Edward Thompson: Wartime C.M.E. Discussion

I think your point about mentality is really important. There is always the danger of groupthink about something, if everyone says a loco is a bad steamer then it will subconsciously influence how people will drive and fire. The question would be why does group think emerge and here there are two things that could influence - management attitudes towards crews in a depot. If confrontational this could induce antagonistic relations. Likewise, a depot with say dominant figures who maybe small c conservative when it comes to locos may well influence how other crews think about locos. (Thinking here about how hierarchical railways can be).

 

I guess if you were a Cardiff fireman used to a narrow firebox Castle fired from the left, and were given a wide firebox Britannia fired from the right, it may not be an enormous surprise if you struggled at first to adapt - and from a driver's point of view, he's got the additional cognitive workload of sighting the signals from what feels inherently the wrong side of the cab while possibly having to coach a fireman who is struggling, when his own firing experience isn't in any case a lot of help. Not surprising in those circumstances if certain crews struggled, and rated them accordingly.

In an alternative universe in which steam had a long future, LH drive probably would have had to come to the Western Region, but you might not have chosen to introduce it on your top link passenger locos ...

Tom

 

Although weren't the A1s RHD and the A3s a mix of LHD and RHD? (Obviously less stress if you are switching within class than between classes).

 

Yes, but (a) Western firemen were a crude and unskilled bunch mostly used to just piling coal in a big mound in the firebox with no finesse and (b) more seriously, the switch from Castle to Britannia was RHD to LHD and narrow long firebox to wide short firebox. I've never been on a big GWR loco, but everything I have read, and experienced on smaller locos, suggests that the firing technique is very different to that of many other locos. Conceivably, had the crews been given Standard 5s in place of Halls as an initial step, they could more easily have adapted to LHD first, and then had the Britannias to make the second switch to wide fireboxes. Doing both at once is a big step.

Other railways made both switches at times, but not simultaneously with the introduction of a new class.

Tom

 

All Pacifics slip, true, but some slipped more than others. The fact that there may be more lightfooted Pacifics on other railways is not relevant to the discussion we are having here. What matters is how the Thompson rebuilds performed on home turf and what the crews thought of them. I recall an article in a magazine many years ago where Scottish driver was quoted as saying the Thompson Pacifics would “slip on Musselburgh Sands” and was on record to say he was glad when they were transferred away from his shed. Now his opinion may not have been universal on his shed but equally it goes to show these locomotives were not universally liked. As for their availability, That is only part of the picture. How they performed the work they were given would obviously influence what people thought of them. If They had greater availability then say a V2 but Performed less well at the same job, then crews would develop a lower regard for them. After all, availability figures would not be their first concern. Ability to do the job in hand would.

 

There is an important point in there, which is about perspective. For footplate crew, your primary concern is how the loco performs out on the road. Availability and other statistics don't come into it: if loco type A has 60% availability but is a brilliant performer when in traffic; and loco type B has 80% availability but is an average performer, the crew will prefer loco A. The Chief Accountant will prefer loco B because they would have to build only 75 of type B to do the same work being performed by 100 examples of type A - so they make a capital saving replacing 100 type A locos with 75 type B locos. But the crew don't care whether, when they climb on a loco in the morning, there are currently 25 or 40 others undergoing repairs; they just rate it on its ability to do the job.

The historiographical point is that historical accounts from footplate crew; and from those interested in performance, are rather more common than those from accountants or workshop staff. Quite possibly that is the part of the root of why some classes are less well regarded - and why I think @S.A.C. Martin delving into the statistical availability is of value - just take it as a different perspective on a definition of "good".

Tom

 

You might be describing something close to the A1 v A4 situation there. There is also the comfort dimension to factor in for the crew perspective.

To describe the whole picture, you would want a set of performance indicators all linked together and all relating to the customs and practices re what gets done on shed, what the heavy maintenance programme is, the duration of works overhauls, the incidence of unscheduled repairs. Then for some types, year round availability of traffic will have been a factor, which loco types (or categories eg 60,000 miles plus) were upgraded from B class work to A class work to cover the peaks. Some practices will be influenced by labour market availability especially in London.

The historian will inevitably be working with partial information and an incomplete jigsaw puzzle. @S.A.C. Martin will have shed light on corners other torches didn't reach. That's a contribution.

 

True but to dismiss the opinions of the crew in favour of opinions of shed masters and or bean counters is to show just part of the story.

 

A very valid point, well made.

 

Peter Smith describes a S&D Fireman having a 'full and frank' conversation with Control about the newly introduced 9F's, his view being that if they wanted super power loco's they had better get some super human fireman.

Ditto from a video interview, I gather Clive Groom's first experience of a 9F wasnt a happy one until he got the knack of getting it to steam properly

 

How far do you take that one negative review to prove a negative on an entire range of locomotives across the whole of their working lives though?

But no one is actually doing that.

If anything we are shining a light on the things which have - never - been otherwise considered when talking about how good or bad locomotives are.

 

On the other hand if one wishes to know why things were done and decisions made then the opinions of the crews may be largely irrelevant, because surely its going to have been the works managers, the bean counters, and possibly to a lesser extent the shed masters who will have most influenced the CMEs decision making.

 

I am somewhat wary of interpretation of statistical type records at Kew, the basis and accuracy of which may be subject to lots of various conclusions, and when they go against the vast body of other contemporary evidence and accounts.

I would also cite an engineering design appraisal and consideration both at the time and subsequently that appears to show (and in this I think we can be pretty objective) that the Thompson pacific rebuilds had 'issues'.

I fail to be persuaded by Simon that the stuff he has unearthed at Kew etc is really as significant as he proposes.

One might construct lots of counter arguments as to why one person's interpretation of the record cards etc might be misplaced and only show a partial "picture".

 

Are you suggesting that the information recorded on the locomotive record cards was falsified?

 

Perhaps you could interpret the statistics in some other way. Perhaps you could quote an engineering appraisal. Perhaps you could construct a counter argument. You’re welcome to keep trying but after umpteen years of this thread I fail to be persuaded that you could do so quite as well as Simon has.

 

The historian Elisabeth Barker warns people to be cautious 'of reading too much into comments scrawled in a hurry by people under pressure.'

People are not objective in documents, people lie in documents all the time, from white lies and exaggerations and minimisations to whoppers. This is a fact of life, people who are party to an event, they may choose to magnify or minimise their role (or the role of others).

'Did you break that?' Asks the foreman.
'Not me it was like that when I got here'
'Is it on the report card?'
'No'

Who broke it? No one knows.

Which is why you have to be cautious with small sample size data. The more data you have the more you can weed out statistical noise caused by people under reporting/over reporting things. Plus, you cross reference with other source material. If you are doing statistical analysis you would control for variables.

I recall a driver who was annoyed that his preferred engine was on standby that day and he had to take out another engine. While it was the end of the season and the loco was not in the best of condition. After grumbling all the way through preparing the loco he basically failed the loco at the first opportunity he could so he could get his preferred loco. He obviously filled in a report about how terrible the loco was. Next day the same loco went out with a different driver. Not the easiest of turns but the loco managed its trips fine.

Nothing was done to the loco between being failed and the next day because the supervisor was of the view that the driver just wanted his preferred loco and he wasn't going to pick a fight about it.

I am not saying that the first driver was lying in his report because he wasn't, and I am not saying that the second driver wasn't reporting stuff. But there were two very different spins on the same loco. The supervisor had a third spin. And here you are getting a fouth spin on the story which is mine.

* An example of data that might be incorrectly recorded, or not recorded or falsely recorded for example are signal box log books. Quite often when there was an accident and it comes down to what time a train passed a certain signal box or was belled to a box, there are comments about times not being recorded, times being crudely corrected, times being wrongly recorded and so on. The point is that we don't know how widespread this was because it is only when there was an accident that it was checked. Likewise, I was reading someone reminiscing about how when he was just starting as a porter, how the signalman showed him how to bell things, work the levers and then the signalman disappeared to the pub most evenings. Clearly, this was a wayside station with not much supervision so you can't read too much into it as evidence. But again, the source might be misremembering, the story might have got exaggerated, maybe 'most' evenings is an add in, maybe it was occasionally, maybe it was 30 minutes, maybe it was 3 hours, maybe there was nothing scheduled anyway.

 

A lot of comments made today, for which I am grateful for the additional discussion.

I would like to make one point however before I retire for bed, and then post a longer response tomorrow.

That point is this: it is actually the contemporary reports of the Thompson Pacifics that show a different light to them than was subsequently written after their withdrawal twenty years later, and even up to books written and published this year.

I submit that engine record cards, statistical LNER data (at present 1942-49) and reports to the board and locomotive committees are not likely to be falsified, or particularly “spun” - especially given the self critical nature of the one man in question - Edward Thompson - and the fact if you do go down the conspiratorial route, you are in fact asking hundreds of different people to falsify records at that time.

How likely is that really?

And if the argument is that the statistics are open for error - even accounting for a potential 10% margin of error change in the recording I’ve done in the records - it’s incredibly unlikely that every single bit of data we have on the Thompson Pacifics is so far out that it changes the story the statistics are telling us.

So I remain to be convinced that there isn’t a level of “heads in sand” thinking here where the statistics are dismissed out of hand.

 

The only thing I would watch for, or think about is incentives to over/under report things. It is the Soviet/Russian economic data problem. Under communism the system encouraged over reporting, once taxation was introduced after the collapse of communism this incentivised under reporting. Hence we have no real data just best guesses. (An extreme example).

 

You obviously don't understand about loco record cards and the various cross-checks that were carried out to ensure their accuracy. Loco mileages were cross-checked against drivers' tickets. The number of days in steam was recorded to ensure that washouts were carried out at the correct intervals. The cumulative mileage was recorded so that any particular loco could have valve & piston exams at the correct intervals, or be put forward for repair/overhaul under a shopping proposal once a certain mileage threshold was achieved.

If drivers kept failing locos without a very valid reason, they would have been lucky to stay in their job. Likewise, it would not be in anyone's interest at an MPD to falsify loco mileage records, and risk their livelihoods if found out. Loco record cards did not have their details 'scrawled in a hurry by people under pressure'.

Keep your conspiracy theories for the West Somerset Railway thread.

Andy

 

Also, if you read Adrian Vaughan's books, sometimes signal box logs aren't kept as updated in real time as would be optimal; ISTR at least once he helped another signal box 'recreate' some missing entries.

Noel