Runaway Train


I decid­ed to add this pro­logue after the first few com­ments to this post. This post uses an inci­dent in India, but is actu­al­ly uni­ver­sal in nature and focus­es on the moral, philo­soph­i­cal, and eth­i­cal deci­sion-mak­ing involved in an emer­gency.

Main Post

Imag­ine you’re trav­el­ing from Mum­bai to Pune by train, which is full to capac­i­ty, as usu­al in India. An addi­tion­al engine is added to the train to climb the ascent of the West­ern Ghats from Kar­jat at sea-lev­el to Lon­avala at a height of 2000 ft. above sea lev­el. Your train trudges labo­ri­ous­ly upwards and reach­es Lon­avala after 1.5 — 2 hours. You enjoy the beau­ti­ful scenery of the Sahyadri ghats. It stops at Lon­avala for a while and every­one gets back on board, ready to pro­ceed.

 Indrayani2 Sud­den­ly the train starts inch­ing back­wards. There are smiles, gig­gles, and wise­cracks about what antics the dri­vers are up to. Some won­der if they’re sim­ply chang­ing tracks or if some engine replace­ment or some­thing had to be done. The ‘inch­ing’ turns into ‘crawl­ing’, and soon enough, omi­nous­ly, the train is now real­ly ‘mov­ing’ back­wards. There is puz­zle­ment all around and you are amused as to what’s hap­pen­ing.

There is no let up how­ev­er, as the train starts get­ting momen­tum, accel­er­ates fur­ther, and starts gain­ing speed. Amuse­ment dis­ap­pears as you and every­one else real­ize that some­thing is seri­ous­ly wrong. The train gains fur­ther accel­er­a­tion and you’re already cruis­ing at a rea­son­able speed. Every­one is peer­ing out the com­part­ment doors and win­dows only to find peo­ple from oth­er com­part­ments doing the same. “Has the dri­ver lost his mind?” you won­der, as peo­ple start voic­ing obscen­i­ties at the train staff.

But, was the staff (dri­ver and guard at oppo­site ends), on the train when it start­ed off at Lon­avala?” some­one asks and nobody real­ly knows. The worst pos­si­bil­i­ty comes to your mind — you’re on a run­away train, down­hill, with no one at the con­trols. Indrayani1

By this time, the train is so fast that it would be dan­ger­ous to jump off. Pan­ic and con­fu­sion all around you. You calm your­self and start think­ing rapid­ly. You visu­al­ize the labo­ri­ous twists and turns of the track as it winds down the moun­tains. You imag­ine a full-speed, no holds barred, run­away train hurtling across those tracks and over­turn­ing into the pic­turesque Sahyadri val­leys. Is this how you were des­tined to die?

Point A: Question 1

At this point, if you jumped off, you assess your chances. Let’s say there’s a 70–80% prob­a­bil­i­ty that you’ll get seri­ous­ly hurt, and a 20–30% pos­si­bil­i­ty that you might die in the process. Will you jump off?

Point A: Question 2

Assume you don’t, and cling on to hope, that there will be some mirac­u­lous inter­ven­tion and that you will be saved. After all, when one lives in a civ­i­lized and mod­er­ate­ly devel­oped soci­ety, it is a ratio­nal expec­ta­tion that there will be sys­tems and process­es in place to deal with such emer­gen­cies.

Some peo­ple are seri­ous­ly doubt­ful how­ev­er. They’re con­tem­plat­ing jump­ing off. Will you dis­cour­age and/or pre­vent peo­ple from doing so?

Mean­while, the train has reached a break­neck speed. The sparks from the wheels are now of alarm­ing pro­por­tions and reach­ing the win­dows. Peo­ple from anoth­er com­part­ment come rush­ing into yours as their com­part­ment catch­es fire. The ghat sec­tion, where the real twists and turns begin, is just around the cor­ner. Peo­ple are scream­ing, women are cry­ing in hys­te­ria.

Point B: Question 1

At this point, there’s an almost 100% prob­a­bil­i­ty of seri­ous injury, includ­ing per­ma­nent hand­i­cap, and a 70% prob­a­bil­i­ty of death. Will you jump?

Point B: Question 2

Assume you don’t, and still have hope that you will be saved. How­ev­er, there are peo­ple who are get­ting ready to jump. Will you discourage/prevent them, just because you have hope even if they haven’t?


The above sit­u­a­tion is not hypo­thet­i­cal. This is what hap­pened to the Indrayani Express in the 1990s, when my cousin broth­er was on the train. Dur­ing a nor­mal return jour­ney from Pune to Mum­bai (down­hill), the train used to descend the height of the ghat sec­tion in approx­i­mate­ly an hour. That day, it ran the same track down­hill in 11 min­utes. The train did not over­turn. Few peo­ple who jumped off were seri­ous­ly injured. There were no major casu­al­ties. My broth­er urged dozens of peo­ple not to jump and end­ed up sav­ing them in the process.

(Pho­to Cred­its: the Indi­an Rail­ways Fan Club)
(Title of Post: Run­away Train by Soul Asy­lum)

This entry was posted in India, Personal, philosophy, psychology, pune, Travel and tagged , , , , , , . Bookmark the permalink.
  • wow it was sen­si­ble of ur broth­er
    final­ly its the indi­an rail

    ps i cant believe that u haven’t been fea­tured on the dash­board!

  • Very inter­est­ing posit…
    I would jump if the grass looks soft and clean… and not if there were a riv­er.…
    most­ly I would be in a fetal posi­tion, near tha fat­test guy/gal in the train!

  • I wouldn’t jump. Tech­ni­cal­ly the tracks have an angle of bank­ing that sup­ports max­i­mum speed of the train, so top­pling ‘ide­al­ly’ wouldn’t be pos­si­ble and I’d stick to this.

    Sec­ond­ly I wouldn’t try to alarm oth­ers either, thats because my ide­al­ist the­o­ry (in which I believe) is quite shaky in real­i­ty (maybe the design was faulty, maybe they don’t main­tain the tracks or car­riages etc.)

    That makes me curi­ous. What exact­ly did your brave cousin say to stop oth­ers? Emo­tion­al appeal, or tech­ni­cal rea­son­ing?

  • I’m just curi­ous about the brake chains that are in each com­part­ment of a bogey. Do they only work when the engines are run­ning? Or did no one think of pulling them?

  • Vora­cious Blog Read­er


    Depends whether I am alone or not. If I am alone, I would jump. If I sur­vive, the oth­ers may decide for their own.

    If I am not alone i.e., with my fam­i­ly. I would rather stay behind.

    Vora­cious Blog Read­er

  • Prax: I’m not so sure about the ‘sen­si­ble’ part, regard­less of what was the out­come in that par­tic­u­lar inci­dent. And no…almost every­one I know has been fea­tured on the Word­Press home­page, but as far as I know, not me.

    Ram­bodoc: Apart from the humor, I would like to have your views about the ethics/moral issues involved espe­cial­ly from the view­point of a per­son engaged in med­ical prac­tice. If there were emer­gency med­ical team mem­bers on board, how would they have respond­ed to this sit­u­a­tion? How would you like med­ical emer­gency per­son­nel (say, who were trav­el­ing as ordi­nary pas­sen­gers on board) to respond in such a sit­u­a­tion?

    While I appre­ci­ate your humor, I wouldn’t want you to miss the grav­i­tas of my post. Because I val­ue your opin­ion high­ly.

  • Priyank: //I wouldn’t jump. Tech­ni­cal­ly the tracks have an angle of bank­ing that sup­ports max­i­mum speed of the train, so top­pling ‘ide­al­ly’ wouldn’t be pos­si­ble and I’d stick to this.//
    I’m not sure how you’re sure about the angle of bank­ing sup­port­ing a run­away train. Any more info would be help­ful. I checked out what­ev­er resources I could, but couldn’t find any. If the ‘max­i­mum’ speed of the train you’re refer­ring to actu­al­ly refers to nor­mal con­di­tions, that is inap­plic­a­ble in this sit­u­a­tion.

    //Secondly I wouldn’t try to alarm oth­ers either, thats because my ide­al­ist the­o­ry (in which I believe) is quite shaky in real­i­ty (maybe the design was faulty, maybe they don’t main­tain the tracks or car­riages etc.)//
    I think you meant “I wouldn’t try to calm oth­ers either”. And that’s exact­ly what my ques­tion is about.

    //That makes me curi­ous. What exact­ly did your brave cousin say to stop oth­ers? Emo­tion­al appeal, or tech­ni­cal reasoning?//
    Think mob con­trol. How does one con­trol a sui­ci­dal mob? We have sui­ci­dal bombers, who’re brain­washed. Think about con­trol­ling a sui­ci­dal mob out to harm them­selves in the inter­est of self-preser­va­tion. Do you think tech­ni­cal rea­son­ing works in such sit­u­a­tions? For­get tech­ni­cal, do you think ratio­nal rea­son­ing will work?

  • Priyank: By the way, thanks for being the first one to answer my ques­tions in the post!

    Amit: For some rea­son, the emer­gency brakes failed. Every­one tried pulling all the chains, includ­ing chants of “1, 2, 3, go”, but it didn’t work.

    Vora­cious Blog Read­er: //Depends whether I am alone or not. If I am alone, I would jump. If I sur­vive, the oth­ers may decide for their own. If I am not alone i.e., with my fam­i­ly. I would rather stay behind.//

    VBR: You raise anoth­er sig­nif­i­cant com­pli­cat­ing fac­tor out here regard­ing fam­i­ly!

    If you were alone, at which point would you jump?

    If you’re with your fam­i­ly, why would you pre­fer to hold on?

  • I have added a Pro­logue to this post after all these com­ments, because I think that I didn’t get through with the ini­tial post.

    The “Ques­tion No. 1“s are of a per­son­al, indi­vid­ual nature. The “Ques­tion No. 2“s are wider in scope.

    The essence of Ques­tions #2 is this: do we have a right to impose our val­ue-judge­ments of a cer­tain sit­u­a­tion on oth­ers? I mean, it is fine if you think you’ll ulti­mate­ly be saved from this dis­as­ter, but when all facts point oth­er­wise, do you have a right to dis­suade oth­ers from choos­ing their own deci­sions regard­ing what risk they would like to take?

    No one has respond­ed to Ques­tion 2 (with the excep­tion of Prax, indi­rect­ly). Con­trast this with Ques­tion 1.

    Come on, you intel­li­gent folks — let’s get your respons­es and dis­cuss! 🙂

  • VRB: I was dis­tract­ed by your point of fam­i­ly so much, that I com­plete­ly dis­re­gard­ed the impor­tant part of your response — I apol­o­gize.

    You said if you were alone you would jump. We have our first response, who says he would have jumped. Thanks! VBR: At which point? A or B?

    If you ask me, (some­how no one has), I would’ve indeed liked to jump off as ear­ly as pos­si­ble.

  • Point A Ques­tion 1: I would jump off. Pri­mar­i­ly because I do not have pre­cise phys­i­cal knowl­edge about the trac­tion lev­els of train wheels and tracks. So I could­nt be sure about the train not derail­ing at some point. The only bit of knowl­edge I have would be the instances where trains have derailed. So I will choose a fair­ly grassy area and jump off

    Point A Ques­tion 2: Very hard ques­tion. But I think I would urge those who look phys­i­cal­ly fit to jump off.

    Point B Ques­tion 1: I would not jump off. While my knowl­edge of rail­way tracks and train wheels is still incom­plete, I at least know for sure that jump­ing will prob­a­bly kill me. So I would rather take a chance with an unknown prob­a­bil­i­ty.

    Point B Ques­tion 2: Again, my knowl­edge about the dan­gers of jump­ing off are clear­er at this point. So I will advise peo­ple (per­haps not force) not to jump off and take their chances. That might be the log­i­cal thing to do.

    Your ques­tion on the cor­rect­ness of impos­ing our jump/no jump opin­ion on oth­ers is very intrigu­ing. I would say that it is cor­rect to let peo­ple know the facts of what you do know and impor­tant­ly what you dont know. So I would tell them that I think its a safer bet not to jump at Point A because of so-and-so facts but I would try and ask for a quick consensus/gut feel­ing of group. Some­times col­lec­tive intel­li­gence works under the right con­di­tions. I think its impor­tant not to impose val­ue judge­ments in such sit­u­a­tions, but instead bring your cards to the table and try to max­i­mize col­lec­tive intel­li­gence. I do real­ize that its hard to do in a crisis/high ten­sion sit­u­a­tion, but I cant think of any­thing bet­ter.

    Or ofcourse, I could ask Scot­ty to beam me and the rest of the guys up to the Star­ship Enter­prise.

  • Ashok: Thanks for your excel­lent com­ment!

    You (once again) echoed my thoughts. I would’ve jumped off at Point A, and if for some rea­son I missed that oppor­tu­ni­ty, I would’ve held on till the end.

    //Your ques­tion on the cor­rect­ness of impos­ing our jump/no jump opin­ion on oth­ers is very intriguing.//
    Thanks. That was one of the main points of this post. I think you’ve expressed my thoughts bet­ter than I would’ve, except for the col­lec­tive intel­li­gence part.

    I would’ve brought my cards to the table and urged oth­ers to do the same, but in life-threat­en­ing con­di­tions, each one should think indi­vid­u­al­ly.

    How­ev­er, in emer­gen­cies peo­ple also tend to aban­don indi­vid­ual judge­ment and adopt a herd men­tal­i­ty. That’s the insid­i­ous nature of ‘pan­ic’ — where you aban­don your judge­ment and sim­ply do what every­one else is doing.

  • True,
    although I think col­lec­tive intel­li­gence is very fre­quent­ly con­fused with herd men­tal­i­ty. Col­lec­tive intel­li­gence is infact the result of the appli­ca­tion of pure­ly indi­vid­ual intel­lect under cer­tain con­di­tions where the side effect tends to be col­lec­tive intel­li­gence. Ill give you an exam­ple. A few years back, a group of guys searched and locat­ed a miss­ing ship off the coast of cal­i­for­nia. They ini­tial­ly assem­bled a group of lost ship loca­tion experts and gave them the assign­ment, and they failed. Then the group leader tried some­thing dif­fer­ent. He assem­bled a larg­er group of indi­vid­u­als, each one with a com­plete­ly dif­fer­ent skill — some in oceanog­ra­phy, some in weath­er, some ship mechan­ics, some even radar and radio experts. And he used Bayesian meth­ods to aggre­gate the guesses/findings of every­body in the group and they man­aged to locate the ship. Now ofcourse, this might not quite work this way on a run­away train, but the basic idea is still to act inde­pen­dent­ly, but pool in everybody’s intel­li­gence, and then make a per­son­al deci­sion. You can read the ship case study in James Surowiecki’s Wis­dom of Crowds

  • Ashok: yes, what you describe is indeed dif­fer­ent than the herd men­tal­i­ty I was refer­ring to. I think the lat­ter is more prob­a­ble than the for­mer in pub­lic cri­sis sit­u­a­tions.

    Your exam­ple is inter­est­ing; nev­er heard of it before. It is quite like how we man­age soft­ware devel­op­ment and trou­bleshoot­ing! 🙂 Thanks for shar­ing!

  • Ashok: When you talk of “each one with a com­plete­ly dif­fer­ent skill”, you’re refer­ring to a sit­u­a­tion where dif­fer­ent peo­ple bring dif­fer­ent domain knowl­edge, i.e. spe­cial­ized knowl­edge to the sit­u­a­tion.

    I very much agree that it can be very valu­able in such cir­cum­stances: for exam­ple, some­one like Priyank may offer knowl­edge about the bank­ing of the rails, some­one may offer knowl­edge of the Indi­an rail system’s emer­gency response sys­tems, and so on. I would pre­fer call­ing it ‘col­lec­tive knowl­edge’, rather than ‘col­lec­tive intel­li­gence’, for obvi­ous rea­sons.

    When you apply the adjec­tive ‘col­lec­tive’ to the terms ‘intel­li­gence’ and ‘knowl­edge’ it has vast­ly dif­fer­ent mean­ings, as might be clear from my pur­suant post. Do you agree?

  • Yes. Col­lec­tive knowl­edge and intel­li­gence are very dif­fer­ent things. I think my wikipedia exam­ple might have mud­dled things a bit. The col­lec­tive putting togeth­er of facts is col­lec­tive intel­li­gence. But the sys­tem of pre­vent­ing van­dal­ism, the neu­tral PoV etc rep­re­sent col­lec­tive intel­li­gence.
    Per­haps the dif­fer­ence is in the result — the share mar­ket or any sort of pre­dic­tion mar­ket is a col­lec­tive intel­li­gence sys­tem — the out­put tends to be some sort of decision/choice etc, as opposed to a sim­ple aggre­ga­tion like an ency­clopae­dia. so in the case of crowd­chess, while there is a col­lec­tive choice, the fact that it is a sim­ple major­i­ty vote kills the col­lec­tive intel­li­gence. Each per­son here does not bring in enough diver­si­ty in terms of knowl­edge (after­all, its just chess rules, right 😉 )

  • For instance, in the e.g i gave ear­li­er, search­ing for ship, the coor­di­na­tor used bayesian equa­tions to fig­ure out a new set of coor­di­nates to search based on mul­ti­ple, diverse people’s inputs. A share mar­ket again, is not a sim­ple sta­tis­ti­cal aver­age kind of sys­tem. The price of a share rep­re­sents the col­letive intel­li­gence of all the investors, as an indi­ca­tion of its true val­ue. Ofcourse, a share mar­ket can be rigged in such a way the col­lec­tive intel­li­gence can be lost, as it hap­pened dur­ing our stock scam of the 90s.

  • Ashok: 🙂 I too was think­ing about equi­ty mar­kets after read­ing your ear­li­er com­ments!

    //The col­lec­tive putting togeth­er of facts is col­lec­tive intelligence.//
    I think you meant col­lec­tive knowl­edge, I under­stand.

    And agree that in the Wiki exam­ple, the sys­tem works because of col­lec­tive intel­li­gence.

    You are so right that the equi­ty mar­kets are not a sim­ple sta­tis­ti­cal aver­age kind of a sys­tem. Absolute­ly.

    When I referred to your col­lec­tive intel­li­gence com­ment in my Fri­day Flea Mar­ket post, I wasn’t equat­ing it rather jux­ta­pos­ing it. Hasn’t this dis­cus­sion enlight­ened us all by clar­i­fy­ing it fur­ther? 🙂

    Thanks for your com­ments, always appre­ci­at­ed.

  • 🙂 Ah well. It isnt often that I get a chance to talk about work relat­ed mat­ters in a blog com­ment thread.