Honking in India

Two weeks ago at around 6 am, I landed in Bangalore after a year in Munich. I was still sleepy and the first 20 kilometres into the city in the taxi were peaceful, so I drifted off to sleep. I soon awoke however to a hallmark of Indian traffic — incessant honking. If you close your eyes anywhere close to an Indian road and listen, you’ll notice that not a second goes by without it being punctuated by a honk. I smiled to myself because it was a reminder that I was back in India.

But this behaviour of ours stops being charming after the first ten minutes in traffic. Why do we honk so much on our roads? The question is not a new one by any means. The accepted theory is that Indian roads are inherently unsafe if you don’t honk because drivers don’t follow traffic rules. This Forbes article, for example, suggests that the “complete disregard for side mirrors” is the reason for the honking. Accepting this theory, it’s hard to imagine the situation changing unless the entire traffic culture in India changes — a near impossibility.

I drove on Indian roads for a good five years before moving to Munich and I’m happy to report that I used my horn a grand total of about ten times. This is what makes me question the theory that honking is just something that is necessary in India. I barely ever honked and I’m still alive to tell the tale, so there must be more to this story. Moreover, if you’ve ever waited at a traffic signal and heard drivers start honking the second the light turns green, you’ll know that it is driven as much by impatience as by uncertainty.

However, Briefcase, a behavioural design firm based in Mumbai, have something very interesting to say about the issue. In his TED Talk, Anand Damani of Briefcase says that honking in India is an “automatic behaviour[…] where one doesn’t consciously think about the action”. Using the terms popularised by Daniel Kahneman in his book Thinking, Fast and Slow, honking in India — like almost all of driving — occurs using System 1 (the fast, automatic and unconscious mode of thinking) instead of System 2 (the slower, more deliberate mode of thinking). So to make drivers think about honking more consciously, they installed a red button on the dashboard that beeps and flashes with a frowning face after every honk. The driver would have to press the button to turn off the beeping, thereby engaging System 2 while giving the driver a taste of their own medicine. They call this solution “Bleep”. In their tests, they claim that the button reduces honking behaviour by about 60% — a substantial improvement. Digging deeper, I was left with more questions than answers. Where does this habit come from? Why does the trigger of feeling unsafe on Indian roads lead to the action of honking?

As we drove through the city, a hypothesis was forming in my head. I watched the taxi driver closely as we waited at a traffic signal. As expected, he let out a short double-burst as soon as the signal turned green. I pounced at the opportunity. “Why did you honk just now?”

His reply made me sit up and open the notes app on my phone:

“See! The fellow started moving as soon as I honked.”

What he said reminded me of a folk tale about an old woman and her rooster. Every morning, the rooster crows and the sun rises immediately after. One day, after a fight with the villagers, she decides to teach said villagers a lesson and kills her rooster. In her mind, she has punished the village by plunging it into eternal darkness, but the run rises as usual, leaving her red-faced and a good rooster short. The moral of this story is made out to be the perils of arrogance but I think the tale is additionally a warning against superstition.

Bleep by Briefcase (Image is a screenshot from the webpage)

Just as the old woman mistook the crowing of her rooster as the reason for the sun rising, so do Indian drivers in most cases mistake the honking as the reason for the car ahead of them moving. The car would have moved regardless of whether the driver had honked or not, but they nonetheless revel in this illusion of control. If we consider the old woman’s behaviour superstitious, isn’t honking at a signal just as superstitious?

While it might appear like there isn’t that much of a difference in thinking about honking in this way, crossing the line from habit to superstition does open some interesting conversational doors. For example, JL Risen gives us this insight into how superstitions are formed or acquired.

Acquiring superstition, Risen 2016.

In our lives, we come across many things that we cannot immediately explain. But we’re not happy with this situation. We like explaining things to ourselves and we like feeling in control, so we often intuitively jump to a conclusion that gives us the illusion of control. When this “magical intuition” of System 1 is triggered, System 2 can either realise the error or let it slip. If System 2 catches the error, the misconception is corrected. But if System 2 doesn’t realise that a mistake has been made, the superstition is confirmed and endorsed. Unfortunately, System 2 is not infallible and it is not helped by System 1, which is remarkably adept at leading System 2 astray. One way it does this, is by substituting a hard question with an easier question that it knows System 2 can answer. In the case of honking, we tend to answer the question “did the car move after I honked?” instead of the much harder “did the car move because I honked?” System 2 is thus fooled into a false causality.

When the superstition has been accepted once, it becomes that much more likely that the superstition will be reinforced the next time. This mechanism is reminiscent of the mechanism of habit-forming that Nir Eyal proposes. His Hook Model involves a Trigger causing an Action; the Action leads to a Variable Reward; this Reward makes us Invest in the action and makes perform the Action the next time the Trigger comes around. But in the case of superstition, instead of a Reward, there is the Desired Result that we think we caused, and instead of Investment, the superstition is reinforced.

Based on this, one “cycle” of my taxi driver’s honking behaviour at a traffic signal might look something like this: 
TRIGGER: The light turns green.
ACTION: He honks.
DESIRED RESULT: The car ahead of him moves. 
REINFORCEMENT: He believes that the car moved because he (the taxi driver) honked. This drives the next cycle of the action.

At this point, I feel like I should back up and address the big assumption that I’ve made before writing all this: Honking doesn’t actually affect the level of safety or certainty on the road. This is heavily influenced by my own experience of Indian traffic. The way I see it, even if honking does have its desired effect to a certain extent, there must be something differentiating me, and others like me who honk very less, and the others who do honk. I might have a much higher threshold for uncertainty than the average Indian does — which means that the average Indian driver is very “honk-happy”.

If we define “honk-happiness” (from trigger happiness) as this threshold that the uncertainty has to cross before the driver honks, I imagine a graph of “road uncertainty” of a particularly eventful drive to look something like this (assuming that drivers on European roads are usually surer about what’s around them than drivers on Indian roads): (see graph)

Additionally, if this is true, based on the reinforcing model of superstition-forming, the superstition should get stronger as the driver becomes more experienced. An interesting question to ask of the data collected by Briefcase is if the beginner drivers and more experienced drivers behaved differently before and/or after the Bleep button was added.

Testing these hypotheses, and quantifying “honk-happiness” and “road uncertainty” appears to be tricky because of the setting of the behaviour and the number of variables involved. A fun experiment to conduct would be to mute a car’s horn on the outside the car, but simulate the sound of the horn inside so that it seems to the driver as if the horn is working, then have them drive a route without telling them about the modification to see if their experience changes in any way. But this is obviously out of the question because of how unsafe it would be if this hypothesis were false. Collecting a lot of data (about where drivers honk the most, when cars move in relation to the timing of honks, etc.) and running some sort of simulation of one entire route with different honking parameters is safer but the technical difficulty of this is just as prohibitive as the safety considerations of a partially muted car.

Here’s what Risen goes on to say about superstition:

[…] If a magical intuition comes to mind — for example, “this is my lucky seat for watching football” — and System 2 does not become engaged, then even a fan who is not explicitly superstitious will sit in the lucky seat and feel more optimistic about winning. If System 2 processes are engaged, however — if, say another person is already sitting in the fan’s lucky seat — he may be forced to confront his magical intuition. Furthermore, if he recognizes that it is irrational, he will override the intuition and sit somewhere else.

Let’s use this as a lens to think about Bleep. In the example above, the superstitious person is made to think consciously BEFORE carrying out the superstition-driven action. But in the case of Bleep, the driver is made to think consciously AFTER he has already honked. Yes, turning off the Bleep button after the previous honk is remembered to a certain extent, but there is no time, or need, to pause between the TRIGGER and ACTION of the next cycle. While Bleep has been really successful, it might build on its success if it is slightly modified to include an extra step pre-honk.

Despite the success of its initial trials, Damani says that Bleep is difficult to implement at a scale where it will actually make a difference. Very few people will volunteer to get it installed in their cars and bikes. The alternative, he says, is to get the government to make the system mandatory in cars, just as the seatbelt now is. I have no idea how much success they’ve had till now in convincing the authorities, but the Indian government’s track record isn’t great when it comes to implementing and enforcing changes like this. Riding motorcycles without wearing helmets has been illegal in India since 1988, but only about 35% of bikers wear helmets even now. This doesn’t inspire much confidence in the power of the authorities to effect change.

Maybe turning our attention to the root causes of this habit might help. It is still not clear why the driver would honk BEFORE it has turned into habit, and in turn, superstition. After all, we’re not born with some biological instinct to reach for the horn if we feel unsafe. It is probably a combination of these factors:

• We are taught by parents and siblings and driving school instructors to aggressively honk at every uncertainty, lowering our uncertainty threshold to around the average value of the traffic system.

• We hear the other drivers honking on the roads and assume that it’s convention.

• It’s too easy to honk, the way steering wheels are often designed. In other words, the horn is right there.

Out of these, the way driving is taught in India might be the epicentre of our honking problem. The student is taught to not have any trust in the traffic system and this every-man-for-himself attitude makes the system worse. Which, in turn, makes the argument for honking stronger — in a classic case of the Pygmalion Effect. By getting driving instructors to teach students to honk less — or only when absolutely necessary — we might nip the habit in its bud. My humble suggestion to Briefcase would be to try and install Bleep systems in the cars used by driving schools. This way, novice drivers as well as driving instructors are made to think consciously about honking. By doing so, drivers will hopefully have the awareness and the discipline to stop themselves from reciprocating the honks of others on the road.

I understand that excessive honking is not India’s biggest problem right now. Even after (exactly) 73 years of independence, there are several economic, social and communal issues that plague the country. But studying and solving the issue of honking might give policy and system designers — and citizens — a better insight into the several other negative actions that are unconsciously and “mindlessly” carried out in the country. After all, a mindful India is a better India.

Sources

• Kahneman, D., & Tversky, A. (1982). On the study of statistical intuitions. Cognition, 11, 123–141. http://dx.doi.org/10.1016/0010-0277(82)90022-1

• Kahneman, D. (2011). Thinking, fast and slow.

• Eyal, Nir 2013. Hooked: How to Build Habit-Forming Products.

• Risen, JL (2016). Believing What We Do Not Believe: Acquiescence to Superstitious Beliefs and Other Powerful Intuitions. Psychological Review, Vol. 123, №2, 182–207. http://dx.doi.org/10.1037/rev0000017

• Wadhwaniya, et al (2016). A comparison of observed and self-reported helmet use and associated factors among motorcyclists in Hyderabad city, India. Public Health, Vol. 144, Supplement, March 2017. [https://doi.org/10.1016/j.puhe.2016.11.025]

Read this article and others on Medium: @bawakulkarni