Menu Close

‘For God’s sake go and stop him’: The Sharnbrook crash, Feburary 1909

Something of a departure for our usual project focus, this week’s blog makes use of an accident report type we don’t usually have reason to include. Our project database so far draws largely from reports issued by the Railway Inspectors appointed solely to investigate accidents to workers (called Sub-Inspectors or Assistant Sub-Inspectors, producing the Appendix B & C reports). At present it excludes cases which were investigated by those Inspectors whose job it was to investigate train crashes: usually they were passenger train crashes, and were a rather different affair (Appendix A reports). However, this week’s accident looks at a cross-over case: a collision which killed 2 staff and injured a third and was investigated by the Inspectors in an Appendix A report (available from the Railways Archive website).

Contemporary postcard of the Sharnbrook accident.

The accident happened on the Midland Railway at Sharnbrook, Bedfordshire, on 4 February 1909. At around 3.45am a stopping goods train was en route from Bedford to Birmingham, on the ‘down’ goods line. This was a line for slower-moving traffic; ‘down’ lines moved away from the relevant railway company’s ‘head quarters’, in this case London, whereas ‘up’ lines moved towards the HQ. The stopping goods service was moved on to what the signalman, Alfred Robins, thought was the down line for faster trains. Unfortunately, it wasn’t.

It was the up fast line. The crew of the goods service realised the mistake and tried to raise the alarm. However, before they were able to do so, the up express goods train from Manchester to London, on the same line, hit the goods service at around 50 miles per hour ‘and a violent head-on collision ensued’ at 3.51am. On the footplate of the express goods service were driver Arthur Coope and fireman John Hawley. Both were killed. The guard of the express service, William Henson, was also severely injured. Both of the engines involved were destroyed, along with 20 wagons, such was the violence of the impact.

Map of the location
Sharnbrook station, c.1900.
Courtesy National Library of Scotland Maps.

If this had been an accident not involving a train crash, there was no guarantee that it would have been investigated. Even if it had been, the investigators would have produced an Appendix B or C report. That would have been around a page long at most. Instead, because of the train collision, it was an Appendix A report, and stretched to nearly 7 pages. Importantly, it means it includes the statements made by the surviving staff involved, so we have some access to the voices of the staff usually missing in the other accident reports. Through it we can see some of the experience of being involved in an accident.

The accident was inquired into by Major JW Pringle. He took evidence from the driver of the stopping goods train, William Alcock. Upon realising the signalman’s error, ‘I at once sent my fireman forward with a red light to stop anything’ on the same line. He sounded the engine’s whistle, and moved his engine back as far as he safely could. Whilst doing this ‘I was shouting and whistling to the signalman, and immediately I brought my engine to a stand, I put a red head light on the front of the engine’ (to make it visible to any oncoming traffic). He the ‘shouted and whistled to my guard to get him to leave his brake [van]. After seeing him get out […] I got off my engine on the goods road [railway line] side. I commenced to scramble up the slope to the down goods line, but slipped back. I then heard the collision take place, and something struck me in the back […] and I am still off duty in consequence.’

The fireman of the same service, Rowland Wallis, noted that driver Alcock said ‘We are wrong’; ‘For God’s sake go and stop him’ [the oncoming service that they’d realised the signals were set for]. Wallis went on: ‘I left my engine near the north end of the platform and ran along the down passenger line exhibiting a red light. I did not get very far before I saw the train approaching on the up passenger line. […] I shouted as loud as I could when the engine approached.’ Sadly it was to no avail.

The guard of the stopping goods train, Frederick Fleetwood, noted that when his train stopped and he saw the light shone by his driver, ‘I got out to go and see what was the matter, and I had only got three or four wagon lengths from the brake [van] when the collision occurred.’ The guard of the express goods train, Henson, was unable to give evidence at the inquiry, as a result of his injuries. However, he gave a statement to the Midland Railway company on 17 March, noting that moments before the collision he sat down. He was then ‘thrown first across the brake handle, then on to the locker, and fell on my back on the floor’. He injured his ribs and chest and right arm; ‘these injuries are progressing satisfactorily, but I suffer a great deal of pain in my head and this does not appear to improve.’ Statements like these give us some sense of the immediacy of the accident, as well as some of the lasting effects.

Signalman Robins gave his evidence, admitting responsibility for the accident; Pringle noted that ‘there is no conflicting evidence on any material point.’ Pringle’s conclusion was that Robins had inadvertently moved an incorrect level in the signalbox, and in the dark failed to spot that the stopping goods train was on the incorrect line. The Midland Railway was advised to consider installing an additional set of safeguards, in the form of some extra signals, which might have prevented the error. Driver Alcock was praised by Pringle for his actions in the course of the 3 minutes or so in which he knew that something was wrong; at the same time he and fireman Wallis forgot ‘in the excitement of the moment’ to take detonators to place on the line to warn the oncoming train that something was wrong, as directed by their rule book. Pringle suggested that engine crews should be reminded of these rules.

Interestingly, Robins does not appear to have been tried at law. Signalmen implicated in causing accidents involving loss of life to passenger trains had, in the preceding years, been taken to court on charges of manslaughter. Even if this wasn’t necessarily reasonable (the 1892 Thirsk accident was a stand-out example), why not in this case? Was it because the dead were ‘only’ staff?

A future blog post will look at an interesting artefact of this – and other – crashes: the ‘disaster postcards’ of the scenes, and the comments that people included on them. If you’ve any postcards of the Sharnbrook crash or any other that you’d be willing to share images of, please let us know (via this page, on Twitter, Facebook, or email: railwayworkeraccidents[at]gmail.com). We’re particularly interested in any messages on the reverse of the postcards – whether or not they talk about the accident. There’s all sorts we’d like to say about how people saw and interacted with accidents at this time … but that’s for another blog post!

Shortly after this blog was posted, we were contacted by Megan Carswell, the Great Niece of John William Hawley, the fireman killed in the Sharnbrook crash. It’s always a privilege to hear from family members and descendants of the men and women discussed in our blogs and found in our project, so we were delighted (and would absolutely encourage anyone in the same position to get in touch with us). In the course of the discussion that followed, we found out more about John and his family, including after his death.

We were therefore really pleased that Megan was happy to put together a blog post for everyone to read – it’s available here, and is well worth the read. It’s very moving, and shows us much more about John as an individual and about the lasting impact of his death, right down to the present day.

1 Comment

  1. Pingback:John William Hawley, 1868-1909: A family tragedy remembered - Railway Work, Life & Death

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.