by then considered a broadband cacophony.17 Of greater significance to our chapter are the challenges that faced those listening not to the sonar echoes but to the ocean noises themselves. By the outbreak of World War II—in which German U-boats were again dictating the terms of submarine warfare—the oceans were no longer silent. Underwater listening devices picked up a pandemonium of sounds: friendly engines, the turbines of one’s own vessel, storms, waves, breaking ice, and marine life. This latter problem, sounds of biological origin, informs our thesis that marine biology and acoustic technologies co-produced each other and also produced
Testing the Underwater Ear 305
the ocean as a sonic space audible to human ears. It was well known that several species, such as the aptly named croaker family (Sciaenidae) of fish, also called “drums,” produced specific sounds, but the exigencies of antisubmarine warfare rendered those sounds into noise to be separated from relevant signals. An example will help illustrate the existential urgency of the problem. In 1942, while testing hydrophones for antisubmarine warfare, researchers at the U.S. Navy Radio and Sound Laboratory (based at the University of California Division of War Research in San Diego) were deafened by an intense high- frequency sound. They identified other occurrences of the same “crackling noise” of “high magnitude” and set out to determine its origins.18 The project was led by Martin W. Johnson (1893–1984), an assistant professor at Scripps Institution of Oceanography and the author of the microbiology portions of The Oceans published that same year—a book that would soon come to be considered a main reference for physical oceanography.19 At first, Johnson and his team thought the noise was an artifact of their technology, but after conducting new tests they decided it was external. Notice here that testing listening devices and classifying hitherto unknown sounds were a single process. After ruling out meteorological and other causes, “identification of noise was accomplished through laboratory tests of various animals and through field studies in many different habitats and over different populations.” This was an expensive and highly elaborate technological setup: “In the laboratory the animals were usually tested in large concrete tanks supplied with running sea water,” and the sounds they emitted were compared to the problematic ones through both audio records and graphic representations of different frequency bands—with a degree of accuracy of “50 cycles wide or a half-octave.”20 The sound was, then, only classifiable when laboratory tests on the animals were combined with analyses of graphic representations of recordings. Snapping shrimp turned out to be the best candidate. This was quite surprising to those who had only heard the sounds of a few individuals at a time: their deafening roar was only possible in large numbers. At this point, such sounds were considered biological noise that interfered with the sounds the military needed to hear, and thus triggered a number of militarily significant ecological studies to determine where to expect shrimp, what minimum numbers were required to produce acoustic interferences with listening technologies, and at what depths and times of the day their noise was most intense.21 Through a comparison of oscillogram images of laboratory and field recordings, the presence of snapping shrimp in areas of military importance was confirmed. The circle was complete: hydrophone testing led to a surprising biological discovery, and laboratory testing of shrimp sounds
306 World as Testbed
produced graphs that were then, with the help of hydrophones, used to test ecological predictions. We argue that although the role of sound for marine life itself was not the object of such research, the main problems that would characterize early bioacoustics emerged in this context. Johnson and his colleagues devoted significant effort to understanding how and why shrimp used their claws to snap. They considered that the shrimp could be using the pressurized water created by their snap as a weapon, either between individual shrimp or collectively against a predatory species, but that the “biological significance of the accompanying loud sound [was] uncertain.”22 As we will see, biological function would shift from an aside to the central inquiry of later research projects. There were other problematic species. The characteristic scream of toadfish during their mating season, for example, interfered with acoustically triggered mines, homing torpedoes, and submarine detection equipment. Such sonic threats prompted the Naval Ordnance Laboratory (NOL) to perform a series of studies in the early 1940s on noise-making marine life to classify and better understand the nature and distribution of what they called “natural background noise.”23 This was the first comprehensive precision measurement of the frequency distribution and intensity of sounds generated by underwater noise-making creatures.24 Over the years, as this research program was extended, it produced not only the highest-quality recordings of underwater background sounds ever made at the time but also the first recordings of several species of noise-making fish. In 1954, the reference file of underwater sounds was made a permanent, continuously updated repository of underwater biological sounds that could be compared to recordings of unknown sounds to determine whether they were human-made or not. The sonification of the ocean co-evolved with new biological knowledge and technologies of submarine detection. But before we delve further into that story, we need to return to the military ear.
Testing Perception With some precedents, such as the telegraph and musical training discussed in this volume, the making of the underwater ear in World War II offers a rare case of forced and rapid standardization of perception. Testing and training thousands of ears allowed the U.S. Navy to define the collective experience of listening below the sea’s surface. In 1941, about fifty sonar operators graduated every month from the Navy’s training schools. At the beginning of 1942, the United States having just entered the war, the graduation rate rose to
Testing the Underwater Ear 307
one hundred per month. By the end of 1942, it was five hundred per month. Rapidly training so many people for a specialized and vital task was a serious undertaking, and a team at the Columbia Underwater Sound Laboratory in New London, Connecticut was appointed to find the best way.25 The group leader was experimental psychologist William D. Neff (1912– 2002). Trained in human and animal physiological psychology, Neff was particularly interested in hearing deficits and perceptual thresholds, for which he combined behavioral and neuroanatomical studies. After World War II, Neff would go on to found programs at the Universities of Chicago and Indiana devoted to an experiment-based understanding of perception.26 His task during the war was to design experimental tests of auditory discrimination for future sonar operators. Neff ’s group spent the first months of 1942 collecting experiences at the sonar operator schools of San Diego, California and Key West, Florida. They came to the conclusion that it was a waste of resources to train people who were not likely to pass the final graduation tests. Accordingly, they developed a series of stricter auditory tests to be employed during the initial trainee selection stage.27 The first test was a simple one that had been in use in other capacities for over a decade: an audiometer with increasing decibels distinguished potential trainees with “normal” hearing capacities from those who had hearing loss.28 Around 80 percent of candidates passed to the next round. A second set of tests of sound perception drew upon Carl Seashore’s The Measurement of Musical Talent (1919). As Viktoria Tkaczyk discusses in this volume, Seashore had developed these tests in the hopes of objectively measuring “musicality” through tests of pitch, time, and intensity discrimination. The pitch discrimination test was a classic of musical training and early experimental psychology.29 It consisted of playing one hundred sets of two tones separated by three seconds. The difference between the tone pairs varied from intervals of five to thirty cents.30 The potential sonar trainee was asked to tell them apart. The intensity and duration discrimination tests were also taken from Seashore and were similar to those conducted for musical education at the time. It became evident to Neff ’s team after a few months of testing potential trainees that the Seashore tests were not ideal for identifying the talents required of sonar technicians after all. In particular, Seashore had explicitly excluded the spatial dimension from his tests, yet this was a crucial skill for locating a submerged enemy. The Seashore tests tested the wrong skills, thereby crafting, through selection, a poor underwater ear. Neff ’s group decided instead to develop its own pitch memory test, “a single improved test which would measure both pitch discrimination and tonal memory.”31 The
308 World as Testbed
critically important ability to hear the changing pitch of the Doppler effect— increasing pitch over time would indicate, say, an approaching submarine and a decreasing pitch would indicate a departing one—motivated Neff to include a test of tonal memory. Neff and his team also added the “relative movement test,” a set of conceptual questions devised to gauge candidates’ spatial imagination. The emancipation of underwater ear testing and training from the musical tradition was, however, short-lived. In the first months of frantic recruitment at the beginning of the war, the new testing protocol was too cumbersome and time-consuming. So the Navy developed a fourth set of tests, essentially a survey titled Inventory of Musical Background, that was sent directly to the docked warships to be distributed among the men already on board. Whereas the previous tests either selected possible trainees or certified the actual trainees to then be posted as sonar technicians across the Navy, the Inventory of Musical Background determined who on board had the most musical training and designated those people as the most likely to successfully complete sonar training. Once the best ears had been selected, they had to be adapted to the underwater world. The training and standardization of those selected, if they were not yet shipping off to war, continued at sonar operator schools. Students performed exercises in tone differentiation and recognition and were then repeatedly subjected to the Navy’s pitch memory test. The selection test was thus now also a training method. Additionally, this continued use of the test served to test the performance of the test itself, and in the next two years of the war, the Navy’s pitch memory test was repeatedly modified in pursuit of better underwater ears. The most important part of sonar operator training was learning to recognize sounds specific to the oceans. This practice can be distinguished from earlier efforts to standardize recognition, mostly in the musical realm, in two notable ways. First, identification was the end goal, not the first step toward recreating the sound. Second, the sounds were confirmed against standard reference files, training records, and manuals that were far removed (physically, temporally, and materially) from their original sounding sources. From 1942 to 1945, a staff of fifty people produced thirteen manuals and five albums with recordings of the sounds sonar technicians should expect to experience. The records varied from individual sounds (a propeller, a cargo ship, snapping shrimp) to a more realistic cacophony of sounds that included an enemy vessel or a sperm whale. Trainees were taught to distinguish, recognize, and memorize the relevant sounds. This was done through repetitive listening.32 Technologies such as loudspeakers and recordings made it
Testing the Underwater Ear 309
possible for the thousands of sonar operator trainees to hear the same sound. Training records, manuals, and tests enabled a collective listening experience of, say, a specific, individual croaker (in part because the records, manuals, and tests presented a limited number of examples). The underwater sounds and the underwater ears were co-created, standardized, and made reliable and portable. Most of these recordings included a final test to assess whether students had acquired the necessary listening skills. Eight to ten sounds were played in a row and prospective sonar technicians were asked to identify them. Animal and human-made sounds recorded through hydrophones were now testing human ears and, by extension, testing the navy’s antisubmarine warfare defense system. The tests also provided data for Neff ’s psychophysical studies of perception on the hearing of pure tones against background noise, the ability to maintain attention under boring or stressful situations, and the perception of sounds of very short duration.33 The U.S. Navy’s tests were not the only ones being developed in this period, and nonhuman underwater ears were also being tested and trained. In the 1930s, German physiologist Karl von Frisch and a colleague performed a series of experiments on fish to determine whether the creatures had a hearing sense.34 Several species were conditioned to respond to a tone, usually a tuning fork, for food. The response varied among species and individuals but was always a somewhat frantic snapping for food (some would dive to the bottom, others would surface). The minnows (Phoxinus loevis) performed the best. Once conditioned, the minnows’ underwater ears were tested. A range of tone frequencies was used on the fish to test the limits of their hearing range. A second set of experiments, generating sounds at various distances from the aquarium, tested the hearing sensitivity of the fish. Lastly, tone differentiation was tested by training the fish to associate food with one tone and punishment (the rap of a rod) with another. The interval between the tones was decreased and the fish were observed distinguishing intervals as small as a half-tone. Experiments were also done on the trained fish to determine the physiological site of their hearing sense by surgically removing first the labyrinths of their ears, then sacculus and lagena structures, and observing their responses, or lack thereof, to tones. The fish with removed sacculus and lagena remained steady in the water (those lacking labyrinths were tippy) but only responded to very low tones, likely felt through their skin. The larger goal of these experiments and similar ones performed in the previous decades was to determine the mechanism of hearing in fish.35 Locating which physiological apparatus prompted reactions to tones seemed to be sufficient to define the nature of fish hearing.
310 World as Testbed
By the 1950s, however, these experiments were critiqued as inadequate because they had not demonstrated that the specific sounds tested were—or sound at all was—biologically significant to the fish.36 For our argument, this is particularly important because it indicates the beginning of a shifting understanding among biologists regarding the defining criteria of the sensory perception of sound. Recall that Johnson had similarly dismissed the significance of the snapping shrimp’s sound to the snapping shrimp (uncertain in its biological importance, he said). But as more sounds became available through military surveillance, new questions about their function arose. The postwar critique of Frisch and his colleagues was that they had only demonstrated the fish’s sensation of sound, not their perception of sound.37 The training and testing of the fish’s underwater ear had become a psychophysical problem. And as with all psychophysical problems, scientific observation of the individual’s subjective experience of sound was tricky. Especially when the individual was a fish. To sum up, an examination of the evolution of underwater ear tests—both human and nonhuman—in the decades in which antisubmarine warfare motivated the sonification of the sea reveals several interesting developments. First, there were ontological consequences. The need to distinguish human- made sounds from “natural” and “background” sounds of the ocean prompted the systematic recording of these nonhuman sounds; they existed for human ears for the first time. New ears were also created—or rather, new types of hearing among sonar operators, circumscribed by the tests themselves. Second, there were epistemological consequences, especially for the sciences. Marine biologists decided that if a fish’s sound-making was determined to be purposeful, then that must mean that within the species, such a sound was perceived.38 We place this shift within the context of the U.S. military purposefully training human ears to hear and act on specific sounds. We cannot claim direct causation, but certainly a correlation exists: once underwater human ears were purposeful, the scientific questions about underwater fish ears shifted to purposefulness as well. The ability and subsequent demand for intentional, directed underwater listening in humans co-developed with the understanding that other, nonhuman inhabitants of the ocean might also be listening very carefully both to their own species and to others.39 As a result, the experimental programs in the 1950s shifted to the systematic study of the sounds and behaviors of sound-making species. Marine biologists’ interests in better understanding animal behavior and animal perception, which pushed the limits of the human-centric conception of hearing, dovetailed with the technological abilities and needs of, say, the ONR reference file of marine animal sounds, to which we now turn.40
Testing the Underwater Ear 311
Listening Together: Separating and Stabilizing Sounds In their discussion of the early history of underwater bioacoustics in the 1970 “Reference File of Biological Underwater Sounds,” Sounds of Western North Atlantic Fishes, Marie Fish and William Mowbray explained that experiments conducted during and immediately after World War II made it clear that “definite identification of an unseen soundmaker require[d]instrumental analyses.”41 This was because the human ear was unilinear and therefore unable to distinguish between slightly different sounds of very short duration.42 And just as some different sounds were indistinguishable, the same sound could be described by human listeners any number of ways—as a bump, a thump, a knock.43 Spectrograms (previously called sonograms— we use actors’ terms) and oscillograms became “the standard analytical method for description and comparison of underwater sounds.”44 Human ears needed visual assistance, the transduction of pressure waves picked up by hydrophones into graphic representations, to improve their ability to hear underwater sounds and standardize their descriptions of them. Creating sonic and visual catalogs of biological sounds gave marine biologists a wealth of new investigative objects that would change the face of the discipline. But it is important not to lose sight of the military origins of this project. Researchers at the NOL performed two main lines of observations that would allow them to make the link between animal noises and animal communication. The first was a series of underwater recordings in 1942 and 1943 in locations that they considered to be particularly intense sonically, from Florida to New Jersey, under a variety of oceanographic conditions, seasons, and times of day. The NOL recorded noises picked up by the hydrophones onto discs, which were made available “for listening and identification”—it is interesting to note that a distinction is made here between listening and identification.45 The recordings were also rendered graphically.46 Researchers explained that the variation in volume and frequency of water noise among the different locations was mostly due to differences in the time the measurement was taken and differences in geographical location (and therefore of local marine life). These discussions of seasonal, diurnal, and geographical variations hinted at the animals’ behavior, both night feeding and the seasonal movement of the croakers common to every location. In the same period, a second line of research at the NOL with captive noise- making fish shows how understanding and archiving biological noises, separated from the context of the open ocean, was also increasingly of military interest. These experiments were done at the Shedd Aquarium in Chicago and the U.S.
312 World as Testbed
Fishery Biological Laboratory in Beaufort, North Carolina. The goals were to determine which species could be induced to make noise (perhaps in the interest of militarizing these creatures) and, more generally, to gain a better understanding of the nature of the noises and their purposes. Such knowledge would allow the navy to sort out the sources of sounds at sea as well as to predict their locations. The fish were put in three-foot-deep enclosures by species, separated by chicken wire, with mud bottoms upon which hydrophones were placed. The lab in Beaufort studied the Atlantic croaker (Micropogon undulatus), the toadfish (Opsanus tau), the hogfish (Orthopristis chysopterus), the spot (Leiostomus xanthurus), the sea robin (Prionotus carolinus), and the sea catfish (Felichthys felis). The croakers made noise spontaneously. The toadfish was “sluggish” and “ill-tempered” and refused to make sound in captivity at all.47 In 1954, the Navy determined that the assembly and integration of this data needed to be more systematic. Nuclear submarines were joining the underwater Cold War, and the strategy of seek and destroy was no longer adequate in times of tense peace. In a context of nuclear deterrence, the navy’s goal was the constant and global surveillance of Soviet submarines.48 With that in mind, the ONR asked researchers at the Narragansett Marine Laboratory to institute and maintain a reference catalog of biological underwater sounds. By 1970, the ONR’s “Reference File of Biological Underwater Sounds,” as the reference file was now called, included three hundred species, mostly fish, submitted by bioacousticians from all over the world. The reference file was to be continually supplemented and updated with new research and recordings. For each species, information on distribution, habits, size, sound production, and sound-making mechanism was included, along with spectrograms and oscillograms of the creature’s sound. The data was available to (and sought by) fishing interests, oceanographic agencies and industries, developers of acoustic devices, and others interested in “possible ambient background sounds at specific sea localities at particular seasons, and other certain circumstances, identification of unknown sounds in the sea.”49 We would like to claim two things here. First, these studies show that “natural background noise” or “biological water noise” was transformed into species-specific sounds. Each noise-making species now had a correlated description of its sound, a graphic representation of its sound’s frequency distribution, and a recording available for dubbing. Natural background noises and sounds were disaggregated, separated out according to their sources. But, second, these understandings of the various species’ sounds were only in the terms important to the Navy’s requirements—the seasonal variation of these species’ noise-making behavior and their frequency distribution. Clearly, this was not the only context in which marine biologists took an interest in sound,
Testing the Underwater Ear 313
but bioacousticians working for the Navy were the better-funded and most visible members of their discipline, and therefore dominated the scholarship in this period. Their experiments produced the information necessary to build appropriate filters on acoustic triggers and provide instructions for sonar operators about regional and seasonal sounds. At the end of the 1950s, maintenance of the ONR reference file was transferred to marine biologists, but it was still originally created by and for the underwater ears of sonar technicians.50 The ONR reference file became the standard against which to test both equipment and ears in the field. When the source of the sound was known, sonar operators on board a submarine could confirm what they had learned from their training records, reinforcing the authority of those sound objects. When they suspected a source but could not directly confirm it, they could compare it (or its graphic representation) to the recordings. When they could not match the new sound with anything previously heard or heard of, they could enrich the ONR reference file. As such, each use became a test of the reference file itself, increasing its credibility. And once embedded in the ONR reference file, the testing continued to standardize the underwater ears of both sonar technicians and marine biologists.
The Speciation of the Underwater Ear: Divergence and Diagrams In the mid-1960s, the human underwater ear speciated. Although technologically assisted underwater listening was pioneered by the military, informing the goals and priorities of how underwater sounds were heard, the marine biologists’ scientific interests and standards eventually diverged and they started to listen to the ocean differently. Scholars have recently begun to examine the relationship between the U.S. military and marine biologists in this period, offering significant insights into the tangled military and environmental goals of the new field of cetology.51 By documenting the role of testing in bringing together the various material, personal, and animal networks at work in the sonification of the ocean, we find not only a tense convergence between the military and marine biologists but also the precise point of divergence. A brief survey of how a single noise-making species was described over time illustrates this divergence of military and scientific underwater ears. Soviet submarines were not the only scary things making noise underwater. The toadfish (Opsanus tau, also called the Plainfish Midshipman) sits on the ocean floor and grumbles to itself, loudly. The toadfish was one of the original subjects of the NOL experiments and therefore one of the original
314 World as Testbed
entries in the marine sounds reference file. In 1947, its sound was described as “an intermittent, low-pitched musical blast of about 1/2-second duration . . . concentrated at the low-frequency end of the spectrum.”52 It was also described by simile, as like “a boat whistle,” a “strange musical toot . . . like a fog horn,” and a “ ‘foghorn’ toot” louder than ever before recorded at that gain setting.53 A 1952 navy report included a discussion of the toadfish, noting that it was actually associated with two kinds of sounds. The common one was “the musical blast” described by Dobrin, but there was also “a raucous grunt of about the same intensity, duration, and pitch, but of different quality.”54 A decade later, William Tavolga described the toadfish’s vocalization as a “so-called ‘boat-whistle’ sound” that began with “a grunt-like sound with a fundamental of about 50 to 75 cps”; he offered two spectrograms as illustrations.55 Not only were the graphic representations of the toadfish’s sound shifting from tidy plotted oscillograms to messier spectrograms, but also the accompanying biologists’ descriptions were changing. Similes were still employed, but the details of the descriptions focused on frequency, duration, and now harmonics—features captured by spectrograms. Indeed, the harmonics visible in one spectrogram and absent in the other confirmed the field observations (of higher-pitched calls in the Florida subpopulation). As was the case for the sonar technicians, the biologists’ experience and understanding of underwater sounds were bound up with the graphic representations of those sounds. This was, however, beginning to change. At the second Marine Bio-Acoustic symposium, a gathering of biologists and navy acousticians, in 1966, William Watkins presented his research on toadfish antiphony.56 His use of a spectrograph image troubled the otherwise mostly placid gathering. Several of the researchers present questioned the use of the spectrograph to represent underwater sound, and especially the ability of experimenters to accurately read pulse tone frequencies and repetition rates on such spectrographs. Others in the audience noted drawbacks with the Vibralyzer machine commonly used, arguing that baseline frequencies were sensitive to temperature changes and that multiple signals could result in cross-modulation. Mowbray, one of the managers of the ONR reference file, insisted that “a sonogram by itself is not enough in a published report to define a sound.” He continued: “The ear in many cases can easily detect the difference, but the reader cannot.”57 Efforts to standardize the graphic representation of underwater sounds were in fact testing the limits of representation. In 1970, the Reference File of Biological Underwater Sounds included a new table, a sort of reference sheet for communicating about sounds without graphic representation (Figure 11.1). It is a nice illustration of the
Marie Fish and William Mowbray, Sounds of Western North Atlantic Fishes: A Reference File of Biological Underwater Sounds (Baltimore: Johns Hopkins Press, 1970), xvii, Table 1. © 1970 Johns Hopkins University Press. Reprinted with permission of Johns Hopkins University Press.
Figure 11.1 Typical characteristics of biological underwater sounds.
316 World as Testbed
many ways in which the sounds of marine life could be described. Notice that the categories vary from acoustic (frequency, duration) to descriptions (“crackle,” “scrape”) to similes (“carpenter” “train”). Also included was an indication of the noise-making mechanism. This table could function like an identification flow chart. One might have a recording of a mysterious pulse sound. Well, is its basic sound a thump or a knock or a click? If it is a click, is it more of a raspy click or a creaky click? And so on. The end result is not necessarily species identification, but it would get the reader far enough to narrow the choices down to the relevant species section of the reference file. The table might also be used to imagine a particular species’ sound. Knowing its sound-making mechanism, the frequency and duration of sound, that it is a growl or a honk, a well-trained person could create the sound in their mind’s ear, supplementing a reading of a graphic representation. We discuss concrete examples of this phenomenon of seeing with one’s ears in the next section, but here it should be noted that this was already possible in 1970. By the standards of science, spectrographic images were unstable and unreliable; the spectrogram obscured the fickleness of the listening technologies behind it.58 For the marine biologists, the spectrographic image did not replace sound—they were not simply listening with their eyes. This subverts the narrative of the ever-increasing dominance of the visual in underwater listening. Rather, visualization technologies only contributed to the sonification of the oceans to the extent that they helped biologists understand what the hydrophones had recorded. This newly skeptical approach to graphic representation was made possible by three larger trends. First, likely informed by the rise of cognitive science as a discipline as well as the rise of ethology and animal communication, there was a growing interest among some biologists to define hearing more broadly.59 The testing of the limits of the hearing sense was, as discussed earlier, bound up with training the ears of fishes. Also, equipment became cheaper, easier to use, and more accessible, releasing the marine biologists from their dependency on the military for data collection. And finally, biology itself was moving away from taxonomy as the basis of analysis to behavior itself. Sounds might well be a means of better understanding behavior, which meant that sounds were analyzed in conjunction with observed behavior (courting, defense, and so on), prioritizing difference between individuals. A single curve representing the sound of an entire species did not advance this project. Thus listening—a very specific kind of listening—remained an important skill for biologists.
Testing the Underwater Ear 317
Listening with the Eyes and Seeing with the Ears Recall that the priority of Cold War submarine strategy was not destruction but surveillance. Though the post–World War II nuclear-powered submarines were faster and could remain submerged for much longer periods of time, they were also louder, and every engine produced unique pitch and rhythm patterns. These “sound signatures” allowed for individual identification, via another reference file designed for positive identification of submarine sounds rather than negative elimination of biological sounds. Aiming for constant global surveillance of each ship, in 1962, U.S. Navy engineers installed the first hydrophone array of the Sound Surveillance System (SOSUS). Two decades later, about thirty-five arrays of hydrophones covered main routes and passage points around the world and, via triangulation of signals, made possible the precise determination of specific sound sources’ positions. To connect every sound signal to a particular vessel, the navy developed a “library of Soviet submarines” classified according to sound signatures: a sound repository that would continuously expand as navies launched their new models. This volume of sound markers exceeded the capabilities of any individual’s memory. Sonar technicians were still integral to onboard tracking and maneuvering, but surveillance on a global scale was clearly beyond their ears. Additionally, while the SOSUS library theoretically could have consisted of sound recordings, the Navy chose to use sonograms instead. For classification purposes, an archive of images seemed easier to move, share, describe, and interpret. Whereas sonar operators often defined underwater listening as “seeing with the ears,” SOSUS operators were hearing with their eyes. The new perceptual possibilities opened by SOSUS’s visual library of sounds did much for the sonification of the ocean beyond antisubmarine warfare. Once the Cold War came to an end, researchers detecting earthquakes or tracking biological sources of sound were sometimes given access to SOSUS stations. These sonograms were often only useful, however, when combined with another sensory modality, namely listening. We conclude this chapter with a brief example from current marine bioacoustics suggestive of the distinctly different way in which listening is presently mobilized for such work. Chris Clark is the head of the bioacoustics lab at the Cornell Lab of Ornithology. He trained with Roger Payne, who, with Katy Payne, was the first to record and analyze the songs of humpback whales.60 Clark grew up in a musical household, full of jam sessions and a mother insisting that the family listen to the Brahms symphony wafting from the radio.61 He sang in a boys’ choir at the Cathedral of St. John the Divine in New York City, and can be described as highly trained in classical music. He told one of us about
318 World as Testbed
an incident in the early 1990s when he was given security clearance to work with some of the SOSUS data. He was even offered a tour of the facility where the arrays print out spectrograms in real time. Walking with an officer “down the beams” of the array, looking at the data, Clark inquired about a strange pattern. The officer waved it off as snapping shrimp, but her response did not make sense in terms of the sound’s duration. Clark guessed that the military was not particularly curious about the signal because it did not look like any thing human-made. He checked the frequency and again looked at the pattern. It had to be a fin whale. Soon, further down the array, he noticed another strange pattern at a different frequency. He describes reconstructing in his mind’s ear what such a pattern would sound like and determined that it was the song of a blue whale.62 For Clark, the exciting feature of the SOSUS system was that he could stand in a room in Newport News, Virginia, and experience a whale singing off the coast of Newfoundland, Canada. But we would like to highlight the role that Clark’s ear played in reading the SOSUS printout. Western classical musical training allows musicians to hear sounds in their mind’s ear as they read graphic notation on a page, before they then use their bodies to make the sounds. It seems that something similar was going on here. Clark’s early musical training and subsequent career in bioacoustics enabled him to further his knowledge about submerged creatures—indeed, to hear entirely new sounds. Of course, we are not arguing that Western classical music training is the only form of ear training that would make such auditory breakthroughs possible, but we find the connection for this particular individual to be especially compelling. Clark’s underwater ear enabled him to hear a whale from an image. Rather than seeing with their ears, as the sonar operators described their tasks, Clark listened with his eyes. We have focused in this chapter on the recursive and reinforcing role that testing played in the development of the underwater ear throughout the twentieth century, and in particular during and after World War II. The demands of submarine warfare motivated the rapid development of new technologies and new listening skills. In turn, sounds never before heard by humans were defined, classified, and standardized. The ocean became noisy. As both the nature of underwater warfare and the goals of biology shifted, the ways to listen below the surface multiplied. Testing, too, multiplied along with the goals of these new tests. Onboard sonar monitoring, global SOSUS arrays, spectrograms, and behavioral conditioning of humans and fish all pushed the transformation of the ocean into a soundscape.63 In this story, the dynamic interplay between signal and noise proved to be richer than in previous understandings. Hydrophone designers during World
Testing the Underwater Ear 319
War I restricted their understanding of signals to the information they received from a known emitter at a particular frequency. Everything else was noise, and any and all of it could be relevant for engaging enemy submarines. As the technology made ocean sounds audible to humans, human-made sounds became relevant signals and new distinctions between noises emerged: “background noises” were separated from “marine-life noises.” As the different marine-life noises were identified, classified, and recorded, and as bioacousticians became increasingly interested in animal behavior, this category itself broke apart. What sonar operators considered noise was meaningful for marine biologists interested in cetacean echolocation or fish communication. Another twist came recently, when cetologists began to understand anthropogenic sound as noise interfering with the signals necessary for marine mammals’ survival.64 The saturation of the ocean by human sounds, not least military sonar, now threatens to decimate its rich polyphony.65 Although we restricted our use of the concept of testing to actors’ categories, its richness has taken us from instruments and equipment to people’s selection and training, and from animal physiology to graphic archives of sound. Tests proved to be particularly malleable practices, full of meanderings and feedback loops. Intake tests became training tests became indicators of the need for new and different tests. Tests were also particularly fruitful in opening the way to new discoveries, as when Johnson and his team, testing different animals to find the source of a characteristic deafening sound, demonstrated that the volume of a snapping shrimp’s snap was magnified by large numbers. Because of their flexibility and productivity, tests could make commensurable the different realities they checked against one another. By deploying the ear to test the hydrophone and the hydrophone to test sound classification, tests laid the basis for new epistemes of underwater acoustics. Tests also played a crucial role in producing the new ontology of the sonic ocean. To be sure, the oceans were never a “silent world,” to return to Cousteau’s appellation. But making the oceans audible for humans was much more a transformation process than it was a discovery. Tests and calibrations made possible and trustworthy the electroacoustic transduction of underwater pressure waves into sounds perceptible to human ears on dry land. The point is further underscored when we consider the frequencies outside our auditory thresholds that populate the oceans and are only rendered audible through modern technology. There is some anthropocentrism in calling certain mechanical waves infra-and ultrasounds, since the designation only exists in reference to the human listener. This nevertheless reminds us that while “sounds,” in order to be heard, require subjects to exist as such, there is no need for these subjects to be human.66 Testing the underwater ear created
320 World as Testbed
and mobilized a plurality of sounds and listening subjects, which subsequently co-evolved, diverged, and clashed. Further materials related to this chapter can be found in the database “Sound & Science: Digital Histories”: https://acoustics.mpiwg-berlin.mpg.de/sets/clusters/ testing-hearing/testing-underwater-ear-camprubi-and-hui.
Notes 1. See Graham Burnett, Sounding of the Whale. 2. M. Norton Wise, “Mediations.” 3. Trevor Pinch, “ ‘Testing—One, Two, Three. . . .” 4. Joeri Bruyninckx, “Trading Twitter.” 5. Jean-Daniel Colladon and Charles-François Sturm, “Mémoire sur la compression des liquides.” 6. Crosbie Smith and M. Norton Wise, Energy and Empire; John Shiga, “Sonar.” 7. Willem Hackmann, Seek and Strike, 3–10. 8. Roland Wittje, Age of Electroacoustics, 96. 9. Hackmann, Seek and Strike; Elizabeth Bruton and Paul Coleman, “Listening in the Dark.” 10. Gary L. Frost, “Inventing Schemes and Strategies”; Shaul Katzir, “Who Knew Piezoelectricity?” For developments on the German side, see Wittje, Age of Electroacoustics, 89–98. 11. Hackmann, Seek and Strike, 139–228. 12. Emily M. Smith and Frederic C. Bartlett, “On Listening to Sounds of Weak Intensity. Part II,” 166; see also Wittje, Age of Electroacoustics, 112, 202. 13. Akiko Saito, Bartlett, Culture and Cognition. 14. Emily M. Smith and Frederic C. Bartlett, “On Listening to Sounds of Weak Intensity. Part I,” 111–13. 15. Marvin Lasky, “Historical Review.” 16. Helen Rozwadowski, Fathoming the Ocean; Sabine Höhler, “Depth Records and Ocean Volumes.” 17. Lasky, “Historical Review”; see also Malcolm Llewellyn-Jones, The Royal Navy and Anti- Submarine Warfare, 23–24, 34. 18. Martin W. Johnson, F. Alton Everest, and Robert W. Young, “Role of Snapping Shrimp,” 122, 137. We thank Alistair Sponsel for pointing us toward this text and for sharing his insights on snapping shrimp and antisubmarine warfare. 19. H. U. Sverdrup, Martin Johnson, and Richard H. Fleming, The Oceans; Jacob Hamblin, “Seeing the Oceans,” 352–56. 20. Johnson, Everest, and Young, “Role of Snapping Shrimp,” 124. 21. Ibid., 133–37. 22. Ibid., 128. Johnson would gain fame in the last months of the war for hypothesizing that the Deep Scattering Layer was of an organic nature. Drawing on his work on microbiology for The Oceans, he predicted that the middepth layer that disturbed sound propagation
Testing the Underwater Ear 321 would have diurnal variations following plankton behavior. John A. McGowan, “Martin W. Johnson.” 23. M. B. Dobrin, “Measurements of Underwater Noise,” 19; Johnson, Everest, and Young, “Role of Snapping Shrimp,” 124. 24. It also covered the wave and wind sound of the sea itself—the “state 1 sea” noise was, for example, the sound of a “typical sea” of waves less than 30 cm high. 25. On the enormous workforce and institutional structure of U.S. World War II underwater research, see Lasky, “Historical Review.” 26. His goal was to determine function (for instance, sound localization) by systematically removing sensory organs and cortical areas. Jay M. Goldberg and Nelson Y.-S. Kiang, “William Duwayne Neff ”; William D. Neff and Wolf Keidel, Handbook of Sensory Physiology, vol. 5. 27. The studies were published as part of postwar publications by the National Defense Research Committee, led by Vannevar Bush and James B. Conant. Bush and Conant, Summary Technical Report. 28. Mara Mills, “Deafening.” 29. Alexandra Hui, “The Bias of ‘Music-Infected Consciousness.’ ” 30. Cents are a measure of interval (not frequency). An equal-tempered semitone spans 100 cents. Currently, just-noticeable-difference of tone is accepted as 5 to 6 cents. 31. Bush and Conant, Summary Technical Report, 4:11–18. 32. “The most important part of your job as operators . . . is that you would be able to recognize and interpret the sounds made by a submarine maneuvering under water. There is only one way for you to develop this ability; that is, to listen to such sounds and study them many, many times.” U.S. Navy Underwater Sound Laboratory, JP Sonar Training Records. Sound files can be found at https://maritime.org/sound/index.htm. 33. William D. Neff and W. R. Thurlow, “Auditory Discrimination in Sonar Operation.” 34. Karl von Frisch and H. Stetter, “Untersuchung über den Sitz des Gehörsinnes”; Karl von Frisch, “Über den Gehörsinn der Fische”; Frisch, “Zur Psychologie des Fisch-Schwarmes”; Frisch, “Sense of Hearing in Fish.” 35. J. P. Froloff, “Bedingte Reflexe bei Fischen I”; Froloff, “Bedingte Reflexe bei Fischen II”; H. O. Bull, “Studies on Conditioned Responses in Fishes. I”; Bull, “Studies on Conditioned Responses in Fishes. II”; F. B. Manning, “Hearing in the Goldfish.” 36. James Moulton, “Influencing the Calling of Sea Robins.” 37. O. Lowenstein and T. Roberts, “Localization and Analysis.” 38. James Moulton, “Eliciting and Suppressing of a Marine Biological Sound”; Moulton, “Swimming Sounds”; James Moulton and Richard Backus, “Annotated References”; Lowenstein and Roberts, “Localization and Analysis.” 39. Tavolga explained that in bioacoustics, “signals need not be voluntary or purposeful in an anthropomorphic sense, but could be the direct result of the normal activity of the animal. The reception of the signals is not in itself sufficient for the situation to be communicative. The receiving animal must react in some way to indicate the signal has been perceived, but this response may not necessarily involve an overt behavior. . . . Hearing, then, remains as the most effective sensory modality for an aquatic animal, and the probability is that if an animal produces any sort of sound, some other animal would be capable of receiving and reacting to the signal.” William Tavolga, Review of Marine Bio-Acoustics, 76.
322 World as Testbed 40. There was also some discussion of how this research might provide the insight necessary to use sound to control certain fish species, we suppose for the fishing industry and possibly military defense (see the snapping shrimp story), but it only worked for certain species, such as sea robins. And who, really, wants to herd sea robins? 41. Marie Fish and William Mowbray, Sounds of Western North Atlantic Fishes, xiii. 42. There was, of course, extensive psychophysical research on precisely this phenomenon in the late nineteenth and early twentieth centuries. 43. Fish and Mowbray, Sounds of Western North Atlantic Fishes, xiii. 44. Ibid. 45. Dobrin, “Measurements of Underwater Noise,” 20. 46. The discs were played through an octave analyzer and into a series of tape recorders (one octave per tape). Spectra were calculated from the tapes and generated plots comparing locations as well as seasonal and diurnal variation at single locations. Dobrin, “Measurements of Underwater Noise.” 47. Experimenters eventually tracked one with a hydrophone in the open water until the sound was at its maximum intensity. A baited crab trap was dropped at this location and when it was pulled up a week later, it had a toadfish inside. Ibid., 21. 48. Gary E. Weir, An Ocean in Common. 49. Fish and Mowbray, Sounds of Western North Atlantic Fishes, xiv. 50. Max Ritts and John Shiga, “Military Cetology.” 51. Burnett, Sounding of the Whale; Ritts and Shiga, “Military Cetology”; Mette Bryld and Nina Lykke, Cosmodolphins; John Durham Peters, Marvelous Clouds, esp. ch. 2, “Of Cetaceans and Ships; or, the Moorings of Our Being.” 52. Dobrin, “Measurements of Underwater Noise,” 21. 53. Milton Dobrin, “Recording Sounds of Undersea Life,” 94. 54. L. H. Rumbaugh, Investigation of Biological Underwater Background Noises, 13. Rumbaugh listened to the toadfish in the navy’s spirit of identifying all underwater sounds. We might read this as evidence of curiosity above and beyond military goals. Although there is some degree of speculation in our argument, Rumbaugh’s choice seems to show he was aware that a single sonogram curve did not fully represent the sounds of an individual fish, let alone an entire species. 55. Tavolga, Review of Marine Bio-Acoustics, 42. 56. On the tense relations between navy scientists and cetologists at the first symposium, in 1963, see Burnett, Sounding of the Whale, 517–646. 57. William Tavolga, Marine Bio-Acoustics; William Watkins, “The Harmonic Interval,” 43. 58. The lengthy discussions of instruments and settings and calibration in every scholarly article at the time reflect this anxiety. 59. Gregory Radick, Simian Tongue. 60. The Paynes’ 1970s LP fundamentally altered public attitudes to whales. As an auditory moment of the American environmental movement, Songs of Humpback Whales was comparable to the silence evoked by Rachel Carson. The LP jacket included a lengthy tutorial on how to analyze the spectrograph of the whales’ vocalizations, a discussion of the whaling industry’s destruction of the species, pen and ink drawings of whales and whaling ships, plots of population collapses, and a single color photograph of a ship deck awash in whale blood. The final page was a set of perforated tear-out cards that could be filled out to indicate one’s commitment to whale conservation and mailed in along with a donation.
Testing the Underwater Ear 323 61. Author interview with Chris Clark, April 11, 2014. 62. Ibid. 63. Stefan Helmreich, “An Anthropologist Underwater.” 64. Ritts and Shiga, “Military Cetology.” 65. Joshua Horwitz, War of the Whales. 66. Lino Camprubí, “Sonic Construction of the Ocean.”
References Bruton, Elizabeth, and Paul Coleman. “Listening in the Dark: Audio Surveillance, Communication Technologies, and the Submarine Threat During the First World War.” History and Technology 32, no. 3 (2016): 245–68. Bruyninckx, Joeri. “Trading Twitter: Amateur Recorders and Economies of Scientific Exchange at the Cornell Library of Natural Sounds.” Social Studies of Science 45, no. 3 (2015): 344–70. Bryld, Mette, and Nina Lykke. Cosmodolphins: Feminist Cultural Studies of Technology, Animals and the Sacred. London: Zed Books, 2000. Bull, H. O. “Studies on Conditioned Responses in Fishes. I.” Journal of the Marine Biological Association of the United Kingdom 15, no. 2 (1928): 485. Bull, H. O. “Studies on Conditioned Responses in Fishes. II.” Journal of the Marine Biological Association of the United Kingdom 16, no. 2 (1930): 615–37. Burnett, Graham. The Sounding of the Whale: Science and Cetaceans in the Twentieth Century. Chicago: University of Chicago Press, 2012. Bush, Vannevar, and James B. Conant, eds. Summary Technical Report of Division 6, vol. 4: Methods and Devices Developed for the Selection and Training of Sonar Personnel. Washington, DC: National Defense Research Committee, 1946. Camprubí, Lino. “The Sonic Construction of the Ocean as the Navy’s Operating Environment.” In Navigating Noise, edited by Nathanja van Dijk, Kerstin Ergenzinger, Christian Kassung, and Sebastian Schwesinger, 219–45. Cologne: Walther König, 2017. Colladon, Jean-Daniel, and Charles-François Sturm. “Mémoire sur la compression des liquides et la vitesse du son dans l’eau.” Annales de chimie et de physique, série 2, 36 (1827): 113–59, 225–57. Dobrin, Milton. “Measurements of Underwater Noise Produced by Marine Life.” Science 105, no. 2714 (1947): 19–23. Dobrin, Milton. “Recording Sounds of Undersea Life.” Transactions of the New York Academy of Sciences 11 (1949): 91–96. Fish, Marie, and William Mowbray. Sounds of Western North Atlantic Fishes: A Reference File of Biological Underwater Sounds. Baltimore: Johns Hopkins Press, 1970. Frisch, Karl von. “The Sense of Hearing in Fish.” Nature 141 (1938): 8–11. Frisch, Karl von. “Über den Gehörsinn der Fische.” Biological Reviews 11, no. 2 (1936): 210–46. Frisch, Karl von. “Zur Psychologie des Fisch-Schwarmes.” Naturwissenschaften 26, no. 37 (1938): 601–6. Frisch, Karl von, and H. Stetter. “Untersuchung über den Sitz des Gehörsinnes bei der Elritze.” Zeitschrift für vergleichende Psychologie 17 (1932): 686–801. Froloff, J. P. “Bedingte Reflexe bei Fischen I.” Pflügers Archiv für die gesamte Physiologie des Menschen und der Tiere 208 (1925): 261–71. Froloff, J. P. “Bedingte Reflexe bei Fischen II.” Pflügers Archiv für die gesamte Physiologie des Menschen und der Tiere 220 (1928): 339–49.
324 World as Testbed Frost, Gary L. “Inventing Schemes and Strategies: The Making and Selling of the Fessenden Oscillator.” Technology and Culture 42, no. 3 (2001): 462–88. Goldberg, Jay M., and Nelson Y.-S. Kiang. “William Duwayne Neff.” In Biographical Memoirs, vol. 89. Washington, DC: National Academies Press, 2006. Hackmann, Willem. Seek and Strike: Sonar, Anti-Submarine Warfare and the Royal Navy, 1914– 54. London: Her Majesty’s Stationery Office, 1984. Hamblin, Jacob. “Seeing the Oceans in the Shadow of Bergen Values.” Isis 105, no. 2 (2014): 352–63. Helmreich, Stefan. “An Anthropologist Underwater: Immersive Soundscapes, Submarine Cyborgs, and Transductive Ethnography.” American Ethnologist 34, no. 4 (2007): 621–41. Höhler, Sabine. “Depth Records and Ocean Volumes: Ocean Profiling by Sounding Technology, 1850–1930.” History and Technology 18, no. 2 (2002): 119–54. Horwitz, Joshua. War of the Whales. New York: Simon and Schuster, 2014. Hui, Alexandra. “The Bias of ‘Music-Infected Consciousness’: The Aesthetics of Listening in the Laboratory and on the City Streets of Fin-de-Siècle Berlin and Vienna.” Journal of the History of the Behavioral Sciences 28, no. 3 (2012): 236–50. Johnson, Martin, F. Alton Everest, and Robert W. Young. “The Role of Snapping Shrimp (Crangon and Synalpheus) in the Production of Underwater Noise in the Sea.” Biological Bulletin 93, no. 2 (1947): 122–38. Katzir, Shaul. “Who Knew Piezoelectricity? Rutherford and Langevin on Submarine Detection and the Invention of Sonar.” Notes & Records of the Royal Society 66, no. 2 (2012): 141–57. Lasky, Marvin. “A Historical Review of Underwater Acoustic Technology 1916–1939 with Emphasis on Undersea Warfare.” U.S. Navy Journal of Underwater Acoustics 24, no. 4 (1974): 597–624. Llewellyn-Jones, Malcolm. The Royal Navy and Anti-Submarine Warfare, 1917–49. London: Routledge, 2006. Lowenstein, O., and T. D. M. Roberts. “The Localization and Analysis of the Responses to Vibration from the Isolated Elasmobranch Labyrinth: A Contribution to the Problem of the Evolution of Hearing in Vertebrates.” Journal of Physiology 114, no. 4 (1951): 471–89. Manning, F. B. “Hearing in the Goldfish in Relation to the Structure of Its Ear.” Journal of Experimental Zoology 41 (1924): 5–20. McGowan, John A. “Martin W. Johnson. Marine Biology: San Diego.” In 1987, University of California: In Memoriam, edited by UC Academic Senate, 47–50. Berkeley: University of California, 1987. Mills, Mara. “Deafening: Noise and the Engineering of Communication in the Telephone System.” Grey Room 43 (2011): 118–43. Moulton, James. “The Eliciting and Suppressing of a Marine Biological Sound.” Bulletin of the Ecological Society of America 36, no. 3 (1955): 80. Moulton, James. “Influencing the Calling of Sea Robins (Prionotus spp.) with Sound.” Biological Bulletin 111, no. 3 (1956): 393–98. Moulton, James. “Swimming Sounds and the Schooling of Fishes.” Biological Bulletin 119, no. 2 (1960): 210–23. Moulton, James, and Richard Backus. “Annotated References Concerning the Effects of Man- Made Sounds on the Movements of Fishes.” Fisheries Circular 17 (1955): 1–8. Neff, William D., and Wolf Keidel, eds. Handbook of Sensory Physiology, vol. 5: Auditory System. New York: Springer, 1974. Neff, William D., and W. R. Thurlow. “Auditory Discrimination in Sonar Operation.” In Human Factors in Undersea Warfare, edited by Committee on Undersea Warfare, 219–31. Washington, DC: National Research Council, 1949. Peters, John Durham. The Marvelous Clouds: Toward a Philosophy of Elemental Media. Chicago: University of Chicago Press, 2015.
Testing the Underwater Ear 325 Pinch, Trevor. “‘Testing—One, Two, Three . . . Testing!’: Toward a Sociology of Testing.” Science, Technology, & Human Values 18, no. 1 (1993): 25–41. Radick, Gregory. The Simian Tongue: The Long Debate About Animal Language. Chicago: University of Chicago Press, 2007. Ritts, Max, and John Shiga. “Military Cetology.” Environmental Humanities 8, no. 2 (2016): 196–214. Rozwadowski, Helen. Fathoming the Ocean: The Discovery and Exploration of the Deep Sea. Cambridge, MA: Harvard University Press, 2008. Rumbaugh, L. H. Investigation of Biological Underwater Background Noises in Vicinity of Beaufort, NC. Naval Ordnance Laboratory Report no. 880, publication no. 66121, 1944. Saito, Akiko, ed. Bartlett, Culture and Cognition. Hong Kong: Psychology Press, 2000. Shiga, John. “Sonar: Empire, Media, and the Politics of Underwater Sound.” Canadian Journal of Communication 38, no. 3 (2013): 357–77. Smith, Crosbie, and M. Norton Wise. Energy and Empire: A Biographical Study of Lord Kelvin. Cambridge: Cambridge University Press, 1989. Smith, Emily M., and Frederic C. Bartlett. “On Listening to Sounds of Weak Intensity. Part I.” British Journal of Psychology 10, no. 1 (1919): 101–29. Smith, Emily M., and Frederic C. Bartlett. “On Listening to Sounds of Weak Intensity. Part II.” British Journal of Psychology 10, no. 2 (1920): 133–68. Sverdrup, H. U., Martin Johnson, and Richard H. Fleming. The Oceans: Their Physics, Chemistry, and General Biology. New York: Prentice Hall, 1942. Tavolga, William, ed. Marine Bio-Acoustics: Proceedings of a Symposium Held at the Lerner Marine Laboratory, Bimini, Bahamas, April 11 to 13, 1963. New York: Pergamon Press, 1964. Tavolga, William. Review of Marine Bio-Acoustics. Technical Report NAVTRADEVCEN 1212- 1. Port Washington, NY: U.S. Naval Training Device Center, 1965. U.S. Navy Underwater Sound Laboratory. JP Sonar Training Records. Audio files retrieved from U.S. Naval History and Heritage Command: http://www.hnsa.org/resources/historic-naval- sound-and-video/jp-sonar/. Last accessed June 27, 2017. Watkins, William. “The Harmonic Interval: Fact or Artifact in Spectral Analysis of Pulse Trains.” In Marine Bio-Acoustics, vol. 2: Proceedings of the Second Symposium on Marine Bio- Acoustics Held at the American Museum of Natural History, New York, April 13–15, 1966, edited by William Tavolga, 15–43. New York: Pergamon Press, 1967. Weir, Gary E. An Ocean in Common: American Naval Officers, Scientists, and the Ocean Environment. College Station: Texas A&M University Press, 2001. Wise, M. Norton. “Mediations: Enlightenment Balancing Acts, or The Technology of Rationalism.” In World Changes: Thomas Kuhn and the Nature of Science, edited by Paul Horwich, 207–56. Cambridge, MA: MIT Press, 1993. Wittje, Roland. The Age of Electroacoustics: Transforming Science and Sound. Cambridge, MA: MIT Press, 2016.
12 This Is Not a Test Listening with Günther Anders in the Nuclear Age Benjamin Steege
More Than a Test “Most of the features that experimental psychology finds in its test subjects arise as products of the experiments themselves,” wrote the philosophical anthropologist and peace activist Günther Anders (1902–1992) in 1965. “They are in a certain sense the red pressure marks that the experimental apparatus has produced, has in some cases permanently left behind. What the experiments make visible is not how someone is, but rather what can be made out of them. That one can make many things of them is of course indisputable. But one does not thereby believe oneself to know something about them.”1 Given this attitude, one might think that Anders would be an unlikely figure for inclusion in a volume on the hearing test. Far from participating or taking direct interest in any specific testing practice, Anders represented a strain of antiscientific thought, to put it in crude terms, deeply and sometimes blindly skeptical of what he felt to be an accelerating technicization and instrumentalization of the human lifeworld. The very concept of the test was suspect for him because, in the form of “human engineering”—which is to say, in particular, the early and midcentury use of standardized testing practices to discover and cultivate individual skills and render the workforce more efficient—it masked its power over the test subject with claims to neutral knowledge, extending unchecked from the laboratory to the workplace and to the home.2 It would be easy to imagine that Anders’s antipathy toward the culture of testing began at home. His father, William Stern, had coined the notion of “psychotechnics” in 1903, as well as the classic psychotechnical notion of the “intelligence” or “mental quotient” in 1912, and had in fact taken the child Günther from birth to age seven as a central subject of his 1914 study in child development.3 One might then comfortably read the son’s stance as taking Benjamin Steege, This Is Not a Test In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0013.
328 World as Testbed
place against an Oedipal background, a reaction against what he saw as paternalistic big science, all the more so as Stern’s foundational description of psychotechnics had made explicit that this was to be a style of research that would directly support the institutions of liberal governance, providing concrete techniques for improving the efficacy of pedagogical, medical, and legal institutions (in the latter area promising surer judgment of witness testimony as well as determining more effective punishments). In the end, he was superseded by other psychologists in setting the agenda for a future psychotechnics and in his later career moved away from this project.4 Yet whatever the nature of the filial psychodrama, the very fact that the test, in Anders’s perspective, was a scene not of stabilizing knowledge but of change and even indeterminacy meant that it leapt straight into the path of history. Though Anders, over a span of decades, was consistently excoriating on the emergent cultural significance of testing and experiment, the basic claim that occasions the present chapter is that he also maintained a countervailing positive evaluation of the possibility of “making something of someone” that he otherwise denounced in the test situation. In other words, the following discussion explores what happens at the moments when Anders ironically inverts his negative evaluation of “human engineering” and admits the potential redirection of its basic cultural ambition—that is, the effort to transform the human person in a certain accordance with changed technological circumstances—toward other ends than those originally intended. I then attempt to follow up this rather theoretical possibility through a thematic description of a culturally proximate aesthetic document: a midcentury electronic composition that directly (and, it must be said, quite unusually) engages Anders at the level both of his own theoretical concerns and of the concrete historical experience of various kinds of testing. Though Anders’s distaste for the experimentalization of life was given form through antipsychologistic attitudes adopted from his teachers Edmund Husserl, Max Scheler, and Martin Heidegger, the clear point of articulation came with the onset of the Cold War. In the spring of 1954, the U.S. military tested a series of hydrogen bombs on Bikini Atoll in the occupied Marshall Islands. For Anders, observing the news of these detonations from Vienna, where he had relocated following wartime exile as a German-Jewish refugee, the “Operation Castle” tests, as they were known, showed that the belief that Hiroshima had brought an end to the waves of escalating violence of the previous decades was precisely the opposite of the truth. These detonations were one thousand times more powerful than the 1945 bombs, and their very unimaginable scale and enduring aftereffects made it delusional to refer to them as “tests” or “experiments” at all. As Anders wrote in his
This Is Not a Test 329
1956 treatise in philosophical anthropology, Die Antiquiertheit des Menschen (“The Obsolescence of the Human”), which might be read as an antinuclear manifesto: That one carries them out “island-fashion” on coral reefs or somewhere else in the ocean, that one thus finally resorts to “insularity” in its original geographical sense in order to preserve the principle of isolation, is to be sure quite characteristic. But these final attempts at isolation—that is, the desperate efforts still to perform “tests” [“Versuche”]—remain futile. However fantastically the tests may succeed, the testing [das Versuchen] fails, because every testing immediately goes off course, becomes more than testing. The effects are so egregious that in the moment of the experiment [Experiment], the “laboratory” becomes coextensive with the globe. But that means nothing else than that the distinction between “rehearsal” and “performance” has lost its meaning, that every “experiment” has become “the real thing.”5
Anders was responding here to the catastrophically mispredicted range of fallout from the first completed test detonation, “Castle Bravo,” which resulted in a massive radioactive coral dust cloud twice as large as what engineers had forecast, contaminating seven thousand square miles of the Pacific Ocean. This error led to widespread miscarriages, birth defects, and thyroid cancer among Marshallese Islanders over subsequent years, and seems also to have led to the premature deaths of up to ten Japanese fishermen who were aboard a tuna boat named Lucky Dragon No. 5 and were caught unawares inside the expanded fallout zone.6 Anders interpreted this event not as simply an isolated scientific miscalculation, but rather as a sign of the essential incalculability of the material result of any engagement of nuclear energy. But he went further. The spectacle of Castle Bravo and of what became known as “The Lucky Dragon Incident” did not just scandalize the idea of the nuclear test; rather, for Anders, it suggested that whatever had once been called a “test,” nuclear or otherwise, would now require significant reinterpretation. As late as the 1980s, he doubled down on his critique of what he considered a falsely innocent notion of experiment, writing that “our entire activity [as a species] is . . . ‘pseudoexperimental’ ” in the sense that “whatever we undertake, our undertakings represent experiments that are irreversible and whose consequences we will not be able to foresee.”7 We might dismiss all of this as a kind of uncritical antimodernism. Anders made no pretense to philosophy of science, and his reflections on the status of the test and the experiment were purposefully and strategically blunt, intended to jar readers out of complacency.8 More specifically, we might wonder,
330 World as Testbed
in the generalizing leap from the specific case of U.S. military nuclear testing in the Marshall Islands to “Testing” with a capital T, what Anders could have meant by such an indiscriminate indictment. For one thing, as the present volume attests, there is good reason not to conflate experimenting in general with testing more specifically as a set of strategies for discovering the capacities, limits, properties, or other salient features of human and nonhuman beings. Moreover, there are obvious differences between psychotechnical tests and field tests (of which the nuclear test is an especially dramatic example). The psychotechnical test was designed to find out how individuals would perform circumscribed tasks that might in principle be translated from a laboratory or testing station to a factory floor or home or battlefield, whereas the field test of necessity takes in a specific environment as part and parcel of what is to be tested: in other words, not just a thing but also its interaction with untranslatable conditions, a feature that allows Anders to generalize outward to an image of the test fully unbounded. My sense of Anders’s deployment of terms and emphasis, however, is that his primary concern was indeed with testing specifically, not with experimentation more generally, though with the qualifying insight that tests share with experiments the potential to generate something new rather than confirming how something is. In fact, the idea of the test had troubled Anders’s thought at key moments from the 1920s on. Yet it appears not to have been particularly, or even primarily, natural-scientific in character; it was rather an element of a background theoretical strain more aesthetic than anything else. Anders had joined in extended reflections, centering around his friends Bertolt Brecht and Walter Benjamin (who was also his cousin), on how the historical emergence of testing in various domains of more or less everyday experience since at least the end of World War I could be seized upon as material for aesthetic reworking and thereby turned to political or ethical advantage. The question then arises, for Anders as much as for us, whether this maneuver, a refunctioning of the test, might still be possible in the context of Anders’s bleak Cold War diagnoses. As I will discuss later, a provisional answer lies in the suggestion that Anders’s aestheticizing conception of the test is partly a matter of formulating technical strategies, techniques, for exercising one’s moral and emotional imagination. Finally, it will emerge that Anders, whose career ambitions prior to around 1930 had been in the philosophy of music, sees hearing—and specifically musical listening—as a model form of such a technique. Listening to music, he would suggest in a crucial yet underdeveloped moment of theoretical optimism, might in principle occasion the sort of work on oneself that could be the only honest response to the existential crisis of impending nuclear violence.
This Is Not a Test 331
An Aesthetic Theory of Testing If the experience of being tested has come to assume a banal ubiquity today, the comparative novelty of, for example, career aptitude testing in the years following World War I was disarming enough to have furnished material for both affirmative and critical reflection in various aesthetic milieus within the Weimar Republic and the early Soviet Union.9 Brigid Doherty has explored Brecht’s and Benjamin’s ambivalent responses around 1930 to a cultural practice that on the face of it would appear to be a matter of rationalizing human potential and converting personal virtues or capacities into discrete, marketable gestures.10 As Doherty observes, the critic and the playwright intuited that the psychotechnical aptitude tests of their historical moment were in a certain sense already quasi- Brechtian: they consisted of a theatricalization of interrupted bodily movements broken down into defamiliarized fragments, as well as Haltungen—postures or attitudes—that could be studied and learned, in theory, to disassemble the person in its normal habits and rebuild it in an Ummontierung or “reassembly,” fashioning a revised persona from the rubble. “This is something new, and what is new about it is that it can be learned,” Benjamin explained in a radio talk in reference to Brecht’s adoption of what may be described as a kind of psychotechnical poetics.11 The larger significance of the analysis of habits of gesture and thought for Brecht and Benjamin was that it would enjoin a manner of detached observation among spectators, mimicking or transposing the position of the test administrator and neutralizing the false satisfaction of merely empathizing with aesthetic events and characters. When Brecht declared in 1931 that “the new school of play-writing must systematically see to it that its form includes ‘experiment,’ ” the connotation of that term was not simply freewheeling poetic invention, casting about for new modes of expression for the sake of innovation, but rather a more specific putting-to-the-test of given figures of contemporary life.12 As Benjamin set out the idea in a 1934 Paris lecture: At the center of [Brecht’s] experiment stands the human being. Present-day man; a reduced man therefore, chilled in a chilly environment. But since this is the only one we have, it is in our interest to know him. He is subjected to tests, examinations. What emerges is this: events are alterable not at their climaxes, not by virtue and resolution, but only in their strictly habitual course, by reason and practice. To construct from the smallest elements of behavior what in Aristotelian dramaturgy is called “action” is the purpose of the [Brechtian] theater.13
Benjamin’s formulation thus involves an important modulation that anticipated the paradox Anders would observe in 1965: the scene of the test—and
332 World as Testbed
its theatrical application makes this more explicit—does not transparently unveil to an observer the aptitudes, abilities, capacities, or virtues of the test subject but can instead, by interrupting the unreflected course of habits and events, compel a recognition of the potential for changing those habits and events. Since such a change is politically neutral, this potentiality becomes the pivot about which a cultural practice that had entered the world as a technique of human engineering could in principle be reoriented as a technique of emancipation.14 Anders picked up this theme in his recollection of a 1941 conversation with an assenting Brecht shortly after they had both arrived in Los Angeles (just months after Benjamin’s suicide). He reports himself commenting to the playwright at some length on how the Brechtian “test” or “experiment,” alternately Versuch and Experiment, takes its spectators to be like observers of a demonstration test at a physics practicum, such that the proceedings on the stage may reveal the “conditions for possible [social] transformations.” The aesthetic experiment should not simply represent life; it should “indicate not only what unfolds physically or chemically, but rather also, if the experiment [Experiment] is to succeed, what one has to do or allow to happen.”15 It does not go without saying that an experiment may be didactically transformative, as opposed to merely exploratory or aleatoric, let alone confirmational. There is, again, some slippage here between experiment and test, but it is worth noting that rather than immediately opening up to the more general category of aesthetic experimentalism (though in a sense, it does this too), the apparent confusion of terms can in fact help to keep our focus on a more specific point. What is relevant about Anders’s rhetoric is the way it seems to characterize the experimental test as falling so evenly between enabling liberatory transformation and opening the person up to manipulations of power beyond its control. This tension had been present in Brecht and Benjamin, but they had optimistically insisted that authors and artists should become conscious and hence critical of their function under prevailing modes of production, so as not to remain complicit in an ongoing manipulation of affect, gesture, and persona. Yet now, after the war, disillusioned, grieving, and numb, would a similar optimism remain possible?
Techniques of Feeling The answer that appeared in “The Obsolescence of the Human” is not encouraging if one is looking to hope for a better world to come. In that work, where Anders rages against the delusion that the nuclear test was anything other
This Is Not a Test 333
than pseudo-experimental as it overflowed the bounds of the test scene, these terms and concepts are irredeemably tainted. Hope itself is ironically diagnosed as a cynical attitude that obscures the severity of the situation. The 1956 book contains no vision for an experimental aesthetics per se. What it does contain, and this is in fact one of its primary theses, is an argument for the recognition of the historical malleability of feeling, by which Anders means an active expansion of emotive imagination—his term is Vorstellung—which is susceptible to historical intervention and experimentation at the level of “technique” (Technik). His philosophical anthropology, articulated in essays going back to the 1920s, was premised on the belief that there is nothing like a fixed human nature to begin with, unless that nature is to be understood as lying precisely in its artificiality: The history of styles and morals is a never-broken chain of enterprises in which humanity has tried to compensate for its own indeterminacy through commitments imposed on itself; to define itself socially and psychologically always anew; always to make something new of itself; always something that, “of its nature,” it had not been; but which, insofar as it wants to be at all, it had to be, because it could only function as a defined society, however artificial this might be.16
Anders really made two noteworthy claims on this score: first, the claim that feeling, like “morals,” had a history in the first place, that we should not assume people have always inhabited just those possibilities of feeling presently held open to them; and second, the claim that at junctures where the pace of historical change suddenly increased, as it seemed to have done in the mid-twentieth century, the capacity for feeling in a manner appropriate to the newly formed environment would not spontaneously keep pace.17 An undeniable and growing gap, disproportion, or Gefälle, between the astonishingly enhanced ability to execute technical aims (killing or dispossessing large numbers of people at a stroke, with comparatively little forethought) and the shameful inability to imagine their consequences made it urgent to recognize opportunities for picking up some of the slack in that failed moral imagination, to begin practicing “techniques of feeling”—or at least to pose a question as to whether any transformation or expansion in feeling would be possible in the first place. Near the end of “The Obsolescence of the Human,” Anders returns to the theme of human engineering, which he had begun the book by scorning. The salutary aspect of “testing” or “experimenting” lies in its capacity not to prescribe a course of action or behavior nor to identify something fixed and innate, but to raise the possibility of imaginative revision, so there must be no inhibition about using the characteristic gestures of an experimental
334 World as Testbed
attitude for other ends. “The weapons of the attacker determine those of the defender,” he writes, signaling a “refunctioning” (Umfunktionierung) in the Brechtian manner. “If it is our fate to inhabit a world (produced by ourselves) that in its excess eludes our imagination and our feeling and thereby mortally threatens us, then we must attempt to catch up with this excess.” We must begin an ongoing “experiment” consisting of “moral stretching exercises,” “hyperextensions of our capacity for fantasy and feeling.”18 Artistic practices are uniquely suited to performing the sort of work on ourselves that would be required in order not to “fall behind,” because art by its nature shows feeling not to be native to a naturalized psyche but constructively available to the worldly, historical person. And what should be the “technique of feeling” par excellence, a tool for pursuing such a repurposed practice of experimental testing, but musical hearing? This is a claim Anders elaborates in an extended appendix, “On the Plasticity of Feelings.” A Bruckner symphony, he proposes with no small measure of poetic irony, is an “apparatus made by ourselves, with whose aid we expand the capacity of our soul.”19
Anders, Eimert, and the Epitaph für Aikichi Kuboyama This proposition may well seem preposterous, and it is certainly a mark of Anders’s wild desperation. If the only hope in the face of self-annihilation is to cultivate intensified aesthetic feeling, then hope must be very slim indeed and his theoretical contribution a laughable self-deception—the more ironic given his steady drumbeat of warning against this very snare. Worse still, he again and again makes a pointed rhetorical strategy of exaggerating the extent to which humans have placed themselves in jeopardy, a strategy so effective that it all but defeats itself, leaving little room for solution. Yet exaggeration, Anders maintained, was necessary precisely because people had so completely forgotten how to feel a fear adequate to the present danger. This very line of thought indicates that the style of “feeling” Anders had in mind has almost nothing to do with inherited aesthetic categories such as consolation, nobility, sentimentality, sadness, joy, and the like.20 Invested as he is in the possibility of novelty, he can hardly name what he has in mind without the risk of getting caught in the status quo, falling farther and farther behind the world as it is continually made, unmade, and remade again. It is also clear that what is being called for is not moral betterment through culture, but rather anything that might increase the state of alarm, heighten alertness, and restore not just breadth but also precision and attentiveness
This Is Not a Test 335
to what he insisted on calling the soul. We have become “illiterates of fear” (Analphabeten der Angst, in one of his memorable formulations), implying that it would be necessary to learn an affective language of horror that could in fact be highly articulate in its own terms.21 Incongruously enough, however, the only music Anders names in this context is Bruckner’s: a lyrical late Romanticism that surely bespeaks a certain soulful expansiveness but that can also appeal to cheap sentimental heroization and self-pity (as in the use of the Seventh Symphony Adagio on Nazi radio). It is the temporal breadth of the Brucknerian idiom, the way it often does not seem to synchronize with the habitual pace of thought or accustomed intensities of shifting affect, that recommends it as an exemplary phenomenon to Anders.22 Yet it is difficult to perceive how one might recuperate anything of value from this repertory if one is after an opportunity for exercising the capacity for anxiety as a defensive device in the manner of a high alert. In light of these qualifications, it seems natural to suppose that we might identify more apt examples of what Anders was demanding. I would suggest that we can discover one in the immediate proximity of “The Obsolescence of the Human”: a composition that takes as its raw material a recorded passage of poetry that appears in a crucial footnote in the text itself. Epitaph für Aikichi Kuboyama (1962) is a tape work by Herbert Eimert, composer, critic, and founder of the electronic music studio at West German Radio in Cologne, who corresponded with Anders in the years immediately following the book’s publication.23 Although “The Obsolescence of the Human” was never translated into English and has remained relatively little known to the Anglo-American public, it was widely read in Europe, especially at this early, pre-détente stage of the Cold War, so it is not surprising that Eimert would have been captured by the book’s call for an experimental practice of perception that engaged far- reaching ethical goals. The twenty-minute Epitaph memorializes the Lucky Dragon’s radio operator, Aikichi Kuboyama, who died a few months after Castle Bravo of complications stemming either from radiation sickness itself or from attempts to treat it.24 Sometimes considered the first postwar casualty of nuclear violence, Kuboyama’s September 1954 death was an event of major international significance, with the fate of the Lucky Dragon loosely providing material for Gojira, released one month later (Americanized in 1956 as Godzilla, King of the Monsters!). In “The Obsolescence of the Human,” Kuboyama received an epitaph printed in the guise of an anonymous public memorial in fact written by Anders himself:25 You little fisherman, we don’t know whether you had merits.
336 World as Testbed (Where would we be if everyone had merits?) But you had worries like us, like us, somewhere the graves of your parents, somewhere, on the shore, a woman who waited for you, and at home, the children who ran to meet you. Despite your worries you found it good to be there. Just like us. And you were right, Aikichi Kuboyama You little fisherman, even if your foreign name does not tell of merit, let us learn it by heart for our brief term Aikichi Kuboyama. As a word for our disgrace Aikichi Kuboyama. As our warning call Aikichi Kuboyama. But also, Aikichi Kuboyama, as the name of our hope: For whether you preceded us in your dying or only departed in our stead— that depends only on us, even today, only on us, your brothers, Aikichi Kuboyama.26
Eimert began preliminary work on his tape piece a year after Anders’s book appeared in print, taking as its basis a recording of the Kuboyama pseudo-epitaph read by a German actor, Richard Münch. Much of this preparatory work, carried out in collaboration with studio co-director Robert Beyer and technician Leopold von Knobelsdorff, consisted of what might be described as acoustical etudes, in which the component sounds of the voice were transformed in a number of ways within the available catalog of tape manipulation techniques: they were slowed down, sped up, reversed, raised, lowered, filtered, and analyzed into constituent phonemes, which were in turn made objects of further manipulation. In short, the text was mined for potential materials with an analytical thoroughness typical of the practices of early electronic composition, especially at Cologne, with the distinction that no single sound appears in the work that did not in some way derive from the source recording.27 The usual way of assessing the significance of the Cologne studio has been to emphasize how it enabled the production
This Is Not a Test 337
and manipulation of sounds from scratch, building them up out of sine tones generated in situ and hence ostensibly under the more direct control of technicians and composers, even when the composition ultimately incorporated other recorded or live sounds. Yet whatever Eimert’s intervention within the immediate history of the studio, what is of interest here is not a production-side ethos of experimentalism in the sense of charting out new ways of organizing musical sound. Rather, the aim of the remaining discussion is to entertain the question of whether, and how, Eimert’s composition in fact might allow for the kind of listening that Anders held out as a compensatory technical act appropriate to inhabiting a long historical moment of perpetually imminent mass violence and fatality. In other words, what recommends this particular piece for consideration under the present heading of the test is not so much the brute circumstance of the laboratory- like environment of the early electronic music studio, but rather the intuition that its engagement with Anders prompts us to hear it somehow in the experimental style theorized by Brecht, Benjamin, and Anders. This is an ambitious goal, however, which can only be gestured toward rather than fulfilled. It will involve a gradual movement over the following pages from some initial concrete aspects of the sonic texture of the piece, through a reflection on the challenges of expression and communication it poses, to end with theoretical proposals concerning the larger action of “feeling” as a historical variable. It is far from clear that the perceptual experiences entertained will in fact show “what must be done” to allow for the sort of “transformation” Anders might have envisioned. A first order of business is to bring into focus the peculiar sonic medium within which an act of feeling might transpire at all. A kind of sonic and affective focalizing is the very gesture that initiates Eimert’s Epitaph itself, where the inclination to take up an attitude of openness toward suggested, but not entirely specified, categories of sensation and affect becomes essential. The opening two minutes (2:43–4:46) of the twenty-one-minute piece make a miniature drama of the transformation from the immediately intelligible linguistic presence of the speaking voice to something far less transparent. At the beginning of the piece, we hear a gently resonant metallic swell with a twenty- second decay that seems to function as a call to auditory focus, followed by extended patches of direct linguistic enunciation from the Anders text (“Du kleiner Fischermann, Aikichi Kuboyama . . . nur von uns, deinen Brüdern,” etc.), which then fades into various shades of obscurity and increasingly nonlinguistic bursts of articulation. Eimert’s own words about the piece highlight two basic poetic concerns: on the one hand, the vestigial, traditional effects of expression and reference, and on the other hand, a blurring between linguistic
338 World as Testbed
and nonlinguistic sound—that is, a mixture of orienting and disorienting effects: In Epitaph, the speech sounds are predominantly acoustic and phonetic in nature, and their transformation into purely musical material often makes it nearly impossible to determine their origin. But words, with their sense of meaning and their expressive weight and referential content, repeatedly rise to the surface and lend the entire weight of their meaning to single words or phrases. . . . In addition, there are idiosyncratic word or speech constructions made from jumbled syllables or parts of words, words spoken backwards, . . . motivic figures which function at a purely musical level but whose origins as spoken words can still be recognized, and sentences so unintelligible that we perceive them as unknown foreign languages.28
To take an example, the opening swell (2:43–3:03), largely unrecognizable as being of linguistic origin, is in fact composed of a threefold layering of isolated and extended vowels from the name “Kuboyama”: U, O, and A. In one regard, a worrying of the edge between word and sound was part and parcel of certain historical peculiarities surrounding the Cologne studio, which had been the site of experimental work on phonetics inspired by the Bonn acoustician and information theorist Werner Meyer-Eppler.29 But it seems appropriate to look beyond a technocentric narrative of the Cologne studio to take in a view of the ethical work being performed there on this occasion, as odd as Anders and Eimert may have been in wearing their moral aspirations so unmistakably on their sleeves. Eimert’s intention of creating “unknown foreign languages”— a widely shared preoccupation in much avant-garde music from the mid-1950s and later, not just at the Cologne studio but also in Milan, Paris, New York, and elsewhere—could be interpreted as an effort to capture how the sheer act of enunciating even an unintelligible concatenation of phonetic material already produces a kind of communicative presence that is different from any semantically and syntactically integrated utterance. In other words, it seems to want to show us “communication” allegorically writ large as a sort of formal dramatis personae, which ironically means departing from any actual achievement of everyday communication. If music like this seems invested in a kind of depersonalization by moving sharply away from a lyrical centering and continuity, it is far from necessary to hear it as thereby straightforwardly representing the annihilation of the person as a source or locus of value, as some classic interpretations of the postwar avant-garde once did.30 In a related context, in which he is comparing the compositional poetics of early electronic music with the compositional poetics of early modernity, Eimert puts this matter of humanness “showing
This Is Not a Test 339
up” in slightly different terms, describing how compositional material can be “brought to speak,” no matter how un-speech-like it is: And once again regarding the “human”: as far as those who talk so eagerly of “humanitas,” you can bet most of them mean nothing more than the violin cantilena on the G string saturated with vibrato à la Tchaikovsky. From the perspective of a heightened espressivo, the older, pre-expressive [late medieval and early Renaissance] music would [likewise] appear subjectively underdeveloped. And it has yet other traits in common with electronic music, most importantly that of a distinctly material character. The material here is brought to “speak” not because it is itself endowed with speech, but because it has been organized by a human being, to be sure with the help of theology, but quite certainly without the compulsively or sufferingly self-determined subject in the modern sense.31
Eimert moves to maintain the value of “speaking” while jettisoning that of “expressivity,” or at least expressivity taken in what he calls the “psychographic” sense of self-revelation. What has come to be taken for the musically “human” is to be seen as nothing other than a historical accumulation of imitations of particular kinds of ostensibly self-revelatory lyrical songfulness. For Eimert, this unperceived artifice becomes patent and hence ridiculous when the new electronic media are then made to continue the chain of imitations—by, say, synthetically recreating the effect of a bassoon melody, which itself has already been historically treated as a kind of sublimated vocal persona—rather than allowing the novel particularities available in the material to “speak” on their own terms, which is to say, to let them “bespeak” the personhood that has acted on and through them. And here is a point of contact with Anders’s thesis concerning the historicity of feeling. As with Eimert’s notion of bringing material to speech, “feeling” can be understood as what is being picked out as the salutary counterweight to expressive individuality. It is a value that can be personal, of the person in the world, without being confined to the interior of a solipsistic self. “Aikichi Kuboyama,” we recall, does not just name an individual but, in its manifold reiteration in the source text, is made to name certain other things, which could be described as specific moral feelings: shame, fear (as prompted by the “warning call” or Warnungsruf), and hope. Eimert does not overtly play on the Schande text or its associated speech sounds, and the structure of shame in this context is probably more complex than can been addressed here.32 Fear and hope, however, are unmistakably at play in the tape work’s poetics. It is almost embarrassing to say outright that something like a feeling
340 World as Testbed
of fear is palpable throughout the piece, though in any case, I would note for the moment that what is at issue is less the possibility of an evocation of fear than the more germane question of how fear is meant in relation to the possibility of technique.33 I will return to this. At the composition’s intensive apex (17:21–18:24), we initially hear a dense passage featuring a collage-like assemblage of overlapping textures and ongoing melodic ideas, including a surprising element of lyrical continuity wrought from the speech material in the musical-saw-like gestures in the upper register. An explosive rush of white noise wipes out the ongoing activity but is itself interrupted by silence. The semantic directness of this passage, as well as the sharp chiseling of its aural images, would seem to leave little need for commentary. A rhetoric of fearsome speed and jarring suddenness sonically frames the call to imagine “where we would be” (Wo kämen wir hin?, at 17:54) in the event of nuclear catastrophe with an explicit sonic suggestion of possible acts of violence. And the closing gesture of the passage executes an elegant speech act in which what is named, the “warning call” (Warnungsruf, at 18:18), does the naming. We have intuited all along that we are listening to some sort of warning, but Eimert indulges in licensed rhetorical excess to allow the ominous to name itself so as to thematize an affective response that might otherwise go unreflected. There would be a programmatic drabness to the notion that all Eimert accomplishes here is a musical paraphrase of Anders’s point that we should be very afraid: at once telling us to feel fear and making us do so. This would be to open up the loosely shared Eimert-Anders agenda to the easy charge of propagandistic manipulation of the sort that had disgusted both of them in the 1930s. And in any case, for all the importance Anders attached to fear in jolting one out of complacency, it could only ever be an initial step on the way toward something else. It is more challenging to articulate the nature of that something. To that end, we might turn to a different moral-affective quality attached to Kuboyama’s name: following “shame” and “fear” in the poem comes “hope.” We have already heard an acoustically masked reference to the poetic line in question, “als Namen unserer Hoffnung,” appearing in the opening minutes of the piece (4:05–4:08). But the text does not recur with any markedness until the very last moments (23:14–23:22), so that Eimert partly follows Anders in withholding the activation of this constellation of ideas until after discharging a range of other, complicating factors. If “hope” in a literal sense gets the last word here, it is nonetheless choked out, barely recognizable as an acoustic image for being registrally smeared from high to low and deprived of its article, “our” (unser), which has been altogether swallowed by distortion. How to recuperate a redemptive element from
This Is Not a Test 341
this is a puzzle, and doing so will require taking seriously Anders’s reflections on the necessity for the historical adaptation of feeling. Like Eimert, Anders rejected the idea that what music does is to translate an internally, instinctively felt emotion into a musical code, which can then be decoded so the same feeling can be felt, again internally, by a listener. Rather, it is the process of art-making itself that produces feelings from scratch, as if in a sort of humanist laboratory: The variant of subjectivistic aesthetics that sees in the artwork nothing other than the reflection or the “expression” of feeling already felt is completely foolish. . . . What the composer feels in composing their piece, they can only feel by means of this piece. That means: artworks produce feelings, and produce them sui generis; feelings which, without the produced objects, could scarcely be realized; which, independent of the structure of the objects, as mere being-in-a-mood, would remain nonexistent. Even the conditions in which an artwork puts us are artful; if you like, they are, in a word: “artworks.”34
This statement reworks one of the core insights of classical phenomenology, in which Anders had been intensively schooled (prior to his friendship with the decidedly nonphenomenological Brecht and Benjamin): namely that we do not have detached consciousness on the one hand and things on the other; rather, we only ever have consciousness of something. Similarly, we do not simply have detached feeling by itself, prior to some act of poiesis, of artistic making. Rather, the poetic act itself is charged with the production of feeling in one and the same gesture. “Feeling,” once again, is not to be understood as originating within a naturalized, self-contained, inner experience.35 That feeling would arise only through an act of making allows for what Anders demanded: the possibility of intervening in the historical structure of feelings, which was the opening for an element of hope. Hope would have to lie precisely in seeing aesthetic experience not as a matter of observing art, or even of participating in it—going along with it as a listener—but of coming into a position of being able to grasp the artificiality, or to parallel Künstlichkeit more directly, the “artfulness” of the situation or condition one is in when one is listening. This is saying more than that there is an “art of listening,” a sort of cultivated skill in the Enlightenment tradition. It is saying that beyond the psychology of feeling, in which one would be reflecting on the state of one’s subjective response to things, there is also a phenomenology of feeling, where feeling is understood to be neither subjective nor objective but part of a shared world of intentional meaning, a matter of worldly personhood.
342 World as Testbed
If we are to take seriously the demand to orient ourselves toward felt objects as an intentional act and task, and hence to see them as matters for ongoing work and critical transformation, then even the explicitly mimetic aspects of the Epitaph, the elements that seem designed to evoke fear and put us on notice, would need to be heard anew. If there are fearsome acoustic objects such as launched rockets, screaming sirens, rushing watery clouds, and so on, attention will need to be focused on moments of novelty in the way these objects show up within the fabric of the electronic composition, rather than isolating them as warmed-over sound effects that are meant to cue predetermined responses. To return to the Epitaph’s final sonic image of a choked-out, half- swallowed hope, the artful indeterminacy of what is being presented or produced there (not “expressed”) would need to be described as such. What is apt about Eimert’s way of ending the piece is that in the reworking of the final Hoffnung text, it manages to yield an opportunity for enacting the sort of hope that can only be imagined precisely because it appears amid a situation of hopelessness that, Anders suggests, must compel us to imagine things otherwise, and it is only in the exercise of practicing some imaginative perceptual expansion that we can even begin to hear the word.
Anruf/Warnungsruf In an evocative passage concerning his central thesis of the Gefälle, the fateful disproportion or gap between the technical and moral-imaginative capabilities of the modern human, Anders inverts the question of the audibility of feeling. He suggests that what we are effectively doing whenever we are in the position of experimenting with the unimagined and the unfelt is to enact a kind of prompting, a Vorsagen, in which we forespeak or foretell by shouting across the gap in an effort to make ourselves heard to ourselves. As the expression “foretelling” or “prompting” [Vorsagen] indicates, this is a kind of call [Anruf]; not, as in the “call of conscience,” one that is primarily heard, but rather one that we call to ourselves. For one calls out over the abyss of this disproportion [Gefälle-Kluft], as if the capacities that lag behind on the other side of the gap were personae. And it is they, imagination and feeling, which should hear, or which we want to “lend us an ear” [“Ohren machen”] in the first place.36
I want to conclude by drawing out two ideas from this admittedly somewhat improbable formulation. The first pertains to the image of feeling. If Eimert’s Epitaph is in some sense calling out over an “abyss,” then the suggestion to
This Is Not a Test 343
be entertained from Anders is that in listening to it, we, too, are calling along with it in an imaginatively participatory, though not necessarily empathetic, way. In good phenomenological style, the “call” goes out as a kind of intentional being-directed-toward (Gerichtetsein-auf) the world, and the composition forms the occasion for that act, rather than merely being there as a thing to be apprehended for itself. The fact that it is so difficult to name or describe the “new feelings” Anders wishes us to engage is a matter of their not simply being available in the “text,” as it were, but needing to be hailed from a distance, with its help. We are as foreign to them as they are to us. The second, final, and inevitably gloomier point pertains to the implicit temporality of this image. The idea of Vorsagen, of saying what will happen before it happens, is of course central to Anders’s philosophical anthropology, which is at all times a matter of anticipating an apocalypse, either through genocide or, more emphatically, through nuclear holocaust (to which one might now want to add the slower, more easily repressed history of climate change).37 But in a certain sense, Anders had good news as well as bad news. The bad news was that, although he did not consider it his business to forecast when and how this would happen, that we lived under apocalyptic conditions was a certainty. The kurze Frist or “brief term” that is broached in the Kuboyama epitaph was a consistent theme in Anders’s postwar work, which was premised on the knowingly exaggerated notion that we had come to live in a historical circumstance of immediate finitude. This was not so much a matter of empirical, factical death as it was a matter of the sheer idea of the bomb now having put us in a situation in which death was imminent, hanging over us at all times.38 But this was also the good news, precisely because the fact of nuclear death was not something that could be either forestalled or quickened. The apocalypse was not just being foretold in a straightforward temporal sense in which “pre-diction” says what the future will bring tomorrow or next year. Instead, it had already happened, was already upon us, simply by virtue of the historical introduction of these devastating new potentialities.39 Hiroshima ist überall, went one of Anders’s slogans. The “call” to ourselves, in which we were always in a position of catching up with ourselves, was endemic or structural, something to be lived with, and it is in this spirit that a work like Eimert’s may be heard to inhabit a disposition of both hope and hopelessness at once, such that they are completely intertwined. Similarly, what Anders seems to be asking of us is at once overly modest and nearly impossible. Reduced to the formulation that we are simply to answer the domination of “human engineering” with a countervailing program of moral exercises or self-tests on the terrain of aesthetic experience, it is impossible to avoid the impression of a quaint humanism unequal to its task even as it becomes aware of its own
344 World as Testbed
“obsolescence.” But then, after all, the action of testing is like this: it is ongoing, does not ask in advance for any particular outcome, acquires substance only in relation to its latest iteration, and hence does not entail a particular hope for its success, yet nonetheless wants to offer us some material possibility for continuing and revising.
Notes 1. Günther Anders, Philosophische Stenogramme, 70. Here and throughout, all translations are my own unless otherwise attributed. 2. On “human engineering,” an English term he consistently deployed untranslated, see Anders, Die Antiquiertheit des Menschen, 1:35–50; translated by Christopher John Müller as “On Promethean Shame,” in Müller, Prometheanism, 40–47. Anders’s critique finds a recent echo in Nikolas Rose, Governing the Soul, esp. 17–20, which sees the testing practices developed during and just after World War I as heralding the onset of an increasingly psychologized form of governmentality. 3. On “psychotechnics,” see William Stern, “Angewandte Psychologie,” 28–33; on “mental quotient,” see Stern, Psychological Methods, 36–42. Günther, along with his two sisters, is intimately discussed throughout Stern, Psychologie der frühen Kindheit. 4. Two key figures displacing Stern’s early formative role were Hugo Münsterberg and Fritz Giese. Giese’s Grundzüge der praktischen Psychologie, 1:3–10, offers a firsthand account of the history of the field. Viktoria Tkaczyk’s “Archival Traces of Sonic Fiction” discusses pertinent work by Giese. Also see Jeremy Blatter, “Psychotechnics of Everyday Life.” 5. Anders, Antiquiertheit des Menschen, 1:260. 6. On Marshallese responses to the legacy of nuclear colonialism, see Jessica A. Schwartz, “A ‘Voice to Sing’ ”; Schwartz, “Matters of Empathy and Nuclear Colonialism.” See also Aya Homei, “The Contentious Death of Mr. Kuboyama,” and the contemporaneous account, Ralph E. Lapp, Voyage of the Lucky Dragon. 7. Günther Anders, Ketzereien, 98. 8. A more recent perspective, contrasting with Anders’s thesis of the uncontainable nature of weapons testing, is Donald MacKenzie, “From Kwajalein to Armageddon?” 9. On the latter, see Margarete Vöhringer, Avantgarde und Psychotechnik. 10. Brigid Doherty, “Test and Gestus in Brecht and Benjamin.” 11. Walter Benjamin, “Bert Brecht,” 366. 12. Bertolt Brecht, “Literarization of the Theatre,” 46; and, more generally, “On Experimental Theatre,” 133–45. 13. Walter Benjamin, “Author as Producer,” 779. 14. Benjamin’s sharpest statement of a retooling of psychotechnical practice appears in a 1930 radio talk in which he identifies the ascendancy of behaviorism as a counterpoint to “the psychology of the individual, which attempts to understand the behavior of the individual essentially through his nature. To the contrary, nature is important to behaviorism only in its malleability. Behaviorism is interested in the profoundly transformative, profoundly invasive effects of the work process on character.” Walter Benjamin, “Carousel of Jobs,” 285. 15. Günther Anders, Bert Brecht, 13–16.
This Is Not a Test 345 16. Anders, Antiquiertheit des Menschen, 1:310. A more developed statement of these ideas appears in Anders, “The Pathology of Freedom,” an essay based on a text that was originally drafted in 1929. 17. Irmingard Staeuble, “ ‘Psychological Man’ and Human Subjectivity,” helpfully contextualizes Anders’s contribution on these points within midcentury social theory. 18. Anders, Antiquiertheit des Menschen, 1:273–74. 19. Ibid., 1:313. 20. In this regard, Anders shares something with the exploration of negative affect by Sianne Ngai, Ugly Feelings, with the difference that he is less concerned with interpreting feeling as a sign of something else (say, social problems) than with wondering how it might change or be changed. 21. Anders, Antiquiertheit des Menschen, 1:265. 22. See the extended discussion in Antiquiertheit des Menschen, 1:313–16, whose poignant inadequacy is all the keener since it revives without significant adaptation ideas first elaborated in his 1930–31 habilitation thesis, “Philosophische Untersuchungen über musikalische Situationen.” For a useful treatment of Anders’s early philosophy of music, see Reinhard Ellensohn, “Von der Musikphänomenologie zur Technikkritik.” 23. The first live presentation took place at the Darmstadt Ferienkurse für neue Musik on July 9, 1962, and the first radio broadcast was on September 4, 1962, via West German Radio Cologne. For further details of Anders’s attitude toward Eimert in comparison with other contemporary composers, see Jason Dawsey, “Limits of the Human,” 433–35; see also Franz Haas, “ ‘Sul ponte di Hiroshima.’ ” 24. No consensus about the cause of death was reached, and the question was intensely politicized. See Homei, “The Contentious Death of Mr. Kuboyama.” 25. Anders and Eimert corresponded about Kuboyama in 1957 and 1958, and it was in one of these letters that Anders revealed himself to be the author of the poem, leaving it up to Eimert to decide whether or not to disclose the authorship or maintain its apparent anonymity; Eimert decided in favor of the latter. It is also worth noting that Anders admitted to Eimert in 1958 that he had initially mischaracterized Kuboyama as a “simple fisherman,” as opposed to the skilled technician Anders later learned him to be. The published attribute is “little,” whose undeniable racializing connotations call out for a more developed interpretation in terms of the text’s broader rhetorical and ethical strategies. See Michael Custodis, Die soziale Isolation der neuen Musik, 131–32. 26. Anders, Antiquiertheit des Menschen, 1:346–47. My translation. A recitation of this text appears on Herbert Eimert, Epitaph für Aikichi Kuboyama, Track 1, at 1:03–2:41, followed immediately in the same track by Eimert’s composition at 2:43. Subsequent track timings refer to this source. 27. This contrasts with other procedures in the Cologne studio, where compositions had typically contained a preponderance of synthetically produced sound as opposed to manipulating preexistent recordings. Eimert’s correspondence with Anders suggests that he had initially considered incorporating recorded sounds of a harmonica. Custodis, Die soziale Isolation der neuen Musik, 131–32. 28. Eimert, Epitaph für Aikichi Kuboyama, liner notes, 26–28. 29. See Oliver Kautny, “Pionierzeit der elektronischen Musik”; Elena Ungeheuer, “Producing, Representing, Constructing”; Jennifer Iverson, “Statistical Form Amongst the Darmstadt School.”
346 World as Testbed 30. Examples of such interpretations are Hans Sedlmayr, Verlust der Mitte; Hugo Friedrich, Die Struktur der modernen Lyrik. For a recent corrective, see Marcelle Pierson, “The Voice Under Erasure.” 31. Herbert Eimert, “Die sieben Stücke,” 10. 32. Shame is a major theme in Anders’s thinking in the 1956 book, albeit as Scham, rather than the Schande of the Kuboyama epitaph, with its connotation of disgrace. See in particular the translated excerpt and related discussion in Müller, Prometheanism, 29–95, which elaborates the thesis that a historically specific feeling of shame arises in the sense of being inferior to one’s own technological products, or of “falling behind” one’s own creations. 33. For one early critic, Eimert’s piece was affectively unambiguous in conveying “an immediately moving music of mourning, rich in images.” “Herbert Eimert,” record review in Der Spiegel 25 (June 13, 1966). Yet the question of “immediacy” was demonstrably less straightforward than the Spiegel reviewer imagined. In a fascinating pair of studies first published in 1966 and 1968 by the music sociologist Vladimír Karbusický, excerpts from Eimert’s Epitaph were played for hundreds of Czech test subjects, who, asked to articulate the associations or indeed Vorstellungen the music called to mind, generated an enormous disparity of responses, ranging from fairy tales, to space travel, to nuclear attack (and this without the benefit of any of the intelligible textual prompts present at the beginning of the composition), to the explicitly hopeful vision of a future society of greater humanity and reason. Karbusický, Empirische Musiksoziologie, 101–2. 34. Anders, Antiquiertheit des Menschen, 1:315. 35. For a recent exploration of this point, see Sara Ahmed, “Happy Objects.” 36. Anders, Antiquiertheit des Menschen, 1:275. This passage stands in complex relation to Heidegger’s analysis of the “call of conscience,” or Gewissensruf, in Being and Time, 336–40, where “good conscience” points forward toward and warns against a guilty deed, while “bad conscience” points back reprovingly to a deed already done. In the case of good conscience, the “voice” summons as it were from a moment of anticipation, as “Dasein ‘is’ ahead of itself in such a way that at the same time it directs itself back to its thrownness” (Being and Time, 337). 37. Babette Babich, “Angels, the Space of Time, and Apocalyptic Blindness,” offers germane reflections on temporality in Anders’s thinking. 38. Or as Jean-Pierre Dupuy formulates it in his reading of Anders, if we stand too close to the fire we get burned, but if we move too far away we forget its existence just as much at our peril. Dupuy, Mark of the Sacred, 175–94. 39. This is a point noted by Michael Hardt and Antonio Negri (Multitude, 18–19), who maintain that implications Anders saw in the general state of ongoing imminent war in 1956 have persisted into the present and constitute a form of “biopower in this most negative and horrible sense of the term, a power that rules directly over death. . . . When genocide and atomic weapons put life itself on center stage, then war becomes properly ontological.”
References Ahmed, Sara. “Happy Objects.” In The Affect Theory Reader, edited by Melissa Gregg and Gregory J. Seigworth, 29–51. Durham, NC: Duke University Press, 2010. Anders, Günther. Die Antiquiertheit des Menschen, vol. 1: Über die Seele im Zeitalter der zweiten industriellen Revolution. Munich: C. H. Beck, 1956.
This Is Not a Test 347 Anders, Günther. Bert Brecht: Gespräche und Erinnerungen. Zurich: Verlag der Arche, 1962. Anders, Günther. Ketzereien. Munich: C. H. Beck, 1982. Anders, Günther. “The Pathology of Freedom: An Essay on Non-Identification.” Translated by Katharine Wolfe. In The Life and Work of Günther Anders: Émigré, Iconoclast, Philosopher, Man of Letters, edited by Günter Bischof, Jason Dawsey, and Bernhard Fetz, 145–70. Innsbruck: Studienverlag, 2014. Anders, Günther. Philosophische Stenogramme. Munich: C. H. Beck, 1965. Anders, Günther. “Philosophische Untersuchungen über musikalische Situationen.” In Anders, Musikphilosophische Schriften: Texte und Dokumente, edited by Reinhard Ellensohn, 15– 140. Munich C. H. Beck, 2017. Babich, Babette. “Angels, the Space of Time, and Apocalyptic Blindness: On Günther Anders’ Endzeit–Endtime.” Ethics & Politics 15 (2013): 144–74. Benjamin, Walter. “The Author as Producer: Address at the Institute for the Study of Fascism, April 27, 1934.” Translated by Edmund Jephcott. In Selected Writings 2, Part 2, edited by Michael W. Jennings, Howard Eiland, and Gary Smith, 768–82. Cambridge, MA: Belknap Press, 1999. Benjamin, Walter. “Bert Brecht.” Translated by Rodney Livingstone. In Selected Writings 2, Part 1, edited by Michael W. Jennings, Howard Eiland, and Gary Smith, 365–71. Cambridge, MA: Belknap Press, 1999. Benjamin, Walter. “Carousel of Jobs.” Translated by Jonathan Lutes. In Radio Benjamin, edited by Lecia Rosenthal, 283–91. London: Verso, 2014. Blatter, Jeremy. “The Psychotechnics of Everyday Life: Hugo Münsterberg and the Politics of Applied Psychology, 1887–1917.” PhD diss., Harvard University, 2014. Brecht, Bertolt. “The Literarization of the Theatre.” In Brecht on Theatre: The Development of an Aesthetic, edited and translated by John Willett, 43–47. New York: Hill and Wang, 1964. Brecht, Bertolt. “On Experimental Theatre.” In Brecht on Theatre: The Development of an Aesthetic, edited and translated by John Willett, 133–45. New York: Hill and Wang, 1964. Custodis, Michael. Die soziale Isolation der neuen Musik: Zum Kölner Musikleben nach 1945. Stuttgart: Franz Steiner, 2004. Dawsey, Jason. “The Limits of the Human in the Age of the Technological Revolution: Günther Anders, Post-Marxism, and the Emergence of Technology Critique.” PhD diss., University of Chicago, 2013. Doherty, Brigid. “Test and Gestus in Brecht and Benjamin.” MLN 115, no. 3 (2000): 442–81. Dupuy, Jean-Pierre. The Mark of the Sacred. Translated by M. B. DeBevoise. Stanford, CA: Stanford University Press, 2013. Eimert, Herbert. Epitaph für Aikichi Kuboyama. Wergo 6773 6. CD. 2012 (remastered version of WERGO LP, WER 600014, 1966). Eimert, Herbert. “Die sieben Stücke.” Die Reihe 1 (1955): 8–13. Ellensohn, Reinhard. “Von der Musikphänomenologie zur Technikkritik: Zur frühen Musikphilosophie von Günther Anders.” Musik und Ästhetik 19, no. 75 (2015): 5–20. Friedrich, Hugo. Die Struktur der modernen Lyrik von Baudelaire bis zur Gegenwart. Hamburg: Rowohlt, 1962. Giese, Fritz. Grundzüge der praktischen Psychologie, vol. 1: Theorie der Psychotechnik. Wiesbaden: Vieweg & Teubner, 1925. Haas, Franz. “‘Sul ponte di Hiroshima’: Günther Anders und die Ästhetik in italienischer Sicht.” In Günther Anders kontrovers, edited by Konrad Paul Liessmann, 103–13. Munich: Beck, 1992. Hardt, Michael, and Antonio Negri. Multitude: War and Democracy in the Age of Empire. New York: Penguin Books, 2004. Heidegger, Martin. Being and Time. Translated by John Macquarrie and Edward Robinson. New York: Harper and Row, 1962.
348 World as Testbed “Herbert Eimert. Epitaph für Aikichi Kuboyama. Sechs Studien (1962).” Der Spiegel 25, (June 13, 1966): 137, sec. Schallplattenspiegel. Homei, Aya. “The Contentious Death of Mr. Kuboyama: Science as Politics in the 1954 Lucky Dragon Incident.” Japan Forum 25, no. 2 (2013): 212–32. Iverson, Jennifer. “Statistical Form Amongst the Darmstadt School.” Music Analysis 33, no. 3 (2014): 341–87. Karbusický, Vladimír. Empirische Musiksoziologie: Erscheinungsformen, Theorie und Philosophie des Bezugs “Musik–Gesellschaft.” Wiesbaden: Breitkopf & Härtel, 1975. Kautny, Oliver. “Pionierzeit der elektronischen Musik: Werner Meyer-Epplers Einfluß auf Herbert Eimert.” In Musik im Spektrum von Kultur und Gesellschaft, edited by Bernhard Müssgens, Oliver Kautny, and Martin Gieseking, 315–37. Osnabrück: epOs-Verlag, 2001. Lapp, Ralph E. The Voyage of the Lucky Dragon. New York: Harper & Brothers, 1958. MacKenzie, Donald. “From Kwajalein to Armageddon? Testing and the Social Construction of Missile Accuracy.” In The Uses of Experiment: Studies in the Natural Sciences, edited by David Gooding, Trevor Pinch, and Simon Schaffer, 409–36. Cambridge: Cambridge University Press, 1989. Müller, Christopher John. Prometheanism: Technology, Digital Culture and Human Obsolescence. London: Rowman & Littlefield, 2016. Ngai, Sianne. Ugly Feelings. Cambridge, MA: Harvard University Press, 2005. Pierson, Marcelle. “The Voice Under Erasure: Singing, Melody and Expression in Late Modernist Music.” PhD diss., University of Chicago, 2015. Rose, Nikolas. Governing the Soul: The Shaping of the Private Self. New York: Routledge, 1990. Schwartz, Jessica A. “Matters of Empathy and Nuclear Colonialism: Marshallese Voices Marked in Story, Song, and Illustration.” Music & Politics 10, no. 2 (2016). http://dx.doi.org/10.3998/ mp.9460447.0010.206 Schwartz, Jessica A. “A ‘Voice to Sing’: Rongelapese Musical Activism and the Production of Nuclear Knowledge.” Music & Politics 6, no. 1 (2012). http:// dx.doi.org/ 10.3998/ mp.9460447.0006.101 Sedlmayr, Hans. Verlust der Mitte: Die bildende Kunst des 19. und 20. Jahrhunderts als Symptom und Symbol der Zeit. Salzburg: O. Müller, 1948. Staeuble, Irmingard. “‘Psychological Man’ and Human Subjectivity in Historical Perspective.” History of the Human Sciences 4, no. 3 (1991): 417–32. Stern, William. “Angewandte Psychologie.” Beiträge zur Psychologie der Aussage 1 (1903–1904): 4–54. Stern, William. Psychologie der frühen Kindheit bis zum sechsten Lebensjahre. Leipzig: Quelle & Meyer, 1914. Stern, William. The Psychological Methods of Testing Intelligence. Translated by Guy Montrose Whipple. Baltimore, MD: Warwick & York, 1914. Tkaczyk, Viktoria. “Archival Traces of Applied Research: Psychotechnics and Language Planning in Interwar Germany.” In “Listening to the Archive: Sound Data in the Humanities and the Sciences,” edited by Carolyn Birdsall and Viktoria Tkaczyk. Special issue of Technology and Culture 60, no. 2, Supplement (2019): 64–95. Ungeheuer, Elena. “Producing, Representing, Constructing: Towards a Media- Aesthetic Theory of Action Related to Categories of Experimental Methods.” In Sounds of Science— Sound im Labor (1800–1930), edited by Julia Kursell, 99–112. Berlin: Max Planck Institute for the History of Science, Preprint 346, 2008. Vöhringer, Margarete. Avantgarde und Psychotechnik: Wissenschaft, Kunst und Technik der Wahrnehmungsexperimente in der frühen Sowjetunion. Göttingen: Wallstein, 2007.
On Testing An Afterword Hans-Jörg Rheinberger
In this brief concluding exposition, I would like to sort out a number of different meanings associated with forms of testing in the sciences, as well as their relation to forms of experimentation. The topic of hearing will remain in the background. I will try to cover the spectrum that testing as a “way of knowing” has assumed in the sciences and their applications as broadly as possible,1 illuminating the different forms primarily with examples taken from the scope of this volume. But in passing, I will also hint at a number of ways of testing that are not specifically addressed in this collection of essays on testing hearing. In doing so, I hope to offer an additional view into the present volume with its deliberate focus on the history of hearing. I will begin with a few remarks on the relationship between testing and experimenting. In the language of everyday life, the two notions are often used more or less interchangeably. The chapters in this book present ample historical material, mainly from the nineteenth and twentieth centuries, on the basis of which a broad range of scientific practices related to “trying out” is laid out before our eyes and invites differentiation. We can approach this spectrum from two directions. If we access it from the perspective of an emphatic concept of experimentation, we can identify one extreme of the spectrum with what historians of science have come to call “exploratory” experimentation.2 It is a form of experimentation that sets out to chart novel territory—to bring particular phenomena into being, so to speak, in a form as yet unobserved. On the other hand, we have what we may call “demonstrative” experimentation. Taken to its extreme, this is a form of experimentation where the outcome is known in advance. Its purpose is no longer to chart new ground, but to convince others of the stability of a particular phenomenon. To highlight the distinction, we could call the first a research experiment, the second a test: in a test, if all conditions are properly arranged, we can predict what will happen when we put it into action. In my Hans-Jörg Rheinberger, On Testing In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0014.
352 Afterwords
study on the history of protein synthesis research, I described the setting-up of an experimental system that was exploratory over a period of fifteen years, then turned into an instrument of corroboration at the point where another, comparable system took over its former exploratory function.3 Clearly, this identification of test with demonstration does not exhaust the different meanings that testing can assume, a point to which the contributions in this volume bear ample witness. So let us approach the spectrum from a more encompassing concept of test. If we do so, we realize that this, too, varies between two extremes. At the one extreme, we find the case, just mentioned, of putting a device or a phenomenon to the test—if everything is correctly arranged, we know what will happen. But down the scale in the other direction, the forms of testing or probing have a much broader meaning. At the other pole is something that we could call, in parallel to exploratory experimentation, “exploratory testing.” There is a point where experimenting and testing meet on the line between the two poles. Thus, if we wish to maintain a sharp distinction between the concepts of “experiment” and “test,” we are confronted with almost insurmountable difficulties: varieties of test-like experiments abound in scientific practice, as do varieties of experiment-like tests. Can we maintain a difference at all? Perhaps we could at least distinguish between two attitudes in scientifically approaching the world around us: one oriented toward containing and thus closure, the other oriented toward roaming and thus aperture. We might call the first a technical, the second an epistemic attitude. In recent historiography of science, these two attitudes can be retraced. One is oriented toward finding out, in the words of Peter Galison’s famous book title, “how experiments end.”4 The other is more interested in finding out how science charts new ground. Looking back to the first part of the twentieth century, the former attitude is epitomized by the work of Ludwik Fleck, the latter by that of Gaston Bachelard.5 My own work on experimental systems sees itself in the second tradition. Let me now, in a first step, briefly scroll through the present collection of papers and highlight a few of these variants of testing with their peculiar mixtures of attempts to close and to open. This might help us find clues for the beginning of a typology. Viktoria Tkaczyk looks at Otto Abraham’s early twentieth-century studies on the absolute pitch discrimination of musical hearers. Abraham’s tests took the form of a questionnaire, which he hoped would make it possible to test individual, syncretic abilities, in contrast to the ability of an average person. Questionnaires are a form of test that allows comparisons on the basis of large sets of data.
On Testing 353
The history of speech audiometry, in particular the phonographic practices it implies, gives Mara Mills the opportunity to highlight a different aspect of testing that became increasingly important in industrial contexts over the twentieth century, namely the quality testing of equipment, products, and parts thereof. A new term arose for this kind of practical activity, in many cases integrated into the production process of commodities: screening, or, to emphasize its most important aspect, “mass” screening. Actually, screening is a selection process. Items fall below or above a certain threshold value and are either kept or discarded accordingly. Screening can again come in many guises and involve either human beings or instruments as judges. Emily Dolan examines, throughout the twentieth century, a kind of test in which the ear of the expert and the instrument reciprocally test each other’s capacity for discrimination. The expert’s ear is required to tell the sound of an old Italian violin from that of a newly crafted violin. What is interesting here is the role of the norm. The norm is supposed to be the old craft that new technology strives to emulate. But the tests and their outcomes show that technology is not just technology here: it forms an inseparable amalgam with aesthetics, acoustics, and aural conventions. Jonathan Sterne, in his history of the attempts to construct digital models of analog signal processors, describes a similar kind of test. The software, as he paradoxically puts it, “passes the test when the user fails it.” Testing assumes the meaning of deciding whether a standard has been matched or not. In a similar vein, Sebastian Klotz thematizes the limits of a reductive standardization of tests. An important concomitant aspect of testing here moves center stage: the role that context plays in testing, that is, the environment in which a test is carried out, in this case the laboratory and the field. Standardization—that is, gauging—appears to be one of the cruxes of testing. Roland Wittje, in his study on acoustic testing in Norway in the interwar period, also follows the issue of standardization and its essential dilemma: gaining comparability and with that, objectivity, but at the same time, and as a result, losing specificity. In Alexander Rehding’s piece on Friedrich Wilhelm Opelt’s mid- nineteenth-century siren, the notion of probing is central. Here, it is a “music- theoretical instrument,” as Rehding calls it, that probes the limits of human hearing. Again, we are confronted with a clash between an aesthetic tradition and a physical Aufschreibesystem that challenges its basic distinctions. Jennifer Hsieh deals with environmental noise as a boundary object and the parties involved in its negotiation. What it means to test in this context is to control noise and regulating it accordingly. This amounts basically to measuring noise according to a standardized procedure and setting limits to be reached according to the negotiation of the needs of the different stakeholders.
354 Afterwords
Through the case of the construction and uses of an artificial head (Kunstkopf) to test spatial hearing, Stefan Krebs shows how a particular object can oscillate between being an epistemic object and being a test device. The artificial head was invented to free the procedure of testing the spatial hearing qualities of musical environments from the testing ear of the expert. But at the same time, it turned out to be a new research object that enabled more to be discovered about spatial hearing and the parameters involved in it. We can introduce the notion of sounding out here, as a process of mutual shaping. Such processes are to be observed in the development of research technologies more generally: the epistemic object to be sounded out is shaped in close interaction with the fine-tuning of the technical instrument being used in the sounding procedure. Joeri Bruyninckx introduces the topic of testing animals. Electroacoustic audiometry, as Wittje also stresses, allowed the human ear to be replaced for test purposes. One of the options this opened up was the use of electro acoustic instruments in place of the human ear. But there was also the option of testing animals’ hearing with the equipment and then using animals for otological purposes. These tests, in turn, revealed the superior hearing capacities of many animals. The example shows once again the interdependency of research, standardization, and the transformation of knowledge into technical tools. That point is reinforced in the study by Lino Camprubí and Alexandra Hui on a test instrument, the hydrophone, and its capacity to discriminate between the sonic realm of human-made underwater noise and ecologically relevant animal sounds. Under wartime and Cold War conditions, research conducted on the former co-produced knowledge on the latter. Finally, Benjamin Steege points beyond testing in two ways. On the one hand, he describes Günther Anders’s reaction to the atomic bomb tests at Bikini Atoll in 1954, where a “laboratory test” on an island acquired a global dimension. On the other, he leads us back to, not testing, but an experiment on hearing in its material entrenchment beyond anything to be captured with and in a test: Herbert Eimert’s musical Epitaph für Aikichi Kuboyama. This survey gives us a number of characteristic forms and variants of testing, each with its particular mix of closure and aperture. Among them are standardizing, testing against a standard, probing, questioning, screening, sounding out, and working by proxy in various forms—to recall just the most conspicuous of the variants encountered in this volume. They cover the whole spectrum from exploratory testing to the mere application of a routine procedure. Standardizing lies at the core of most testing procedures. Testing means comparing, and comparing involves a tertium comparationis. Arriving at such
On Testing 355
a measure, however, is itself a matter of negotiation and triangulation and is thus in flux, up for revision. The possibility of testing against a standard is the result of such a process. It involves an item that embodies a particular standard in a stabilized form, making it convenient to compare particular cases with the standard and measure the degree of coincidence or deviation. Probing refers to a different testing attitude. Here, insecurity is an integral part of the game, and standards are of little help in deciding which actions should be taken in the process. The enjeu is deciding on the aptitude of a particular object for a particular task. The questionnaire, invented for statistical survey purposes in the nineteenth century, is a testing instrument that usually works to accomplish data aggregation over larger populations. In the context investigated here, it receives a particular twist: it serves to determine individual constellations rather than building averages. Instead of lumping, it helps to split. Screening refers to a testing procedure in which a particular feature of an object is searched for among a multitude of objects, such as mass screening for symptoms in medicine or traces of contamination or damage in industrial products. Sounding out is about the scope and range of phenomena. It means testing how far something reaches or how widespread it is. And then we have the wide field of working by proxy, a common form of which is extremely widespread in pharmacology and the medical realm: the use of test animals to check the effects of particular substances or treatments. All these different forms of testing come with their own, idiosyncratic auras. The boundaries between them are not neat and clear; nevertheless, we can distinguish them as epistemic practices with their own kernels of rationality and practical requirements. But all of them require, as we might generally put it, control in one form or another. The last piece in the volume, however, reminds us that tests can also run out of control, with disastrous consequences. As well as the forms of testing just enumerated, all addressed in the contributions to this volume, there are quite a number of further forms of testing, some of which I would like to briefly mention in closing. One of them is troubleshooting. Here, a defect—generally of a technical device—is the starting point, and the testing activity follows a protocol to find out where that defect might be located. In the biological research literature, but also in other areas of research, we encounter a form of versatile experiment that can oscillate between the functions of research activity proper and testing activity, called an assay. Assaying is thus located halfway between an exploratory experiment and a standardized form of testing. In its more standardized and commercially available form, it is often also called a kit. The kit contains the ingredients of an experiment/test in standardized quality and amounts, but they still have to be mixed and put to work under particular circumstances.
356 Afterwords
Diagnostic kits abound in the health sector. Finally, at the intersection between chemistry and biology we find an interesting form of testing known as biological testing. Here, parts of animals are used to test the enrichment and efficiency of certain biological substances—vitamins, hormones, enzymes— in trace amounts that otherwise would be hard to measure or assess. The biological parts thus function as indicators. This last example also shows us that testing comes with a historical index. Biological testing of this kind reached its climax in the first half of the twentieth century. Ways of testing come into use, but they can also disappear again and be replaced by other forms in keeping with the development of the technologies upon which they are based. We could continue to explore the permutations of testing to build a—very preliminary—typology of testing. The present volume, with its focus on testing hearing, offers a fascinating and rich array of forms of testing, but it by no means exhausts these forms. As varieties of scientific practice, they cannot be detached from the materials upon which they act and through which they are realized, and will thus necessarily take different forms in different areas of science.
Notes 1. See John V. Pickstone, Ways of Knowing. 2. Richard M. Burian, “Exploratory Experimentation”; C. Kenneth Waters, “Nature and Context of Exploratory Experimentation”; Friedrich Steinle, Exploratory Experiments. 3. Hans-Jörg Rheinberger, Toward a History of Epistemic Things. 4. Peter Galison, How Experiments End. 5. Ludwik Fleck, Genesis and Development of a Scientific Fact; Gaston Bachelard, New Scientific Spirit.
References Bachelard, Gaston. The New Scientific Spirit. Translated by Arthur Goldhammer. Boston: Beacon Press, [1934] 1984. Burian, Richard M. “Exploratory Experimentation and the Role of Histochemical Techniques in the Work of Jean Brachet, 1938–1952.” History and Philosophy of the Life Sciences 19, no. 1 (1997): 27–45. Fleck, Ludwik. Genesis and Development of a Scientific Fact. Translated by Frederick Bradley and Thaddeus J. Trenn. Chicago: University of Chicago Press, [1935] 1979. Galison, Peter. How Experiments End. Chicago: University of Chicago Press, 1987. Pickstone, John V. Ways of Knowing: A New History of Science, Technology, and Medicine. Chicago: University of Chicago Press, 2001.
On Testing 357 Rheinberger, Hans-Jörg. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Stanford, CA: Stanford University Press, 1997. Steinle, Friedrich. Exploratory Experiments: Ampère, Faraday, and the Origins of Electrodynamics. Translated by Alex Levine. Pittsburgh, PA: Pittsburgh University Press, 2018. Waters, C. Kenneth. “The Nature and Context of Exploratory Experimentation: An Introduction to Three Case Studies of Exploratory Research.” History and Philosophy of the Life Sciences 29, no. 3 (2007): 275–84.
Coda Testing and Why It Matters Trevor Pinch
It is late fall in upstate New York. My daughter, a student at Cornell University, has just told me about the weird sounds she has heard emanating from the Cornell McGraw Clock Tower, which houses the Cornell Chimes. The bells usually play the Cornell alma mater: “Far Above Cayuga’s Waters.” They are used sometimes to play popular tunes. It is the most distinctive soundmark on the Cornell campus. “Oh, that will be Annie and Sarah rehearsing for their concert next week,” I say, trying to sound knowledgeable. My daughter replies, “Cool. I thought I saw Annie coming out of the tower.” Annie Lewandowski, a Cornell experimental musician, used to be her piano teacher; Sarah Hennies is a percussionist. One week later I am at the base of the tower with a sizable crowd including my naturalist friend Katy Payne. Katy is famous for first identifying the sounds of elephants and the song of the hunchback whale (from a recording made by a navy engineer). Her record of hunchback whale songs, released with her husband, Roger Payne, has played a role in the environmental movement and the conservation of whales almost akin to that of Rachel Carson’s Silent Spring. It is a beautiful evening and the wind is gently rustling the ivy creeper on the side of the tower. Another friend, Bill McQuay, who used to be NPR’s top sound engineer and the supervisory sound engineer at the Natural Sound Library at the Cornell Lab for Ornithology, tells us where to stand to get good sound. He is reputed to have the best ears in the business. Annie Lewandowski and percussionist Sarah Hennies are at the top of the tower and about to perform their duet, “Cetus: Life After Life,” Annie’s composition for whale song and chimes. Annie has spent the last few months working with Katy, poring over sonograms and recordings of whale songs. Sarah starts to play the chimes. We hear a new sound: it is Katy Payne’s recording of the humpback whale rendered through giant loudspeakers affixed to the top of the tower. We listen Trevor Pinch, Coda In: Testing Hearing. Edited by: Viktoria Tkaczyk, Mara Mills, and Alexandra Hui, Oxford University Press (2020). © Oxford University Press. DOI: 10.1093/oso/9780197511121.003.0015.
360 Coda
transfixed. The irony of hearing the magnificent song of our fellow mammals who live deep below the waters, interspersed with the clamorous sound of the chimes used to play “Far Above Cayuga’s Waters,” is lost on no one. Actually it is lost on one passerby who, perplexed by the strange sounds, is heard to remark, “Sounds like a cat is trapped up there!” Dissonance. How is this story related to this book? Read on. * * * In the late 1970s and early 1980s, there was much interest in the detail of the practices through which scientific facts were established. Inspired by the work of Thomas Kuhn and by moves toward a post-empiricist philosophy of science, studies of laboratories and scientific controversies showed the contingent basis for what became known as the “social construction of scientific knowledge.”1 Scholars zeroed in on the crucial role played by experiments in establishing scientific facts, and the importance of experimentation as a system emerged.2 It was clear that facts in the technical world are not established in isolation, and an important strand of literature emerged on the theme of metrology and standardization.3 Standardization raised the question of how commensuration is practically achieved in science and technology. Commensuration is also at the basis of testing. A test is a localized way of achieving commensuration between different measurements, entities, or contexts. It provides a framework whereby things can be compared to the same standard. By the late 1980s, science and technology studies (STS) scholars were arguing that testing for engineers was in many ways equivalent to experimentation for scientists.4 Testing is the means by which technological facts are established. If experiments could be shown to involve social and cultural processes in the making of scientific facts, then could not testing be subject to similar processes? It has been pointed out by historians of technology that the testing of particular technologies entails machines, traditions, and practices local to specific engineering cultures. For example, the Prony Brake, used to determine the power of water turbines (and later automobile engines), was a technology peculiar to the culture of testing turbines.5 The traditions of knowledge, skills, and practices that form part of testing are located within particular laboratories and engineering disciplines and are learned and practiced in situ. Specific test rigs and pieces of laboratory equipment are often involved. Tacit knowledge and craft skills are built around testing instruments and within particular laboratories. Thus, at the heart of testing are communally acquired and passed-on traditions of knowledge and practices.
Testing and Why It Matters 361
Such “traditions of testing,” as Viktoria Tkaczyk refers to them (Chapter 2), can be subject to wider social, cultural, and economic processes and legal, ethical, and regulatory regimes, as new technologies and materials emerge, different sets of interests intersect, and new spaces for testing technologies arise.6 The establishment of such a new space for audio testing in Norway is the subject of Roland Wittje’s study (Chapter 9). The sociology of testing has shown that some of the “hardest facts” about how technologies work, such as the accuracy of guidance systems for nuclear missiles, are replete with social interests.7 For years, the U.S. manned-bomber lobby refused to accept facts about the accuracy of nuclear weapons established by inertial guidance testing on a target in the Marshall Islands because, for the air force, dropping nuclear weapons by manned bomber was the only surefire way to obtain accuracy. This terrifying argument rested on the assumption that the Marshall Islands were not similar enough to the target of Moscow in a real nuclear war. Benjamin Steege returns to the sites of these nuclear explosions in the Marshall Islands (Chapter 12), noting that Günther Anders, in his radical critique of testing, refused to designate what occurred there as tests or experiments at all. For test results to be applied to different contexts, as in the actual performance of an aircraft compared to tests in a wind tunnel, similarity assumptions have to be projected between these different contexts. Similarity between contexts has been shown to be at the core of what counts as an experimental replication in science.8 Failure to replicate in science can always be put down to important differences between the original and the replicating experiment. Tests, like experiments, are performances. Who gets to witness such performances and under what circumstances is a key part of the social, cultural, literary, and technical apparatus of science and technology.9 Tests of new ships, such as the turbine-driven Turbinia discussed by Edward Constant, or of aircraft can turn into public spectacles where failure is potentially very costly.10 The media themselves can provide a forum for judging the validity of tests, as Nelly Oudshoorn argues in her work on the testing of male contraceptives.11 Historians of technology have revealed the key role of testing in providing a site for the integration of complicated components in, say, the manufacture of an advanced jet fighter.12 The testing of automobiles has proven a particularly salient research site, with scholars showing how the test equipment to monitor brake performance eventually morphed into today’s antilock braking systems, and that sonic testing is a key aspect of modern auto engineering.13 Three types of testing can be distinguished: “prospective,” “current,” and “retrospective.”14 An example of prospective testing is Stefan Krebs’s
362 Coda
discussion of the tests on the feasibility of the Kunstkopf artificial head as a research tool for spatial listening (Chapter 8). Examples of current tests are those carried out by U.S. Navy sonar operators on board submarines (discussed by Lino Camprubí and Alexandra Hui in Chapter 11), who constantly listen for new sounds in the ocean to update the U.S. Office of Naval Research (ONR) reference file, the standard against which both equipment and ears are tested. Retrospective testing is used to determine the properties of the special varnish used on Stradivari violins, as discussed by Emily Dolan (Chapter 4). Sociologists of technology have been concerned with the role of users in technology.15 Users are implicated in tests of technologies such as phones, personal computers, and consumer electronics. If a technology fails, is that the result of designers failing to understand and test for user practices (testing the user) or some incipient fault in the technology itself? A failure to test how users of software actually use novel features can be seen as a failure of design if the features are then routinely rejected by users. Jonathan Sterne’s chapter on modern digital audio testing is replete with examples of such user tests (Chapter 6). Modern digital music technologies are continuously being tested by offering listeners choices between similar sounds. As Sterne points out, the user’s failure to distinguish between the sounds in such tests is actually a success for a digital technology that posits equivalence between an old analog sound and a digital emulation. The performative aspect of tests is crucial for audio technologies and musical instruments. It can also lead to new music, as Alexander Rehding shows in his chapter on F. W. Opelt and his use of the siren (Chapter 5). The routine testing of audio technologies such as microphones, loudspeakers, and PA systems at concerts is accomplished with the performance of stock phrases such as “Testing . . . one, two, three, testing!” Tests of instruments are carried out in an empty concert hall before the show and are predicated on the assumption of enough similarity between the test performance and the live performance. In the early days of analog synthesizers, such assumptions did not hold up, as the presence of the audience changed the air temperature and detuned sensitive oscillators.16 This testing failure also led to new music, as the rock musician Keith Emerson discovered in concert, when the drifting out of tune of his modular Moog synthesizer created a psychedelic “freak-out” moment with his band Emerson, Lake, and Palmer. Sound studies, an emerging interdisciplinary field, arrived in the late 1980s and 1990s. Interestingly, one of the most influential early studies of sound technologies involved testing: Edison famously used “tone tests” as a way of marketing the new phonograph.17 The collective skills of listeners were at play, the test being whether you could distinguish a phonograph recording
Testing and Why It Matters 363
from a live performer. Both would be on stage, and with one hidden from view, audiences were astounded to find that they could not distinguish between phonograph and performer. This not only helped market and sell phonographs and later Victor gramophones but also established audio fidelity as the benchmark for gramophone performance and listening. The chapters in this book concern many facets of audio technologies and musical instruments and their fascinating histories. Testing plays a role in all these stories. Indeed, testing is a particularly apt research site for sound studies precisely because subjectivity in general, and the essence of hearing as a sense in particular, is shaped, defined, and constituted in these tests along with the technologies or sonic devices being tested. Sonic tests index and configure listening, but at the same time it is listening that does the indexing and configuring of sonic technologies. As Viktoria Tkaczyk notes (Chapter 2) in her discussion of the role of musical instruments and phonographs in the testing traditions of hearing impairment at the end of the nineteenth century, the tools used for testing were as much an object under test as was the human ear. If testing is about establishing similarity or equivalence or commensurability between different things, as the authors of these chapters argue, then it is particularly powerful when it does so for humans and technologies. Attributes of human hearing (or animal hearing; see Bruyninckx in Chapter 10 and Camprubí and Hui in Chapter 11) and properties of technologies or materials are established together in testing, in a process known in STS as co-production or co-construction. Testing is thus at the heart of the politics of sonic technologies. That politics is evident in Mara Mills’s chapter on the emergence of standardized speech sentences used both for testing communication equipment such as telephones and for testing human hearing (Chapter 1). This occurred in a context where new workers’ compensation laws were taking effect. Particular phonetic tests first developed by Bell Telephones became standardized, such as Auditory Test No. W-22, which was marketed to audiologists by the Central Institute for the Deaf and has become one of the most common speech hearing tests today. Mills argues that the testing of mass-produced communication devices was coextensive with mass testing of American ears and that the widespread categorization of the deaf and hard of hearing, rather like medical screening in general, set the limits of permitted variability in both humans and machines. In short, testing was a technocratic means to establish the parameters of the normal. By ostensibly producing objective facts with a standardized procedure, speech testing and its associated technologies provided the very tools for normalizing the range of human capabilities18 and at the same time for standardizing new technologies of communication
364 Coda
and promoting “average” spoken English. In this case, testing enabled the co- production of hearing, technology, and language. The testing of auditory perception is involved not only in the development of new audio technologies and musical instruments but also in the creation of new musical theories. Alexander Rehding examines the history of the siren and the part it played in the now obscure and almost forgotten musical theories of Friedrich Opelt (Chapter 5). The nineteenth-century siren, with its unusual glissando sound, seemed to defy the usual understanding of how musical tones were produced and broke down the conventional distinction between musical rhythm and pitch.19 As Rehding writes, “there is a point in human auditory perception, which was successfully identified with a siren, when pulsations are no longer perceived as individual events, but fuse into a continuous pitch.” It is this idea that discrete sounds—rhythm—can merge into continuous sounds—pitch—that the siren demonstrated. Opelt built complicated polyphonic sirens that blended rhythms and pitches together and allowed for harmonic progression. The many holes in the discs served as a kind of digital record of the music (rather like the punchcard in early computers or the piano roll in the player piano). Opelt’s siren was a recursive or self-exemplifying device—it demonstrated the musical principles built into it. It served as a kind of “music-theoretical instrument.” The piano, too, can be thought of as such an instrument. Producing only notes based on the twelve intervals of the scale, it embodies a certain musical theory and can teach and make manifest or give epistemic weight to that musical theory every time it is used. Opelt’s siren did something similar, the only trouble being that there was almost no music for which it could be used. Rehding describes it as a music that “could have been,” that is to say, “a kind of music that tests and plays with the specific features of human hearing.” This way of testing hearing, unlike others discussed in this book, never took off. Commensuration requires hard work. The new episteme, the instruments, and the music that might have been have vanished into obscurity not because they weren’t possible, but because the sociotechnical work needed to make them manifest was never performed. The role played by testing ears in marketing and selling is echoed in Joeri Bruyninckx’s (Chapter 10) discussion of ultrasound in commercial pest control technologies. Salespeople developed demonstrations where they would produce a screeching blast of audible sound, making prospective customers cover their ears as it was explained to them that still higher frequencies, inaudible to humans, would “zap” pests away. As Bruyninckx puts it, “these demonstrations made ultrasound’s potency to the animal testable through the human ear.” Although the control of animals by sound turned out to be more difficult than these commercial products promised, the very categories
Testing and Why It Matters 365
of “pest” and “nonpest” (a normal human hearing subject) were co-produced through the testing of the audio technologies. Noise control and abatement is an important topic for sound studies.20 It can become a matter for the state and a way of projecting a modernizing state to its citizens. Jennifer Hsieh (Chapter 7) shows this process at work through her examination of the Kuomintang regime’s use of test measurements of noise to establish average acceptable standards throughout Taiwan. Public noise monitoring systems were built, such as a giant display screen at Taipei’s main station that, like a monster traffic light, flashed red if the sound level reached a certain number of decibels.21 Rather than appealing to their own ears as arbiters of noise, citizens learned to depend on the state to take that role. Hsieh describes this as a process of co-production of the state and its citizens. The tests rendered certain sounds as noise, thus making noise legible (and audible) to the state and at the same time establishing what counted as normal hearing for its citizens. The importance of “traditions of testability” is highlighted by Viktoria Tkaczyk in Chapter 2, which examines Otto Abraham’s 1901 test of “absolute tone consciousness” (what we would today call absolute pitch). Abraham, a physicist, physician, and research assistant to Carl Stumpf at the Institute of Psychology in Berlin, was interested in the nuances of individual subjects’ experiences and degrees of skill with musical tones. With his questionnaire method that catered to subtle differences in musical experiences, he departed from the standard experimental methods of hearing tests administered in laboratory sciences. This, argues Tkaczyk, is an important break with nineteenth- century traditions of testability. Abrahams in effect transferred the practices of hearing tests from the natural sciences to the applied human and social sciences. His emphasis on testing for different forms of musicality led the way toward tailoring education and training programs to spot “musical talent” and supporting the new sorts of musical professions and identities that arose with the recording, radio, and film industries. Roland Wittje’s chapter on the successful efforts of Johan Peter Holtsman to establish the first acoustic laboratories at the Norwegian Institute of Technology in Trondheim (Chapter 9) points to a fundamental tension in testing. Many engineering tests of materials are carried out in the circumscribed world of the laboratory, where the best equipment can be built and maintained and repeatable standardized results can be established. But such laboratories do not exist to produce knowledge for its own sake, as in physics, but rather to understand the properties of materials as they are actually used in the world of practice. This dilemma is illustrated nicely by Holtsman’s activities, which were split between the cloistered world of his
366 Coda
laboratory and the messy world of applying his knowledge in the construction of buildings, radio studios, cinemas, and the like. Holtsman and his colleagues needed a way of translating their skills and knowledge from the laboratory to the field. As Wittje shows, they tried to do this by physically moving the laboratory to the field, building special portable versions of testing instruments that had to be robust, compact, and easy to set up. Holtsman’s field practices are best described as “consulting” rather than science and entailed negotiating the complex sociotechnical world of architects, civil engineers, and building users involved in, say, the construction of the Norwegian Broadcasting Corporation (Norsk Rikskringkasting, NRK) broadcasting house. This form of testing, which did not offer the same standardization and replication as was possible in the laboratory, nevertheless still produced robust and credible knowledge. By separating his consulting from his science, and his testing in the lab from his testing in the field, and moving back and forth between the two worlds, Holtsman seems to have skillfully solved the problem of projecting laboratory testing knowledge into actual use. The issue of establishing a similarity relationship between test site and the context of use is at the heart of Stefan Krebs’s chapter on the use of artificial heads in the development of binaural audio technologies in Germany between 1967 and the early 1980s (Chapter 8). The use of the Kunstkopf projected a similarity relationship between real listeners and artificial listeners and enabled researchers to bring the concert hall into the laboratory, which, notes Krebs, rendered subjective listening tests more objective. The fascinating, almost creepy, pictures of these heads point also to the implicit biases of gender in this projected similarity relationship, as the German scientists often modeled the heads on themselves or their male friends (head- related recordings were optimized for male listeners rather than for the generally smaller female heads and ears). The program’s failures are partly due to this bias, but also, as Krebs points out, to the fact that the different German groups were deeply rooted in divergent experimental cultures and local laboratory contexts. As a result, listening tests and measurements often turned out to be incommensurable—a paradoxical outcome, given that testing is about commensuration. The topic of Sebastian Klotz’s chapter is what happens when hearing tests enter the field, such as in anthropological studies of remote populations (Chapter 3). Klotz examines the Cambridge Anthropological Expedition to Torres Straits (CAETS), which began in April 1898, as well as a comparative study in a remote part of Aberdeenshire, Scotland. The expedition was multidisciplinary, but at the heart of its methods was the administration of hearing tests. These involved portable devices, such as the Galton whistle.
Testing and Why It Matters 367
Klotz describes the hearing tests administered by psychologist Charles Myers as “mediating a whole series of translations,” including translations from the laboratory to the field, from anthropology to psychology and eventually to sociology, and from discursive paradigms of genealogy and racial inequality to quantitative measurements including statistics. He points out that the hearing tests not only tested the hearing of specific actors in the Torres Straits but also tested the convictions that informed the CAETS. The researchers had come armed with logics of scientific measurement, initially hoping to establish through hearing that “primitives” had different hearing abilities, but they also experienced the reflexivity of the research situation, the limitations of their own methods, the difficulty of finding “culturally pure” islanders untainted by modernity and commerce, and the reality that many of the people they were studying had hearing severely damaged by diving yet still produced a sophisticated musical culture, which Myers found himself “tested” to hear and transcribe. Hearing tests were changing in this period, too, as tone tests began to be informed by phenomenology and philosophy. Klotz’s conclusion is that the hearing tests administered outside the laboratory in these complicated and novel social, cultural, methodological, and technical situations tell us “more about the dynamics of these interactions and research ideologies than about their actual topic, human hearing.” Lino Camprubí and Alexandra Hui’s chapter, on the history of how “technological listening” transformed the ocean from a quiet space to a sonically rich one, shows again the multifaceted nature of testing, with tests of one thing morphing into tests of another, and the key role played by commensuration (Chapter 11). The military and, in particular, the U.S. Navy play a crucial part in this story, beginning with the need for trained sonar operators during World War II. Special hearing tests were developed for naval personnel to identify those most likely to be skilled at operating sonar equipment. The navy started off using the aptly named “Seashore test” of musical talent, discussed by Tkaczyk, but found it inadequate for testing the special directional auditory skills that made for a good sonar operator. Camprubí and Hui also discuss tests of machine listening, animal sounds, transducers, and various visual representations of acoustic phenomena. They tell the famous stories of the discovery of one of the loudest ocean noises that disrupted sonar listening—the roar of snapping shrimp22—and of how the songs of whales were heard for the first time by Chris Clark and Katy Payne when asked to analyze sonograms of unexplained sounds detected in tests by the U.S. Navy’s global sonar surveillance system. Camprubí and Hui argue that testing was the mediator between these different sonic realms. For instance, the testing of hydrophones enabled the sounds of marine species such as the snapping
368 Coda
shrimp to be identified; at the same time, animal laboratory tests enabled new interpretations of hydrophone signals. As they put it: Once the connections between these sonic realms were established through testing, these realities came to function as proxies of one another, enabling the calibration, standardization, and mobility of sonic technologies and phenomena across the world’s oceans. By rendering alien realities commensurable, . . . [t]esting performed the dual epistemological and ontological work of making the underwater world knowable by sonifying it.
Jonathan Sterne explores the modeling of analog signal processing for commercial music in the digital domain (Chapter 6). He focuses on the digital modeling of analog spring reverb devices: “the ways in which digital models at once test the hearing of machinery . . . and use the machinery to test the hearing of users.” Both are at stake in the sorts of audio testing Sterne discusses. But, as he points out, in the actual use of these devices, sound alone is only part of the story—the look of devices, in particular their skeuomorphic properties; the feel of an instrument; and the identity of the user as a musician may all be tied together in a way that sonic testing alone fails to capture. For example, a famed plate reverb unit from the Muscle Shoals studio is supposed to have contributed to the unique sound of that studio because of the humid swamp conditions of its operation—but no one knows for sure, and anyway most listeners cannot tell. The mystique of this treasured artifact, however, assures a value that is beyond testing. Hard-to-define aesthetic values are at play. Tests are performances where politics and value are part of what is established. Sterne is a pioneer of sound studies, and his work exemplifies his program to research what he calls the “politics of transduction: how cultural, historical, and economic relations are rendered in the sonic realm, and how dimensions of sound and sound technologies come to have value.” For Sterne, the moment of testing listening—whether it is how a signal processor receives and renders a signal or how a listener responds to a digital model of an analog device— is crucial because this is where the politics of transduction are “negotiated, refracted, and come to life.” The process of building a digital model of an analog device is an iterative one of defining, testing, refining, and redefining with listening tests at every stage. Tests are performances that establish and consecrate equivalence. Politics are always there: differences that are important may be set aside at the moment of testing. As Sterne reminds us, “who gets to signal process and under what conditions is a central question of media theory, and the very question that is left aside at the moment of the listening test.”
Testing and Why It Matters 369
That the construction of facts about sonic technologies is not solely a matter of following the best test procedures, knowledge, and equipment is shown in Emily Dolan’s chapter on testing the violin (Chapter 4). This most mythical of instruments has been subject to more scientific interest and tests than almost any other musical instrument. Yet despite investigations by acousticians, chemists, and climate scientists showing that the famed properties of old violins are often rather mundane, and despite every test revealing that skilled violin players cannot distinguish old instruments from modern ones, or even that they prefer modern ones, the myth of the ancient Stradivari as musical perfection is only enhanced. As Dolan points out, the diverse forms of testing to which the violin has been subjected have to be understood as partly performative and aesthetic acts. It seems in this case that the test results themselves are robust; no one casts doubt on the particular protocols and procedures followed. It is rather that the tests tout court fail to make a difference to the perception of the violin, or paradoxically even enhance its reputation by making it an object of interest to science. * * * Given the critique of testing offered by this book as a whole, are there any alternatives to testing, and what would they look or sound like? Are other goals than similarity, standardization, equivalence, and commensuration possible? The first thing to note is that by excavating and reframing testing— by showing how it works and what is at stake—the authors in this collection are doing politics. Showing what is at stake in testing offers the possibility of doing it in a different way. We can argue for different sets of equivalences, for including social groups and actors who are currently excluded from test sites, and for collective decision making as to when and under what circumstances it is appropriate to run a test. Testing programs can be resisted. But are there even more radical possibilities, such as using tests as an intervention into aesthetics and art to free human capabilities from the mantle of standardization, commensuration, and the incipient enframing of technology as pointed out by Heidegger? Benjamin Steege (Chapter 12) is one of the few authors to consider such a possibility, examining the writings of Günther Anders on testing. It seems Anders, inspired by Walter Benjamin and Bertolt Brecht, wanted to radically recast tests as an opportunity for poiesis—not only to resist the standardization of humans but also to produce a transformation in humans and bring forth artistic and aesthetic feelings. Steege examines this unlikely
370 Coda
prospect through a composition by Herbert Eimert (the founder of the famous electronic music studio at West German Radio in Cologne), Epitaph für Aikichi Kuboyama. This takes as its raw material a poem about the radio operator on a Japanese fishing boat, the Lucky Dragon, that became caught up in the fallout from a nuclear test on the Marshall Islands. Kuboyama received much international attention in the 1950s as the first victim of the Marshall Islands nuclear testing. The Eimert work ends with what Steege hears as a call for hope, despite the prevailing feeling of hopelessness of living in the shadow of Hiroshima and the nuclear bomb. Paradoxes abound: A work like Eimert’s may be heard to inhabit a disposition of both hope and hopelessness at once, such that they are completely intertwined. Similarly, what Anders seems to be asking of us is at once overly modest and nearly impossible. Reduced to the formulation that we are simply to answer the domination of “human engineering” with a countervailing program of moral exercises or self- tests on the terrain of aesthetic experience, it is impossible to avoid the impression of a quaint humanism unequal to its task even as it becomes aware of its own “obsolescence.” But then, after all, the action of testing is like this: it is ongoing, does not ask in advance for any particular outcome, acquires substance only in relation to its latest iteration, and hence does not entail a particular hope for its success, yet nonetheless wants to offer us some material possibility for continuing and revising.
Most testing is indeed mundane, routine, and ongoing, yet as the chapters in this book show, it can be critiqued. Testing, because it inevitably involves cultural and social assumptions and politics, must deal with unruly humans as well as nonhumans. Testing can be unpredictable, subject to multiple interpretations, and seems to find its least promising territory in art and aesthetics. There are practices in testing technology that are built on a more radical politics. The tests by the “circuit benders” who try to repurpose sonic technologies for new aesthetic experiences may be such an example.23 Also, the search in valuation studies for moments of dissonance is a form of analysis that tries to resist commensuration.24 The possible paradoxical outcomes of testing bring us back to the story with which I started this Coda. Whale songs may have been discovered from sonograms at the heart of a U.S. military beast, but the sounds of the whales still maintain their magnificence as they reverberate through the oceans and from Ivy League clock towers to produce moments of dissonance. They defy and resist equivalence.
Testing and Why It Matters 371
Notes 1. On laboratories, see Bruno Latour and Steve Woolgar, Laboratory Life; Karin Knorr-Cetina, Manufacture of Knowledge. On scientific controversies, Andrew Pickering, Constructing Quarks; Harry Collins, Changing Order; Trevor Pinch, Confronting Nature. 2. David Gooding, Trevor Pinch, and Simon Schaffer, Uses of Experiment; Peter Galison, How Experiments End; Hans-Jörg Rheinberger, Toward a History of Epistemic Things. 3. Simon Schaffer, “Late Victorian Metrology”; Joseph O’Connell, “Metrology”; Bruce Hunt, “The Ohm Is Where the Art Is”; Stefan Timmermans and Marc Berg, “Standardization in Action”; Graeme Gooday, Morals of Measurement; Karen Rader, Making Mice. 4. Donald MacKenzie, “From Kwajalein to Armageddon”; Trevor Pinch, “ ‘Testing—One, Two, Three, Testing . . . !’ ” 5. Edward Constant, Origins of the Turbojet Revolution; Constant, “Scientific Theory and Technological Testability.” 6. Benjamin Sims, “Concrete Practices.” 7. MacKenzie, “From Kwajalein to Armageddon.” 8. Collins, Changing Order; Pinch, “ ‘Testing—One, Two, Three, Testing . . . !’ ” 9. Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump. 10. Constant, Origins of the Turbojet Revolution; Constant, “Scientific Theory and Technological Testability.” 11. Nelly Oudshoorn, “On Masculinities, Technologies and Pain.” 12. Glen Bugos, “Program Management and the Manufacture of Certainty.” 13. Ann Johnson, Hitting the Brakes; Eefje Cleophas and Karin Bijsterveld, “Selling Sound.” 14. Pinch, “ ‘Testing! One, Two, Three, Testing . . . !’ ” 15. Nelly Oudshoorn and Trevor Pinch, How Users Matter. 16. Trevor Pinch and Frank Trocco, Analog Days. 17. Emily Thompson, “Machines, Music and the Quest for Fidelity”; Thompson, Soundscape of Modernity. 18. See also Allan Hanson, Testing, Testing. 19. Myles W. Jackson, Harmonious Triads. 20. Thompson, Soundscape of Modernity; Karin Bijsterveld, Mechanical Sound. 21. This is reminiscent of the “SoundEar” monitoring technology developed for classroom use in Denmark, discussed in Trevor Pinch and Karin Bijsterveld, “New Keys to the World of Sound.” 22. See also Stefan Helmreich, “Anthropologist Underwater.” 23. Trevor Pinch, “ ‘Bring on Sector Two!’ ”; Lauren Flood, “Building and Becoming.” 24. Trevor Pinch, “Moments in the Valuation of Sound”; Ariane Berthoin Antal, Michael Hutter, and David Stark, Moments of Valuation.
References Antal, Ariane Berthoin, Michael Hutter, and David Stark, eds. Moments of Valuation: Exploring Sites of Dissonance. New York: Oxford University Press, 2015. Bijsterveld, Karin. Mechanical Sound: Technology, Culture, and Public Problems of Noise in the Twentieth Century. Cambridge, MA: MIT Press, 2008.
372 Coda Bugos, Glen. “Program Management and the Manufacture of Certainty in the 1950s.” Paper presented to the annual meeting of SHOT, Sacramento, CA, 1989. Cleophas, Eefje, and Karin Bijsterveld. “Selling Sound: Testing, Designing, and Marketing Sound in the European Car Industry.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 102–24. Oxford: Oxford University Press, 2012. Collins, Harry. Changing Order: Replication and Induction in Scientific Practice. London: Sage, 1985. Constant, Edward. The Origins of the Turbojet Revolution. Baltimore, MD: Johns Hopkins University Press, 1980. Constant, Edward. “Scientific Theory and Technological Testability: Science, Dynameters, and Water Turbines in the Nineteenth Century.” Technology and Culture 24, no. 2 (1983): 183–98. Flood, Lauren. “Building and Becoming: DIY Music Technology in New York and Berlin.” PhD diss., Columbia University, 2016. Galison, Peter. How Experiments End. Chicago: University of Chicago Press, 1987. Gooday, Graeme. The Morals of Measurement: Accuracy, Irony and Trust in Late Victorian Electrical Practice. New York: Cambridge University Press. Gooding, David, Trevor Pinch, and Simon Schaffer, eds. The Uses of Experiment: Studies in the Natural Sciences. Cambridge: Cambridge University Press, 1987. Hanson, Allan. Testing, Testing: Social Consequences of the Examined Life. Berkeley: University of California Press, 1993. Helmreich, Stefan. “An Anthropologist Underwater: Immersive Soundscapes, Submarine Cyborgs, and Transductive Ethnography.” American Ethnologist 34, no. 4 (2007): 621–41. Hunt, Bruce. “The Ohm Is Where the Art Is: British Telegraph Engineers and the Development of Electrical Standards.” Osiris 9 (1994): 48–63. Jackson, Myles W. Harmonious Triads: Physicists, Musicians, and Instrument Makers in Nineteenth-Century Germany. Cambridge, MA: MIT Press, 2006. Johnson, Ann. Hitting the Brakes: Engineering Design and the Production of Knowledge. Durham, NC: Duke University Press, 2009. Knorr-Cetina, Karin. The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science. Oxford: Pergamon, 1981. Latour, Bruno, and Steve Woolgar. Laboratory Life: The Social Construction of Scientific Knowledge. London: Sage, 1979. MacKenzie, Donald. “From Kwajalein to Armageddon: Testing and the Social Construction of Missile Accuracy.” In The Uses of Experiment, edited by David Gooding, Trevor Pinch, and Simon Schaffer, 409–36. Cambridge: Cambridge University Press, 1989. O’Connell, Joseph. “Metrology: The Creation of Universality by the Circulation of Particulars.” Social Studies of Science 23, no. 1 (1993): 129–73. Oudshoorn, Nelly. “On Masculinities, Technologies and Pain: The Testing of Male Contraceptives in the Clinic and the Media.” Science, Technology, & Human Values 24, no. 2 (1999): 265–89. Oudshoorn, Nelly, and Trevor Pinch, eds. How Users Matter: The Co-Construction of Users and Technology. Cambridge, MA: MIT Press, 2003. Pickering, Andrew. Constructing Quarks. Edinburgh: Edinburgh University Press, 1984. Pinch, Trevor. “‘Bring on Sector Two!’ The Sounds of Bent and Broken Circuits.” Sound Studies 2, no. 1 (2016): 36–51. Pinch, Trevor. Confronting Nature. Dordrecht: Reidel, 1986. Pinch, Trevor. “Moments in the Valuation of Sound: The Early History of Synthesizers.” In Moments of Valuation: Exploring Sites of Dissonance, edited by Ariane Berthoin Antal, Michael Hutter, and David Stark, 15–36. New York: Oxford University Press, 2015. Pinch, Trevor. “‘Testing—One, Two, Three, Testing . . . !’ Toward a Sociology of Testing.” Science, Technology, & Human Values 18, no. 1 (1993): 25–41.
Testing and Why It Matters 373 Pinch, Trevor, and Karin Bijsterveld. “New Keys to the World of Sound.” In The Oxford Handbook of Sound Studies, edited by Trevor Pinch and Karin Bijsterveld, 3–35. Oxford: Oxford University Press, 2012. Pinch, Trevor, and Frank Trocco. Analog Days: The Invention and Impact of the Moog Synthesizer. Cambridge, MA: Harvard University Press, 2002. Rader, Karen. Making Mice: Standardizing Animals for American Biomedical Research. Princeton, NJ: Princeton University Press, 2004. Rheinberger, Hans-Jörg. Toward a History of Epistemic Things: Synthesizing Proteins in the Test Tube. Palo Alto: Stanford University Press, 1997. Schaffer, Simon. “Late Victorian Metrology and Its Instrumentation: A Manufactory of Ohms.” In Invisible Connections: Instruments, Institutions, and Science, edited by Robert Bud and Susan E. Cozzens, 23–56. Bellingham, WA: SPIE Optical Engineering Press, 1992. Shapin, Steven, and Simon Schaffer. Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life. Princeton, NJ: Princeton University Press, 1985. Sims, Benjamin. “Concrete Practices: Testing in an Earthquake-Engineering Laboratory.” Social Studies of Science 29, no. 4 (1999): 483–512. Thompson, Emily. “Machines, Music and the Quest for Fidelity: Marketing the Edison Phonograph in America, 1877–1925.” Musical Quarterly 79, no. 1 (1995): 131–71. Thompson, Emily. The Soundscape of Modernity: Architectural Acoustics and the Culture of Listening in America, 1900–1933. Cambridge: MA: MIT Press, 2003. Timmermans, Stefan, and Marc Berg. “Standardization in Action: Achieving Universality Through Local Medical Protocols.” Social Studies of Science 27, no. 2 (1997): 273–305.
Index Figures are indicated by f following the page number. 1176 limiting amplifier, 167–68 Aachen Head, 228 Abbey Road, 160 Abel, Jonathan, 168, 171, 177–79, 181, 182n20 Aberdeenshire, Scotland, 79, 86–88, 366 Ableton Live, 165, 169 Abraham, Salomon Otto, 12–13 absolute tone consciousness and, 49–68, 70n45, 95, 365 acoustics and, 52, 55, 61–63 animals and, 49 anthropology and, 60–62, 67, 70n58 audiology and, 49, 53, 67 auditory acuity and, 50, 62 Beethoven and, 49, 61 frequency and, 50, 64 Hornbostel and, 55, 62, 68, 72n95 Institute of Applied Psychology and, 96 localization and, 55, 57 loudness and, 50 music and, 49–56, 59–68, 69n21, 70n45, 352 neuroscience and, 55, 57, 67 otherness testing and, 61–66 parrot of, 49, 61 phonetics and, 52, 54, 65 phonograph and, 51, 53, 62, 68n11 physiology and, 50–51, 54, 56, 66–67 pitch and, 49–56, 59–68, 69nn20–21, 70n45, 70n52, 352, 365 politics of the testing of many and, 66–68 Preyer and, 51–53, 67 psychology and, 50, 53–68, 70n45, 365 radio and, 65 Seashore and, 50, 63–64, 68 statistics and, 50, 59–60, 70n58 stimulus–response experiments and, 49, 54, 67 Stumpf and, 50–51, 54–56, 59, 61–63, 67–68, 70n45, 365
surveys and, 49–50, 53, 56, 59–61, 67, 70n58 talent and, 50, 52, 59–65, 68 timbre and, 51, 53, 64 tone tests and, 50–53, 62, 64 tuning forks and, 52, 59, 63, 68n11 violins and, 50–52, 68n7 absolute pitch, 49–56, 59, 61, 63, 66–67, 352, 365 absolute tone consciousness Abraham and, 49–68, 70n45, 95, 365 Beethoven and, 49, 61 consciousness of sensation and, 55–56 cultural nurturing and, 56, 67 epistemological issues and, 53–59 laryngeal memory and, 55, 59 perfect pitch and, 9, 49–52, 56, 67 relative pitch and, 49, 53, 66–68 speech and, 50–52, 59, 68n11, 69n20 statistics and, 50, 59–60, 70n58 stimulus–response experiments and, 49, 54, 67 subjective experience and, 53–54 talent and, 50, 52, 59–65, 68 tone tests and, 50–53, 62, 64 two-culture divide and, 54 absorption, 14, 246–47, 255, 262 A/B tests, 162, 171, 174 A/B/X tests, 162, 174–75 Academic Radio Club, 249 acoumeters, 8, 30, 31f, 83–84 Acoustical Society of America, 116, 248, 250, 286 Acoustic Audio, 177 acoustics Abraham and, 52, 55, 61–63 absorption and, 14, 246–47, 255, 262 Anders and, 336, 338, 340–42 architectural, 252–56 attacks using, 292, 296n66
376 Index acoustics (cont.) auditory thresholds and, 273–92, 293n20, 295n53, 296n67, 354 Catgut Acoustical Society (CAS) and, 113 Chladni patterns and, 119 cinemas and, 167, 248, 252, 255, 262, 366 concert halls and, 5, 14, 115, 134, 216–19, 222, 231–32, 247–48, 255, 262, 265n52, 362, 366 cross-cultural studies and, 82–83, 88, 91, 98 Demolition and, 219–20 digital models and, 159, 168, 173, 177, 353 dummy-heads and, 14, 234n1, 234n10, 236n38, 237n62, 238n78 electroacoustics and, 1, 10–12, 25, 27, 35, 40–41, 159, 214–16, 219, 226, 228–29, 243–50, 256, 258, 260–61, 319, 354 environmental, 189, 191, 193, 200–204 Institute of Technical Acoustics and, 214, 239n89 lowered quality of, 262–63 materials and, 243–63, 264n22, 264n25, 264n35, 264n37, 265n43, 265n48, 265n51, 266n56, 266n70, 353, 365 motion pictures and, 243, 246–48, 252, 261 music and, 1, 10–15, 35, 52, 55, 61–63, 82– 83, 88, 98, 111, 133–34, 138, 140, 145, 149–51, 159, 168, 214, 219, 222–23, 226, 229–33, 243–46, 249, 255, 317–18, 338, 340, 367, 369 perfection and, 111–16, 119, 123–24, 369 Pohl and, 133–34, 138–39, 149, 151, 152n12, 154n49 psychoacoustics, 82, 88, 91, 225, 229, 232–33, 284–88 radio and, 14, 25, 35, 243–51, 261–62 sirens and, 131–40, 145, 147, 149–51, 152n12, 152n16, 154n49 Sixth International Congress on Acoustics and, 235n18 soundwaves and, 113, 136, 143 spatial hearing and, 214–19, 222–33, 235n18, 236n46, 239n89 speech and, 25, 27–28, 35, 38, 40–41 Taiwan study and, 189, 191, 193, 200–204 testing and, 1, 7–15, 17n21 transmission and, 14, 247, 255, 258, 262 underwater studies and, 302–6, 311–20, 321n39, 367
Western Electric (WE) and, 25–27, 31–32, 34–35, 37–38, 40, 46n79 aesthetics Anders and, 328, 330–34, 341, 343 Benjamin and, 331–32 Brecht and, 331–32 Cogdell and, 32 cross-cultural studies and, 82, 93 digital models and, 159, 162, 168, 170, 172, 180–81, 370 Eimert and, 370 empirical, 4 Fechner and, 60 golden ratio and, 60 materials and, 247, 254 music and, 9, 11, 13–15, 54, 82, 110, 123, 125, 134, 149, 159, 162, 168, 170, 172, 180–81, 330, 353, 368–69 perfection and, 110, 123–25, 353 speech and, 32 Stumpf and, 54 Taiwan and, 191–92 underwater studies and, 301 AKG BX20, 161, 168–73, 177–80 Akiyama, Mitchell, 274, 293n4 Allegemeine Theorie der Musik (Opelt), 139, 142f, 148 Amati, 113, 117 American Association for the Advancement of Science (AAAS), 273 American Association for the Hard of Hearing, 31 American Bureau of Ethnology, 90 American Guild of Violinists, 116 American League for the Hard of Hearing, 34 American Ontological Society, 24, 276 American Telephone and Telegraph (AT&T), 12, 25, 26f, 32–35, 37, 276 amplifiers 1176 limiting, 167–68 analog signal processing and, 10, 13, 177, 244 audiology and, 35, 43 auditory thresholds and, 293n17 digital models and, 159, 161, 163, 167–77, 179, 183n29, 213 LA-2A, 167–68 materials and, 244, 258, 260 vacuum tube, 10, 177, 244 analog signal processing
Index 377 amplifiers and, 10, 13, 177, 244 black-boxing and, 167, 170, 179 concept of, 166–67 digital models of, 13, 159–81 guitar amplifiers and, 172–76 historical perspective on, 159, 167–72 reverbation and, 161, 163, 167–73, 178, 180, 368 revival of, 160–61, 167–68 testing and, 162–67 vibe of, 161 Anders, Günther acoustics and, 336, 338, 340–42 aesthetics and, 328, 330–34, 341, 343 anthropology and, 327, 329, 333, 343 antipathy toward testing of, 327–28 background of, 327–29 Benjamin and, 330–32, 337, 341, 344n14, 369 Brecht and, 330–32, 334, 337, 341, 369 Die Antiquiertheit des Menschen and, 329, 332–33, 335, 345n22 Eimert and, 334–43, 345n23, 345nn26–27, 346n33, 354, 370 Epitaph für Aikichi Kuboyama and, 334–43, 345n25, 346n32, 354, 370 “human engineering” and, 327–28, 332– 33, 343, 344n2, 370 military and, 328, 330 modulation and, 331–32 music and, 330, 334–41, 345n22, 346n33 noise and, 340 nuclear energy and, 15, 329–35, 340, 343, 354 Operation Castle and, 328 poiesis and, 341, 369 psychology and, 327–28, 333, 341, 344n2, 344n14 standardization and, 327, 369 Stern and, 327–28, 344n4 techniques of feeling and, 332–34 tones and, 337 anechoic chambers, 221–22, 224, 227, 244, 262, 282, 290 Anfänge der Musik, Die [The Origins of Music] (Stumpf), 61 Animal Hearing Laboratory, 276, 280 animals Abraham’s parrot, 49, 61 audiometry and, 30, 277, 354
auditory thresholds and, 273–92, 293n20, 294n28, 295n53, 354, 363–64 birds, 49, 61, 282, 288–91, 296n57, 302 cats, 276–80 dogs, 276–80, 285 electric shock and, 277 elephants, 359 fish, 293n20, 294n28, 301, 305–6, 309–16, 318–21, 322n40, 322n47, 322n54 food and, 275–77, 285, 309 music and, 49, 61, 363, 367 Noise Control Act and, 190 rats, 201, 273–76, 279–80, 290, 295n56 snapping shrimp, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 testing and, 3, 14, 190, 200–201, 204, 275–80, 354–56 underwater studies and, 301–2, 305–13, 316–19, 321n39, 322n60, 359, 367–68, 370 whales, 301–2, 308, 317–18, 322n60, 359, 367, 370 whistles and, 80 Anleitungen zur psychologischen Untersuchung primitiver Menschen [Introductions for the psychological investigation of primitive people] (von Hornbostel), 96–99 anthropology Abraham and, 60–62, 67, 70n58 Anders and, 327, 329, 333, 343 Anleitungen and, 96–99 Cambridge Anthropological Expedition to Torres Straits (CAETS) and, 13, 77–104, 366–67 digital models and, 182n9 Fischer and, 192 Galton and, 60 multidisciplinary approach and, 95–99 nature–nurture debate and, 54 Notes and Queries on Anthropology and, 83, 91, 97 Seashore and, 15, 50, 63–64, 68, 72n92 Taiwan study and, 192 testing and, 5, 10–13, 15 Thurnwald and, 96–99, 103nn88–89, 103nn92–93 Virchow and, 70n58 Wundt and, 60 Anthropometric Laboratory, 80
378 Index Antiquiertheit des Menschen, Die [The Obsolescence of the Human] (Anders), 329, 332–33, 335, 345n22 Appunn, George, 8, 51 Archives of Otolaryngology journal, 35 articulation cross-cultural studies and, 94, 96 Opelt studies and, 141 perfection and, 115 speech and, 27, 36–39, 43, 45n61, 46n70, 337 Aschoff, Volker, 224 Asia-Oceania Association of Otolaryngology Head and Neck Surgery, 195 assaying, 355 Association of Ear, Nose, and Throat Doctors, 195 Atkinson, R. C., 219 audiology Abraham and, 49, 53, 67 auditory thresholds and, 280 History of Audiology and, 25 speech and, 25, 28 testing and, 4, 7, 9, 15, 17n21 audiometry animals and, 30, 277, 354 AT&T and, 12, 25, 26f, 32–35, 37 auditory thresholds and, 276–79, 290, 293n13, 293n17 Beltone and, 194 decibels and, 19, 34, 189–93, 198–205, 206n5, 208n36, 250, 264n35, 307, 365 detectability and, 25, 36–38 Feldman on, 24–28 geopolitics and, 193–96 hearing norms and, 11 hydrophones and, 301–6, 309–12, 316–19, 322n47, 354, 367–68 intelligibility and, 25, 27–28, 36–38, 40, 45n47, 255, 265n52 noise and, 10, 27, 35, 42, 191, 193–96, 203–4, 277 pain threshold and, 131 phonographs and, 23–27, 29–30, 31f, 36, 43n5, 51, 53, 62, 68n11, 89–90, 96, 98, 102nn65–66, 102n68, 353, 362–63 radio and, 5 (see also radio) reference zero and, 25, 32 speech and, 24–40, 42, 43n9, 44n37, 44n39, 45n42, 45n45, 353
systematic approaches to, 28 Taiwan and, 103–6, 194, 208n36 telephones and, 11–12 (see also telephones) testing and, 7–12, 354 underwater studies and, 307 Western Electric (WE) and, 25–27, 31–32, 34–35, 37–38, 40, 46n79 Audiophone, 30 auditoriums, 116, 218, 222, 236n27 auditory acuity Abraham and, 50, 62 cross-cultural studies and, 78, 82, 84–86, 92–94, 98 frequency and, 8–9, 50 loudness and, 2, 8–9, 13, 50, 62, 78, 82, 84–86, 92–94, 98 testing and, 2, 8–9, 13 Auditory Test No. W-22, 41, 363 auditory thresholds acoustics and, 273–92, 293n20, 295n53, 296n67, 354 amplifiers and, 293n17 anechoic chambers and, 282, 290 animals and, 273–92, 293n20, 294n28, 295n53, 354, 363–64 audiology and, 280 audiometry and, 276–79, 290, 293n13, 293n17 bandwidth control and, 283–86 deafness and, 287, 289, 293n5, 295n51 echolocation and, 280 extraordinary sensitivity and, 289–92 food and, 275–77, 281, 285 frequency and, 14, 273–92, 293nn4–5, 293n9, 294n25, 294n28, 295n51, 295n53, 296n57, 364 Gould and, 276, 279–80 hearing aids and, 289 hearing loss and, 289 imaginary sensitivity and, 289–92 imagination and, 289–92 loudness and, 9–10, 24, 42, 286 Maier and, 273, 275, 293n8 military and, 14, 275, 280–82, 289, 292, 295n46, 295n53, 296n61 Moe and, 284, 294n40, 295n46 Morgan and, 274–76, 279–80, 282, 293n8 music and, 291, 292n2, 295n51 neuroscience and, 273, 280, 282, 295n51
Index 379 noise and, 277, 280, 282, 286, 288, 290–91, 292n3, 295n56, 354 pain and, 131 physiology and, 273, 276, 281, 290 pitch and, 8–9 psychology and, 273–80, 284, 287, 290 radio and, 286, 291, 293n13, 295n46 seizures and, 273–76, 280, 282 sirens and, 132–33, 280–84, 296n57 snapping shrimp and, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 speech and, 42, 276, 285, 289 spondees and, 41 standardization and, 274, 276, 291, 354 stimulus–response experiments and, 277–78, 293n20 telephones and, 279, 285, 295n44 tones and, 276–78, 282–83, 287–88, 291, 296n57 tuning forks and, 277 turbojets and, 280, 294n28 ultrasound and, 14, 80, 274–75, 279–92, 293n4, 293n6, 294n28, 294n40, 295n46, 295n48, 295nn53–54, 296n66, 364–65 whistles and, 281, 284 Auerbach, Felix, 66–67 aurality, 2, 4, 7, 9, 11–16, 49 AV-Alarm and Transonic, 289 Bachelard, Gaston, 352 Bacon, Lloyd, 248 Bądarzewska-Baranowska, Tekla, 189 Baer, Karl Ernst von, 154n48 Bagenal, Hope, 262–63 bandwidth, 283–86, 290 Bárány, Ernst, 35 Bartlett, Frederic C., 304 Bartók, Béla, 55 Beethoven, Ludwig van, 49, 61, 112, 149, 189 Bell, Alexander Graham, 36–37 Bell Telephone Laboratories, 25–27, 32, 38, 40–41, 45n61, 46n79, 213–14, 222, 264n36, 266n58, 276, 363 Beltone, 194 Benjamin, Walter, 330–32, 337, 341, 344n14, 369 Berg, Reno, 249–51 Berger, Richard, 246–47 Berlin Academy of Music, 64–65
Berlin Institute of Psychology, 50, 54–56, 63–64, 67–68, 69n35, 82, 365 Berners, Dave, 168, 171–72 Beyer, Robert, 336 Bezold, Friedrich, 34 Bijsterveld, Karin, 190 Bikini Atoll, 328, 354 binaural sound Blauert and, 216, 224–29, 233, 235n18, 237n62, 239n90 concert halls and, 216–19, 222, 231–32, 362 constructing average human listener and, 220–24 cross-cultural studies and, 83 Damaske and, 220–22, 224, 230, 233, 236n39, 236n46 de Boer and, 213–14, 234n4 Demolition and, 219–20 early studies of, 234n7 Genuit and, 228–29 Hammer and, 213–14 hearing aids and, 14, 213–14, 223, 228–29, 231, 233 increasing testing reliability and, 215–20 KEMAR and, 223 Kürer and, 215–19, 231, 233, 235n23 Laws and, 225–29, 233, 237n59 localization and, 214, 217–21, 224–32 Mellert and, 221–23, 233, 236n46 Platte and, 225–29, 233, 237n66 Plenge and, 215–19, 222, 227, 231, 233, 235n14, 236n27 Snow and, 213–14, 234n1 spatial hearing and, 213–23, 227–33, 234n1, 234n4, 234n7, 234n10, 236n36, 236n39, 238n81, 239n91, 366 underwater studies and, 303–4 van Urk and, 213–14 Wagener and, 220–22 Wilkens and, 215–19, 222, 227, 231, 233, 235n14, 235n21, 236nn27–28, 237n47 See also Kunstkopf Biographie universelle (Féti), 140, 149, 153n34 Bird-E-Vict, 289–91 birds, 49, 61, 282, 288–91, 296n57, 302 Bjørnstad, Eilif, 250 black box, 3, 5, 28, 167, 170, 179, 215 Blauert, Jens, 216, 224–29, 233, 235n18, 237n62, 239n90
380 Index Blaukopf, Kurt, 101n29 Blume, Stuart, 26 Blunt, Anne, 122 Boas, Franz, 55, 92, 96 bone conduction, 8, 83 Borgmann, Albert, 166–67 Borneo, 79 Borsarello, Hugues, 115 Bose, Fritz, 55 Boucher, Alexandre-Jean, 120 Brahms, Johannes, 317 brass instruments, 111 Brecht, Bertolt, 330–32, 334, 337, 341, 369 Bristol, William, 30 Bristolphone, 30, 35 Brooks, Evan, 180–81 Bruckner, Josef Anton, 334–35 Brüel & Kjaer, 223, 226, 230 Bruner, Frank, 92–93, 98 Bruyninckx, Joeri, 10, 273–99, 354, 364–65 Bryant, William Sohier, 24, 29–30, 31f Bunch, Cordia C., 39 Burkhard, Mahlon D., 223 Burnham, Scott, 112 Cabot, Samuel, 246 Cage, John, 1 Cagniard de la Tour, Charles, 13, 135–43, 152n16, 153n28 Cambridge Anthropological Expedition to Torres Straits (CAETS) Borneo and, 79 dynamics of modern science and, 99–100 Galton whistle and, 78, 80, 83, 86, 87f–88f, 92, 99, 366 Haddon and, 77, 80–81, 102n67, 104n103 Hornbostel and, 86–88, 91, 96–98, 102n60, 102n64, 103n76, 103n101 impact of, 77, 90–91, 95–99 McDougall and, 77, 96 multidisciplinary approach and, 95–99 Murray Island and, 83–91, 94 music and, 77–91, 96–98, 101n26, 102nn64–65 Myers and, 62–91, 94–98, 100, 102n64, 102n68, 367 Ray and, 77 Rivers and, 77, 84–86, 96 Seligman and, 77 stimulus–response experiments and, 78, 82–84, 93, 95
survey of, 78–83, 91, 94, 98, 102n68 Wilkin and, 77 Cambridge Psychological Laboratory, 304 Campbell, George, 37 Camprubí, Lino, 11, 14–15, 64, 301–25, 354, 362–63, 367–68 cancer, 6, 329 Cannone, Il, 123 Carson, Rachel, 322n60, 359 Castle Bravo, 329, 335 catfish, 312 Catgut Acoustical Society (CAS), 113 cats, 276–80 Cattell, James McKeen, 81, 92 cellos, 112, 116, 127n45 Center for Research in Computer Music and Acoustics, 168 Central Institute for the Deaf, 41, 45n61, 363 “Cetus: Life After Life” (Lewandowski and Hennies), 359 Chanot, François, 118, 120–21, 127n43, 127nn45–46 Charcot, Jean-Martin, 57 Chélonomie, La (Sibire), 117, 121 Chen, Guang-Yue, 113 Chiang Ching-kuo, 197 Chiang Kai-shek, 197 Chicago Medical Society, 23 China acoustic attacks and, 292 An Investigation on the Nuisance of Noise Among Chinese and, 201 Cultural Revolution and, 123 KMT and, 189, 195, 197, 206nn3–4, 207n32, 365 noise and, 201 Taiwan and, 189–91, 197, 201, 206n3, 207n32, 208n34 Chladni, Ernst, 119, 136, 149 chords, 51, 53, 63, 95, 111, 135, 141, 142f, 144, 146, 147f, 150 cinemas, 167, 248, 252, 255, 262, 366 Clark, Chris, 317–18, 367 classical music, 291, 317–18 Cobbett, Walter Willson, 116 Coffin, Sean, 171, 177–78 Cogdell, Christina, 32 Cold War, 204, 208n34, 239n87, 301–2, 312, 317, 328, 330, 335, 354 Colladon, Jean-Daniel, 303 Cologne Institute of Musicology, 229
Index 381 Columbia Underwater Sound Laboratory, 307 Columbia University, 92 concert halls acoustics and, 5, 14, 115, 134, 216–19, 222, 231–32, 247–48, 255, 262, 265n52, 362, 366 etiquette of, 134 Kunstkopf and, 216–19, 222, 231–32, 362 materials and, 247–48, 255, 262, 265n52 reverberation and, 247 violins and, 115 consonants, 8, 28, 37–38, 40 Constant, Edward, 361 convolution, 170–71, 177 cookie-bite hearing loss, 43n9 Cornell University, 317, 359 Council of Physical Therapy, 38 Cousteau, Jacques, 301, 319 Cowell, Henry, 146–48 Cremonese instruments, 112, 116, 123–24 croakers, 305, 309, 312 Cross, Charles R., 51 cross-cultural studies Aberdeenshire and, 79, 86–88, 366 acoustics and, 82–83, 88, 91, 98 aesthetics and, 82, 93 Anleitungen and, 96–99 articulation and, 94, 96 auditory acuity and, 78, 82, 84–86, 92–94, 98 binaural sound and, 83 Borneo and, 79 Cambridge Anthropological Expedition to Torres Straits (CAETS) and, 13, 77–104, 366–67 deafness and, 84 dynamics of modern science and, 99–100 ethnology and, 77–78, 81, 90–91, 96–99, 102n60, 103n88, 103n92 evolution and, 79, 81–82, 95 frequency and, 80, 86, 88 Galton whistle and, 78, 80, 83, 86, 87f-88f, 92, 99 Haddon and, 77, 80–81, 102n67, 104n103 hearing loss and, 83 Hornbostel and, 86–88, 91, 96–98, 102n60, 102n64, 103n76, 103n101 loudness and, 80 McDougall and, 77, 96
military and, 92 Murray Island and, 83–91, 94 music and, 77–91, 96–98, 101n26, 102nn64–65 Myers and, 62–91, 94–100, 102n64, 102n68, 103n101, 104n103, 367 neuroscience and, 82 noise and, 83, 93 phonographs and, 89–90, 96, 98, 102n65, 102n68 physiology and, 79, 81–82, 97–98 pitch and, 89 psychology and, 77–100, 102n68, 103n88, 103n93, 103n95 racial inequality and, 78, 81–82, 367 Ray and, 77 Rivers and, 77, 84–86, 96 screening and, 77–78, 82, 92–93, 99 Seligman and, 77 standardization and, 90, 93, 99 statistics and, 78–81, 90, 367 stimulus–response experiments and, 78, 82–84, 93, 95 St. Louis World’s Fair and, 91–93 Stumpf and, 81–82, 86, 90–91, 94–98, 101n29, 102n67 surveys and, 78–83, 91, 94, 98, 102n68 telephones and, 93 Thurnwald and, 96–99, 103nn88–89, 103nn92–93 timbre and, 82, 95 tones and, 78, 80, 88, 93, 96, 98, 100 tuning forks and, 83, 86, 89 violins and, 83 whistles and, 78, 80, 83, 86, 87f-88f, 92, 99 Wilkin and, 77 Crystal Palace, 111 Cuba, 292 Culler, Elmer, 276, 278 Curtin, Joseph, 114 Damaske, Peter, 220–22, 224, 230, 233, 236n39, 236n46 Darwin, Charles, 61, 70n63 Davis, Hallowell, 41, 45n61, 276, 280 deafness auditory thresholds and, 287, 289, 293n5, 295n51 biological causes of, 194 Central Institute for the Deaf and, 41, 45n61, 363
382 Index deafness (cont.) cross-cultural studies and, 84 education and, 2, 27–31, 34–36, 40–43, 194 hard of hearing and, 27, 31, 34–35, 363 high-tone losses and, 43n9 Lexington School for the Deaf and, 30 music and, 295n51 sign language and, 28, 35 speech and, 27–31, 34–36, 40–43 Taipei School for the Blind and Deaf and, 194 Taiwan and, 194 underwater studies and, 305, 319 U.S. National Institute on Deafness and, 42 deaf schools, 28, 30, 40, 43 de Boer, Kornelis, 213–14, 234n4 decibels audiometry and, 19, 34, 189–93, 198–205, 206n5, 208n36, 250, 264n35, 307, 365 hearing loss for speech and, 34, 39 Norwegian Broadcasting Corporation and, 250, 264n35 pain threshold and, 131 Taiwan’s urban noise and, 189–93, 198–205, 206n5, 208n36, 365 underwater studies and, 307 Decot, 131 Deep Scattering Layer, 320n22 Demolition (binaural radio drama), 219–20 DeNora, Tia, 112 detectability, 25, 36–38 Dewey, Godfrey, 38, 40 dialect ear, 52 “Dialect Test” (Ellis), 52 Digital Audio Workstations (DAW), 168 digital models Abel and, 168, 171, 177–79, 181, 182n20 A/B tests and, 162, 171, 174 A/B/X tests and, 162, 174–75 acoustics and, 159, 168, 173, 177, 353 aesthetics and, 159, 162, 168, 170, 172, 180–81, 370 AKG BX20 and, 161, 168–73, 177–80 algorithms and, 165, 170–72, 174, 177 amplifiers and, 159, 161, 163, 167–77, 179, 183n29, 213 analog signal processing and, 13, 159–81 anthropology and, 182n9 black-boxing and, 167, 170, 179
Center for Research in Computer Music and Acoustics and, 168 Coffin and, 171, 177–78 commercial/academic context for, 182n8 concept of, 166–67 convolution and, 170–71, 177 Digital Audio Workstations (DAWs) and, 168 distortion and, 159, 173, 175 echoes and, 159–60 Echoplate III and, 169 equalizers and, 213, 230, 238n67 ethnology and, 161, 182n9 forms of, 163 frequency and, 165, 171 guitar amplifiers and, 172–76 historical perspective on, 160 increased use of, 159 microphones and, 168, 175 MIDI data and, 168 modulation and, 168–69, 173 Moog and, 165–66, 362 Mowitt and, 159, 163 music and, 159–63, 166–76, 180, 182nn7–8, 183n29, 362, 368 noise and, 159, 182n9 perfection and, 160 reverberation and, 161, 163, 167–73, 178, 180, 368 Roland RE-201 and, 160, 178–79 Roland SH-101 and, 163, 164f software and, 13, 160–64, 168, 170, 174–81, 182n20, 353, 362 sonic signature and, 13, 160, 169 stereo sound and, 169 studio environments and, 160 synthesizers and, 71n87, 163–66, 177–78, 180, 362 testing and, 162–67, 176–81 timbre and, 160, 170 tones and, 175 Universal Audio and, 161, 167–68, 170–72, 177–80, 182n20 violins and, 160 volume and, 169, 173, 175 Dilthey, Wilhelm, 54 distortion, 37, 40, 159, 173, 175, 238n67, 341 Dobrin, M. B., 314 Doegen, Wilhelm, 65 dogs, 276, 277–80, 285
Index 383 Doherty, Brigid, 331 Dolan, Emily I., 9, 109–30, 160, 353, 362, 369 Donders, Cornelius, 28 Doppler effect, 308 Dove, Heinrich Wilhelm, 131–32, 140 dummy-heads, 14, 234n1, 234n10, 236n38, 237n62, 238n78 ear canals, 203, 216, 225–26, 235n26 eardrums, 2, 8, 86, 216, 222, 226–27 earphones, 43 Ebbinghaus, Hermann, 54, 96 Echard, Jean-Philippe, 115–16 echoes anechoic chambers and, 221–22, 224, 227, 244, 262, 282, 290 auditory thresholds and, 282, 290 digital models and, 159–60 materials and, 244, 252, 262 spatial hearing and, 221–22, 224, 227 underwater studies and, 303–4, 319 echograms, 235n15 echolocation, 280, 319 Echoplate III, 169 Edison, Thomas, 23, 29, 51, 53, 63, 362 Egan, James P., 40–41 Egerer, Heinz, 256 Eidsheim, Nina Sun, 175, 183n29 Eimert, Herbert aesthetics and, 370 Anders and, 334–43, 345n23, 345nn26–27, 346n33, 354, 370 Epitaph für Aikichi Kuboyama and, 334–43, 345n25, 346n32, 354, 370 music and, 140, 150–51, 155n67 Opelt and, 150–51 sirens and, 342 speech and, 338–40 West German Radio and, 140, 335, 345n23, 370 Eitelberg, Abraham, 59 electron scattering, 243 Elektron, 170 elephant, 359 Ellis, Alexander J., 52–53, 67, 69n20 Emerson, Keith, 362 English Chamber Orchestra, 222 Epitaph für Aikichi Kuboyama (Eimert), 334–43, 345n25, 346n32, 354, 370 equalizers, 213, 230, 238n67
ethnology American Bureau of Ethnology and, 90 cross-cultural studies and, 77–78, 81, 90–91, 96–99, 102n60, 103n88, 103n92 digital models and, 161, 182n9 Institute of Psychology and, 54 multidisciplinary approach and, 95–99 music and, 54, 77, 79, 81, 90, 96–98, 101n26, 102n60, 102n64 Phonogramm-Archiv and, 69n35 eugenics, 32, 60, 64, 71n90, 79–80 evolution, 79, 81–82, 95, 109–10 Ewing, J. Alfred, 51 Falisse, Auguste, 116 Fan, He, 191 “Far Above Cayuga’s Waters” (Cornell alma mater), 359–60 Fechner, Gustav, 28, 59–60 feedback, 8, 13, 55, 115, 169, 171, 173–75, 183n29, 319 Feldmann, Harald, 17n21, 24–28 Fels, Janina, 239n89 Fétis, François-Joseph, 111, 120, 127n50, 140, 146, 149, 153n34 FilterFreak, 164–65 Fink, Gottfried Wilhelm, 140 fish auditory thresholds and, 293n20, 294n28 underwater studies and, 301, 305–6, 309–16, 318–21, 322n40, 322n47, 322n54 Fish, Marie, 311 Fiske, George, 23–24, 30, 35 Fleck, Ludwik, 352 Fletcher, Harvey, 34–35, 37, 234n10 food animal studies and, 275–77, 285, 309 auditory thresholds and, 275–77, 281, 285 sterilization of, 281 testing and, 4–5 underwater studies and, 309 Foucault, Michel, 6 Fourier, Joseph, 137–38, 226–27 Fowler, Edward Prince, 31–32 Free University Berlin, 24 frequency Abraham and, 50, 64 auditory acuity and, 8–9, 50
384 Index frequency (cont.) auditory thresholds and, 14, 273–92, 293nn4–5, 293n9, 294n25, 294n28, 295n51, 295n53, 296n57, 364 bandwidth and, 283–86, 290 cross-cultural studies and, 80, 86, 88 digital models and, 165, 171 Doppler effect and, 308 frequency and, 302–6, 309–19, 321n30 Helmholtz and, 8, 28 hertz unit for, 293n9 imaginary sensitivity and, 289–92 materials and, 247, 258 modulation and, 135 music and, 132, 135–40, 143–45, 148 Noise Control Act (NAC) and, 190 octaves and, 51, 63, 65, 71n82, 111, 117, 141–43, 147f, 305, 322n46 pitch and, 1, 8–9, 29, 50, 132, 135, 138–40, 144–45, 148, 236, 314 spatial hearing and, 216–17, 221–22, 225, 235n18, 236n39 speech and, 24–29, 32, 36–40, 45n45 Taiwan studies and, 190, 194 tones and, 1, 8, 24–25, 32, 64, 80, 86, 88, 138, 143, 148, 258, 276–78, 283, 287–88, 291, 296n57, 309, 314 ultrasound and, 14, 80, 274–75, 279–92, 293n4, 294n28, 295n46, 295n48, 295n54, 296n66 underwater studies and, 302–6, 309–19, 321n30 fricatives, 8, 29 Friedner, Michele, 293n5 Frings, Hubert, 281–84, 288–89, 294n38, 296n58 Frings, Mable, 282–84, 288–89 Fritz, Claudia, 114, 122, 126n19, 126n21, 128n60 Fryxell, Robert E., 113 Galambos, Robert, 280, 294n25 Galison, Peter, 352 Gallup International, 195–96 Galpin Society Journal, 109 Galton, Francis, 60, 80 Galton whistle, 15, 78, 80, 83, 86, 87f-88f, 92, 99, 281, 366 Garson, John George, 83–84 Gellé, Marie-Ernest, 29
gender, 78, 85f, 132, 175–76, 194, 201, 222, 232, 366 Genuit, Klaus, 228–29 German Acoustical Society, 227, 229 German Association of Hearing Aid Acousticians, 229 Gestalt theory, 54, 86, 95, 103n93, 231 Giedion, Sigfried, 244–45 Giese, Fritz, 65 Girard, François, 123 Glen Mills school, 35 Godzilla, King of the Monsters! (film), 335 Gold, Thomas, 197 golden ratio, 60 Goodman, Steve, 274, 293n6 Gottlob, Dieter, 222, 224, 233, 236nn45–46 Gould, James, 276, 279–80 gramophones, 30–31, 65, 258, 288, 363 Gray, Elisha, 303 Griffin, Donald, 280, 294n25 Griffiths, I. D., 201 Grove Music magazine, 111–12 Guarneri, Bartolomeo Giuseppe (del Gesù), 110, 113–14, 117, 120–24 guitars acoustic, 173 amplifiers for, 172–76 Chanot and, 118, 127n45 distortion and, 173, 175 electric, 160–63, 172–76, 180 Fender, 180 Stratocaster, 160 Gustavino, Raphael, 247 Gutzmann, Hermann, 29 Haddon, Alfred Cort, 77, 80–81, 102n67, 104n103 Hall, Brian, 208n34 Hammer, Karl, 213–14 hard of hearing, 27, 31, 34–35, 363 harmony, 13, 49, 67, 123, 146, 152n12 Harvard, 27, 40, 51, 243, 275–76, 279–80 Harvard sentences, 38–41, 46n74 Hayles, N. Katherine, 163 HEAD Measurement System I, 228 headphones, 16n2 spatial hearing and, 216, 219, 221, 226, 228, 230, 234n1, 234n10, 238n67, 238n69 speech and, 41
Index 385 testing, 1 (see also Kunstkopf) Health Education journal, 195 hearing aids audiometry and, 213 auditory thresholds and, 289 binaural, 14, 213–14, 223, 228–29, 231, 233 German Association of Hearing Aid Acousticians and, 229 Kunstkopf and, 213–14, 223, 228–29, 231, 233 spatial hearing and, 213–14, 223, 228–29, 231, 233 speech and, 41 Taiwan and, 194, 203 hearing loss auditory thresholds and, 289 cross-cultural studies and, 83 speech and, 24, 26–27, 31, 34–37, 40, 42, 43n9 Taiwan’s urban noise and, 193–95, 201, 203 underwater studies and, 307 Heidegger, Martin, 328 Heinrich Hertz Institute, 214–15, 248, 251 Helmholtz, Hermann von Abraham and, 51–52, 56 phonetics and, 28 sensation of tone and, 56 sirens and, 132–33, 137–38, 152n7 tuning forks and, 81 vibrating strings and, 113 vowel theory of, 51 Helmreich, Stefan, 293n5 Hennies, Sarah, 359 Hermann, Ludimar, 51 Heron-Allen, Edward, 111, 120 hertz, 293n9 Herzog, George, 55 Hewlett Packard loudness analyzer, 221 high-fidelity recordings, 220, 222, 231 Hippocrates, 7 Hiroshima, 343, 370 Hirsh, Ira, 25, 28 History of Audiology (Feldmann), 25 Hoffmann, E. T. A., 149 hogfish, 312 Holbein, Hans, the Younger, 60 Holtsmark, Johan Peter Acoustical Society of America and, 248 acoustic laboratory of, 245–52, 365–66 architectural acoustics and, 252–56
Heinrich Hertz Institute and, 251 laboratory testing and, 256–61 materials testing and, 243–62, 263n1, 264nn35–37, 265nn43–44, 266n56–58 NRK and, 35, 243, 246, 250–52, 254, 261, 264n34, 366 reputation of, 245–46 research papers of, 250 scholarship plans of, 251–52 Homefree, 286–87 Hope, Daniel, 124 Hornbostel, Erich Moritz von Abraham and, 55, 62, 68, 72n95 cross-cultural studies and, 86–88, 91, 96– 98, 102n60, 102n64, 103n76, 103n101 Hsieh, Jennifer, 10, 14, 189–212, 365 Huang Chyan-chyuan, 196, 198 Hui, Alexandra, 1–19, 64, 301–25, 354, 362–63, 367–68 Husserl, Edmund, 328 Hutchins, Carleen, 113 hydrogen bombs, 328 hydrophones, 301–6, 309–12, 316–19, 322n47, 354, 367–68 Ikeda, Ryoji, 1–2 Imaginary Landscape no. 1 (Cage), 1 Indians, 92–93 in-head localization (IHL), 218–19, 226, 232 Institute for Broadcasting Technology (IRT), 230–31 Institute for Sound and Heat Research, 247, 258 Institute of Applied Psychology, 96 Institute of Electrical and Electronics Engineers, 41 Institute of Electrical Communication Engineering, 214, 224, 229, 231, 237n55, 238n84 Institute of Technical Acoustics, 214–16, 239n89 Instruments of Science: An Historical Encyclopedia (Blume and Regeer), 26 intelligence tests, 5, 92, 99 intelligibility, 25, 27–28, 36–38, 40, 45n47, 255, 265n52 International Conference on Pitch, 52 International Radio Exhibition, 219–20 International Telecommunications Union (ITU), 228 Inventory of Musical Background, 308
386 Index Investigation on the Nuisance of Noise Among Chinese, An (report), 201 Itard, Jean Marc Gaspard, 28 Jackson, Samuel, 123 James, William, 28 Japan, 37 Godzilla and, 335 Hiroshima and, 343, 370 Ikeda and, 1–2 Lucky Dragon No. 5 and, 329, 335, 370 noise and, 190, 195, 200–202, 207n18, 208n36 jazz, 222 Jenkin, Fleeming, 51 Johannesen, Oddvar, 250 Johnson, Martin W., 305, 320n22 Jones, Norah, 169 Journal of Comparative Psychology, 273 Journal of Sound and Vibration, 201 Journal of the Acoustical Society of America, 250, 286 Journal of the American Musical Instrument Society, 109 Journal of the Violin Society of America, 112–13 “Judgment of Paris” test, 115 Julsrud, Erik, 250 Kant, Immanuel, 149 Karlin, John, 45n61 KEMAR (Knowles’ Electronics Manikin for Acoustic Research), 223 keyboards, 69n21, 109–11, 125n4, 150, 170 Keys to Play (Moseley), 125n4 Kittler, Friedrich, 145, 155n66 Klotz, Sebastian, 9, 13, 62, 77–106, 117, 202, 353, 366–67 Knoblauch, Oscar, 246–47 Knowles Electronics, 223 Koenig, Rudolf, 140–41 Köhler, Wolfgang, 51, 54 Kolinski, Mieczyslaw, 55 Krämer, Sybille, 145 Krebs, Stefan, 11, 14, 213–42, 354, 361–362, 366 Kresge Auditorium, 116 Kuboyama, Aikichi, 335–40, 343, 346n32, 354, 370 Kuhn, Thomas, 360 Külpe, Oswald, 96
Kunst, Jaap, 55 Kunstkopf as artificial head, 213 Blauert and, 216, 224–29, 233, 235n18, 237n62, 239n90 concert halls and, 216–19, 222, 231–32, 362 constructing average human listener and, 220–24 Damaske and, 220–22, 224, 230, 233, 236n39, 236n46 de Boer and, 213–14, 234n4 Demolition and, 219–20 different roles of, 215 Genuit and, 228–29 Gottlob and, 222, 224, 233, 236nn45–46 Hammer and, 213–14 hearing aids and, 213–14, 223, 228–29, 231, 233 increasing testing reliability and, 215–20 industrial agendas and, 215 KEMAR and, 223 Kürer and, 215–19, 231, 233, 235n23 Laws and, 225–29, 233, 237n59 localization and, 214, 217–21, 224–32 Mellert and, 221–23, 233, 236n46 Neumann microphones and, 216, 218–19, 224, 227–30, 235n19, 235n26, 236n33, 237n55, 237n62, 238n75, 238n78, 239n90 Originalkopf (original head) and, 227–29 Platte and, 225–29, 233, 237n66 Plenge and, 215–19, 222, 227, 231, 233, 235n14, 236n27 Schroeder and, 222, 224, 236n46 Siebrasse and, 222, 224, 233, 236n46 Snow and, 213–14, 234n1 spatial hearing and, 213–33, 234n4, 234n10, 236n32, 239n87–89, 354, 362, 366 standardization and, 228, 230, 238n83 van Urk and, 213–14 Wagener and, 220–22 Wilkens and, 215–19, 222, 227, 231, 233, 235n14, 235n21, 236nn27–28, 237n47 Kürer, Ralf, 215–19, 231, 233, 235n23 LA-2A amplifier, 167–68 Langdon, F. J., 201 Langevin, Paul, 303 laryngeal memory, 55, 59 Lashley, Karl, 275
Index 387 Laws, Peter, 225–29, 233, 237n59 Lefebvre, 120 Leibniz, Gottfried Wilhelm, 137, 144, 151, 154n49 Lewandowski, Annie, 359 Lewin, Kurt, 54 Lexington School for the Deaf, 30–31 Lichtwitz, Leopold, 29 Line 6, 161, 170, 173–77 linguistics, 53–54, 77, 337–38 Lipmann, Otto, 103n88 localization Abraham studies and, 55, 57 in-head (IHL), 218–19, 226, 232 noise and, 192 spatial hearing and, 14, 214, 217–18, 220–21, 224–25, 227–28, 230–31 London Science Museum, 80 Lorenz, Carl, 60, 70n45 loudness Abraham and, 50 auditory acuity and, 2, 8–9, 13, 50, 62, 78, 82, 84–86, 92–94, 98 cross-cultural studies and, 80 Hewlett Packard analyzer and, 221 noise and, 10, 14, 29, 42, 189, 191, 203, 221, 225–26, 230, 291, 313, 367 perceived, 14, 24 perfection and, 115 pitch and, 8–9, 28–29, 42, 50 snapping shrimp and, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 spatial hearing and, 217, 221, 224 speech and, 24, 28–29, 36–38, 42, 44n19 standardization and, 24 thresholds of, 9–10, 24, 42, 286 whispers and, 8, 23, 29, 59 loudspeakers Audiophone system and, 30 home computers and, 150 materials and, 248–50, 256, 258, 260 music and, 150, 362 spatial hearing and, 214, 219, 222, 225–27, 230–31, 234n1, 236n41, 236n46 speech and, 30 Taiwan’s urban noise and, 189 testing, 1, 362 underwater studies and, 308, 359 Lucky Dragon No. 5 (boat), 329, 335, 370 Lupot, Nicholas, 121 Luschan, Felix von, 96, 103n93
MacFarlan, Douglas, 24, 31, 34–36, 38–39, 44nn38–39, 45n45 machine signaling, 1 Madonna (Holbein the Younger), 60 Maier, Norman R. F., 273, 275, 293n8 Marshall amplifier, 174 Marshall Islands, 328–30, 344n6, 354, 361, 370 materials, 5 absorption and, 14, 246–47, 255, 262 acoustics and, 243–63, 264n22, 264n25, 264n35, 264n37, 265n43, 265n48, 265n51, 266n56, 266n70, 353, 365 aesthetics and, 247, 254 amplifiers and, 244, 258, 260 anechoic chambers and, 244, 262 architecture and, 252–56 concert halls and, 247–48, 255, 262, 265n52 echoes and, 244, 252, 262 electron scattering and, 243 frequency and, 247, 258 Giedion and, 244–45 Holtsmark and, 60, 65, 243–62, 263n1, 264nn35–37, 265nn43–44, 266n56–58 laboratory testing and, 256–61 loudspeakers and, 248–50, 256, 258, 260 lowered quality of, 262–63 microphones and, 249–50, 254, 256, 258, 260 music and, 243–46, 249, 255, 265n52 noise and, 245–47, 250, 254, 260, 263 Norges Tekniske Høgskole (NTH) and, 14, 243, 245, 248–62, 264n26, 264n35, 265n44, 266n57, 266n60, 266n65 NRK and, 35, 243, 246, 250–52, 254, 261, 264n34, 366 quantum physics and, 243, 248 Quilt, 246 radio and, 243–51, 261–62 reverberation and, 244, 247, 250, 252–55, 258, 260, 262, 264n37 standardization and, 243–49, 255–58, 261–63 Stark effect and, 243 timbre and, 252, 255 transmission and, 14, 247, 255, 258, 262 ultrasound and, 295n46 violins and, 113 World War I era and, 243, 245, 249, 264n25 Mathew, Nicholas, 112
388 Index Maupertuis, Pierre Louis, 119 May, Ernst, 245 McDougall, William, 77, 83, 96 McKellar, Don, 123 McQuay, Bill, 359 Measurement of Musical Talent (Seashore), 71n87, 307 Medwin, Herman, 116 Meintjes, Louise, 167 Mellert, Volker, 221–23, 233, 236n46 Mercedes Benz, 228 “Methods of Measuring Children’s Hearing” (Fletcher), 34 metronomes, 9, 89 Meyer, Erwin, 222, 250–51, 265n40 Meyer-Eppler, Werner, 338 Michaelis, Christian Friedrich, 149 microphones Brüel & Kjaer, 223, 226, 230 digital models and, 168, 175 gold standard for testing, 41 Kunstkopf and, 213 (see also Kunstkopf) materials and, 249–50, 254, 256, 258, 260 Neumann, 216, 218–19, 224, 227–30, 235n19, 235n26, 236n33, 237n55, 237n62, 238n75, 238n78, 239n90 Opelt and, 362 PB sentences and, 41 Schoeps, 223 Sennheiser, 220 spatial hearing and, 213–32, 234n1, 235n26, 237n55, 238n67 speech and, 41, 43 Telefunken, 221 testing, 362 underwater studies and, 301–6, 309–12, 316–19, 322n47, 354, 367–68 MIDI (Musical Instrument Digital Interface), 168 military Anders and, 328, 330 auditory thresholds and, 14, 275, 280–82, 289, 292, 295n46, 295n53, 296n61 combat noise and, 27 cross-cultural studies and, 92 field communications and, 40 hearing loss and, 31 hydrogen bombs and, 328
nuclear energy and, 328, 330 Operation Castle and, 328 PAL and, 27, 45n61 Seashore and, 64, 72n92 sound location and, 245 Taiwan and, 190, 197, 208n34 testing and, 2, 9, 11, 27, 34, 42, 45n61, 64, 72n92, 370 underwater studies and, 301–2, 305–6, 310–13, 316–19, 322n40, 322n54, 367, 370 Mills, Mara, 1–19, 23–49, 82, 162, 285, 353, 363 MIMO (Musical Instrument Museums Online), 127n45 MIT, 116, 246 Moby, 148 modulation Anders and, 331–32 cross, 314 digital models and, 168–69, 173 frequency, 135 tones and, 80 underwater studies and, 314 Moe, Lowell, 284, 294n40, 295n46 Möller, Fredrik, 250 monochords, 135, 150 Moog, 165–66, 362 Morgan, Clifford, 274–76, 279–80, 282, 293n8 Morgan, Jane, 274–75 Morrel-Samuels, Palmer, 114 Moseley, Roger, 125n4 motion pictures, 243, 246–48, 252, 261 Motown, 160 Mowbray, William, 311, 314 Mowitt, John, 159, 163 Mozart, Leopold, 116–17 Mozart, Wolfgang Amadeus, 222 Münch, Richard, 336 Murray Island, 83–91, 94 Muscle shoals, 160, 169, 368 music Abraham and, 49–56, 59–68, 69n21, 70n45, 352 acoustics and, 1, 10–15, 35, 52, 55, 61–63, 82–83, 88, 98, 111, 133–34, 138, 140, 145, 149–51, 159, 168, 214, 219, 222–23, 226, 229–33, 243–46, 249, 255, 317–18, 338, 340, 367, 369
Index 389 aesthetics and, 9, 11, 13–15, 54, 82, 110, 123, 125, 134, 149, 159, 162, 168, 170, 172, 180–81, 330, 353, 368–69 Anders and, 330, 334–41, 345n22, 346n33 animals and, 49, 61, 363, 367 auditory thresholds and, 291, 292n2, 295n51 Bartók, 55 Beethoven and, 49, 61, 112, 149, 189 Brahms and, 317 brass instruments, 111 Cage and, 1 Cagniard and, 13, 135–43, 152n16, 153n28 cello, 112, 116, 127n45 Center for Research in Computer Music and Acoustics and, 168 chords and, 51, 53, 63, 95, 111, 135, 141, 142f, 144, 146, 147f, 150 classical, 291, 317–18 concert halls and, 5, 14, 115, 134, 216–19, 222, 231–32, 247–48, 255, 262, 265n52, 362, 366 consciousness of sensation and, 55–56 cross-cultural studies and, 77–91, 96–98, 101n26, 102nn64–65 culture and, 2, 4, 9–12, 50–53, 56, 59–63, 66–68, 77, 81, 88–90, 97–98, 110, 123, 134, 175–76, 182, 220, 230, 233, 291, 367 deafness and, 295n51 digital models and, 159–63, 166–76, 180, 182nn7–8, 183n29, 362, 368 Eimert and, 140, 150, 155n67 electronic, 1–2, 10–15, 25, 35, 64–65, 131, 148, 150, 159, 173, 183n29, 214, 229, 231, 243, 246, 249, 335, 337–39, 362, 370 ethnology and, 54, 77, 79, 81, 90, 96–98, 101n26, 102n60, 102n64 experimental, 1–2, 11, 359 frequency and, 132, 135–40, 143–45, 148 guitars, 118, 127n45, 160–63, 172–76, 180 harmony and, 13, 49, 67, 123, 146, 152n12 hearing technologies and, 131–51, 152n10, 154n47, 154n59 high-fidelity, 220, 222, 231 Ikeda and, 1–2 instruments and, 1–2, 4, 9–13, 15, 35, 51– 52, 62–69, 98, 109–17, 120–25, 127n45, 132, 135, 143, 145–46, 150, 154n47, 159–60, 168, 170, 183n29, 249, 255, 265n52, 353–54, 362–64, 368–69
intervals and, 49, 52, 62–67, 89, 98, 138–44, 321, 364 jazz, 222 keyboards, 69n21, 109–11, 125n4, 150, 170 loudspeakers and, 150, 248–50, 256, 258, 260, 362 materials and, 243–46, 249, 255, 265n52 metronomes and, 9, 89 MIDI and, 168 MIMO and, 127n45 Mozart and, 222 Myers and, 62, 77, 83–84, 86, 88–90, 96, 98, 102n61, 367 noise and, 131–35, 143, 145, 153n41, 154n45, 353 octaves, 51, 63, 65, 71n82, 111, 117, 141–43, 147f, 305, 322n46 Old Italians and, 112–17, 120–21, 127n43, 127n50, 353 Opelt and, 13, 139–51, 152n1, 155n68 perfection and, 109–17, 120–25, 126n10, 127n45, 369 physiology and, 132, 138, 140, 144 pitch and, 1, 13, 49–56, 59, 61, 63, 66–70, 89, 132–41, 144–48, 155n68, 307, 314, 352, 364–65 Preyer and, 51–53, 67 prized older equipment and, 182n7 probing and, 133, 150, 353 psychology and, 144 radio and, 14, 65, 140, 236n33, 365 recorded, 1, 14–15, 23, 35, 51, 62, 65, 69n35, 89–90, 98, 102n60, 140, 148, 159–62, 168–73, 180, 216, 219–23, 229–33, 236, 243, 308, 314, 317, 335, 337, 346n33, 364–65 rhythm, 13, 49, 53, 62–67, 90, 138–48, 154n52, 155n68, 317, 364 Riemann and, 53, 63, 67 rock, 362 Savart and, 113, 118–22, 127n43, 127n50, 138–39, 141, 152n16, 153n41 Schünemann and, 64–66 Seashore and, 15, 50, 63–64, 68, 307, 367 singing, 8, 49–50, 53, 61, 63, 65, 89–90, 102n61, 318 sirens and, 131–34, 138–51 sonic signature and, 13, 160 soundwaves and, 113, 136, 143
390 Index music (cont.) spatial hearing and, 213–16, 219–23, 226, 229–33, 236, 354 studio environments and, 160 Stumpf and, 50–51, 54–56, 59, 61–63, 67–68, 81–82, 86, 90–91, 96–98, 102n64, 365 surveys and, 149 synthesizers and, 71n87, 163–66, 177–78, 180, 362 Taiwan and, 189 talent and, 50, 52, 59–65, 68, 71n87, 98, 307, 365, 367 timbre and, 64, 137, 148 tone and, 1–2, 9–12, 24–25, 35, 43n6, 49–56, 59–68, 70n45, 82, 84, 88–89, 96, 98, 111, 114, 117, 132–33, 136, 138, 143, 147–48, 230, 291, 307, 337, 364–65, 367 tuning forks and, 9, 18, 24, 63, 83, 89, 156 underwater studies and, 306–8, 314, 317–18, 367 violas, 127n45 violins, 137, 146 (see also violins) vocal, 51, 134, 265n52, 314, 339 volume and, 131, 135 Wagner and, 133–34, 152n11 whistles and, 9, 63 wind instruments, 109–11 Myers, Charles Aberdeenshire survey of, 79, 86–88, 366 background of, 79 Cambridge Anthropological Expedition to Torres Straits (CAETS) and, 62–91, 94–98, 100, 102n64, 102n68, 367 cross-cultural studies and, 77–91, 94–100, 102n64, 102n68, 103n101, 104n103 Galton whistle and, 80, 83, 86 Hornbostel and, 86–88, 91, 96–98, 102n60, 102n64, 103n76, 103n101 Institute of Applied Psychology and, 96 Murray Island and, 83–91, 94 music and, 62, 77, 83–84, 86, 88–90, 96, 98, 102n61, 367 psychology and, 62, 77–91, 94–100, 102n64, 102n68, 103n101, 104n103, 367 Runne’s clock and, 83–84, 85f Text-Book on Experimental Psychology and, 91 Nagyvary, Joseph, 113 Narragansett Marine Laboratory, 312
National Defense Research Committee, 40 National Taiwan Normal University, 194, 198 Native Instruments, 162, 170 Natural Sound Library, 359 Naval Ordnance Laboratory (NOL), 306, 311, 313 Nazis, 66, 335 Nebula, 177 Neff, William D., 307–9 Neue Zeitschrift für Musik, 131 Neumann, 216, 218–19, 224, 227–30, 235n19, 235n26, 236n33, 237n55, 237n62, 238n75, 238n78, 239n90 neuroscience Abraham and, 55, 57, 67 auditory thresholds and, 273, 280, 282, 295n51 Charcot and, 57 cross-cultural studies and, 82 spatial hearing and, 231, 239n93 testing and, 304 underwater studies and, 307 New England Conservatory of Music, 246 New York League for the Hard of Hearing, 31 Nicholson, Alexander, 32 Nixon, Richard, 197 noise Anders and, 340 audiometry and, 10, 27, 35, 42, 191, 193– 96, 203–4, 277 auditory thresholds and, 277, 280, 282, 286, 288, 290–91, 292n3, 295n56, 354 cross-cultural studies and, 83, 93 digital models and, 159, 182n9 hearing loss and, 193–95, 201, 203 Japan and, 190, 195, 200–202, 207n18, 208n36 localization and, 192 loudness and, 10, 14, 29, 42, 189, 191, 203, 221, 225–26, 230, 291, 313, 367 materials and, 245–47, 250, 254, 260, 263 music and, 131–35, 143, 145, 153n41, 154n45, 353 perfection and, 120 Preliminary Noise Control Project and, 198, 200 schoolchildren’s hearing and, 193–95, 201, 203–4 spatial hearing and, 214, 216, 221–26, 230, 233 speech and, 27, 29, 35, 37, 40, 42–43
Index 391 Taiwan and, 14, 189–205, 206n3, 206n10, 207n18, 207nn22–23, 207n26, 208n46, 365 underwater studies and, 301, 304–6, 309, 311–13, 316, 318–20, 321n24, 367 white, 1, 226, 288, 340 Noise Control Act (NCA), 190, 192–93, 196–97 Noise Policing Act, 190, 196–97 Norges Tekniske Høgskole (NTH): 14, 243, 245, 248–62, 264n26, 264n35, 265n44, 266n57, 266n60, 266n65 Norsk radio journal, 249 Norton, Charles, 246–47 Norwegian Broadcasting Corporation (NRK), 35, 243, 246, 250–54, 261, 264n34, 366 Norwegian Institute of Technology, 256 notches, 25, 43n9 Notes and Queries on Anthropology, 83, 91, 97 NPR, 359 nuclear energy Anders and, 15, 329–35, 340, 343, 354 Castle Bravo and, 329, 335 colonialism and, 344n6 Eimert and, 346n33 Hiroshima and, 343, 370 hydrogen bombs and, 328 Lucky Dragon No. 5 and, 329, 335, 370 Marshall Islands and, 328–30, 344n6, 354, 361, 370 Operation Castle and, 328 radioactive fallout and, 329, 335 underwater studies and, 15, 312, 317 nuclear physics, 246, 251, 262, 266n60 Oceans, The (Johnson), 305 octaves, 51, 63, 65, 71n82, 111, 117, 141–43, 147f, 305, 322n46 Ohm, Georg Simon, 137–38, 143–44 Old Italians, 112–17, 120–21, 127n43, 127n50, 353 “On the Sensitiveness of the Ear to Pitch and Change of Pitch in Music” (Ellis), 52 Opelt, Friedrich Wilhelm Allegemeine Theorie der Musik and, 139, 142f, 148 background of, 139 Eimert and, 150–51 microphones and, 362 music and, 13, 139–51, 152n1, 155n68
notation of, 139, 142, 144–45 pitch and, 13, 139–41, 144–46, 155n68, 364 probing and, 150, 353 rhythm and, 13, 139–47, 154n52, 155n68, 364 sirens and, 13, 139–51, 152n1, 153n31, 153n36, 153n40, 154n48, 154n52, 154n59, 155n68, 353, 362, 364 Über die Natur der Musik and, 139, 143, 154n47, 154n57 Operation Castle, 328 ophthalmology, 23 oscillators, 64, 279, 293n17, 362 Oslo Municipal Sound Insulation Committee, 258, 260 otolaryngology, 31, 35, 193, 195, 208n36 Oudshoorn, Nelly, 361 overtones, 8–9, 51, 80, 133, 175 Paganini, Niccolò, 123 pain threshold, 131 Pantalony, David, 141 parrots, 49, 61 patents, 118, 177, 239n91, 284, 292, 294n40, 295n46, 295n53, 296n61 Payne, Katy, 301, 317–18, 322n60, 359, 367 Payne, Roger, 301, 317, 322n60, 359 PB sentences, 40–41 Pearson, Karl, 32 Penn State University, 281, 284, 288, 296n61 perception testing, 306–10 perfection acoustics and, 111–16, 119, 123–24, 369 aesthetics and, 110, 123–25, 353 articulation and, 115 digital models and, 160 equalizers and, 213, 230, 238n67 Guarneri and, 110, 114, 122 Heron-Allen on, 111, 120 Kunstkopf and, 223 (see also Kunstkopf) loudness and, 115 music and, 109–17, 120–25, 126n10, 127n45, 369 noise and, 120 Old Italians and, 112–17, 120–21, 127n43, 127n50, 353 statistics and, 114 Stradivari and, 110–17, 120–24, 126n10, 160, 362, 369 surveys and, 110, 120 timbre and, 112, 120
392 Index perfection (cont.) tones and, 111–12, 114–15, 117, 121, 123 violins and, 13, 109–24, 125nn4–5, 126n10, 126n21, 127n41, 127n45, 353, 362, 369 perfect pitch, 9, 49–52, 56, 67 Pest Control journal, 282 Petiscus, Johann Conrad, 117 Philips Research Laboratory, Eindhoven, 213–14 Philosophical Magazine, 37 phonemes, 8, 29, 35, 38, 336 phonetics Abraham and, 52, 54, 65 balance word lists and, 38–41 Bell Telephone and, 363 Eimert and, 338 speech and, 27–28, 38–41, 52, 54, 65, 338, 363 phono-audiometers, 26–27, 30–31, 34–36 Phonogramm-Archiv, 62, 64, 69n35, 100n6, 102n60 “Phonograph in Testing Hearing, The” (Fiske), 23 phonographs Abraham and, 51, 53, 62, 68n11 Bryant and, 29–30 cross-cultural studies and, 89–90, 96, 98, 102n65, 102n68 Edison and, 23, 29, 51, 53, 63, 362 Fiske and, 23–24, 30 gramophones and, 30–31, 65, 258, 288, 363 speech and, 23–31, 36, 43n5, 353 Standard Phonograph Company and, 29 tone tests and, 362–63 WE 4A audiometer and, 26–27, 35 physiology Abraham and, 50–51, 54, 56, 66–67 auditory thresholds and, 273, 276, 281, 290 cross-cultural studies and, 79, 81–82, 97–98 music and, 132, 138, 140, 144 spatial hearing and, 231, 237n59 speech and, 28, 30 testing and, 4, 8–10, 15 underwater studies and, 304, 307, 309, 319 pianos, 25, 51, 111, 150, 359, 364 Pierce, George W., 279–80 Pinch, Trevor, 172, 359–73
pinnae, 216, 221–24, 227–28, 230, 232, 235n26, 237n50, 239n90 pitch Abraham and, 49–56, 59–68, 69nn20–21, 70n45, 70n52, 352, 365 absolute, 49–56, 59, 61, 63, 66–67, 352, 365 auditory thresholds and, 8–9 cross-cultural studies and, 89 Doppler effect and, 308 frequency and, 1, 8–9, 29, 50, 132, 135, 138–40, 144–45, 148, 236, 314 International Conference on Pitch and, 52 loudness and, 8–9, 28–29, 42, 50 music and, 1, 13, 49–56, 59, 61, 63, 66–70, 89, 132–41, 144–48, 155n68, 307, 314, 352, 364–65 octaves, 51, 63, 65, 71n82, 111, 117, 141–43, 147f, 305, 322n46 Opelt and, 13, 139–41, 144–46, 155n68, 364 perfect, 9, 49–52, 56, 67 phonemes and, 8, 29, 35 relative, 49, 53, 66–68 spatial hearing and, 236n39 speech and, 28–29, 35, 42 standardization and, 52, 308 tone differential devices and, 52 underwater studies and, 307–8, 314, 317 volume and, 5, 29, 135, 307 Planning for Good Acoustics (Bagenal and Wood), 262–63 Platte, Hans-Joachim, 225–29, 233, 237n66 Plenge, Georg, 215–19, 222, 227, 231, 233, 235n14, 236n27 Poggendorffs Annalen, 137 Pohl, Richard, 133–34, 138–39, 149, 151, 152n12, 154n49 Pohlman, Augustus, 36 poiesis, 341, 369 Poitevineau, Jacques, 114 Politzer, Adam, 17n21, 83 pollution, 189, 196–97, 207n26 Preliminary Noise Control Project, 198, 200 prevention, 6, 34, 192, 201 Preyer, William Thierry, 51–53, 67 Princeton Applied Research, 221 Principles of Physiological Psychology (Wundt), 81 probing, 133, 150, 302, 352–55 pronunciation, 52, 69n20 Prony Brake, 360
Index 393 ProTools, 180 Prussian State Library, 65–66 Psycho-Acoustic Laboratory (PAL), 27, 40–41, 45n61, 45n63 psychoacoustics, 82, 88, 91, 225, 229, 232–33, 284–88 psychology Abraham and, 50, 53–68, 70n45, 365 Anders and, 327–28, 333, 341, 344n2, 344n14 Anleitungen and, 96–99 applied, 4, 50, 96, 98 auditory thresholds and, 273–80, 284, 287, 290 Berlin Institute of Psychology and, 50, 54–56, 63–64, 67–68, 69n35, 82, 365 Cambridge Psychological Laboratory and, 304 consciousness of sensation and, 55–56 cross-cultural studies and, 77–100, 102n68, 103n88, 103n93, 103n95 Gestalt, 54, 86, 95, 103n93, 231 Jena, 51 Journal of Comparative Psychology and, 273 multidisciplinary, 95–99 music and, 144 Myers and, 62, 77–91, 94–100, 102n64, 102n68, 103n101, 104n103, 367 Preyer and, 51–53, 67 speech and, 28, 40, 42 Stumpf and, 50–51, 54–56, 59, 61–63, 67–68, 70n45, 81–82, 86, 90–91, 94–98, 101n29, 102n64, 102n67, 365 Taiwan study and, 207n23 testing and, 4–6, 9, 12, 14–15 tones and, 12, 42, 50, 53, 55–56, 60–68, 70n45, 78, 80, 82, 88, 96, 98, 100, 276–77, 287, 307, 365, 367 two-culture divide and, 54 underwater studies and, 304, 307 Wundt and, 70n45 psychophysics, 1, 4, 9, 28, 65, 81–82, 89 psychotechnics, 64–65, 327–28, 330–31, 344n4, 344n14 Public Opinion Poll Quarterly, 197 Pure Data, 166 Pythagoreans, 140–41, 143, 151, 153n40 quantum physics, 243, 248 Quartet Romantic (Cowell), 146–47 Quilt, 246
racial inequality, 78, 81, 82, 367 radio, 5 Abraham and, 65 acoustics and, 14, 25, 35, 243–51, 261–62 auditory thresholds and, 286, 291, 293n13, 295n46 Benjamin and, 331, 344n14 Demolition and, 219–20 Holstman and, 366 International Radio Exhibition and, 219 Kuboyama and, 335 Kunstkopf signals and, 219–20 materials and, 243–51, 261–62 music and, 14, 140, 236n33, 365 Nazis and, 335 pure tones and, 25 spatial hearing and, 219–20, 228, 236n33, 236n36 speech and, 25, 35 underwater studies and, 305, 317 West German Radio and, 335, 345n23, 370 radiology, 72n106 rats, 201, 273–76, 279–80, 290, 295n56 Rat-X, 286 Ray, Sidney, 77 “Recommended Practice for Speech Quality Measurement” (IEEE), 41 Red Cross, 31 Red Mendelssohn, 123 Red Violin, The (film), 123 “Reference File of Biological Underwater Sounds” (ONR), 306, 310–17, 362 reference zero, 25, 32 Regeer, Barbara, 26 Regener, Erich, 258 Rehding, Alexander, 9, 13, 131–57, 353, 362, 364 Reiher, Hermann, 247, 258 Relativ Frequency of English Speech Sounds (Dewey), 38 relative pitch, 49, 53, 66–68 Repertorium der Physik, 132 reverberation AKG BX20 and, 161, 168–73, 177–80 concert halls and, 247 digital models and, 161, 163, 167–73, 178, 180, 368 Holtsmark and, 250, 252, 255, 260, 264n37 materials and, 244, 247, 250, 252–55, 258, 260, 262, 264n37 spatial hearing and, 217, 219, 222
394 Index reverberation (cont.) speech and, 29 tape recording and, 222 underwater studies and, 370 reverse slope, 43n9 Rheinberger, Hans-Jörg, 3, 351–57 Rhineland Westphalian Technical University (RWTH), Aachen, 214–15, 224–30, 233, 237n55, 239n89 rhythm music and, 13, 49, 53, 62–67, 90, 138–48, 154n52, 155n68, 317, 364 Opelt and, 13, 139–47, 154n52, 155n68, 364 record for fastest, 148 underwater studies and, 317 Riemann, Hugo, 53, 63, 67 Rinne, Heinrich Adolf, 8 Rion SF-5 noise generator, 201 Rivers, William Halse, 77, 84–86, 96 Robel, Ernst, 140 Rochester University, 273, 275–76 rock music, 362 Rodgers, Tara, 180 Roland RE-201, 160, 178–79 Roland SH-101, 163, 164f Royal Norwegian Society of Science and Letters, 250 Ruben, Robert, 42 Ruhr University Bochum, 225 Rumbaugh, L. H., 322n54 Runne’s clock, 83–84, 85f Ryle, Marcus, 174 Sabine, Wallace, 247 Santa Rita Technology, 289 Saunders, Frederick, 113 Savart, Félix, music and, 113, 118–22, 127n43, 127n50, 138–39, 141, 152n16, 153n41 violins and, 113, 118–22, 127n43, 127n50 wheel of, 138–39, 141, 152n16, 153n41 Schaefer, Karl Ludolf, 17n21, 140 Scheibler, Johann Heinrich, 69n21 Scheler, Max, 72n95, 328 Schelleng, Joseph C., 113 Scherer, Paul, 224 Schilling, Harold, 281 Schoeps, 223 Schroeder, Manfred, 222, 224, 236n46 Schubert, Johann Friedrich, 117
Schumann, Friedrich, 96 Schünemann, Georg, 64–66 Schwabach, Dagobert, 29 science and technology studies (STS), 6–7, 11, 360 screening, 7, 12 Bryant and, 29–30 cancer, 6 classroom hearing tests and, 11 cross-cultural studies and, 77–78, 82, 92–93, 99 malingerer’s valve and, 30 mass, 42, 77–78, 82, 353–55 newborn, 5 quality and, 32–33 Shewhart and, 32–34, 37 speech and, 26–38, 42–43, 44n31, 353, 363 Scripps Institution of Oceanography, 305 sea robin, 312 Seashore, Carl Emil Abraham and, 50, 63–64, 68 anthropology and, 15, 50, 63–64, 68, 72n92 Measurement of Musical Talent and, 71n87, 307 “Measures of Musical Talent” and, 63 military and, 64, 72n92 music and, 15, 50, 63–64, 68, 307, 367 Secrets of the Violin (film), 124 Seebeck, August, 132, 137–38, 143 Seligman, Charles G., 77 semitones, 49, 111, 321n30 Sennheiser, 220 Sensing Sound (Eidsheim), 175 serviceable hearing, 12, 38–40 Shanks, Will, 172 Shannon, Claude, 32 Shedd Aquarium, 311 shell shock, 100n9 Shewhart, Walter, 32–34, 37 Shiffrin, R. M., 219 shrimp, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 Sibire, Antoine, 117, 121 Siebrasse, Karl Friedrich, 222, 224, 233, 236n46 sign language, 28, 35 Silent Spring (Carson), 359 Silent World, The (Cousteau), 301 sine tones, 337 singing, 8, 49–50, 53, 61, 63, 65, 89–90, 102n61, 318
Index 395 Singing Fool, The (film), 248 sirens, 15 acoustics and, 8, 131–40, 145, 147, 149–51, 152n12, 152n16, 154n49 air-raid, 296n57 auditory thresholds and, 132–33, 280–84, 296n57 Cagniard and, 13, 135–43, 152n16, 153n28 challenge of, 135–39 Cowell and, 146–48 Dove and, 131–32, 140 Eimert and, 342 electric, 131 emblematic status of, 131 Greek mythology and, 13 Helmholtz and, 132–33, 137–38, 152n7 mechanism of, 137–39, 145–46 music and, 131–34 Ohm and, 137–38, 143–44 Opelt and, 13, 139–51, 152n1, 153n31, 153n36, 153n40, 154n48, 154n52, 154n59, 155n68, 353, 362, 364 pain threshold and, 131 Pohl and, 133–34, 138–39, 149, 151, 152n12, 154n49 Seebeck and, 132, 137–38, 143 ultrasonic, 281–84 Smith, Emily M., 304 Snellen, Hermann, 23 Snow, W. B., 213–14, 234n1 software Ableton Live and, 169 Digital Audio Workstations (DAWs) and, 168 digital models and, 13, 160–64, 168, 170, 174–81, 182n20, 353, 362 Nebula, 177 ProTools, 180 simulation and, 11 (see also digital models) sonar, 301–10, 313–14, 317–19, 362, 367 Songs of the Humpback Whale (Payne and Payne), 301, 322n60 Sonic Technology, 286 sonification, 302–3, 306, 310, 313, 316–17 sonograms (spectrograms), 302, 311–18, 322n54, 359, 367, 370 sounding rods, 8 Soundscape of Modernity, The (Thompson), 26
Sounds of Western North Atlantic Fishes (Fish and Mowbray), 311 Sound Surveillance System (SOSUS), 317–18 SoundToys, 164 soundwaves, 113, 136, 143 Soviet Union, 15, 312–13, 317, 331, 361 Space Echo, 160, 178–79 spatial hearing acoustics and, 214–19, 222–33 anechoic chambers and, 221–22, 224, 227 binaural sound and, 213–23, 227–33, 234n1, 234n4, 234n7, 234n10, 236n36, 236n39, 238n81, 239n91, 366 Blauert and, 216, 224–29, 233, 235n18, 237n62, 239n90 concert halls and, 5, 14, 115, 134, 216–19, 222, 231–32, 247–48, 255, 262, 265n52, 362, 366 constructing average human listener and, 220–24 Damaske and, 220–22, 224, 230, 233, 236n39, 236n46 de Boer and, 213–14, 234n4 Demolition and, 219–20 dummy-heads and, 14, 234n1, 234n10, 236n38, 237n62, 238n78 Fourier transforms and, 226–27 frequency and, 216–17, 221–22, 225, 235n18, 236n39 Genuit and, 228–29 Hammer and, 213–14 headphones and, 216, 219, 221, 226, 228, 230, 234n10, 236n41, 238n67, 238n69 hearing aids and, 213–14, 223, 228–29, 231, 233 increasing testing reliability and, 215–20 KEMAR and, 223 Kunstkopf and, 213–33, 234n4, 234n10, 236n32, 239n87–89, 354, 362, 366 Kürer and, 215–19, 231, 233, 235n23 Laws and, 225–29, 233, 237n59 localization and, 14, 214, 217–21, 224–32 loudness and, 217, 221, 224 loudspeakers and, 214, 219, 222, 225–27, 230–31, 234n1, 236n41, 236n46 microphones and, 213–32, 234n1, 235n26, 237n55, 238n67 music and, 213–16, 219–23, 226, 229–33, 236, 354
396 Index spatial hearing (cont.) Neumann and, 216, 218–19, 224, 227–30, 235n19, 235n26, 236n33, 237n55, 237n62, 238n75, 238n78, 239n90 neuroscience and, 231, 239n93 noise and, 214, 216, 221–26, 230, 233 physiology and, 231, 237n59 pitch and, 236n39 Platte and, 225–29, 233, 237n66 Plenge and, 215–19, 222, 227, 231, 233, 235n14, 236n27 poiesis and, 341, 369 radio and, 219–20, 228, 236n33, 236n36 reverbation and, 217, 219, 222 Snow and, 213–14, 234n1 speech and, 214, 226–27, 230 standardization and, 228, 230, 238n83 statistics and, 237n59 stereo sound and, 213–14, 216, 223–24, 229, 231–32, 234n1, 234n4, 234n10, 236n41, 236n46, 237n55, 238n74 surveys and, 225, 237n50 telephones and, 213, 228 timbre and, 225 tones and, 215, 218, 230–31, 234 van Urk and, 213–14 volume and, 222 Wagener and, 220–22 Wilkens and, 215–19, 222, 227, 231, 233, 235n14, 235n21, 236nn27–28, 237n47 speech absolute tone consciousness and, 50–52, 59, 68n11, 69n20 acoustics and, 25, 27–28, 35, 38, 40–41 aesthetics and, 32 articulation and, 27, 36–39, 43, 45n61, 46n70, 337 AT&T and, 12, 25, 26f, 32–35, 37 audiology and, 25, 28 audiometry and, 22–24, 43n9, 44n37, 44n39, 45n42, 45n45, 353 auditory thresholds and, 42, 276, 285, 289 Bryant and, 24, 29–30, 31f consonants and, 8, 28, 37–38, 40 deafness and, 27–31, 34–36, 40–43 detectability and, 25, 36–38 Dialect Test and, 52 Eimert and, 338–40 Feldmann and, 24–28 Fiske and, 23–24, 30, 35 frequency and, 24–29, 32, 36–40, 45n45
fricatives and, 8, 29 Harvard sentences and, 38–41, 46n74 hearing aids and, 41 hearing loss and, 24, 26–27, 31, 34–37, 40, 42, 43n9 intelligibility and, 25, 27–28, 36–38, 40, 45n47, 255, 265n52 linguistics and, 53–54, 77, 337–38 loudness and, 24, 28–29, 36–38, 42, 44n19 loudspeakers and, 30 MacFarlan and, 24, 31, 34–36, 38–39, 44nn38–39, 45n45 microphones and, 41, 43 noise and, 27, 29, 35, 37, 40, 42–43 otolaryngology and, 31, 35, 193, 195, 208n36 phonetics and, 27–28, 38–41, 52, 54, 65, 338, 363 phonograph and, 23–31, 36, 43n5, 353 physiology and, 28, 30 pitch and, 28–29, 35, 42 pronunciation and, 52, 69n20 psychology and, 28, 40, 42 radio and, 25, 35 reverberation and, 29 screening and, 26–27, 29–36, 38, 42–43, 44n31, 353, 363 serviceable hearing and, 12, 38–40 Shewhart and, 32–34, 37 spatial hearing and, 214, 226–27, 230 spondees and, 25, 38–41 standardization and, 24–28, 36, 41, 363 statistics and, 32, 37–38, 42 stimulus–response experiments and, 23, 35–36 surveys and, 34 syllables and, 25, 31, 37–38, 40–41, 338 telephones and, 25, 27, 31–33, 36–38, 40–43, 45n47, 45n61, 46n76, 46n79 timbre and, 24, 28 tuning forks and, 29, 35 volume and, 24, 29–32, 38 vowels and, 8, 28, 37–38, 40, 51, 68n11, 338 Western Electric (WE) and, 25–27, 31–38, 40, 46n79 whispered, 8, 23, 29, 59 speech banana, 39–40 Spencer, Herbert, 82 Sperber, Erik, 35 spondees, 25, 38–41 spot, 312
Index 397 Sprachlaute, Die [Speech Sounds] (Stumpf), 51 Stainer, 117 standardization Anders and, 327, 369 assaying and, 355 auditory thresholds and, 274, 276, 291, 354 certification and, 244, 257 cross-cultural studies and, 90, 93, 99 Kuhn and, 360 loudness and, 24 materials and, 243–49, 255–58, 261–63, 365 pitch and, 52, 308 spatial hearing and, 228, 230, 238n83 speech and, 24–28, 36, 41, 363 Taiwan’s urban noise and, 190–91, 200, 204, 365 testing and, 4, 7, 12, 15–16, 353–55, 360, 363–69 tones and, 1, 12, 24–25, 52, 93, 258, 291, 304, 308, 314 underwater studies and, 302, 304, 306, 308–14, 318 Standardized Tests of Musical Intelligence (Wing), 64 Standard Phonograph Company, 29 Stanford University, 168 Stark effect, 243 State University of Iowa, 63 statistics Abraham and, 50, 59–60, 70n58 cross-cultural studies and, 78–81, 90, 367 perfection and, 114 spatial hearing and, 237n59 speech and, 32, 37–38, 42 surveys and, 10–11, 50, 59–60, 70n58, 80, 355 Taiwan’s urban noise and, 192, 205 testing and, 3–6, 9–12, 15, 355, 367 Steege, Benjamin, 9, 15, 327–48, 354, 361, 369–70 Steinberg, J. C., 213 stereo sound digital models and, 169 hearing norms and, 11 Scherer and, 224 spatial hearing and, 213–14, 216, 223–24, 229, 231–32, 234n1, 234n4, 234n10, 236n41, 236n46, 237n55, 238n74 Sterling, 131 Stern, William, 103n88, 327–28, 344n4
Sterne, Jonathan, 11, 13, 122, 159–85, 353, 362, 368 stethoscope, 30 Stevens, Stanley Smith, 40, 276 Stewart, John L., 289 stimulus–response experiments Abraham and, 49, 54, 67 absolute tone consciousness and, 49, 54, 67 auditory thresholds and, 277–78, 293n20 cross-cultural studies and, 78, 82–84, 93, 95 spatial hearing and, 234n5 speech and, 23, 35–36 St. Louis World’s Fair, 91–93 Stockhausen, Karlheinz, 148 Stradivari, Antonio Echard and, 115–16 modern violins and, 115 perfect violin and, 110–17, 120–24, 126n10, 160, 362, 369 Red Mendelssohn and, 123 varnish of, 113, 115, 121, 123–24, 362 wood source of, 124 Stratocaster guitars, 160 Sturm, Charles-François, 303 Stumpf, Carl Abraham and, 50–51, 54–56, 59, 61–63, 67–68, 70n45, 365 Berlin Institute of Psychology and, 50, 54– 56, 63–64, 67–68, 69n35, 82, 365 consciousness of sensation and, 55–56 cross-cultural studies and, 81–82, 86, 90– 91, 94–98, 101n29, 102n67 Die Anfänge der Musik and, 61 Die Sprachlaute and, 51 Dilthey and, 54 laryngeal memory and, 55, 59 music and, 50–51, 54–56, 59, 61–63, 67–68, 81–82, 86, 90–91, 96–98, 102n64, 365 psychological studies of, 50–51, 54–56, 59, 61–63, 67–68, 70n45, 81–82, 86, 90–91, 94–98, 101n29, 102n64, 102n67, 365 Thurnwald and, 96 tone memory and, 55–56 Tonpsychologie and, 55, 95 submarines, 5, 301–6, 308, 313, 317–18, 321n32 Supersonic Industries, 286 surveys Abraham and, 49–50, 53, 56, 59–61, 67, 70n58
398 Index surveys (cont.) cross-cultural studies and, 78–83, 91, 94, 98, 102n68 Gallup, 195–96 music and, 149 perfection and, 110, 120 spatial hearing and, 225, 237n50 speech and, 34 statistical, 10–11, 50, 59–60, 70n58, 80, 355 Taiwan’s urban noise and, 190–96, 200–204 testing and, 7, 10–11, 354–55 underwater studies and, 308, 313 synthesizers, 71n87, 163–66, 177–78, 180, 362 Taipei School for the Blind and Deaf, 194 Taiwan audiometry and, 103–6, 194, 208n36 China and, 189–91, 197, 201, 206n3, 207n32, 208n34 cultural aesthetics of, 191–92 deafness and, 194 decibel measurement of urban sounds in, 189–93, 198–205, 206n5, 208n36, 365 environmental acoustics and, 189, 191, 193, 200–204 frequency studies and, 190, 194 geopolitics and, 193–97 hearing aids and, 194, 203 KMT and, 189, 195, 197, 206nn3–4, 207n32, 365 military and, 190, 197, 208n34 music and, 189 noise and, 14, 189–205, 206n3, 206n10, 207n18, 207nn22–23, 207n26, 208n46, 365 Noise Control Act (NCA) and, 190, 192–93, 196–97 Noise Policing Act and, 190, 196–97 pollution and, 189, 196–97, 207n26 Preliminary Noise Control Project and, 198, 200 schoolchildren’s hearing and, 193–95, 201, 203–4 Social Maintenance Laws and, 196 standardization and, 190–91, 200, 204, 365 statistics and, 192, 205 surveys and, 190–96, 200–204 testing hearing in, 189–93
United Nations and, 197 urban statistics and, 192, 205 Wang studies and, 194–96, 198, 201–3, 207n18, 207n22, 208n36, 208n41 Taiwan Environmental Protection Administration (TEPA), 191, 196, 202–4 Taiwan Public Opinion Association, 195–96 Tandberg, Vebjørn, 249–50, 253, 255 Tao, Fan-Chia, 114 Tarisio, Luigi, 121 Tavolga, William, 293n20, 314, 321n39 Technical University Berlin, 214, 216, 218, 235n23 Teenage Engineering, 170 Telefunken, 221 telegraphs, 306 telephones AT&T and, 12, 25, 26f, 32–35, 37 auditory thresholds and, 279 Bell Telephone Laboratories and, 27, 32, 38, 213, 266n58, 276, 363 cross-cultural studies and, 93 difficulty of conversations over, 37 hearing norms and, 11 spatial hearing and, 213, 228 speech and, 25, 27, 31–33, 36–38, 40–43, 45n47, 45n61, 46n76, 46n79 underwater studies and, 303 testing A/B, 162, 171, 174 A/B/X, 162, 174–75 acoustics and, 1, 7–15, 17n21 animals and, 3, 14, 190, 200–201, 204, 275–80, 354–56 audiology and, 4, 7, 9, 15, 17n21 audiometry and, 7–12, 354 auditory acuity and, 2, 8–9, 13 biological, 356 Cambridge Anthropological Expedition to Torres Straits (CAETS) and, 13, 77–104, 366–67 cancer, 6 current, 361–62 demonstrative experimentation and, 351–52 digital models and, 162–67, 176–81, 353 double-blind, 11, 114 dummy-heads and, 14, 234n1, 234n10, 236n38, 237n62, 238n78
Index 399 dynamics of modern science and, 99–100 experience of, 26 explanatory experimentation and, 351–56 food and, 4–5 hearing norms and, 11 historical perspective on, 2–11 importance of, 359–70 increasing reliability and, 215–20 intelligent, 5, 92, 99 meanings associated with, 351–56 military and, 2, 9, 11, 27, 34, 42, 45n61, 64, 72n92, 370 multidisciplinary approach and, 95–99 nuclear, 15, 328–35, 340, 343 otherness, 61–66 PB sentences and, 40–41 physiology and, 4, 8–10, 15 politics of, 66–68 prevention and, 6, 34, 192, 201 prospective, 361–62 psychology and, 4–6, 9, 12, 14–15 retrospective, 361–62 screening and, 6 (see also screening) standardization and, 4, 7, 12, 15–16, 353–55, 360, 363, 365–66, 368–69 statistics and, 3–6, 9–12, 15, 355, 367 stimulus–response experiments and, 23, 35–36, 49, 54, 67, 78, 82–84, 93, 95, 234n5, 277–78, 293n20 St. Louis World’s Fair and, 91–93 surveys and, 7, 10–11, 354 transforming context/actors of, 93–95 troubleshooting and, 355 underwater studies and, 301–20 violins and, 113–16 test pattern [times square] (Ikeda), 1–2 Text-Book on Experimental Psychology (Myers), 91 Theile, Günther, 231–32 Thibout, Jacques-Pierre, 127n43 Third Physical Institute, 214, 222 Thompson, Emily, 26, 190 “Thousand” (Moby), 148 Thurnwald, Richard, 96–99, 103nn88–89, 103nn92–93 Tien, Hung-mao, 207n32 timbre, 9 Abraham and, 51, 53, 64 cross-cultural studies and, 82, 95
digital models and, 160, 170 materials and, 252, 255 music and, 64, 137, 148 perfection and, 112, 120 spatial hearing and, 225 speech and, 24, 28 underwater studies and, 304 Times Square, 1 Titchener, Edward B., 81 Tkaczyk, Viktoria, 1–19, 49–76, 82, 95, 162, 352, 361, 363, 365 toadfish, 306, 312–14, 322n47, 322n54 tone differential devices, 52 tone generators, 2, 258, 260 tones Abraham and, 49–68, 70n45, 95, 365 absolute memory and, 55 absolute tone consciousness and, 49–67, 70n45, 95, 365 Anders and, 337 auditory thresholds and, 276–78, 282–83, 287–88, 291, 296n57 Bristolphone and, 35 Cage and, 1 calibration and, 1–2, 5 color and, 51, 53, 114, 148, 218 cross-cultural studies and, 78, 80, 88, 93, 96, 98, 100 digital models and, 175 frequency and, 1, 8, 24–25, 32, 64, 80, 86, 88, 138, 143, 148, 258, 276–78, 283, 287–88, 291, 296n57, 309, 314 Helmholtz on, 56 high-tone losses and, 43, 43n9 Ikeda and, 1–2 modulation and, 80 music and, 1–2, 9–12, 24–25, 35, 43n6, 49–56, 59–68, 70n45, 82, 84, 88–89, 96, 98, 111, 114, 117, 132–33, 136, 138, 143, 147–48, 230, 291, 307, 337, 364–65, 367 octaves and, 51, 63, 65, 71n82, 111, 117, 141–43, 147f, 305, 322n46 overtones, 8–9, 51, 80, 133, 175 perfection and, 111–12, 114–15, 117, 121, 123 Preyer and, 51–53, 67 psychology and, 12, 42, 50, 53, 55–56, 60–68, 70n45, 78, 80, 82, 88, 96, 98, 100, 276–77, 287, 307, 365, 367
400 Index tones (cont.) pure, 1, 8, 10, 12, 24–25, 42, 43n6, 51, 63, 68, 215, 230–31, 260, 277, 282, 287–88, 291, 304, 309 semitones, 49, 111, 321n30 series and, 8, 25, 50, 60, 64, 115, 123, 136, 138, 277, 309, 367 sine, 337 spatial hearing and, 215, 218, 230–31, 234 standardization and, 1, 12, 24–25, 52, 93, 258, 291, 304, 308, 314 underwater studies and, 304, 307–9, 314, 321n30 volume and, 1, 5, 9, 32, 62, 78, 88, 95, 175, 307 tone tests, 367 Abraham and, 50–53, 62, 64 Edison and, 362–63 function of, 1 harmonic series and, 123 phonographs and, 362–63 volume/pitch calibration and, 5 tonometers, 69n21, 89, 98 Tonpsychologie (Stumpf), 55, 95 Torres Straits. See Cambridge Anthropological Expedition to Torres Straits (CAETS) transmission, 10, 14, 41, 43, 213, 234n1, 247, 255, 258, 262 troubleshooting, 355 tuning forks Abraham and, 52, 59, 63, 68n11 auditory thresholds and, 277 bone conduction and, 8 cross-cultural studies and, 83, 86, 89 electric, 8 Helmholtz, 8 music and, 9, 18, 24, 63, 83, 89, 156 speech and, 29, 35 tonometers and, 69n21, 89, 98 underwater studies and, 309 Weber test and, 8 Turbinia (ship), 361 turbojets, 280, 294n28 two-culture divide, 54 Über die Natur der Musik (Opelt), 139, 143, 154n47, 154n57 U-boats, 303–4 Ultrasonic Pied Piper, 284, 290–91
Ultrasonics Panel, 281 ultrasound acoustic attacks and, 292, 296n66 auditory thresholds and, 14, 80, 274– 75, 279–92, 293n4, 293n6, 294n28, 294n40, 295n46, 295n48, 295nn53–54, 296n66, 364 bandwidth control and, 283–86 Galton whistle and, 80, 281 harm from, 273–76, 280–86, 291–92, 294n28, 294n40, 295n53, 296n66 materials testing and, 295n46 Moe and, 284, 294n40, 295n46 multiple uses for, 281 Pierce and, 279–80 repelling pests with, 274–75, 281–92, 293n6, 295n46, 295n48, 295n53, 296n66, 364–65 riot control and, 295n53 sirens and, 281–84 turbojets and, 280, 294n28 underwater studies and, 319 underwater studies acoustics and, 302–6, 311–20, 321n39, 367 aesthetics and, 301 animals and, 301–11, 313, 316–19, 321n39, 322n60, 359, 367–68, 370 audiometry and, 307 binaural sound and, 303–4 Cold War and, 1–2, 312, 317 Cousteau and, 301, 319 deafness and, 305, 319 decibels and, 307 Deep Scattering Layer and, 320n22 echoes and, 303–4, 319 equipment testing and, 302, 306, 313, 316, 319 fish and, 301, 305–6, 309–16, 318–21, 322n40, 322n47, 322n54 food and, 309 frequency and, 302–6, 309–19, 321n30 hydrophones and, 301–6, 309–12, 316–19, 322n47, 354, 367–68 loudspeakers and, 308, 359 microphones and, 301, 303 military and, 301–2, 305–6, 310–13, 316–19, 322n40, 322n54, 367, 370 modulation and, 314 music and, 306–8, 314, 317–18, 367
Index 401 Naval Ordnance Laboratory (NOL) and, 306, 311, 313 Neff and, 307–9 neuroscience and, 307 noise and, 321n24, 367 nuclear energy and, 15, 312, 317 perception tests and, 306–10 physiology and, 304, 307, 309, 319 pitch and, 307–8, 314, 317 psychology and, 304, 307 radio and, 305, 317 reference file and, 306, 310–17, 362 reverberation and, 370 rhythm and, 317 separating sounds and, 311–13 silent world and, 301, 303–6, 319 snapping shrimp and, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 sonar and, 301–19, 322n54, 362, 367 SOSUS and, 317–18 speciation and, 313–16 stabilizing sounds and, 311–13 standardization and, 302, 304, 306, 308–14, 318, 368 submarines and, 5, 301–6, 308, 313, 317–18, 321n32 surveys and, 308, 313 telephones and, 303 timbre and, 304 tones and, 304, 307–9, 314, 321n30 tuning forks and, 309 ultrasound and, 319 U.S. Navy and, 302, 305–9, 312–14, 317, 321n32, 322n54, 362, 367 U.S. Office of Naval Research (ONR) and, 302, 310, 312–14, 362 volume and, 311, 317, 319 whales and, 301–2, 308, 317–18, 322n60, 359, 367, 370 whistles and, 314 World War I era and, 301–4 World War II era and, 302, 304, 306–7, 311, 318, 367 UNESCO Courier, 286 United Nations, 197, 286 Universal Audio, 161, 167–68, 170–72, 177–80, 182n20 University of Berlin, 54 University of California, 305 University of Chicago, 307
University of Göttingen, 214–15, 217, 220– 30, 233, 235n14 University of Indiana, 307 University of Oldenburg, 223–24, 233 University of South Dakota, 36 urban planning, 5, 15, 202 U.S. Air Force, 223, 280–82, 289, 294n28, 294n31 U.S. Army, 281, 284 U.S. Census Bureau, 32 U.S. Federal Trade Commission, 287 U.S. Fishery Biological Laboratory, 311–12 U.S. National Institute on Deafness and other Communication Disorders (NIDCD), 42 U.S. Navy reference file of, 306, 310–17, 362 SOSUS and, 317–18 underwater studies and, 302, 305–9, 312–14, 317, 321n32, 322n54, 362, 367 U.S. Office of Naval Research (ONR), 302, 310, 312–14, 362 U.S. Public Health Service, 31 vacuum tubes, 10, 25, 29–30, 32, 63 van Urk, Arend, 213–14 varnish, 113, 115, 121, 123–24, 362 Verberkmoes, Geerten, 125n3 Vermeulen, Roelof, 213 Vibralyzer machine, 314 Victor gramophones, 363 Vincent, Alfred, 116 violas, 127n45 violins Abraham and, 50–52, 68n7 Amati and, 113, 117 Borsarello and, 115 Boucher and, 120 Il Cannone and, 123 Catgut Acoustical Society (CAS) and, 113 Chanot and, 118, 120–21, 127n43, 127nn45–46 Cobbett Competition and, 116 concert halls and, 115 copying, 121–25 as core of orchestra, 109 Cremonese, 112, 116, 123–24 cross-cultural studies and, 83 Curtin and, 114 digital models and, 160
402 Index violins (cont.) Echard and, 115–16 Eimert and, 339 in film, 123–24 Fritz and, 114, 122, 126n19, 126n21, 128n60 Guarneri and, 110, 113–14, 117, 120, 122–24 Hope and, 124 influence of, 109–10 Klotz and, 117 Lefebvre and, 120 lira da bracchio and, 109 Lupot and, 121 materials testing and, 113 Medwin and, 116 models of, 121–25 Mozart on, 116–17 musical hearing and, 137, 146 Old Italians and, 112–17, 120–21, 127n43, 127n50, 353 Paganini and, 123 perfection and, 13, 109–24, 125nn4–5, 126n10, 126n21, 127n41, 127n45, 353, 362, 369 rebec and, 109 Red Mendelssohn and, 123 Renaissance fiddle and, 109 Savart and, 113, 118–22, 127n43, 127n50 Stainer and, 117 Stradivari and, 110–17, 120–24, 126n10, 160, 362, 369 string making and, 9 testing of, 113–16 varnish of, 113, 115, 121, 123–24, 362 Vincent and, 116 Vuillaume and, 111, 121–23, 127n50 Violin Society of America, 112–13 Virchow, Rudolf, 70n58 volume auditoriums and, 222 digital models and, 169, 173, 175 equipment damage and, 1 loudness and, 8–10, 14, 24, 28–29, 36, 38, 42, 44n19, 50, 80, 115, 217, 221, 224, 286 music and, 131, 135 pitch and, 5, 29, 135, 307 snapping shrimp and, 302, 305–6, 308, 310, 318–19, 320n18, 322n40, 368 spatial hearing and, 222
speech and, 24, 29–32, 38 tones and, 1, 5, 9, 32, 62, 78, 88, 95, 175, 307 underwater studies and, 311, 317, 319 von Frisch, Karl, 309–10 von Karajan, Herbert, 218 von Knobelsdorff, Leopold, 336 vowels, 8, 28, 37–38, 40, 51, 68n11, 338 Vuillaume, Jean-Baptiste, 111, 121–23, 127n50 Wagener, Bernhard, 220–22 Wagner, Richard, 133–34, 152n11 Wang Lao-teh, 194–96, 198, 201–3, 207n18, 207n22, 208n36, 208n41 Watkins, William, 314 wave filters, 25, 32 WE 4A audiometer, 26–27, 35, 36 Weber, Ernst Heinrich, 8 Weber, Reinhard, 223, 237n50 Wegel, Robert, 37 Wendt, Klaus, 224 Wertheimer, Max, 54, 96 Western Electric (WE), 25–27, 31–40, 46n79, 276–77, 293n13 West German Radio, 140, 230, 335, 345n23, 370 whales, 301–2, 308, 317–18, 322n60, 359, 367, 370 whispers, 8, 23, 29, 59 whistles acoustic properties and, 8 adjustable, 8 auditory thresholds and, 281, 284 cross-cultural studies and, 78, 80, 83, 86, 87f-88f, 92, 99 Galton, 15, 78, 80, 83, 86, 87f-88f, 92, 99, 281, 366 music and, 9, 63 railway, 60 toadfish and, 314 underwater studies and, 314 White, Jack, 169 Wigley, Mark, 295n48 Wilder, Thomas, 126n10 Wilkens, Henning concert halls and, 216, 218–19, 222, 231 Kuntskopf and, 215–19, 222, 227, 231, 233, 235n14, 235n21, 235n26, 236nn27–28, 237n47
Index 403 Wilkin, Anthony, 77 wind instruments, 109–11 wine tasting, 115 Wing, Herbert D., 64 Wittje, Roland, 10, 14, 243–69, 361, 365–66 Wolke, Christian Heinrich, 8 Wood, Alex, 262–63 Woodworth, Robert Sessions, 92, 94 World War I era, 331 Anders and, 330, 344n2 auditory thresholds and, 277 disability and, 44n10 materials testing and, 243, 245, 249, 264n25 shell shock and, 100n9
underwater studies and, 301–4 World War II era auditory thresholds and, 274–75, 280, 294n28, 294n31 speech testing and, 27, 38 underwater studies and, 302, 304, 306–7, 311, 318, 367 Wright Field, 281, 289, 294n31 Wundt, Wilhelm, 56, 59–60, 70n45, 70n63, 81, 92 Wu Wang-ji, 197 Young, Thomas, 149 Zhuang, Jing-yuan, 206n10