Chanakya, or at least the Arthashastra tradition credited to him, would have understood your clutch, because its rule on applying force is the rule of the friction zone. Book 1, Chapter 4 says a ruler who punishes too severely is hated, one who punishes too mildly is despised, and one who punishes as deserved is respected. A clutch obeys the same shape: engage it too hard and the engine stalls or the bike lurches; slip it too long and the plates overheat; engage it in proportion to the load and you move smoothly.
- Book 1, Chapter 4: severe punishment is hated, mild is despised, deserved is respected.
- T = μFnr: clutch torque capacity; about 108 N·m in the worked example, 45 N·m at the low friction bound.
- Dump = stall or lurch; ride = slip and heat; feather to lock-up = proportion.
- ½mv²: the heat of a launch equals the kinetic energy gained, however gently you engage.
- 4.2 kW of heat while slipping 20 N·m at 2,000 rpm: long slips, not launches, burn plates.
Calculated in this post with these example inputs (20 N·m of slip torque, a 2,000 rpm speed difference, 209 rad/s) and power = torque × angular velocity (Wikipedia: Torque): 20 × 209 = about 4.2 kW.
The physics adds a twist the text never needed: how gently you let the lever out barely changes the heat of a launch, but how long and how fast you slip changes it a lot.
This is a "collision" post. One half is a text on statecraft compiled roughly two thousand years ago; the other is the multi-plate wet clutch in an Indian motorcycle. I will solve the clutch properly, read the text carefully, and then judge how far the link really holds.
What does the Arthashastra actually say about applying force?
It says force must be proportionate: too much alienates people, too little earns contempt, and force fitted to the case earns respect. The passage sits in Book 1, Chapter 4, on dandaniti, the science of government by the rod.
In 1905 R. Shamasastry, working at the Mysore Oriental Library (today the Oriental Research Institute in Mysuru), found it among a heap of palm-leaf manuscripts. A pandit from Tanjore district had handed the bundle to the library; it was written in early Grantha script.
Shamasastry transcribed and edited the Sanskrit text, published it in 1909, and brought out the first English translation in 1915. Indologists such as Julius Jolly, Moriz Winternitz and A. B. Keith took it up, and it was soon translated into French and German. Every quotation of "Chanakya" on punishment, including the one this post rests on, traces back to that bundle.
Wikipedia: R. ShamasastryThe chapter treats danda (punishment, the rod) as the support of the other branches of knowledge. Its central rule is a three-way contrast: a king who imposes severe punishment becomes repulsive to the people, one who imposes mild punishment becomes contemptible, and one who imposes punishment as deserved becomes respectable. Without any punishment, the chapter warns of the "law of the fishes", where the big fish swallows the small. Limit: a 1915 translation of a text with several authors and layers; wording differs across translations, and later critical editions number the verses differently.
Read the chapter (free)The text also lists four upayas, or means, at 2.10.47, conventionally discussed in this order of rising pressure:
- Sama: conciliation.
- Dana: gifts.
- Bheda: division.
- Danda: force, the last resort.
That ordering is often called graduated escalation, which is where the clutch comparison begins.

A short scholarly comment on how the four means are used and misused. Gautam notes that in Sanskrit literature the fourth upaya is invariably danda, argues that the means should be applied with context rather than as a mechanical template, and criticises scholars who attribute ideas to Kautilya that are not in the text, giving the example of a claim about wearing rebels down through delay. He recommends labelling one's own reading as a commentary (bhashya). Limit: an opinion piece on internal security, not a textual study.
Read the comment (free)Gautam's warning applies here. The Arthashastra says nothing about machines.
What follows is my bhashya: a comparison, not a claim that Kautilya anticipated friction physics.
How does a motorcycle clutch carry torque?
A wet multi-plate clutch carries torque through friction between a stack of plates squeezed together by springs, and its capacity is the product of four numbers. Marcin Bąk's 2021 paper in Scientific Reports gives the standard formula.
- \( \mu \): the friction coefficient between a friction plate and a steel plate.
- \( F \): the spring clamping force.
- \( n \): the number of friction surfaces.
- \( r_m \): the mean radius of the friction ring.
Pull the lever and \( F \) falls toward zero; let it out and \( F \) rises. The friction zone is the range of lever travel where \( F \) is high enough to transmit some torque but too low to lock the plates together.
Tests of a multidisc wet clutch with 10 to 18 friction surfaces (discs of 80 to 115 mm radius), over 200 engagements. Bąk uses \( T = r_m n \mu F_a \) and bounds wet friction coefficients between 0.05 and 0.12. Friction in the splines that let plates slide means the plates farther from the pressure plate are squeezed less, so adding plates gives diminishing returns; a published correction (Osiński) scales capacity by 1.00 at 3 surfaces down to 0.76 at 11, though Bąk found that correction underestimates measured torque. Limit: one disc design in a test rig, larger than a motorcycle clutch.
Read the paper (free)See the data as a table
| Correction factor on ideal torque (Osiński) | |
|---|---|
| 3 | 1.0 |
| 4 | 0.97 |
| 5 | 0.94 |
| 6 | 0.91 |
| 7 | 0.88 |
| 8 | 0.85 |
| 9 | 0.82 |
| 10 | 0.79 |
| 11 | 0.76 |
Source: Bąk, Scientific Reports (2021), Table 1 · x-axis: number of friction surfaces. Bąk reports that measured torque exceeded this correction.

A worked capacity estimate. Assume (my numbers for a small motorcycle clutch) \( \mu = 0.12 \), spring force \( F = 1{,}500\ \text{N} \), \( n = 12 \) friction surfaces and \( r_m = 0.05\ \text{m} \).
With Bąk's lower friction bound of 0.05, the same clutch carries only 45 N·m. Anything that changes \( \mu \), such as oil condition or glazed plates, changes the whole capacity in proportion.
Why does dumping the clutch stall the engine?
Because the clutch suddenly demands more torque than the engine can supply at low revs, so the engine slows to a stop. This is the "too severe" case.
If the clutch clamps fully in a fraction of a second, its torque capacity (above, around 100 N·m) far exceeds what an idling engine produces. The clutch then tries to bring the whole bike and rider up to engine speed almost instantly.
Newton's second law for rotation, \( T_{\text{engine}} - T_{\text{clutch}} = J_e\,\dot\omega_e \), says the engine's speed \( \omega_e \) must fall when clutch torque exceeds engine torque. If it falls to zero before the bike catches up, the engine stalls; if the engine has more revs, you get a lurch, a wheelie or a spin instead.
Kautilya's word for the result of severe punishment is repulsive, and anyone who has been on the pillion of a clutch-dumper will agree.
Does a gentle clutch make less heat?
Surprisingly, no, not for a launch at steady engine speed: the heat made while the clutch slips equals the kinetic energy the bike gains, however slowly you let the lever out. This is the part the analogy does not predict.
Assumptions. Engine held at constant speed \( \omega_e \); clutch transmits a constant torque \( T_c \) while slipping; the bike's inertia, referred to the clutch, is \( J \); the bike starts at rest; drag is ignored.
- Time to lock up. The driven side accelerates at \( T_c/J \) until it reaches \( \omega_e \): \( t_s = J\omega_e/T_c \).
- Energy from the engine. The engine side turns at \( \omega_e \) the whole time: \( E_{\text{in}} = T_c\,\omega_e\,t_s = J\omega_e^2 \).
- Energy into motion. \( E_{\text{kin}} = \tfrac12 J\omega_e^2 \).
- Heat in the clutch. \[ Q = E_{\text{in}} - E_{\text{kin}} = J\omega_e^2 - \tfrac12 J\omega_e^2 = \tfrac12 J\omega_e^2 = E_{\text{kin}} \]\( T_c \) has cancelled out: the gentleness of engagement does not change the total heat, only how long it takes.
- Numbers. A 200 kg bike and rider (assumed) reaching 15 km/h (4.17 m/s) at lock-up: \( Q = \tfrac12 \times 200 \times 4.17^2 = 1{,}740\ \text{J} \). If that heat went only into 0.6 kg of steel plates (assumed; specific heat 490 J/kg·K), they would warm by \( 1{,}740/(0.6\times490) \approx 6\ \text{K} \). Oil carries much of it away.
Calculated in this post with these example inputs (200 kg bike and rider, 15 km/h at lock-up, 0.6 kg of steel plates at an assumed 490 J/kg·K; Wikipedia's table gives 466 J/kg·K for steel in general) and kinetic energy ½mv² (Wikipedia).
What does change the heat is the speed at lock-up and how long you hold the clutch half-in. Slipping at a constant speed difference is a pure heater:
That is 20 N·m (assumed) at a 2,000 rpm slip, held, say, while creeping uphill in traffic. Ten seconds of it is 42 kJ, enough to heat 0.6 kg of steel by about 140 K if the oil could not carry it away.
Calculated in this post with these example inputs (20 N·m, 2,000 rpm, 10 seconds, 0.6 kg of steel at an assumed 490 J/kg·K) and power = torque × angular velocity (Wikipedia: Torque): 4.2 kW × 10 s = 42 kJ, which warms the steel by 42,000 / (0.6 × 490) = about 140 K.
That is "too mild" in Kautilya's terms: you are not committing, you go nowhere, and you pay for it in burnt plates. My own 2017 NS200 has had its clutch plates replaced over its life, and my KTM 390 Adventure has shown clutch judder; both are reminders that friction material is a consumable.
Use this to check what you just read. Compare a clean launch with a long slip.
Clutch heat: clean launch vs riding the clutch
Where does the Kautilya comparison hold, and where does it break?
| Arthashastra | Clutch | How far it holds |
|---|---|---|
| Severe punishment: the people find the king repulsive | Dumped clutch: torque spike, stall or lurch | Holds: excess force causes a violent reaction |
| Mild punishment: the king is despised | Ridden clutch: slip without progress, heat | Holds: too little commitment costs energy and gains nothing |
| Punishment as deserved: respected | Force matched to load: smooth lock-up | Holds: proportionality is the design rule |
| Four upayas, rising force, danda last | Release from light contact to full clamp | Partly: escalation fits, but a clutch has one input, not four kinds |
| Law of the fishes without danda | No clamp: engine spins, wheel does not | Weak: a disengaged clutch is neutral, not chaos |
| Gentleness reduces harm | Launch heat equals ½mv² regardless of gentleness | Breaks: physics says total heat is fixed; only timing and slip duration change |
The last row is where engineering corrects intuition. A gentle hand does not save energy on a launch; it spreads the same heat over more time. The real damage comes from slipping for long periods or at high revs, which is closer to dithering than to mildness.
Kautilya's principle of proportion survives; the moral of "gentle is always kinder" does not. The text's broader worldview sits with other early Indian thought in the oldest Indian texts on the mind and its ethics of action with Stoicism and the Gita. For the hardware itself, the motorcycle side is covered properly in how a motorcycle clutch works.

What would a Kautilyan riding rule look like?
Match the clamp to the load, commit quickly once the bike moves, and never sit in the friction zone for long. In numbers, from the model above:
- Launch at low revs. Heat in a launch scales with the square of lock-up speed; locking up at 10 km/h instead of 15 km/h cuts it from about 1,740 J to 770 J.
Calculated in this post with these example inputs (200 kg, lock-up at 15 km/h or 10 km/h) and kinetic energy ½mv² (Wikipedia): 1,740 J and 770 J.
- Keep slips short. Slip power is torque times speed difference; at 4.2 kW, every second costs about 4 kJ.
Calculated in this post with these example inputs (20 N·m, a 2,000 rpm difference) and power = torque × angular velocity (Wikipedia: Torque): 4.2 kW is 4.2 kJ per second.
- Use the oil your manual specifies. Capacity is proportional to \( \mu \), and in a wet clutch the oil is part of the friction pair; Bąk's 0.05 to 0.12 range shows how much the coefficient can vary.
- Escalate in order. Like sama before danda: take up slack, find the bite, then add clamp. Do not jump straight to full force.
Engineers have now automated this judgement. Honda's E-Clutch, launched in India on the CB650R and CBR650R in May 2025, controls the clutch of a manual gearbox automatically for starts and shifts, while letting the rider use the lever if they want.
Kautilya's rule that severe force is hated, mild force despised and deserved force respected maps cleanly onto stall, slip and smooth lock-up in a clutch with capacity \( T = \mu F n r_m \). But the physics adds a correction the text could not: for a launch at steady revs, the heat made equals the bike's kinetic energy, about 1,740 J at 15 km/h for 200 kg, however gently you engage. What burns clutches is long or high-rev slipping, 4.2 kW in my example. Proportion is the shared law; "gentle is harmless" is not.
Calculated in this post with these example inputs (200 kg, 15 km/h, 20 N·m, 2,000 rpm) with kinetic energy ½mv² (Wikipedia) and power = torque × angular velocity (Wikipedia: Torque): 1,740 J for the launch; 4.2 kW for the slip.
What's new
- MP-IDSA issue brief argues the popular image of Kautilya as a ruthless schemer needs deconstruction MP-IDSA, Jan 2025
- Honda brings E-Clutch, automatic clutch control for a manual gearbox, to India on the CB650R and CBR650R Honda Motorcycle and Scooter India, May 2025
- How the Arthashastra paired harsh penalties with limits on royal power ThePrint, Sep 2024
What if next?
- A traffic signal is a meditation timer Where you will spend most of your clutch-in time, and what it could do for your nervous system.
- Why worry and rust are the same process Slow, constant damage at the surface, just like a slipping clutch.
- Kabaddi is a lesson in breath control Another Indian lesson in committing at exactly the right moment.
Questions people ask
Did Chanakya write the Arthashastra?
The text names its author as Kautilya and Vishnugupta; the identification with Chanakya comes from later tradition. Many modern scholars, including Patrick Olivelle, see a multi-layered text compiled between roughly 150 BCE and 300 CE.
What does the Arthashastra say about punishment?
Book 1, Chapter 4 says severe punishment makes a ruler repulsive, mild punishment makes him contemptible, and punishment as deserved makes him respected. Without punishment, it warns, the strong devour the weak like fish.
Does slipping the clutch damage it?
Long or high-rev slipping does, because slip turns engine power straight into heat: about 4.2 kW at 20 N·m and a 2,000 rpm difference. Brief slipping to pull away is what the clutch is designed for.
Calculated in this post with these example inputs (20 N·m, a 2,000 rpm difference) and power = torque × angular velocity (Wikipedia: Torque): about 4.2 kW.
What are the four upayas?
Sama (conciliation), dana (gifts), bheda (division) and danda (force), listed at Arthashastra 2.10.47, with danda traditionally the last resort.
Sources
- Kautilya Arthashastra, Book 1, Chapter 4, trans. R. Shamasastry (wisdomlib) (source for severe, mild and deserved punishment and the law of the fishes)
- Wikipedia: Upayas (diplomacy) (source for 2.10.47 and the order of the four upayas)
- Wikipedia: R. Shamasastry (source for the story: the 1905 find at the Mysore Oriental Library, the Tanjore pandit, the Grantha script, the 1909 edition, the 1915 translation and the scholars who took it up)
- Wikipedia: Arthashastra (source for authorship, Olivelle's dating, the 1905 manuscript and Shamasastry)
- Gautam P.K.: Understanding Kautilya's Four Upayas, IDSA Comment (2013) (source for careful attribution and the bhashya point)
- Bąk M.: Torque capacity of multidisc wet clutch, Scientific Reports (2021) (source for T = μFnr, μ 0.05-0.12, Osiński coefficients)
- Mishra S.: Imaging Kautilya beyond popular stereotypes, MP-IDSA Issue Brief (2025)
- Honda Motorcycle and Scooter India: 2025 CB650R and CBR650R with E-Clutch (2025)
- ThePrint: How the Arthashastra guided ancient kings on crime and dissent (2024)
More on the science of the mind on the Manthan world page.
