Miami airport, Florida, 2014. Photo by Harry Gruyaert/Magnum


Acoustic naturalism

Our movies and offices are engineered to sound natural based on what rang false in the theatres of 18th-century Paris

by Joseph L Clarke + BIO

Miami airport, Florida, 2014. Photo by Harry Gruyaert/Magnum

A low-level hum of talking and typing punctuated by the occasional warble of a ringing phone: the ambient murmur of a large open-plan office is, to a surprising degree, an intentionally orchestrated one. Not only do office designers use carpeting and absorptive ceiling tiles to reduce reverberation; in some cases they even distribute loud employees and machines around the office deliberately to mask one another and create a blanket of low-level noise in which individual sounds don’t stand out. The goal is to create an acoustic mise en scène that seems natural for office work. As the British architect Francis Duffy puts it, with too little background babble, ‘the environment as a whole seems “dead”.’

Improving how things sound in buildings – making certain sounds audible and attenuating others – is the aim of a multibillion-dollar acoustics industry. Paradoxically, one of the guiding objectives of architectural acoustics over the past 250 years has been to make built spaces and the acts of communication that take place in them seem as natural and unmediated as possible. Hence, the Dutch engineer Roelof Vermeulen once remarked wistfully that ‘the better [an acoustician] does his work, the more natural the result will appear and the less thanks he can expect.’

If you’re a middle-class Westerner, chances are you have internalised expectations for how a room should ‘naturally’ sound. You probably expect its ambient acoustic properties to match its function and visual character. The length and intensity of reverberation, the tendency for sound to resonate at particular frequencies, the background noise characteristic to that particular space – all these factors give clues about the size, form and contents of the environment. Moreover, when you hear a sound, you likely take for granted that it reaches your ears with minimal distortion, and that people located in different parts of the room perceive it in more or less the same way. Finally, you probably assume that your eyes and your ears will give consistent accounts of what the room is like and where sounds are coming from.

In most small buildings, no special effort is required to achieve these results. But large rooms of all kinds – restaurants, airports, hospitals, offices – are painstakingly engineered to meet our acoustic expectations. Making buildings sound natural is not the same as improving the clarity of sound, a goal recognised in the ancient world and discussed in the earliest surviving architectural treatise, by the Roman architect Marcus Vitruvius Pollio. This objective persists today, of course, but the goal of clarity is sometimes in tension with that of naturalness – a distinctly modern ideal arising from an elusive promise of fuller, freer communication and better mutual understanding.

A helpful analogy for the work of an acoustician is the creation of a movie soundtrack. Much of the sound we hear in a typical movie is either created or substantially altered in post-production. If the sound were simply recorded along with the picture and left at that, the microphone would pick up too much background noise and not enough dialogue. An engineer therefore adjusts the intensity of the audio at different frequencies, adds or filters out ambient sound, and introduces a subtle amount of reverb, so the sound matches the spatial environment shown onscreen. When we watch the resulting scene, our brains match up the visual and auditory tracks to suggest a complete illusory world.

The sonic conventions that govern most movies were worked out in early 20th-century Hollywood. Technicians argued over the best way to make a film sound believable – to achieve ‘artificial naturalness’, as audio engineer Wesley C Miller put it in 1929. For example, some initially felt that every time the camera cut between a wide-angle shot and a close-up of someone speaking, the loudness of the sound should change too. Eventually, though, most engineers agreed that if the audio kept shifting in this way to match the image, the soundtrack would draw too much attention to itself and would come across as unnatural. Today, cinematic sound tends to maintain the same ‘perspective’ even when the camera’s viewpoint changes. Ironically, letting visual and audio tracks subtly diverge like this yields a more natural-seeming result.

Movie soundtracks became an object of historical study in the 1980s through the work of Rick Altman and other film scholars. Altman was one of the first to use the term ‘sound studies’, which has since blossomed into an interdisciplinary area of research. But many of the concerns it has raised about sonic media and space have much older roots in architectural theory. When Hollywood technicians debated how to make movies sound natural, they were unwittingly following a trail blazed by 18th-century architects, who spent decades working out the acoustic conventions of modern theatres.

Eighteenth-century Europe is often said to have elevated the epistemic status of vision above that of the other senses – hence the optical metaphor of ‘enlightenment’ – but it also made sound an object of empirical research, commerce and artistic exploration. The public’s preoccupation with new auditory experiences was especially evident in the music world. The flourishing of the sonata (Italian for ‘sounded’) typified the new popularity of music without words. Alongside melody and harmony, audiences began focusing on timbre, which Jean-Jacques Rousseau – who was not only a philosopher but a composer, too – described as a sound’s ‘je ne sais quoi’.

What truly brought acoustic concerns to the fore, though, was the building of theatres, for both operas and spoken plays. Theatre stood at the centre of prerevolutionary French culture, a key site for philosophical reflection, political struggle and the construction of the self. It was also big business. The end of the Seven Years’ War in 1763 triggered a boom in domestic projects such as the construction of performance halls, which might contain 2,000 seats or more. Urban stages were graced by celebrity actors such as David Garrick and Clair Josèphe Hippolyte Leris (known as La Clairon), whose larger-than-life reputations were built up through published theatre criticism.

Audience members now saw themselves as consumers in an increasingly diversified entertainment market and demanded to hear every nuance of the plays they attended. When an actor onstage speaks a line, some acoustic pressure waves from his mouth travel directly out to the audience, but many more are reflected off the walls, floor and ceiling, and keep bouncing around the auditorium until they either dissipate or find their way to someone’s ears. If these reflected waves arrive much later than the original sound – as might happen in a large hall – the listener perceives them as a discrete echo and they can interfere with the intelligibility of the dialogue but, if they arrive quickly, her auditory cortex seamlessly groups the two sounds together. Chances are she will not register the reflection consciously, though it will alter the sound quality and provide subliminal signals about the architectural character of the room.

This process was only dimly understood in the 18th century. Because sound is invisible and intangible, designers and their clients had long regarded acoustics as one of the most obscure branches of architectural theory. The classical orders established an elaborate canon of correct proportions and ornamental forms, but nobody thought to codify how a building ‘should’ sound. To be sure, the ancient Greeks built theatres renowned for propagating actors’ voices with extraordinary clarity. What made these structures significant, however, was how unnatural their acoustics were. Premodern writers sometimes described other buildings where sounds carried in astonishing ways – for example, when Pliny the Elder remarked on the wondrous sevenfold echo from a portico in the city of Olympia, or when Athanasius Kircher described a round room in Heidelberg where two people standing at opposite ends could converse in whispers. Architectural acoustics was a set of techniques for conveying voices over improbable distances and producing marvellous effects.

Passionate, inarticulate cries seemed to appeal to listeners’ moral empathy better than arid discursive speech

This changed in the 18th century, as part of a widespread European reaction against baroque decadence. Manuals of politesse began instructing ladies and gentlemen to comport themselves with studied ease. Women’s clothing was redesigned to look less contrived and to convey a fashionable informality. Picturesque landscapes or jardins à l’anglaise, like the one Marie Antoinette had put in at Versailles, were planned to look unplanned, with irregular topography and meandering paths. The philosophical basis for these shifts was articulated around the middle of the century by Rousseau, who held up nature as the ideal pattern for society and human development. He eventually developed this conviction into a critique of language: he wanted people to rediscover the ostensible origins of communication in passionate, inarticulate cries, which seemed to appeal to listeners’ moral empathy better than arid discursive speech.

For the oxymoronic task of remaking culture in the image of nature, theatre served as an important laboratory. European dramatists began writing fewer plays about the deeds of kings and heroes, and more about the emotional complexities of domestic life. The protagonists of these plays were frequently commoners, with whom audiences felt they could sympathise more readily. For example, in Françoise de Graffigny’s hit sentimental drama Cénie, premiered in 1750, a young woman vacillates between marrying the suitor chosen by her father and the man she really loves. When it emerges that she was switched in infancy and is actually the daughter of her own lowly governess, she is ashamed and resolves to join a convent.

The plot of Cénie hinges on coincidences that would strike most of us as implausible today, but period audiences marvelled at the authenticity of the moral choices and ambivalent feelings it portrayed. Central to this new kind of theatre was a quest for dissimulated artifice, an aspiration reflected in the reform of stage space. Spectators increasingly watched actors move about in ordinary-looking rooms separated from them by an invisible plane, a convention later formalised as the fourth wall. The highest criterion for a play was that it seem true to lived reality. The philosopher Denis Diderot – himself a dramatist – described this new kind of stage work as a drame bourgeois, linking the changes in French theatre with the tastes and mores of the capitalist class.

Graffigny, like some other young playwrights, wrote in prose instead of the verse used by earlier French dramatists such as Jean Racine and Pierre Corneille. Prose seemed to suit the new style of acting, in which performers no longer recited their lines in a stately, booming voice – a neoclassical technique inherited from the ancient art of oratory – but instead sought to evoke a character’s inner motivation faithfully without breaking character. For an actor to pull off a spontaneous-sounding vocal delivery took all the more discipline and calculation, however. For example, Diderot explained how a bereaved mother’s onstage lament was carefully practised to tug on the audience’s heartstrings:

What of these touching and sorrowful accents that are drawn from the very depth of a mother’s heart and that shake her whole being? Are these not the result of true feeling? Are these not the very inspiration of despair? Most certainly not. The proof is that they are all planned; that they are part of a system of declamation; that, raised or lowered by the 20th part of a quarter of a tone, they would ring false.

The kind of acting Diderot described was eventually challenged by Konstantin Stanislavski’s method of psychological realism, in which performers endeavoured to ‘live the part’ by immersing themselves in their characters, and Bertolt Brecht’s technique of estrangement or Verfremdung, which deliberately tried to subvert spectators’ sympathetic involvement with the play. These approaches reacted in different ways against the pursuit of affected naturalness enshrined in the mid-18th century as the hallmark of great theatre. The key tenet of this method was that, for a performer to seem carried away with emotion, she must always maintain inner composure: Diderot called this ‘the paradox of the actor’.

Theatre’s newfound naturalism depended on architectural acoustics to convey the subtle emotional inflections of these performers’ voices. The artist and critic Charles-Nicolas Cochin, a sometime collaborator of Diderot, lamented that, in conventional theatres, actors were obliged to shout, coming across as ‘forced and unnatural’. He argued that the architecture of the theatre should aid in the propagation of the voice, enabling the audience to ‘hear the faintest articulations as distinctly as possible’.

The importance ascribed to naturalness convinced theatre architects that their work should be directly informed by physics, the study of nature. The most extensive acoustic theory put forward in the 18th century was written by the architect Pierre Patte in 1782. Patte had carefully studied numerous theatres and, although he never built one of his own, his book included a model design (below). ‘Today we have returned to the simple, to the natural,’ he wrote. ‘It is now understood that true music does not consist in astonishing the ears, and that, on the contrary, to sing well it is essential to know how to moderate the voice, to support it, and to direct it by means of imperceptible nuances.’ Patte’s key argument was that the ‘natural propagation’ of actors’ voices was not sufficient to make a hall sound natural: rather, the building’s interior surfaces must be shaped to reflect sound toward the audience in a highly artificial way.

Acoustic diagram of a theatre, from Pierre Patte’s Essai sur l’architecture théatrale (1782). Courtesy Wikimedia Commons

Three criteria for a natural-sounding room can be distilled from Patte’s treatise: uniformity, audiovisual coordination, and intimacy. First, performances should sound the same from every seat. Many French theatre critics had complained about halls where the performance was barely audible beyond the front rows. In a good theatre, Patte explained, ‘not only would everyone be able to see the entire stage and all the set decorations, but the actors, with no need for great vocal efforts, would be heard uniformly by everyone: their voices would be supported in every corner, and would always seem full and sonorous.’

Second, Patte observed that a theatre building had a ‘dual objective’ to facilitate audiences seeing and hearing the performance. While he analysed acoustics and sightlines separately, he believed that, with proper coordination, both criteria could be satisfied. This was not a self-evident proposition. One critic had grumbled that in most French theatres ‘the best place for vision [was] the worst for hearing, and vice versa’. Patte understood that, while the audience looked directly at the stage, much of the sound they heard reached their ears only after bouncing off the hall’s interior surfaces. If the various perceptual design considerations were not properly resolved, there might be seats where one could hear actors’ voices but not tell where the sound came from, observe actors’ gesturing without catching the dialogue, or perceive sounds as coming from the wrong location.

A final concern for Patte was that if sound were not reflected in the proper way, it would seem ‘thin and exhausted’ and would lack ‘intimacy’. For him, an acoustically intimate hall was one where performers ‘could maintain the tone of an ordinary conversation and none of it would be lost’. In other words, audience members sitting amid the vast commercial apparatus of a Parisian theatre should still feel they were receiving the sound of a performance individually, as if the actors were right in front of them. This ideal of intimate sound was eventually inherited by the 20th- and 21st-century music industry. In popular music especially, the combination of close mic recording and the use of stereo headphones often heightens listeners’ sense of being in the immediate presence of the performers. Meanwhile, in the design of concert halls for orchestral music, intimacy is prized more than ever as the acoustic holy grail.

Patte’s design strategies for improving theatre acoustics failed. His main suggestion – shaping the theatre as a perfect ellipse, so sound would bounce off the curved walls in an orderly way – proved totally inadequate. When a massive theatre built on this premise opened in Berlin in 1802, it was widely criticised for a distracting echo. Gradually, acoustic design became more empirically grounded, and designers realised that the sound of a theatre – or any other kind of building – depends on numerous factors that Patte had barely considered, including the size of the hall and the materials it is made of. Then, around the turn of the 20th century, the practice of acoustics was transformed by the rise of scientifically trained consultants, the invention of new sound-absorbing building materials, and audio engineers’ efforts to evoke virtual spaces in recorded sound. The historian Emily Thompson argues that these developments reinforced one another to establish a new regime of ‘modern sound’.

Nevertheless, for all these important technical and cultural changes, the imperative to make architecture sound natural remains essentially unchallenged. In the 1960s and ’70s, for example, with the expansion of open-plan offices, engineers were at pains to make them sound like offices ought to sound. ‘Every space should have an acoustical environment that is natural to the activity of the organisation that occupies it,’ advised the furniture company Herman Miller. The company suggested that office workers be carefully arranged so the sounds they each produced would complement one another. The objective was to create a pervasive ‘activity bustle’ in which a sense of ‘“who we are, what we are doing” is preserved and communicated’.

We want to forget that sound is mediated, and to imagine that we are communicating directly and naturally

Occasionally there have been creative experiments aimed at disrupting acoustic naturalism. One of the earliest movies to defy the emerging conventions of cinematic sound was Under the Roofs of Paris (1930), the first sound film by auteur director René Clair. At one point, the camera peers into a tavern through a glass storefront when a girl runs inside to convey some urgent news. We briefly hear sounds from inside the bar, but only for the split second the door is open. After it closes behind her, we see the girl’s excited gesturing but can only guess at what she says. In another scene, as two men fight on a street corner at night, the camera follows the brawl closely but the sound is drowned out by a serendipitously passing train. As the train chugs into the distance, another character shoots out the streetlight; suddenly the sensory channels are reversed, and we can hear the fight scene but without seeing it. Such startling effects defamiliarise the film’s soundtrack, drawing attention to it as an autonomous element. Gestures like this are rarely crowd-pleasers. Despite the popularity of Under the Roofs of Paris, Clair’s challenge to the emerging Hollywood consensus around auditory naturalism was largely ignored by later filmmakers.

Architecture, too, has sometimes challenged acoustic norms in order to produce powerful spatial effects. In 1958, the Philips electronics company sponsored a pavilion at the World’s Fair designed by the architect Le Corbusier to demonstrate its new electronic systems for modifying a room’s acoustics. Through the dynamic simulation of reverberation and other effects, the pavilion was ‘to seem at one instant to be narrow and “dry”, and at another to seem like a cathedral,’ as a Philips engineer put it. These effects were applied to a jarring eight-minute piece by the avant-garde composer Edgard Varèse that remains a landmark of electroacoustic music. Le Corbusier hoped that the pavilion’s use of sound technology would encourage architects’ broader experimentation with acoustics, but its influence has largely been confined to the more rarefied world of sound art installations.

The foregrounding of acoustic artifice thus remains the exception rather than the rule. Something deeply rooted in modern culture makes us want to forget that sound is mediated, and to imagine that we are communicating directly and naturally. Writing about movie soundtracks, the theorist Michel Chion argues that an instinctive desire to believe in ‘human unity, cinematic unity, unity itself’ compels us to resolve a film’s image and its sounds into an integral representation of space. This urge seems to grow only more powerful as communication becomes more mediated. Recently, the production of natural-seeming audio has become a key area of research and development for videoconferencing software. It won’t be enough to eliminate the bizarre echoes, background noises and sound-image disjunctions that still frequently haunt online meetings: the goal is to produce a fully convincing ‘virtual’ acoustic space.

As long as such platforms are measured against the acoustic norms of physical spaces, they will be dogged by the same contradictions that first emerged in 18th-century theatre design. How can demands for clearer sound be satisfied when they conflict with the naturalistic ideal? How much ‘artificial naturalness’ can listeners tolerate before the reality effect breaks down? What happens when perceptual standards evolve and yesterday’s natural sound rings false to today’s listeners? Ultimately, the acoustic vexations of spatial environments – whether the spaces are real or virtual – are not empirical problems to be solved by technology, but philosophical tensions stemming from the dream of unmediated, uninhibited exchange that haunts modern communication.