Should electric vehicles make a noise? If so, what, why and how?

Ed. This piece was first published at cityofsound.com on May 13th 2009, responding to an article in The Economist. It is of its time, and so are some of the links. I’m republishing a decade on due to a July 2019 Guardian report on the Transport for London’s new policy on sounds for electric buses. Subsequently, some further thinking about noise, health, safety and streets occurred in these projects.

A recent article in The Economist (May 7th 2009) suggests that electric cars should generate a noise to compensate for the loss of combustion engine noise, as they are so quiet.

Despite noting there is little research (though I’ll share some below), The Economist says “Some drivers say that when their cars are in electric mode people are more likely to step out in front of them. The solution, many now believe, is to fit electric and hybrid cars with external sound systems.” Their subtitle – “Sound generators will make electric and hybrid cars safer” – suggests that this is their position too.

Where to start?

Let’s quickly deal with the safety issue. People adapt easily enough. We’ve adapted to numerous successive modes of transport in the past without the need to artificially increase the noise that mode of transport generates. (The first automobiles required a man with a flag walking in front of them. Is this not the aural equivalent of that, and so equally likely to fade away?)

One of the numerous reasons why bicycles are a more civic mode of transport is that they do not make much noise. Even at the speeds cyclists can get up to, this near-silent mode is apparently still safe enough not to warrant a pedal-powered drone, say. A bell suffices, and after that it’s about taking due care, attention and respect, on both sides. As bikes again become the dominant mode of personal transport in cities, this shouldn’t change. Some lycra-clad, generally male, faux-motorists aside, cyclists have to rely on individual responsibility to a greater extent than motorists, partly due to their relative fragility. This is not a bad thing necessarily—that respect and responsibility can form a thin undulating layer of civic substrate.

This first aspect of The Economist’s article is borne of auto-centric thinking, and so the concomitant desires for speed and freedom … and often irresponsibility. Speed and freedom are not intrinsically problematic, but they can be. Cars moving at speed in urban areas remain highly dangerous—they are responsible for truly horrifying numbers of fatalities and injuries, and it’s a bit rich to suggest the solution to that particular problem is fake engine noise. Presumably, if people had genuinely wanted to solve this problem they’d have tried a little harder before, rather than apparently relying on the side effect of a noisy carburettor.

On these broader linked systemic challenges of cars, streets and qualities of environmental and public health, I subsequently did further applied research, and prototypes, about these relationships between noise, health, safety and streets in these street retrofit projects in Sweden, from 2019-22.

A user comment from The Economist article declares:

As a Prius owner offended many times daily by excessive noise of motorcycles, trucks, booming car radios, horns, rude cell phone users, etc., I applaud the quiet! I am a responsible person and expect others to be … end of discussion.

Quite. And another comment, pointing to interesting-sounding research:

The National Highway Traffic Safety Administration held a hearing in Washington DC June 23, 2007 to gather the facts and data. It turns out that in spite of having over 500,000 Prius in the USA starting from 2000, there is no accident data showing an unusual risk to pedestrians. The Prius has the same pedestrian accident rate as ordinary gas vehicles.

On more general safety, the congestion that cars cause also limits their average speed in cities (currently down to about 30km/h in Sydney at the moment). There’s an increasingly clear argument that cars, and other motor vehicles, should move more slowly than this. Someone will suggest a GPS-enabled limiter fitted to cars at some point i.e. enforcing a low top speed when the GPS indicates it’s in particularly built-up areas (e.g.). However, the whole point of cars is freedom rather than inhibition, and I’d prefer to see these issues solved through shared space strategies, such as those espoused by Hans Monderman, and traffic reduction in general. Here, drivers are responsible for negotiating urban space alongside others, with few if any demarcations or regulations of space between cars, pedestrians, bikes etc. It’s been proven to make streets both safer and more effective. Crucially, that must include car traffic reduction, as well as a more diverse and active mobility mix—most of which will be slow, safe, clean, enjoyable and quiet.

Such streets might feel closer to this film of George Street, Sydney in 1906. (The version I’ve uploaded here allows you to first compare it with the Sustainable Sydney 2030 strategy for a pedestrianised, light-railed George Street.)

Without wishing to romanticise aspects of that 1906 film (the bubonic plague had been lurking in that same city a few streets to the north only a few years before that film was shot) that shared space model does indicate a more progressive system, based on interdependent and responsive actors negotiating space and relationships in real-time far more fluidly than the averaging effects of mid-20th-century road design, where everyone eventually comes off worse. This is hardly a cityscape without noise, and note how these blurred lines of George Street enable pedestrians, bikes, trams and carts to occupy the same spaces, relying on multi-sensory feedback but essentially with shared responsibility for being aware. Remaining alert and riding the horn might become more relevant than the constant (and therefore less useful) hum of engine noise.

Regarding shared space and horns, a passage in Geoff Dyer’s typically enjoyable latest Jeff in Venice, Death in Varanasi reminds us that not all urban traffic and noise is the same. “Jeff/Geoff” is in India:

The din of horns rendered use of the horn simultaneously superfluous and essential. The streets were narrow, potholed, trenched, gashed. There was no pavement, no right of way – no wrong of way – and, naturally, no stopping. The flow was so dense that we were rarely more than an inch from whatever was in front, beside or behind. But we never stopped. Not for a moment. We kept nudging and bustling and bumping our way forward. Given the slightest chance – a yard! – Sanjay went for it. What, in London, would have constituted a near-miss was an opportunity to acknowledge the courtesy of a fellow road-user. There were no such opportunities, of course, and the idea of courtesy made no sense for the simple reason that nothing made any sense except the relentless need to keep going. From the airport to the hotel, Sanjay had used the horn excessively; now that we were in the city proper, instead of using it repeatedly, he kept it going all the time. So did everyone else. Unlike everything else, this did make sense. Why take your hand off the horn when, a split-second later, you’d have to put it back on?
—Jeff in Venice, Death in Varanasi by Geoff Dyer

Back in ‘Western’ cities, private car use will likely drop anyway, or rather must drop. (I can see that the loss of engine noise might be an issue for blind people, but we must look for a specific solution, in work led by visually-impaired designers and researchers, which may include a diversity of non-visual alerts as cars approach.

Ed. I would later work on the Cities Unlocked project, with Future Cities Catapult, Guide Dogs for the Blind, Microsoft Research and others, where I would learn a lot more about this question of urban interactions for visually-impaired people.

Beyond autocentric, the second aspect of The Economist’s article is borne of what Juhani Pallasmaa would call ocularcentric thinking: an inability to perceive the city, or much at all, in terms of non-visual senses. If the safety issue resolves itself—through fewer cars, and people adapting—and understanding that engine noise is hardly keep the streets safe in the first place, let’s move on to two more interesting implications.

One, if naturally quiet cars should generate a noise, what should that be? And two, if that doesn’t happen, what might increasingly quiet cars do for the urban soundscape?

On the first point, The Economist quotes a Dr. Rosenblum who is researching this area:

What sort of noise should electric-powered cars make? They could, perhaps, beep as some pedestrian crossings do, or buzz like a power tool. Having worked with blind subjects, Dr Rosenblum is convinced of a different answer: “People want cars to sound like cars.” The sound need not be very loud; just slightly enhancing the noise of an oncoming electric vehicle would be enough to engage the auditory mechanisms that the brain uses to locate approaching sounds, he adds.
—‘Electric cars should make a noise’, The Economist

Leaving aside the spurious idea of giving people what they claim to want, reproducing the sound of the internal combustion engine would be ridiculous. It would be a skeuomorph too far; a design feature that nods to an earlier functional (carbon age) incarnation, with absolutely no need to. At some point a function has to replaced, and slowly takes its idioms and by-products with it. The car industry is traditionally loath to do this of course. One of the most exciting features of the MIT CityCar project is that in suggesting a new driving experience. it implicitly indicates how little has changed about interface design of cars – ignition, accelerator, throttle, brake, steering wheel etc; all remain essentially unchanged for decades (save a few brave attempts from Citroën et al). This is not an issue of icon design—as with an old telephone handset representing the function to make a call on the iPhone—but an entirely new functional mode. These are new forms of mobility, potentially, and suffuse with possibility. Unnecessarily tying them to vestiges of the previous mode may prevent them realising their potential.

Doing this with sound would generate aural externalities that simply don’t warrant that level of intrusion. A floppy disk icon still meaning ‘Save’ in Windows 7 is anachronistic and doesn’t augur well for the Microsoft brand, but it hardly changes the essence of the immediate urban area.

Electric or hybrid cars do make a sound of course. It’s just a different noise to the combustion engine. It’s a whine, a hum, a whoosh. Even the ugly Prius is a joy to hear in comparison, if not to see, noiselessly reversing out of a drive. As one of the comments on The Economist article brilliantly points out, there is potential for a rather more progressive sound choice here:

“My vote is for a somewhat high-pitched humming noise ala the Jetsons. After all, when I was a kid, this is what I expected 21st century vehicles would sound like.”

(I’m imagining that as the vehicles in Woody Allen’s magnificently silly Sleeper. This too is a form of nostalgic projection, though.)

Yet if electric cars do have to make a specifically designed, generated noise, let’s at least explore that a little. Brian Eno once suggested that horns in cars should have a little more variation in their noises – that they could play a variety of audio signatures, depending on context. The car is the same. Just as the lovely Honda Puyo concept car suggested its bodywork could glow different colours to indicate different states, so the audio signature of the car could be malleable and responsive. Akin to an instant messaging status indicator, the car’s noise would indicate modes or states that the user wishes to convey, or change in tone as it passes the phone of a friend in the street (admittedly, a feature that would need an off button, for sure.)

This is akin to the ‘I Crossed Your Path’ Facebook app from MIT’s SmartBiking project, but in real-time. Perhaps the sound is a filtered rendition of the music playing in the car—RJDJ externalised rather than internalised—or is simply the music playing directly across the bodywork (one of the more appealing sounds associated with cars in cities is that of a crunching, throbbing sub-bass so impossibly distracting that one looks across and notices that the back seats have been surgically removed to create a giant bass-bin, with the entire chassis becoming a sound-generating devices. These cars also often have a glowing UV light under their skirts, and thus we can only assume the drivers are clearly amongst the most safety-conscious on the streets, announcing their imminent arrival to the visually- and hearing-impaired alike!)

Whether the sound of the streets is improved or further diminished by more clearly hearing this collective cacophony will depend on the musical literacy of your city.

While most signals are necessary for the driver only—battery life indicators, engine performance, personal messages—and so best directed inwards, there may be some possibility in cars as broadcasters of something rather more enriching than the dull roar of internal combustion.

As these cars will be located, addressable and responsive (sooner or later) there’s possibility of creating an interplay between their sounds and the urban environment. To some extent, this is in the spirit of each of Tokyo’s subway stations having a different audio signature, broadcast as trains arrive at the platform. Cars could communicate with each other in real-time, as they pass, and so shift their sounds in response to each other to create discordant atonalities or shimmering consonant harmonies. As you drive across 110th Street in Harlem, your car cheesily fades up into the bassline from ‘Across 110th Street’, with a passing Fiat joining in on percussion while two Nissans emulate the horns and electric guitar. (Pedestrians hanging on the corner are destined to suffer the most annoyingly intermittent cover version imaginable). An array of pipa and guan strike up as you drive through Chinatown, sounds commissioned by the local tourist board. Kyoto’s pedestrian crossings are scored with the engaging knock-knock of doppler’d shishi-odoshi. (These suggestions are deliberately pastiche, knowing that this requires a far more sophisticated response!)

Better, some urban areas commission sound designers to ‘prime’ their streets with latent compositions, which are then performed by passing cars. SND score Sheffield as a series of pulsing, jittery staccato tones; cars pausing at a stop-light in Ginza are suddenly part of a DJ Signify tune; Steve Roden pins up a series of aleatoric triggers across Echo Park; Janek Schaefer creates fields of static and broadcast fragments aurally hung across car park exits throughout West London in homage to JG Ballard, marking up the Westway and its concrete islands, whereas Burial positions a layered series of sub-bass tones along Hackney Road; Steve Reich re-scores City Life for city streets, cars chattering back and forth to each other in fragments of conversation, strings and piano; Filastine sees cars as an intercontinental echo chamber between Barcelona, Kyushu and Marseilles, bodywork rippling with live feeds from distant city streets; Juana Molina plants her sinuous sounds across Buenos Aries, activated as cars drive through her invisible urban space. Drivers begin to follow the threaded patterns through the streets, attempting to stay ‘in tune’ …

Sound is so affecting—often far more distracting than visual interrupts—that its use and abuse should be of primary concern. It is certainly another arena of urban informatics that could be mishandled by a pervasive surveillance culture, even one trying to affect behavioural change ‘for the better’. (‘The pitch of the car’s electric whine shifts depending on the collective energy or water consumption of the area it’s driving through, and so residents receive constant, nagging aural reminders of their performance. The nuances in local crime levels are played out, a form of Oakland Crimespotting with cars generating aural heatmaps, incidentally increasing the nervousness of all within earshot. Perhaps the noise of the car changes if the driver is talking on their mobile, the driver’s speech patterns triggering exact echoes in the car’s hum (so you can tell an Italian driver from a New York driver from an Indian driver ..) Perhaps the car’s whine increases in pitch if the driver has had a drink or two…’)

Given half a chance, we won’t miss the noise of cars (as we know it) in our cities at all. When we (Arup) design new cities, and are able to design without private car use, our city models and simulations indicate noise levels that are far more appealing. I don’t mean quiet, as cities are always noisy—as people are, as the birds are—but that it was possible to hear more, in more detail, and over a wider range.

When Geoff Manaugh interviewed Arup’s Neill Woodger (in Dwell, June 2008) about new cities and the SoundLab aural modelling tool, Woodger said, “These cities are an opportunity to think about a new urban sound experience, including the ability to bring sounds back into cities. People haven’t really known that they can change the sounds of a city …” Masdar, outside Abu Dhabi and predicated on light rail, personal rapid transit and no private cars, affords the same possibilities as the Dongtan design. (I’ve previously speculated about the kind of urban SoundLab approach).

Cities should not be quiet, or only replete with so-called ‘natural’ sounds—whatever that means post-nature, and post-industrialisation—but the urban soundscape is something that could use a little more room for manoeuvre, dynamically. To be clear, I’m not averse to cars per se, or car noise. Some car noises are hugely appealing. They are just best experienced as a distinct note and timbre in a richer, more dynamic city symphony, as opposed to the pervasive ambient roar of thousands of combustion engines. They are to be used sparingly, as mobility and noise, just as a kettle drum in an orchestra is probably best positioned as subservient to the violins and woodwind—but can certainly have an effect when it steps forward, due to this dynamic and relative balance. The combustion engine has a totalising and suppressing effect on urban sound, akin to the scourge of overusing the compressor in contemporary music production. If everything is loud, nothing is distinctively loud, and nothing has any meaning.

Buses—the public transport mode that car-based cities tend to accept — are often the worst offenders. Sydney buses are particularly egregious, amongst the loudest I’ve heard in any city. As I’m that way inclined, I’ve taken to sporadically measuring the decibel level on city streets using the promising but currently flawed iPhone app WideNoise (see also NoiseTube), and find levels well over 100dB when a bus or two roar by, even on an open street corner. This is akin to standing in a sheet metal workshop, and you can watch people actually grimace, subconsciously feeling how unpleasant it is. Put simply, guidelines and research indicates that such levels are highly detrimental for peoples’ health and wellbeing. It leads to iPod users turning the volume up further as a form of aural arms race (a lose:lose scenario). It also flattens the possibility of diverse urban sounds.

That buses are allowed to be this way is due to an endemic lack of understanding of sound. It simply isn’t valued by many policy-makers and so rarely measured. In the case of public transport planning and procurement, travel times is seen as far more important than experience. Again, this is the outcome of an ocularcentric culture to some degree, but also a culture that suffers from a paucity of understanding of the urban experience in general. City and state government officials need not be conversant with the works of John Cage, but basic qualitative probes into the urban experience are surely important.

(Programmes like the London Ambient Noise Strategy are unusual, yet even when they do exist they are usually about noise abatement rather than ‘positive soundscapes’.)

When Jan Gehl’s team were focused on Sydney’s CBD, with perhaps predictable results, they also came to the conclusion that the city was particularly noisy, and due to the combination of buses and urban form (tight canyons). In an article in the Sydney Morning Herald’s glossy (sydney) magazine last year focusing on noise, Gehl said “Sydney has tremendous noise levels in most streets and squares … The main cause is the buses that create a tremendous roar when they accelerate and a shrieking sound when they brake.” The Herald measured decibel levels in several places in Sydney’s CBD and also managed to record over 100 decibels outside the Queen Victoria Boulevard, noting “any exposure to noise above 85dB can permanently damage your hearing—any exposure above 120dB, however brief, can have far greater consequences. High noise levels are also associated with hypertension, stress, heart damage and depression.”

Oh joy. However, this focus on volume (and decibels) as a measure of sound is a little crude, leading naturally to noise abatement rather than a more expansive palette of sound. How high and low frequencies might interact, or more qualitative, descriptive aspects of sound, are rarely discussed or devised.

So with heavily car-scaled cities like Sydney, or Los Angeles say, it’s almost impossible to imagine how different these streets might sound without cars.

With a new suburb like Vauban in Freiburg, Germany, which has been planned to effectively function sans autos, the aural possibilities should be fascinating. Typically, this New York Times article on Vauban makes very little reference to how different it might sound. There is only the tantalising line: “When I had a car I was always tense. I’m much happier this way,” said Heidrun Walter, a media trainer and mother of two, as she walked verdant streets where the swish of bicycles and the chatter of wandering children drown out the occasional distant motor.”

It would be interesting to explore how a city’s sound might be articulated, either naturally or by design, without the presence of pervasive engine noise. If conversation is as loud as, say, 55db, should an electric car be about the same? Or should a car’s engine be effectively silent, so our streets become defined more by the sound of an espresso being made, chitter chatter, whistling, a parakeet, trees in the breeze, the grind and whirr of contemporary industrial machinery, lapping water in the harbour, chimes of ringtones, the rumble of trains and the foghorns of distant ships, a record shop or a practicing violinist, an art installation and the bells of a clocktower, prayer calls and pub arguments?

The Positive Soundscapes research project indicates the range of noises that people may find appealing is actually far broader than this – “car tyres on wet, bumpy asphalt, the distant roar of a motorway flyover, the rumble of an overground train and the thud of heavy bass heard on the street outside a nightclub, a baby laughing, skateboarders practising in underground car parks and orchestras tuning up.” And though I note “the distant roar of the flyover in that list”, I’d rather hear more about the results of their research than Dr. Rosenblum’s.

I don’t think we’ll miss the noise of cars much, particularly as we reduce their numbers anyway. This enables them to become something special. And cars can be something special in the urban environment (as I hope my decision to illustrate this piece with snippets of M Sasek’s illustrations make clear). Yet cars are essentially about freedom not transit. They are for fun, not for the daily grind. They may increasingly be seen as out of place in a busy city on a Tuesday morning at 08.30 as a horse would be now. The idea of them as mass transit, for most people, given ever-increasing urbanisation, is faintly ludicrous. Instead, they’re for casual use, for the sheer enjoyment of the driving experience. Like in the adverts. Something for the weekend, if you will. Again, like a horse is now. (Note that actual horse-power, rather than the petrol-driven analogue used to be the mainstay of urban mobility).

In that respect, those more infrequent sounds could be considered as something special too. We might more fully appreciate the throaty purr of a 1969 Ferrari Daytona or the brawny roar of a 3.5 litre 1978 Ford Capri or the lawnmower rattle of a 2CV or the saucy throb of an old DS, lifting skirts and all, just as we’d appreciate the sizzle and hiss of tyres on wet road—if it was a rare sound, rather than a pervasive thrumming backdrop to everything.

(Or, as I would note in a subsequent piece, where would the infamous Bullitt car chase be without the sound of Steve McQueen’s Ford Mustang?)

View at Medium.com

The corollary of this is that we won’t particularly miss the sound of a 2002 Mazda 323 or a 2007 Honda Jazz or a 2004 Holden Barina or 1998 VW Golf. These kinds of cars are, after all, by far the most prevalent on our roads, and their sound is thus the most prevalent noise in our cities.

Peter Cusack’s Your Favourite London Sounds also lists a few traffic noises (“16th floor up, London roar from the top of a tower block, Holloway Road, on a damp evening”; ”Taxis waiting at Euston Station, squeaky black taxi brakes”; ”Under the flyover, Hackney Wick”). They’re by far in the minority in his collection, but in real London they drown out the others.

Removing cars would enable the other sounds to be picked out more clearly, also accentuating urban difference, in that cars tend to be a somewhat homogeneous globalising force—due to their high production costs, they are essentially the same across the world; the platform for a VW Golf not only services the Golf, but the Skoda Octavia, Seat Leon and Audi A3.

Other sounds are also global in provenance of course, but many more sounds are local. Note how Cusack picks this out in his thoughts on his Favourite Beijing Sounds project:

“So what does the city sound like? The answer is that it’s amazing. Central Beijing has an astounding soundscape. Its shear (sic) scale envelopes you immediately and its variety constantly surprises. This may not last. The older uniquely traditional sounds are fast disappearing, as newer, more globally familiar, ones take their place. Peak traffic is already at high volume. But at the moment the old and new co-exist. Amongst the loud and brash, there are still places of the utmost quiet, where a breath of air touching a dead leaf will catch your ear. Elsewhere people talk, hum and sing loudly, not minding who listens. Music, live and recorded, plays anywhere. It is a city of sound loops. Ubiquitous loud hailers blast out advertising slogans that endlessly repeat in or out of sync with music from the shop next door. Pigeons fitted with bamboo whistles create eerie chords above your head when they fly. Buses screech, shop assistants yell and clap their hands, taximeters talk and woks sizzle. Street cries are commonplace. And in the parks older people sing revolutionary songs in choirs hundreds strong, while others engage in caged-bird singing contests, ballroom dance or practice t’ai chi.”
—From Favourite Beijing Sounds, Peter Cusack

Toyotas are largely the same in each city; pigeon whistles are not. Yet rather than position this as old (local) versus new (global), it may be that the ‘electric car as noise generator’ discussed above provides an opportunity to create new local sounds.

Scoring the city is an interesting idea, whether via discrete car-based sounds or taking advantage of the absence of car-based sounds. Strong urban places already have their own signature, through their behaviour, a point made by William H. Whyte in his 1980 book (and film) Social Life of Small Urban Places, when he and his team rendered the patterns of movement through the plaza at Seagram’s in New York as a form of graph.

He noted that this could be perceived as “music of sorts”:

Since the Seagram’s chart looked so like a player-piano roll, I wondered what the sound would be if all the dots and dashes could be played. A composer friend was fascinated: with the right tonal scale, he said, the roll could be orchestrated and it would be music. I hope one day it will be: A Day in the Life of the North Front Ledge at Seagram’s, Adagio.
—From William H. Whyte’s ‘Social Life of Small Urban Places’.

The opportunity to genuinely explore the sound of the city without this blanket of private cars is compelling, whether through sculpting sound through active intervention or simply through enjoying a level aural playing field for the everyday sounds that already conjure the city.

At first glance, taking The Economist to task for suffering from a severe lack of creative imagination might seem a little like admonishing Cristiano Ronaldo for not spending his Sundays reading Žižek. But let’s at least discuss how sound and the city should best intersect given the emergence of this new mode. We can slowly fade down the volume on that wall of noise – what might we want to hear its stead?

Ed. This piece was first published at cityofsound.com on May 13th 2009, responding to an article in The Economist. It is of its time, and so are some of the links. I’m republishing a decade on due to a July 2019 Guardian report on the Transport for London’s new policy on sounds for electric buses. The title of this piece refers to the songs Cars and Are Friends Electric? by Gary Numan and Tubeway Army. 

Over a decade later we discover that, above 50km/h, electric cars produce the same noise levels as combustion-engine cars anyway, as tire noise is louder than engine noise at this speed. Yet this broader point remains an issue in cities, where speeds should arguably be less than 25-30km/h, and cars should be a far less-dominant element in the mix. We are still left with these questions. On these broader challenges, I subsequently did further applied research about these relationships between noise, health, safety and streets in these street retrofit projects.