Category: Wikipedia

Once more unto the Wikipedia breach, dear friends. Looking at my watchlist the other day, an edit to SMS Baden caught my eye so I had a little browse (I must have edited it long ago). This paragraph intrigued me:

The gunnery school HMS Excellent ran loading trials on the main battery guns. It was found that the guns could be prepared to fire in 23 seconds, 13 seconds faster than in the Queen Elizabeth-class battleships.^[28] The ship’s watertight bulkhead and underwater protection systems also particularly interested the inspection team; they paid close attention to the ship’s pumping and counter-flooding equipment.^[34] Commander W M Phipps Hornby, who lived on board Baden for weeks during the examination, wrote to the naval historian Arthur Marder in 1969 that it was his “considered opinion—which I know coincided with that of others engaged on the same job—that, considered as a fighting machine, anyhow on balance the Baden was markedly in advance of any comparable ship of the Royal Navy”.^[35]

I have been looking a lot at gun-loading times recently for my book, so the claim that the Baden class had a swifter loading cycle interested me. Naturally, however, I never trust a source implicitly. So, I followed footnote 28 to the late Bill Schleihauf’s article on ‘The Baden Trials’ in Warship 2007. He wrote:

Subsequently, the gunnery school HMS Excellent ran trials of the loading arrangements in Baden’s 15in turrets (23 seconds from firing to ready-to-fire vs 36 seconds in Queen Elizabeth) and ignited full 15in propellant charges in the gunhouses of ‘B’ and ‘X’ turrets to test the anti-flash arrangements.⁶

Note 6 reads, ‘CB1594 Progress in Gunnery Material 1921, (TNA, ADM 186/251), pp. 42-45.’ Let’s see what this actually says:

37. Loading Trials in ‘Baden’s’ 38 cm. Turret.–Loading trials have been carried out by H.M.S. ‘Excellent.’ These trials confirmed the loading times obtained from Germany.
The following table shows the comparative loading times for ‘Baden’ and ‘Queen Elizabeth.’
Times in gun-house only are shown.

The underlined text was surprisingly omitted by Schleihauf, and is a rather important qualifier. The times in the table add up to the same 23 and 36 second times give by him, and is immediately followed by the statement:

It should be noted that gun-house times do not necessarily govern the rate of continuous fire.
While generally, the cycle in magazines and shell rooms of ‘Baden’ and ‘Queen Elizabeth’ corresponds with that of gun-house, the rate of continuous fire would probably depend upon shell-room supply in the latter and an additional 3 seconds would be required in superimposed turrets of ‘Baden’ for main cage cycle.

This comparison is therefore based on only one part of the turret and loading cycle. Wikipedia’s claim that ‘that the guns could be prepared to fire in 23 seconds’ is therefore clearly not the whole story.

The sentence on ‘ship’s watertight bulkhead and underwater protection systems’ demands no comment, apart from the fact that technically the section of the source, a 16 March 1921 paper given by Goodall to the Institution of Naval Architects printed as ‘The Ex-German Battleship Baden’, dealing with this can be said to begin on page 22 and not page 23.

Now we turn to the Hornby quote, taken from Arthur Marder’s From the Dreadnought to Scapa Flow, volume V, p. 311. The passage in question is a footnote:

The D.N.C. and other Admiralty experts, having made a careful examination of the raised Baden, concluded (1921) that in the principal features of design they had little to learn from their late enemy. This was going a mite too far. Commander W. M. Phipps Hornby, who lived on board the Baden for weeks, employed on salving her, got to know her internal arrangements as well as those of his own ship, the Ramillies. It is his ‘considered opinion-which I know coincided with that of others engaged on the same job-that, considered as a fighting machine, anyhow on balance the Baden was markedly in advance of any comparable ship of the Royal Navy.’ Perhaps, as he suggests, the D.N.C. and others unconsciously were loath to concede that the young German Navy had much to teach them. Commander Phipps Hornby’s memorandum for the author, June 1969.

Using final ranks without qualification is always an invidious practice, as the reader may infer that with higher rank comes greater experience. When examining the Baden in 1919, Phipps Hornby had been a Lieutenant for roughly a year. Specialising in torpedo duties, he was automatically promoted to the rank of Lieutenant-Commander in 1925 after eight years. After nearly seven years as a Lieutenant-Commander, almost out of the promotion zone (he had repeatedly not been recommended, and was considered to lack leadership qualities), he retired in 1932 at his own request. His promotion to Commander on the Retired List came automatically upon reaching the age of 40. It’s also worth noting that he was directly involved in the mutiny of a Royal Fleet Reserve battalion employed during a strike at Newport in 1921 which ended the career of Captain Edward C. Kennedy (funnily enough not mentioned in his Wikipedia article), father of the late Sir Ludovic Kennedy. As to the veracity of Hornby’s claims, I’ve looked through the relevant part of Marder’s personal papers at the University of California, Irvine (specifically Box 3, folders for May and June 1969) and found no trace of this memorandum, despite there being a number of letters from Hornby to Marder, who was very fond of relying on decades-old recollections for his writing, regardless of their accuracy. As shown in that single footnote quoted, it takes a courageous (Yes, Minister fans take note) historian to take the word of a septuagenarian junior officer over the considered opinions of contemporary qualified naval architects.

And so, courtesy of flawed secondary sources, a single Wikipedia paragraph is equally flawed and misleading.

Back at the end of February I was idly looking at the talk page for the Wikiproject Military History when my eye spotted “HMS Topaze (1903)” amongst the pages being reviewed as a “Good Article nominee”. The criteria for a Good Article are that it should be:

• Well written.
• Verifiable with no original research.
• Broad in its coverage.
• Neutral.
• Stable.
• Illustrated, if possible.

As I do with any page, article or book I immediately looked at the notes and sources. I was surprised, therefore, to see “‘Fire Control in H.M. Ships’. The Technical History and Index: Alteration in Armaments of H.M. Ships During the War. 3 (23). 1919.” Note 3, “Fire Control in H.M. Ships 1919, p. 29.”, allegedly supporting the claim “The ship was equipped with voice pipes but no formal fire control system or range finder.” The reasons I was surprised were:

1. Generally speaking the only place that you will find a volume of The Technical History and Index (TH&I), an Admiralty history on technical matters published in parts (list here) after the First World War, is in an archive (my copies come from The National Archives and the National Maritime Museum), so this automatically qualifies as original research and violates the Good Article criteria.
2. Alteration in Armaments of H.M. Ships During the War was not part of the series title but the name of another part of the TH&I.

I checked p. 23 of both “Fire Control in H.M. Ships” and “Alteration in Armaments of H.M. Ships During the War” and there was nothing to do with Topaze on these pages. So not only was the article apparently relying on original research, but that research couldn’t even be verified! Automatic Good Article fail.

Why would someone invent a source, I wondered? How could they even invent a source like this? There aren’t that many publications which even use the TH&I, and certainly not with regard to specific ships. And then it hit me, as it should have from the beginning. I co-edit a website on fire control! One of the few places on the internet which uses the TH&I! Topaze was a member of the Gem class of cruisers, so I looked at The Dreadnought Project’s entry for the class. There are several sources given in the Armament section:

1. Admiralty Weekly Orders. 28 Feb, 1913. The National Archives. ADM 182/4.
2. The Technical History and Index, Vol. 3, Part 23. p. 29.
3. Handbook for Fire Control Instruments, 1909. Plate 53.
4. Handbook for Fire Control Instruments, 1909. p. 51, Plate 53.
5. Handbook for Fire Control Instruments, 1914. p. 67.
6. Handbook for Fire Control Instruments, 1909. p. 51.
7. absent from list in Handbook of Capt. F.C. Dreyer’s Fire Control Tables, p. 3.
8. Handbook for Fire Control Instruments, 1914. p. 67.
9. Handbook for Fire Control Instruments, 1909. p. 51.

It would take all these sources, and nearly a dozen paragraphs, to support the claim in the Wikipedia article that “The ship was equipped with voice pipes but no formal fire control system or range finder.” Not just one single source (see note 2) which technically does not support it.

Initially I thought this was a new phenomenon, but I checked the user’s contributions to be sure. Take, for example, HMS Teazer (1917). Created in February 2017, information on fire control was added in September 2018. “Fire control included a single Dumaresq and a Vickers range clock.” Source? “‘Fire Control in H.M. Ships’. The Technical History and Index: Alteration in Armaments of H.M. Ships during the War. 3 (23): 31. 1919.” Let’s look at The Dreadnought Project page on the “R” class destroyer (1916), last edited in April 2018. The notes for the fire control section are:

1. The Technical History and Index, Vol. 3, Part 23. p. 31.
2. Progress in Naval Gunnery, 1914-1918. p. 35.
3. The Technical History and Index, Vol. 3, Part 23. pp. 31, 32.

In this case the user has actually used the correct footnote to support the text, whilst still using the made up series title. In fact, if one used that title in the Wikipedia search engine one got 81 results. 81 articles containing a useless source covering up the use of material from The Dreadnought Project without attribution over the course of several years. This is infuriating, as at the bottom of every single page is the notice: “Content is available under Attribution-NonCommercial-NoDerivs 2.5 unless otherwise noted.” This states:

You are free to:

Share — copy and redistribute the material in any medium or format

The licensor cannot revoke these freedoms as long as you follow the license terms.

Under the following terms:

Attribution — You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

NonCommercial — You may not use the material for commercial purposes.

NoDerivatives — If you remix, transform, or build upon the material, you may not distribute the modified material.

On 1 March I wrote on the user in question’s talk page:

Dear #####. It would appear that since 2018 at the latest and up to the present time you have been copying fire control information from The Dreadnought Project and incorporating it into Wikipedia articles without attribution, whilst at the same time citing (often inaccurately and incompletely) our primary source references. This practice needs to stop. We don’t mind our work being used on this website so long as it is properly attributed to The Dreadnought Project. Nor, for that matter, does WP:MILHIST. Regards,

The user’s response? To delete the offending details on the Topaze page (so now there are 80 articles with the bogus sourcing, see image from 2 March at top) and to reply:

Please WP:AGF as I meant no harm. Thank you for your feedback. I am always learning and appreciate any help. I would be honoured if you would edit the references where I am in the wrong so that they are from the correct verified sources. I hope you enjoy Wikipedia.

AGF is “Assume Good Faith”. They meant no harm, but apparently thought that plagiarising the fruit of others’ work, badly, was quite alright. It was also very nice of them to suggest that I edit dozens of pages to fix their mischief. The problem rapidly solved itself, however. Rather than properly utilise The Dreadnought Project as a source for fire control details, the user simply deleted every offending reference to the made up source, including the text it supported. By 3 March the number featuring it was already down to 39, by 4 March to 19, and by 7 March it was eight. Now there are none, which is a shame in a way because it robs many Wikipedia articles of some much-needed detail.