If nobody in India honked for a day, a busy road would get 2 to 7 dB quieter: horns carry roughly 40% to 80% of the sound energy and the street would sound about a quarter to a third less loud. The catch is that a honk is a real message ("I am here, I am passing"), and without it drivers who cannot rely on mirrors and lanes would have to slow to around 17 to 36 km/h near other road users to stay safe. A silent day is possible, and Mumbai has tried, but it only works if the road supplies the information the horn used to.
- 2 to 5 dB(A): the extra noise honking adds on Nagpur roads, measured by CSIR-NEERI (Vijay et al., 2015).
- 79%: the share of sound energy from horns on a Nagpur highway in the morning peak, from the same data.
- 63 honks in 15 minutes: the morning count on that highway, about one every 14 seconds (900 s ÷ 63, calculated here).
- 85 dB: the threshold of Mumbai Police's 2020 "Punishing Signal", which reset the red light when honking crossed it.
- 17 to 36 km/h: the speed range at which a driver can stop for someone cutting in 15 to 30 m ahead without warning them first.
This is the mirror image of another question on this blog, what if all of India honked at once. That one was about energy. This one is about information: what the horn says, how much of the street's noise it makes, and what the road would need to say instead. The everyday background is in why Indians honk so much.
Has any Indian city tried a day without honking?
Yes, and the first attempt failed within hours. Mumbai's traffic police held the city's first No Honking Day in April 2008.
On a Monday in April 2008, Mumbai's traffic police asked the city's roughly 15 lakh vehicles to stay silent for a day. They had enlisted automobile associations, schools and colleges to spread the word, and the city already had no-honking zones around hospitals and schools. Drivers honked anyway. Residents told reporters that on crowded roads it was simply impossible to move forward in silence, and the police conceded it had been a poor start, while promising to keep trying. The notified quiet zones around hospitals and schools were ignored too.
Mail & Guardian (AFP): Mumbai drivers hoot through no-honking day, April 2008Twelve years later Mumbai tried engineering instead of appeals. On 31 January 2020 the traffic police wired decibel meters into signal timers at five junctions, including CSMT, Marine Drive, Pedder Road, Hindmata in Dadar, and Bandra. If the noise at a red light crossed 85 dB, the countdown reset, under the slogan "Honk more, wait more". The police called it awareness more than punishment, and noted honking cases booked had risen from 149 in 2018 to 460 in 2019 (The Tribune, January 2020).

How much of India's street noise is honking?
Often most of it. The best measured answer comes from CSIR-NEERI, which recorded noise on three Nagpur roads with and without honking and found horns added 2 to 5 dB(A) to the average level.
The team measured peak-hour noise with a Type-I sound level meter on a six-lane national highway, a four-lane major road and a narrow minor road in Nagpur during 2010, and counted honks minute by minute. The highway saw 63 honks in a 15-minute morning window. Equivalent levels with honking were 71.2 to 76.6 dB(A); removing the honk events lowered them by 2 to 5 dB(A) overall, and by up to 7.3 dB(A) on the minor road in the morning. Honk energy sat mainly in the 500 Hz, 2 kHz and 4 kHz octave bands. The limit: three sites, one 15-minute window per peak, and some internally inconsistent ranges.
Read the paper (free)| Nagpur road (2010) | Honks in 15 min (am / pm) | Leq with honking, dB(A) | Leq without honking, dB(A) | Horn share of sound energy |
|---|---|---|---|---|
| National highway, morning | 63 | 76.6 | 69.8 | 79% |
| National highway, evening | 57 | 74.7 | 71.9 | 48% |
| Major road, morning | 50 | 72.4 | 68.0 | 64% |
| Major road, evening | 37 | 71.4 | 69.0 | 42% |
| Minor road, morning | 57 | 73.6 | 66.3 | 81% |
| Minor road, evening | 38 | 71.2 | 67.3 | 59% |
The levels are from the paper's statistical method; the horn share is my calculation from them, shown next.
The worked derivation: decibels do not add like rupees
Sound levels are logarithmic, so you add the energies, not the decibel numbers. Two independent sources, traffic and horns, combine as:
- Energy addition.
\[ L_{total} = 10\log_{10}\!\left(10^{L_{traffic}/10} + 10^{L_{horn}/10}\right) \]
- Solve for the horn level alone, using the highway morning numbers \( L_{total} = 76.6 \), \( L_{traffic} = 69.8 \):
\[ L_{horn} = 10\log_{10}\!\left(10^{7.66} - 10^{6.98}\right) = 10\log_{10}\!\left(4.57\times10^{7} - 0.95\times10^{7}\right) = 75.6\ \text{dB(A)} \]The horns on their own are louder than every engine, tyre and exhaust on that road put together.
- Horn share of sound energy.
\[ s = 1 - 10^{-(L_{total}-L_{traffic})/10} = 1 - 10^{-0.68} = 0.79 \]
- Sensitivity. The paper's second method (frequency analysis) gives 72.2 dB(A) without horns, so \( s = 1 - 10^{-0.44} = 0.64 \). Across all six road-and-time cases, the share runs from 42% to 81%.
See the data as a table
| With honking (dB(A)) | Without honking (dB(A)) | |
|---|---|---|
| Highway am | 76.6 | 69.8 |
| Highway pm | 74.7 | 71.9 |
| Major am | 72.4 | 68.0 |
| Major pm | 71.4 | 69.0 |
| Minor am | 73.6 | 66.3 |
| Minor pm | 71.2 | 67.3 |
Source: Vijay et al., J Environ Health Sci Eng 2015 (CSIR-NEERI) · Equivalent continuous levels in peak hours, 2010; without-honking values from the statistical method.
Use this to check what you just read, with readings from your own street or a published survey.
How much of the noise is horns?
How much quieter would a no-honk day feel?
About a quarter to two-fifths less loud on the roads in the Nagpur study. Your ear does not judge energy directly. The named rule is Stevens' power law for loudness: perceived loudness grows roughly as sound pressure to the power 0.67, which works out to roughly a doubling of loudness for every 10 dB.
So the highway would sound 38% less loud, even though 79% of the energy vanished. On the evening highway, where the drop is only 2.8 dB, the street would sound just 18% quieter. That gap between energy and perception is one reason no-honk drives feel underwhelming even when they work.
Calculated in this post from the levels in Vijay et al. (2015), Table 2: 76.6 − 69.8 = 6.8 dB gives 1 − 2−0.68 = 38%, and 74.7 − 71.9 = 2.8 dB gives 1 − 2−0.28 = 18%, using the doubling-per-10-dB rule of thumb.

What does a honk actually say?
A honk is a one-bit broadcast: "I am here." Its job on Indian roads is to substitute for things the road does not reliably provide.
- A mirror substitute: on roads where two-wheelers filter through gaps narrower than a car, many riders never check behind them.
- An indicator: "I am overtaking on your right", on undivided roads where indicators are used inconsistently.
- A blind-corner warning: on ghat roads, where "Blow horn" signs ask for exactly this.
- A priority claim: at unsignalled junctions, where the louder or bigger vehicle tends to go first.
- Impatience: at a red light or in a jam, where it carries no safety information at all.
Why constant honking carries less information
Claude Shannon's information theory gives a clean way to see this. The information in a signal is its surprisal: the rarer it is, the more it tells you.
On the Nagpur highway, 63 honks in 900 seconds means a honk in any given second with \( p \approx 0.07 \), so each honk carries \( -\log_2 0.07 \approx 3.8 \) bits. On a quiet road where \( p = 0.001 \), the same honk carries about 10 bits. The more everyone honks, the less any single honk means, so drivers honk louder and more often to be noticed. That is a textbook arms race and a tragedy of the commons: each driver gains a little attention, and everyone pays in noise.
A horn that is always sounding is just more noise. The signal only works when it is rare.The working of this post
What would replace the horn on a silent day?
Lower speed, mainly. If you cannot tell the rider ahead to hold their line, you must be able to stop if they drift into your path. That turns the horn question into a stopping-distance problem.
The worked derivation: the no-horn passing speed
Assumptions: a road user ahead may move into your path at a gap \( d \); you need perception-reaction time \( t \) and then brake at deceleration \( a \). AASHTO design practice uses \( t = 2.5 \) s and \( a = 3.4\ \text{m/s}^2 \); I also test an alert city driver at \( t = 1.5 \) s.
- Stopping distance must not exceed the gap.
\[ v\,t + \frac{v^2}{2a} = d \]
- Solve the quadratic for the largest safe speed.
\[ v = -a\,t + \sqrt{(a\,t)^2 + 2\,a\,d} \]
- Plug in \( d = 15 \) m, \( t = 2.5 \) s, \( a = 3.4 \):
\[ v = -8.5 + \sqrt{72.25 + 102} = -8.5 + 13.2 = 4.7\ \text{m/s} \approx 17\ \text{km/h} \]
- Sensitivity. With \( d = 30 \) m it rises to 29 km/h; with the alert 1.5 s driver it is 22 km/h at 15 m and 36 km/h at 30 m.
| Gap to the road user ahead | Safe speed, 2.5 s reaction | Safe speed, 1.5 s reaction |
|---|---|---|
| 10 m | 12 km/h | 16.5 km/h |
| 15 m | 17 km/h | 22 km/h |
| 30 m | 29 km/h | 36 km/h |
The horn does not shorten your stopping distance. What it does is ask the other road user to not move, which lets you pass faster than your own stopping distance would justify. Remove it, and passing speed falls back to what physics allows. In dense city traffic that changes little, because speeds are already low; on a two-lane highway with overtakes past cyclists and tractors, it changes a lot. More on the road design side is in how speed breakers are designed in India.
What the road would need to say instead
- Mirrors on every two-wheeler, actually used. They exist to tell you who is behind you, which is the job the horn does now.
- Indicators and lane discipline, so intent is visible rather than audible.
- Physical separation of pedestrians and slow vehicles, so fewer surprises happen inside the 15 to 30 m zone.
- Signal timing that removes the reason to honk at red lights, which is what the Punishing Signal tried in reverse.

Would one quiet day help anyone's health?
One night would, measurably; one day mostly would not. The long-term evidence links road noise to heart disease, and short experiments show even a single noisy night affects blood vessels.
This review pulls together the epidemiology and the lab work. It reports the WHO expert panel's estimate that ischaemic heart disease risk rises by a relative risk of 1.08 (95% CI 1.01 to 1.15) per 10 dB(A) of road traffic noise, starting at 53 dB(A), and that at least 1.6 million healthy life-years are lost each year to traffic noise in Western Europe. In acute studies, healthy volunteers exposed to 30 or 60 train passes in a night (average 52 to 54 dB(A)) slept worse and showed impaired flow-mediated dilatation of the brachial artery compared with a 33 dB(A) control night. The limit: nearly all the data are European, at lower levels than Indian roads.
Read the review (free copy)Putting the numbers together
- Long-term, if the quiet lasted: a 4 dB drop gives \( 1.08^{-0.4} \approx 0.97 \), about 3% lower heart-disease risk from road noise for people living above the 53 dB threshold (calculated from the relative risk of 1.08 per 10 dB(A) in Münzel et al., 2021). Treat this as rough: the dose-response was built for Lden, not peak-hour Leq.
- One night: horns are already banned at night in residential areas under Rule 5A of the Noise Pollution (Regulation and Control) Rules; enforcing that one rule is the cheapest health gain in this post.
- Ears: Awaaz Foundation's Sumaira Abdulali told NPR in 2025 that horns can reach 120 dB or more; at that level hearing damage is about exposure time, not daily averages.
Need to complain about a specific horn or a night-time offender? The steps are in how to file a noise or honking complaint.
So could India go a day without honking?
Step by step, the case looks like this:
- Acoustics: horns are 42% to 81% of measured road noise energy, so a no-honk day would cut 2 to 7 dB.
- Perception: that sounds 18% to 40% quieter (calculated here with Stevens's power law), noticeable but not dramatic.
- Information: most honks today carry under 4 bits each, but some still prevent crashes.
- Safety: replacing those warnings means passing at roughly 17 to 36 km/h near other road users.
- Precedent: Mumbai's 2008 appeal failed; its 2020 signal-reset design targeted only the useless honks at red lights.
No law of physics needs the horn: a no-honk day is possible, and it would remove up to about 80% of the sound energy on a busy road. It fails in practice only when drivers keep normal speeds without the mirrors, indicators and lane discipline that make the horn's message redundant. Slow to about 17 to 36 km/h when passing people, as the stopping-distance maths requires, and the silence is safe. The cheapest permanent version is enforcing the existing night-time and red-light bans, where honks carry almost no information.
Calculated in this post from a 2.5 s perception-reaction time and 3.4 m/s² deceleration (design values described on Wikipedia: Stopping sight distance) and assumed gaps of 15 to 30 m, as worked above.
What's new
What if next?
- What if chai had never reached India? The roadside stall would still exist; what would be in the glass is the interesting part.
- What if aliens landed on an Indian highway? The horn would be the first word of road language they had to learn.
- What if India were still an island? No Himalaya, a very different monsoon, and a different everything after that.
More of these questions live on the Kyun page.
Questions people ask
Is honking illegal in India?
Not in general. Every vehicle must have a working horn, but the Noise Pollution (Regulation and Control) Rules ban horns in notified silence zones and at night in residential areas except in a public emergency, and police can fine unnecessary or excessive honking.
How much does honking add to traffic noise?
In CSIR-NEERI's Nagpur measurements, 2 to 5 dB(A) on average and up to about 7 dB(A) on a narrow road at peak time, which is 40% to 80% of the sound energy.
Did Mumbai's Punishing Signal work?
It reset the red light whenever noise at the junction crossed 85 dB, starting in January 2020 at five junctions. The only effectiveness figure published came from the campaign's own award entry, so there is no independent measurement of its effect.
Why do Indian drivers honk so much?
Because the horn substitutes for mirrors, indicators and lane discipline in mixed traffic, and because constant honking lowers the information in each honk, which pushes drivers to honk more. The longer answer is in the culture post linked above.
Would banning horns make roads more dangerous?
Only if speeds stayed the same. The stopping-distance maths says passing speeds near other road users would need to fall to roughly 17 to 36 km/h, which is already typical in Indian city traffic.
Sources
- Vijay et al., J Environ Health Sci Eng: Assessment of honking impact on traffic noise in Nagpur (2015) (source for all Nagpur levels and honk counts)
- Münzel, Sørensen and Daiber, Nature Reviews Cardiology: Transportation noise pollution and cardiovascular disease (2021) (source for RR 1.08 per 10 dB and the train-noise night study)
- Mail & Guardian (AFP): Mumbai drivers hoot through no-honking day (2008) (source for the story)
- The Tribune: Honk more, wait more (2020) (source for the Punishing Signal and 85 dB)
- Noise Pollution (Regulation and Control) Amendment Rules, 2010: Rule 5A (source for the horn restrictions)
- Wikipedia: Stopping sight distance (source for the 2.5 s and 3.4 m/s² AASHTO design values)
- NPR: India's honk-happy drivers are switching to even louder horns (2025) (source for 120 dB horns and 27 honks a minute)
- Free Press Journal: Gadkari proposes musical-instrument horns (2025)
- Wikipedia: Stevens's power law (source for the loudness exponent)
- Wikipedia: Information content (Shannon surprisal) (source for the information formula)

