Category talk:Railway accidents by type

Major expansion and restructuring
At Wikipedia talk:WikiProject Trains I have proposed a major expansion of this category (before I knew it existed). Read that for background. As that got a couple of favourable comments, I've produced an outline suggestion to start the discussion. It needs work still, including agreement about names - I've listed topics rather than category names. Many have examples, but where they don't it just means I couldn't think of any off the top of my head and I've run out of time to continue looking through the lists of accidents articles. I might have missed some topics, and others may not have enough to fill a category, but everything is up for discussion.

At the route this category would be renamed Category:Rail accidents by type (see my comments linked above about BE/AmE). The lists below are proposed sub-categories of this category. The first list is "caused by" and the second "involving", but there may be overlap? A possible third is by type of location. The fourth is by type of train, and I'm really not certain about that one.

Those that are linked have already existing categories, which may need to be renamed, as sub categories of this category.


 * Caused by:
 * Mechanical failure (including those caused by maintenance issues)
 * Track (e.g. Hatfield)
 * Points (e.g. Grayrigg)
 * Rolling stock (e.g Howden rail crash)
 * Brakes
 * Coupling (e.g. Round Oak)
 * Boilers
 * Wheels (e.g. Eschede)
 * Axles (e.g. Bullhouse Bridge)
 * OHLE
 * Earthworks (embankments, cuttings, etc.)
 * Bridges (e.g. Tay Bridge)
 * Tunnels
 * Metal fatigue (e.g. Versailles)
 * Maintenance activities (directly causing accident, e.g. Staplehurst)
 * Signalling issues If this category doesn't include SPADS, then a suitable name for it might be "signalling failure". However, it might be appropriate to include in this category accidents caused by signal operator error (ie the equivalent to "Aviation accidents and incidents caused by air traffic controller error". Bahnfrend (talk) 08:21, 19 July 2013 (UTC)
 * SPADS (e.g. Lewisham) [moved to under Driver error]
 * Failure of signals/signalling systems (including things like timed interval dispatch) (e.g. Quintinshill)
 * Signaller error
 * Driver error The word "driver" is British / Commonwealth English ("engineer" is the US English equivalent). Bahnfrend (talk) 08:21, 19 July 2013 (UTC)
 * Excessive speed (e.g. Salisbury)
 * Alcohol / drugs (e.g. Eltham Well Hall)
 * SPADS (e.g. Lewisham) [moved from under Signalling issues]
 * Suicide (or should this just be involving?) (e.g. Ufton Nervet)
 * Terrorism There is already a Category:Terrorist incidents on railway systems and a Category:Train bombings (which could simply be sabotage rather than terrorism). The categories for airliners include "Airliner bombings", "Aircraft hijackings" "Airliner accidents and incidents caused by sabotage" and "Airliner accidents and incidents involving in-flight explosions"  Bahnfrend (talk) 08:21, 19 July 2013 (UTC)
 * Weather / natural disasters (e.g Queen of the Sea)
 * Fog (e.g. Lewisham)
 * Ice (e.g. Abbots Ripton)
 * Wind (e.g. Rimutaka Incline; List of wind-related railway accidents)
 * Landslides (e.g. Ais Gill)
 * Heavy rain (e.g. San Raefael River)
 * Unknown cause (e.g. Moorgate)


 * Involving
 * Derailments (e.g. Brétigny-sur-Orge)
 * Runaway trains' (e.g. Lac-Mégantic
 * Road vehicles (e.g. Great Heck)
 * Level crossings (e.g. Lockington)
 * RRVs
 * Buses
 * Lorries (e.g. Hixon)
 * Fire (e.g. Summit Tunnel)
 * Explosions (or should this be combined with fire?) (e.g. Lac-Mégantic The categories for airliners categorise "explosions" and "fires" separately. Bahnfrend (talk) 08:21, 19 July 2013 (UTC)
 * Collisions between rail vehicles (e.g. Southall)
 * Head-on collisions (e.g. Ladbroke Grove)
 * Rear end collisions (e.g. Lewisham)
 * Three or more trains (e.g. Harrow and Wealdstone)
 * Buffer-stop collisions (e.g. Cannon Street)
 * Bridges (where not a cause, e.g. Eschede)
 * Opening bridges (e.g. St-Hilaire)
 * Animals
 * Pedestrians / other people
 * Telescoping of rail vehicles (e.g. Great Revere Train Wreck of 1871)
 * Train or passengers falling in to water (needs a better name, e.g. San Raefael River)


 * Location
 * Stations (e.g. Potters Bar)
 * Junctions (e.g.
 * On bridges (e.g. Tay bridge)
 * In tunnels (e.g. Sutton Tunnel)


 * Type of train
 * Passenger
 * High Speed
 * Subway
 * Light rail
 * Tram
 * Mixed passenger/goods
 * Goods
 * Fuel
 * Coal
 * Oil
 * Animals

Feel free to comment and add examples as you see fit. If you change anything in the above list, please make it clear what you have changed (e.g. use strikeout rather than deletion and italicise inserts) and include your reasoning below to aid discussion. Thryduulf (talk) 09:18, 15 July 2013 (UTC)
 * SPADs are not really problems with the signalling. They are driver errors: if a signal is at danger (or is showing an ambiguous indication, see GERT8000-S2 Iss 1.pdf Rule Book Module S2, section 7), the driver must not pass it without authorisation. If he passes a signal at danger, and there is then an accident which would not have happened if the train had remained behind the signal, it's not a SPAD but something else. -- Red rose64 (talk) 18:02, 15 July 2013 (UTC)
 * Good point about it being a driver issue rather than a signalling issue, I've moved it. As for it not technically being SPAD if there was an accident - that's the first time I've heard that? I think that in non-technical usage it would still be thought of as a SPAD and so worthy of including as such? Thryduulf (talk) 20:36, 15 July 2013 (UTC)
 * No that's not what I meant. What I meant was that if the driver is authorised to pass a signal at danger (see Rule Book Module S5, part A, section 1.2), it is not a SPAD, since the authority from the signalman/pilotman/etc. counts as a "caution" signal (section 3.1). Hence any accident which occurs as a direct consequence cannot itself be a SPAD accident. -- Red rose64 (talk) 21:15, 15 July 2013 (UTC)
 * Ah right, got it now. Something to be careful about when actually categorising. Thryduulf (talk) 22:22, 15 July 2013 (UTC)

As there have been no other comments about this, I'll draft actual category names in a day or so. Thryduulf (talk) 07:23, 19 July 2013 (UTC)
 * I have added some comments above in italics. More generally, I draw your attention to two important matters.  First, that there's a Category:Accidents and incidents involving airliners (equivalent to "Accidents involving trains") and a separate Category:Aviation accidents and incidents by type (equivalent to "Railway accidents [not necessarily involving trains] by type"), each with numerous subcategories with wording that can be adapted to either "trains" or "railway accidents" as appropriate.  Secondly, another important issue is whether the names of the categories should include just "accidents" as has been suggested, or "accidents and incidents" as per airliners/aviation.  At first glance, the difference between an "accident" and an "incident" might seem to be that whereas an "accident" involves damage or injury and, in the context of rail transport, is likely to be notable, an "incident" does not involve damage or injury, and, in the context of rail transport, is therefore unlikely to be notable or the subject of an article in Wikipedia.  However, if you want to be technical about these things, "accident" really means an event with an unexpected or unintended result (such as the Lac-Mégantic runaway/derailment) and "incident" may mean either an event that doesn't involve damage or injury, or (more importantly) an event with an intended result (such as the sabotage of a train, or the bombing of a station).  Thus, it appears that "accidents or incidents" would be a better form of words than just "accidents", for trains and railways as well as for airliners and aviation. Bahnfrend (talk) 08:21, 19 July 2013 (UTC)
 * Yes, "Rail accidents and incidents" does seem appropriate. Even if most incidents are not encyclopaedic some certainly are (e.g. 7 July 2005 London bombings).
 * On "railway vs train", as very few notable railway accidents/incidents don't involve trains I don't really see the value in two separate heirarchies, but just adjusting the category name to suit, e.g. Category:Rail accidents and incidents caused by signals passed at danger vs Category:Level crossing accidents and incidents (as they are road accidents too), I don't anticipate getting the name of each category right first time!
 * Your signaller error comment is a good one and I've added that to the proposal.
 * The British/American language issue is the trickiest as I noted when I originally floated the idea. Where there is a neutral term I propose to use that (e.g. "rail accidents" vs "railway accidents" or "railroad accidents"), but where no such term exists I think we're just going to have to go with one or the other and have a Category redirect at the other title (I've heard it said that rail terminology is where the two varieties are most different!). I speak British English and so it's natural for me to think of these using BE terminology (and there are likely cases I don't know that a term isn't universal) - that doesn't make it better of course! My gut feeling is that there will be value in being consistent in the use of BE or AmE category titles where there isn't a neutral term. Do say if you have any better ideas though! Thryduulf (talk) 13:12, 19 July 2013 (UTC)