Studies in the Zulu War Volume III

Hi,

I was asked to let everyone know when this volume would be available and the nature of the content so here are the details.
Julian Whybra

STUDIES IN THE ZULU WAR 1879: vol. III
Will be available from Brecon Museum from next Wednesday 9th November.
Copies may be pre-ordered (either by phone or e-mail from Stephen Farish on 01874 613310 / [You must be registered and logged in to see this link.]).

Essays included are:

The Letters of Colour-Sergeant William Edwards 1/24th Julian Whybra

The Anglo-Zulu War Memorial in Lichfield Cathedral M. Paul Bryant-Quinn

The Defence of Helpmekaar 22nd January 1879 Graham Alexander

Zabange – Pure Fiction Julian Whybra

The Prisoner Escort at Rorke’s Drift Julian Whybra

The Chard Reports: A Stylometric Analysis David I. Holmes

Description of each:

“The letters of Colour-Sergeant Edwards have recently been donated by a descendant to what is now the Regimental Museum of the Royal Welsh. With an accurate transcription and accompanying notes they deserve being made available to a wider audience.”

“A study of reactions in Britain to the Zulu War shows that caricatures of the Zulu – ranging from the naïvely romanticized or racially stereotypical to the unremittingly negative - were already forming in the national discourse even while the war was still being waged. While the Zulu name may have been remembered, the people themselves became inexorably and simplistically equated with their army and with the events of 1879. Yet, if the default perception of the Zulu inculcated in the public consciousness was one of violence, memories both of the name and of the caricature have proved surprisingly enduring. What was it about this short colonial war and its African protagonists which brought them to such notice in 1879 and which subsequently caught and retained the public imagination? How and why this came about is explored through the medium of Lichfield Cathedral’s Anglo-Zulu War Memorial.”

“A few miles from the slaughter of Isandhlwana and the besieged garrison at Rorke’s Drift lay Helpmekaar. The arrivals at and departures from it and the events of the night of 22nd January which helped to shape its defence are comprehensively covered in a meticulous piece of research.”

“Some years ago an account of Isandhlwana by a Zulu named Zabange was being advocated and used as a genuine account despite its having been shown to be a work of fiction. The proof, revised and updated, has been expanded in a variety of ways including the quotation of extracts from the account itself.”

“The origin of the existence of the Prisoner Escort at Rorke’s Drift deserves to be explained in its own right. Clarification has always been required for the presence of the miscellaneous group of soldiers who were involved in the defence of the post yet were not members of B coy 1/24th or hospital patients. This essay accounts for the presence of six of them.”

“Suspicions that Lieutenant Chard might not have written his post-Rorke’s Drift official report but that someone else may have penned it instead have recently been aired. Using stylometric analysis methods this claim is examined resulting in some definitive conclusions.”

Thanks Julian. Order placed and looking forward to a good read.

Steve

I forgot, the book may not be on the museum shop website yet, but I was told that people can simply e-mail Stephen Farish at the museum to pre-order a copy.

Bonsoir Mr Whybra,

Awaited study!!!
Same echo (Steve)

Kind regards
Frēdéric

“Suspicions that Lieutenant Chard might not have written his post-Rorke’s Drift official report...

Cantwell..Julian.?.

Don't tell him Pike!

Steve

xhosa
Or allaying those suspicions...?

Thank's Julian.. will be interesting to see the source/ 's/

New edition arrived in this morning's post. Very nicely produced and substantial. Could not resist turning to the conclusions of the analysis of the Chard reports and who wrote them - fascinating - but I won't give it away! Well done Julian, another important addition to the canon.

Steve

rusteze wrote:

New edition arrived in this morning's post. Very nicely produced and substantial. Could not resist turning to the conclusions of the analysis of the Chard reports and who wrote them -  fascinating - but I won't give it away!  Well done Julian, another important addition to the canon.

Steve

Bonjour,
I don't yet receipt my copy...
Cheers.
Frédéric

Hi Frederic

Posted in Brecon on Friday, so probably bobbing about on La Manche.

Steve

rusteze wrote:

Hi Frederic

Posted in Brecon on Friday, so probably bobbing about on La Manche.

Steve

Thanks Steve,
So, I should soon receipt it ... if there is no strike by the dockers, postal workers, SNCF (railway network), air traffic controllers, road hauliers, or an invasion of locusts.
In short, the little peripates that regularly shake my country.

Amitié.

frédéric

Tue Nov 08, 2016 1:36 pm Topic: Re: query
Yeah mate all is good my end.. i'm glad your trip
was more than you expected but like you i have
never read any of the Clery connection regarding
Chard, but we have to keep in mind Chard's con-
dition post the defence, i suspect he was done in
with exhaustion and terrified of the attention he
was receiving.. he was badgered to write his
report, then of course he went down with enteric.
but his report to Queen Victoria was a masterpiece
of understatement.. he sets out the sequence of
events in a calm logical manner which i'm sure he
did not feel in the week's immediately following the
defence.. and what would Clery a staff officer have
to do with an officer of engineers?. we know Clery
was a ' busy ' man.. in the sense that he poked his
nose into everything..


The above is an extract to another member.. my
point now after reading Julian's latest is...why did
he write it at all!. there is nothing definitive there, it
all seems to add up to precisely nothing, or as usual
am i missing a trick?.

So what you saying?

The above is an extract to another member.. my
point now after reading Julian's latest is...why did
he write it at all!. there is nothing definitive there, it
all seems to add up to precisely nothing, or as usual
am i missing a trick?.
....Hiya John.. always good to hear from you
mate, i was saying the above, i'm sure Julian will
explain it so that even a thicko like wot i am will under-
stand. cheers xhosa

Xhosa
The conclusion behind David Holmes's analysis is that the author who wrote the first Chard report was the same man who wrote the second Chard report.
The author was not Clery.
The author was not Bourne.
Given the different circumstances, chronological, geographical and psychological, in which the two reports were written Holmes concludes that it could only have been Chard who was the author (and Ian Knight agrees with him).
Those who might disagree with these conclusions will have to prove scientifically (using stylometric analysis) that Chard was not the author (a virtual impossibility in Holmes's opinion), the only alternative being, that an officer, as yet unknown, who had access to Chard on both separate occasions, was the author (an extremely unlikely, virtually impossible scenario).
I.e. Chard wrote both the reports.
I have summarized Holmes's conclusions on pp. 78-9.

Yes Julian.. i do ' get that ', what i'm struggling with is why the piece
was written!.. to what end?. who are these people saying that the
reports were written by different hand's. where was the ' clamour '
for enlightenment?.

And the ' conclusion ' ended up concluding nothing at all!!!. it basically
say's that Chard did indeed write both report's because there was no
evidence to the contrary.. and till such time there is/ might be in the
future we must be content with the ' finding's '... again i'm struggling
to see what this piece achieve's.. and again i'm very sorry if i'm missing
a trick here, i just wanted to comment is all, certainly no personal
criticism implied or intended.

Bonsoir Mr Whybra,
The essay about the letters of Colour-Sergeant William Edwards is particulary touching. A bit of humanity just before the disaster of Isandhlwana...
Bravo!

Cheers

Frédéric

Holmes wrote it to counter the arguments expressed in a recent history and elsewhere outlined on page 67, 2nd paragraph and to prove those authors mistaken.

Your so patient as always Julian, i will have a leisurely
reread.. many thank's.

My pleasure

Now that a number of people have read the article, I thought I might add my two-penneth.

I have no difficulty with what prompted the study (its probably the only book on the AZW I have since disposed of). But I was left with a nagging doubt about the stylometric analysis. Having spent a long career drafting official reports of one kind or another, I am only too aware that they go through many hands and numerous adjustments before they see the light of day. Although they carry the signature of the originator they are the result of extensive collaboration. They are not in any one man's style by the time they emerge. You make the point that Chard would have elicited the help of NCOs, Bourne in particular, when it came to details of names etc and I am sure that must be so. If Clery was there I see no reason why he too might not have steered the text in certain directions. While I can see that the text is shown not to be entirely Bourne, or entirely Clery, I think common sense says that Chard would have conferred and adjusted extensively with an eye to the eventual reader and their requirements. I don't buy the objections re. paper shortage etc. as these could be overcome if the request was from the staff. So yes the reports are both originated by Chard but probably owe a lot to inputs from the others.

Steve

Interestingly no one has mentioned input from Bromhead? It was his 'report' after all that pointed the way to the medals.

Frank

Bonjour Frank,
It seems to me that you were aware a few months ago of something "under fire" regarding the Bromhead's report?

Cheers

Frédéric

Indeed Frederic and still working on it.

Oh, oH!!!.

Frank, are you optimistic about the result of your research about this subject?

Cheers

Frédéric

Not very Frederic, run into a bit of a brick wall with the Hicks Beach archives, a couple of items are missing. from where they should be. But theres always hope.

Frank

I understand...
Impossibility (actually) of proving a point...
But, as you say rightly, "there 's always hope", sometimes in another direction.

Courage, mon ami!

Frédéric

rusteze
I tend to agree about chard receiving input from others (and certainly Bromhead) however the fact remains that from a stylometric analysis of both reports, the STYLE of writing is Chard's.  Some of the names, and recall of incidents and special mentions may belong to others, BUT the use of language and style of writing is not - and thereby hangs the authorship.  I note that Holmes's will not even admit to the text being "partly Clery or partly Bourne" but wholly Chard.
P.S.  As I recall there were a couple of articles and book which cast a question mark over the authorship.  A sensible suggestion given the circumstances and certainly one worth Holmes's examining.

I recall a certain Home Secretary making a statement in the House of Commons which elicited a tricky question from the opposition. He rose some ten minutes or so later to give a fulsome and detailed reply which no doubt appears in Hansard under his authorship. Sitting in the officials box we knew different.

Steve

"A stylometric foray into the Anglo-Zulu War of 1879

David I Holmes​​​and​​Elizabeth D. Johnson
The College of New Jersey, USA ​​ Wilkes University, USA

​Abstract

The most celebrated feat of arms during the reign of Queen Victoria was arguably the Anglo-Zulu War of 1879 and in particular, the heroic defence of the mission station at Rorke’s Drift. The senior officer, Lieutenant John Chard, was asked to write a report of the battle in its immediate aftermath which he duly submitted, yet suspicions have arisen concerning the true authorship of this ‘official’ report, which deflects the blame for the horrifying tragedy at Isandlwana that same day away from the Commander-in-Chief Lord Chelmsford. This article describes how stylometry can shed light on a possible Victorian conspiracy.

1. Background

During Queen Victoria’s sixty-four-year reign, the red-coated British soldier was engaged in many campaigns throughout the British Empire, fighting for Queen and country. Of all these conflicts it was the Anglo-Zulu War of 1879 that most seized the popular imagination. Why the fascination and why do so many people still take the trouble to visit these faraway battlefields? Perhaps it is because on one day, Wednesday 22nd January 1879, Britain suffered her greatest battlefield defeat in her entire colonial history yet that evening, at a small mission station named Rorke’s Drift, a brave little band of British soldiers held out all night against an immense Zulu attack. That same day, two officers died in an attempt to save the Queen’s Colours and became the first men in history to earn posthumous Victoria Crosses. A few months later who should die whilst out on patrol with the British but the Prince Imperial of France. Who would have thought that it would have been the Zulus who brought an end to the Napoleonic dynasty? This campaign was also the first to be covered in detail by the illustrated British press, bringing into the homes of the British public the tale of the clash between two empires.
​At the time of the Anglo-Zulu War, Britain was using the proven policy of Confederation as a means of administering her numerous colonies. In South Africa this involved merging a number of neighbouring colonies under one central and stable administration. The crushing of Zulu King Cetshwayo and his warriors was seen by the local officials, without the sanction of the central British government, as the solution to the problem of knitting together the colonies into a workable confederation. The High Commissioner to South Africa, Sir Bartle Frere, his Secretary for Native Affairs, Sir Theophilus Shepstone, and the Commander in Chief of the British forces in South Africa, Lord Chelmsford, believed that a quick campaign was all that was needed and used minor border incidents to inflame public opinion against King Cetshwayo. Eventually, on 11th December 1878, Zulu chiefs were summoned to the Natal bank of the Tugela River, which runs along the border between Zululand and Natal, and presented with an ultimatum which, among many stringent requirements, included the disbanding of the Zulu army. Knowing the Zulus would and could not comply with the ultimatum, a British invasion force under Lord Chelmsford was already advancing towards the borders of Zululand.
​Three large independent columns were prepared for the invasion, which was scheduled to take place on 11th January 1879. The Central Column, which was accompanied by Lord Chelmsford and his staff, was made up of some 4,700 officers and men and crossed the Buffalo River into Zululand by the former mission station of Rorke’s Drift, now commandeered and converted into a store and hospital. By 21st January the column was encamped by the large crag of Isandlwana mountain. Mistakenly believing that his scouts had located the Zulu army to the south-east, in the early hours of 22nd January Lord Chelmsford left the Isandlwana camp with over half his force leaving some 1,700 men behind in the essentially undefended camp. What followed that day, while Chelmsford was off chasing shadows, was a defeat which was to shake Britain’s military establishment to the core as the camp was utterly overwhelmed by the main Zulu army of some 25,000 warriors attacking from the north-east, leaving only a handful of survivors to escape before the net was closed. A definitive account of this battle may be found in Anglo-Zulu War historian Ian Knight’s book Zulu Rising (2010).
​The far right wing of the Zulu army did not take part in the main engagement at Isandlwana. Against the explicit orders of King Cetshwayo, these regiments proceeded to run ten more miles to cross the Buffalo River into Natal and engage the supply depot and hospital at Rorke’s Drift so that they, too, could wash their spears in British blood. Thus it was that, in the late afternoon of that fateful day, a 4,000 strong force of Zulus reached the little mission station, now fortified by walls of mealie bags and biscuit boxes but only manned by an effective defensive force of 100 soldiers.
​The senior officer that afternoon at Rorke’s Drift was 31-year-old Lieutenant John Chard. Chard was a Royal Engineer who had only arrived in South Africa on January 5th and whose duties at Rorke’s Drift were to supervise the construction of a redoubt overlooking the river and to build ponts. His immediate superior, Major Spalding, had gone to seek reinforcements, leaving Chard and Lieutenant Gonville Bromhead, the officer in charge of the garrison’s B Company of the second battalion of the 24th Regiment to take care of the station. Chard’s peace was disturbed when at about 3.30pm he noticed two horsemen galloping towards the drift from the direction of Isandlwana bringing with them the dreadful news of the massacre. Within an hour, the attack on Rorke’s Drift had begun.
​Throughout the evening and on into the night, wave after wave of Zulus reached the improvised walls only to be repulsed by the line of waiting soldiers with their blood-covered bayonets. Individual feats of heroism that night led to the awards of eleven Victoria Crosses, more than in any other single action in history. After midnight the Zulus were clearly more exhausted than the British, their attacks became sporadic and by 5am the Zulu force had vanished. Only 17 soldiers lost their lives; around 400 Zulu bodies were found in and around the mission station. One of the great ‘last stands’ in British military history had become the stuff of legend.

2. The Chard Reports

Within two days of the battle Chard submitted a perfectly sequential report of the engagement that was complete in extraordinary detail. This first ‘Chard Report’ became the ‘official’ report concerning the attack on the mission station and it was this report that Chelmsford forwarded to the War Office. The many suspicions arising from the preparation of this report are covered in detail in Anglo-Zulu War historian Adrian Greaves’ book Rorke’s Drift (2002).
​Greaves finds it remarkable that Chard could obtain a sufficient supply of clean undamaged paper when all available paper had been burnt in the hospital fire or destroyed during the fighting. Indeed, there is documented evidence that soldiers had to make do with scraps of scorched paper to pen brief notes home. The fact that amidst the chaos there were incessant downpours of rain makes the preparation of a clean report even more remarkable. Greaves also questions the accurate timings, precise locations and detailed names of the soldiers involved given the fact that Chard had only just arrived at the station and could not have known all the participants. Yet, there is no known record of any NCO assisting Chard with the report. Chard had a reputation for slothfulness and both Captain Walter Parke Jones and Lieutenant Henry Curling, whose contribution to this research is covered in the next section, have provided scathing anecdotes concerning his character.
​It would clearly have been in Lord Chelmsford’s interest to promulgate a dramatic report of the victory at Rorke’s Drift which would deflect those critics seeking to humiliate him for the catastrophic defeat at Isandlwana earlier that day. As Greaves (p. 178) comments, in the minds of the British public “an inglorious defeat could be offset by a glorious victory.” Although Chelmsford had not lingered at Rorke’s Drift for more than a few hours, there is evidence that he was expecting a report concerning the victory in addition to issuing instructions for a formal enquiry to be conducted into the Isandlwana defeat. Greaves believes that Chelmsford’s staff took the initiative in ensuring that Chard signed a suitably impressive report to offset the serious repercussions about to be unleashed following the losses at Isandlwana.
​One of Lord Chelmsford’s senior staff officers, Major Francis Clery, had remained behind at Rorke’s Drift after Chelmsford and his staff departed on 24th January. Chard’s ‘official’ report is dated the following day. Clery was a confidant of Chelmsford and an experienced report writer. He was a staff college graduate, a former Professor of Tactics at Sandhurst Military College, self-assured, observant and prone to making gossipy judgments about his colleagues. Greaves comments that Clery had also been culpable in the unfortunate decision-making process that led to the defeat at Isandlwana and he, of all people, would have realized that a dramatic report from Rorke’s Drift might deflect the criticism that would undoubtedly be unleashed upon Chelmsford and his staff. Greaves believes that Clery is a strong candidate for authorship of this first ‘Chard Report’.
​Later in 1879 and now safely back in England, Chard was required to submit a further report at the request of Queen Victoria. He was required to include greater detail, although nothing is known of the preparation or research that went into this second report other than the fact that apologies for a delay were given to Queen Victoria as Chard claimed to have ‘lost his notes’. No notes have ever surfaced. This second ‘Chard Report’ was submitted to Her Majesty at Windsor Castle on 21st February 1880. Major Clery would not have been present to advise on this report.
The key questions that this research will address are:
I. Are there any stylistic differences between the first and second Chard Reports?
II. Are there similarities in style between the first Chard Report and Clery’s writings?

3. Sampling and Textual Preparation

A number of control texts are necessary for this analysis, in the same genre and era but by writers who do not feature as contenders to the disputed Chard Reports. Letters from men writing home during the Anglo-Zulu campaign, not just officers, but NCO’s and private soldiers whose letters, in the custom of the time, were passed on for publication in local newspapers, are available in The Red Soldier (Frank Emory, 1977). Three men in particular, Captain Edward Robert Prevost Woodgate, Captain Walter Parke Jones and Lieutenant William Weallens provided letters of sufficient length to qualify as suitable control texts, with each batch of letters being split into two to facilitate internal comparison.
​Another excellent source of textual material is the book The Curling Letters of the Zulu War (Ed. Brian Best and Adrian Greaves, 2004). Lieutenant Henry Curling was one of only five officers who survived the slaughter at Isandlwana. A family man and prolific letter writer, his detailed accounts of the invasion of Zululand make gripping reading. Two textual samples were taken from this book to add to the control texts.
​Major Francis Clery was an experienced report writer. His letters are part of the papers of Sir Archibald Alison, who was Chief of Intelligence in the British Army and are published in Zululand At War 1879: The Conduct of the Anglo-Zulu War (Sonia Clarke, 1984). Three textual samples were taken from this book.
​Both the first and second Chard Reports are available, unabridged, in Appendix A of Rorke’s Drift (Adrian Greaves, 2002). The second report is reprinted there by kind permission of HM The Queen. The first report at 1,301 words is too small to be split, but the second report at 6,462 words was divided into four approximately equal-sized samples. Regrettably, apart from a brief 312 word letter written in September 1879 to Sir John Stokes (provided by Ian Knight from The Royal Engineer Journal), no usable authentic written material from John Chard is known to exist.
​All these samples are listed in Table 1, the samples being either typed or scanned into machine-readable form. The choice of text size in stylometric studies is always problematic. Smaller units are too short to provide opportunities for stylistic habits to operate on the arrangement of internal constituents, while larger units are insufficiently frequent to provide enough examples for reliable statistical inference. Forsyth and Holmes (1996) found the median text block size in a selection of stylometric studies to be around 3,500 words. In their study of the Book of Mormon, Jockers et al. (2008) claim that even the smallest chapters are of adequate size for stylometric analysis, finding no correlation between the correct assignment of an author and the length of text sample. In all the following analyses, the occurrence rates of words are measured as percentages of the total sample size. Thus for this study, differences in sample size are not critical provided we adhere to a stylometrically desirable minimum threshold of 1,000 words.












Table 1: Textual Samples
Author Title Sample N(sample size in words)

Henry Curling​ Letters​​1​​3112
​​​​​2​​3035
Walter Parke Jones Letters​​1​​1168
​​​​​2​​1336
William Weallens Letters​​1​​1396
​​​​​2​​1564
Edward Woodgate Letters​​1​​1380
​​​​​2​​1379
Francis Clery Letters​​1​​2559
​​​​​2​​2324
​​​​​3​​3047
First Chard Report ​​1​​1301
Second Chard Report​​1​​1687
​​​​​2​​1632
​​​​​3​​1698
​​​​​4​​1445
​​​​​

​​​​​
4. Stylometric Methodology

The pioneering work of Mosteller and Wallace (1964) on the use of function words in authorship attribution was continued by J. F. Burrows (1992). Since then multivariate statistical analyses involving large sets of non-contextual high-frequency function words have met with astonishing success in attributional problems in a wide variety of authors and genres. See, for example, the investigation into the authorship of the so-called ‘Pickett Letters’ of the American Civil War (Holmes et al., 2001) and the new look at the authorship of the Book of Mormon (Jockers et al., 2008). The ‘Burrows’ approach essentially picks the N most common words in the corpus under investigation and computes the occurrence rate of these N words in each text or text-unit, thus converting each text into an N-dimensional array of numbers. Multivariate statistical techniques, most commonly principal components analysis and cluster analysis are then applied to the data to look for patterns. The former aims to reduce the dimensionality of the problem by transforming the N variables to a smaller number (usually 2) of new variables and the latter technique provides an independent and objective view of any groupings amongst the textual samples by means of a tree-diagram or dendrogram. The ‘Burrows’ approach has become the first port-of-call for attributional problems and will be the technique adopted in this investigation.
​The value of N used varies by application and genre but typically lies between 50 and 75, the implication being that these words should be among the most common in the language and that content words should be avoided. A value of N set at 60 is used as a rule-of-thumb heuristic throughout this analysis, being an appropriate value for these sized text samples. Appendix A lists these sixty most common function words, taken from the corpus of texts in Table 1.





5. Hierarchy of Analyses

5.1 Controls: Curling, Jones, Weallens and Woodgate.

The first phase in this investigation is designed to test the validity of the proposed technique. For the purposes of this study, it is required that known texts can be shown to be internally consistent and separate from each other. The occurrence rates of the sixty words listed in Appendix A were computed for the individual textual samples from the letters of Lieutenants Curling and Weallens and Captains Jones and Woodgate. These were used as input to both a principal components analysis and a cluster analysis. The positions of the samples in the space of the first two principal components, which together explain 52.9% of the variation in the original data, are shown in Figure 1. An alternative analysis of the controls may be provided by conducting a cluster analysis on the textual samples, using the sixty word rates as variables and Ward’s method as the clustering algorithm. Figure 2 shows the resulting dendrogram.
​The results with these two methods of analysis are mutually supportive, with samples forming clusters on the basis of authorship. Our writers are internally consistent as regards their usage of these sixty words, yet are distinguishable from each other.



Fig. 1 Principal components plot: Curling, Jones, Weallens and Woodgate.









Fig. 2 Dendrogram: Curling, Jones, Weallens and Woodgate.


5.2 Clery, Curling, Jones, Weallens and Woodgate.

We now add the three textual samples from the letters of Major Francis Clery into the stylometric mix. The occurrence rates of the sixty most frequently occurring function words were once again used as input to both a principal components analysis and a cluster analysis. The positions of the samples in the space of the first two principal components, which together explain 45.6% of the variation in the original data, are shown in Figure 3. An alternative analysis of the controls was provided by conducting a cluster analysis on the textual samples, using the sixty word rates as variables and Ward’s method as the clustering algorithm. Figure 4 shows the resulting dendrogram. Both these plots show excellent internal consistency for the three Clery samples. They are also quite distinct from the samples taken from our three controls, providing validation for the ‘Burrows’ approach on works of known authors.



Fig. 3 Principal components plot: Clery, Curling, Jones, Weallens and Woodgate.



Fig. 4 Dendrogram: Clery, Curling, Jones, Weallens and Woodgate.
5.3 The Chard Reports and Clery.

Having successfully established the internally consistent style in the letters of Major Francis Clery concerning the usage of the sixty function words it is now time to focus on both Chard Reports. Discarding Jones, Weallens and Woodgate, who have done their duty, we now take the first Chard Report and the four samples from the second Chard Report, and add them into the mix with the three Clery samples. Figures 5 and 6 show the principal components plot and the dendrogram, respectively, from similar multivariate statistical analyses on the sixty function words. The former plot explains 50.4% of the variation in the original data.
​Two important conclusions may be drawn from these clear and mutually supportive plots. First, there appears to be no difference in style between the first and second Chard Reports, suggesting single authorship. Secondly, the Chard samples cluster quite distinctly and separately from the samples of the writings of Clery. Our strong contender for authorship of at least the first Chard Report appears not to be a match. Might there be another contender amongst the British soldiers serving in the campaign?





Fig. 5 Principal components plot: Chard Reports and Clery.








Fig. 6 Dendrogram: Chard Reports and Clery.


6. Colour Sergeant Frank Bourne

For many people, the outstanding character in the iconic 1964 film “Zulu” was Colour Sergeant Bourne, as portrayed by Nigel Green. The real Frank Bourne was born in in 1854 and volunteered for the army in 1872. He was short and in his own words painfully thin, but upon arrival in South Africa rapidly ascended the promotion ladder. Bourne’s first duties at Rorke’s Drift on 22nd January 1879 would have been to supervise the taking down of the bell tents to give a clear field of fire, to post lookouts on the higher ground and then to lead a skirmishing line to intercept the advancing Zulu force.
​In penning his first report after the action, John Chard would certainly have elicited help from NCO’s present at the defence of the post since he would not have seen everything nor known everyone’s names. Frank Bourne was literate and acted as ‘unpaid private secretary’ to those soldiers who could barely read or write, deciphering and answering their letters home. Might Bourne be the author of the first Chard report and might Chard have drawn on Bourne’s notes for the second report?
​Bourne died in May 1945 and it is believed that he was the last surviving member of the Rorke’s Drift garrison. Yet in December 1936 he made a BBC radio broadcast concerning the battle for a series entitled ‘I was there’. It generated enough interest for 350 people to write to him. Regrettably the BBC scrapped the recording during the 1950’s as being of insufficient interest but a transcript of this broadcast is available in Appendix E of Rorke’s Drift (Adrian Greaves, 2002). This ‘Bourne Report’ has 2,246 words and was divided into two approximately equal-sized samples for analysis. The Bourne Report samples were added to those from both Chard Reports and the Clery letters and the occurrence rates of the sixty most frequently occurring function words were once again used as input to both a principal components analysis and a cluster analysis. The positions of the samples in the space of the first two principal components, which together explain 49.7% of the variation in the original data, are shown in Figure 7. Figure 8 shows the associated loadings plot from the principal components analysis, which helps to explain the groupings in the main plot. One can imagine superimposing this graph on top of the principal components plot. Figure 9 shows the dendrogram, again using Ward’s method as the clustering algorithm.





Fig. 7 Principal components plot: Bourne, Chard Reports and Clery.


Fig. 8 Loadings plot: Bourne, Chard Reports and Clery.


Fig. 9 Dendrogram: Bourne, Chard Reports and Clery.
7. Conclusion

From Figures 7 and 9 we see that the two halves of the Bourne Report show excellent internal consistency but, although slightly closer in style to the Chard Reports than Clery’s writings, cannot really be considered to be a stylometric match. Figure 8 shows how words on the right such as ‘for’, ‘you’, ‘is’, ‘be’ and ‘that’ have high usage by Clery, whereas words on the left such as ‘were’, ‘and’, ‘the’ and ‘his’ are words of high usage in the Chard Reports. Favoured words in the Bourne Report are words such as ‘our’ and ‘their’.
We are left then with the following conclusions:
• Both the first and second Chard Reports appear to be by the same hand.
• That hand is not that of Francis Clery, despite Clery being mooted as a strong contender for author.
• Neither do the reports appear to be by Frank Bourne. Bourne may well have added details but the slightly closer similarity here may reflect the fact that Bourne was very familiar with Chard’s first report by the time he gave his own account.
Anglo-Zulu War historians Adrian Greaves and Ian Knight, in private correspondence, have both sent interesting comments pertaining to these conclusions. Greaves still believes that Chard could not have written the first ‘official’ report, it being too accurate and of too academic a nature for him. Greaves remains in firm belief that Chard was neither clever nor erudite enough to pen this report and, if it is not by Clery, then the author could have been any erudite officer who had been able to piece together the sequence of events.
​Knight agrees that, if some other officer was the author of the ‘official’ report then Clery was by far the obvious suspect since he was always complaining that everything was left for him to deal with. But, while being in no doubt that Chard was ‘leaned upon’ to provide a glowing report, Knight stresses that this is not the same thing as letting someone else write it. Knight believes that, although the ordinary soldier might not have had access to paper immediately after the Rorke’s Drift battle, Clery could easily have given Chard some sheets when he passed on Chelmsford’s request to write the report. Given Lieutenant Gonville Bromhead’s reluctance to write a report himself, Knight thinks it more than likely that he would have offered up names and incidents for Chard to include, so the fact that Chard himself did not know the names of the private soldiers at Rorke’s Drift is of little relevance. In short, the above results confirm Knight’s belief that Chard himself was the hand behind his reports.
This research project now lies somewhat tantalizingly on hold, awaiting any future discovery by Anglo-Zulu War historians of new textual material from officers who were present at Rorke’s Drift.

Acknowledgements
We wish to thank Anglo-Zulu War historians Dr. Adrian Greaves and Ian Knight for their invaluable advice throughout this project and for their help in providing textual material. We also wish to thank Dr. Richard Forsyth of the University of Leeds, UK, for the specialist computer software used and The College of New Jersey students Chris Millelot and Claudia Beard for their help in textual preparation.




References

Best, B. and Greaves, A. (Ed.) (2004). The Curling Letters of the Zulu War. England: Pen and
​Sword.

Burrows, J.F. (1992). Not unless you ask nicely: the interpretive nexus between analysis and
​information. Literary and Linguistic Computing, 7: 91-109.

Clarke, S. (1984). Zululand At War 1879: The Conduct of the Anglo-Zulu War. England:
​Brenthurst Press.

Emory, F. (1977). The Red Soldier. London: Hodder and Stoughton.

Forsyth, R.S. and Holmes, D.I. (1996). Feature-finding for text classification. Literary and
​Linguistic Computing, 11: 163-74.

Greaves, A. (2002). Rorke’s Drift. London: Cassell.

Holmes, D.I., Gordon, L.J. and Wilson, C. (2001). A widow and her soldier: stylometry and
​the American Civil War. Literary and Linguistic Computing, 16: 403-20.

Jockers, M.L., Witten, D.M. and Criddle, C.S. (2008). Reassessing authorship of the ‘Book of
​Mormon’ using delta and nearest shrunken centroid classification. Literary and Linguistic
​Computing, 23: 465-91.

Knight, I. (2010). Zulu Rising. London: Macmillan.

Mosteller, F. and Wallace, D.L. (1964). Applied Bayesian and Classical Inference: The Case of
​The Federalist Papers. Reading, MA: Addison-Wesley."
​

Source: [You must be registered and logged in to see this link.]

I'll try and post a better version of this, as there are some very go diagrams showing possible writers of the report. Don't know why the link posted doesn't go direct to what I'm seeing?

is it not possible that Chard and perhaps Bromhead dictated the report to someone who had the ability to put the words in to some order that gives an overview of the events. In which case it wouldn't be in Chards hand writing.

Didn't the news correspondent Noggs Newman arrive at RD with Chelmsford, writing stories for newspapers would put him in a very good position as the writer of the report, and I'm sure Noggs would have had paper with him. Just food for thought!

LH
That still wouldn't explain the later handwriting in the second report. Whoever wrote the first had to be around to write the second one.

Frank

As I understand it this is not about comparing handwriting in the traditional sense. It is about the use of particular words and how they are deployed mathematically throughout a text. So it would work equally well on a typed version of the two texts. Frank's point about the originator of both "Chard" texts being shown to be the same person based on stylometric analysis is the crucial point. JW's article adds context to the previously published version of the Holmes study. It wasn't Clery and it wasn't Bourne; if Noggs had been around to write the second version for QV it might be him, but it does't seem likely. As JW says, we need fresh evidence of who else might have ticked the boxes for authorship on both occasions. At the moment the money is still on Chard himself.

Steve

Ockham's Razor.

Neil

rusteze and Neil
Exactly. And the content has so much specific detail that ONLY Chard could know. The source can only be him.
Personally, for my money, I think Chard not having the ability to write a report and being dull-witted is a nonsense. As an officer he must have written hundreds of reports in his career and these were just two more. The reports of there being no paper about are also so much twaddle - maybe for the ordinary soldier - but for an official report on an action sanctioned by LC himself, there was paper.

The Engineers were the inteligensia of the army. The reports of Chard being a dullard need really to be put into the context of PTS, likewise with Bromhead. His report on the conduct of various men and his recomendation is neat and concise. I would believe that sitting around over the next couple of days with Bromhead etc would also have been a contributing factor of what happened and to whom. The ORIGINAl report, broken down into the timed elements and order of events is as goos as a report that one would wish to see really.

Whilst it was kind of Littlehand to post Dr. Holmes's original article, I must point out that Holmes's essay in Studies in the Zulu War III updates his original article and leans even more to the view that Chard was in fact the author of both reports (hence the altered title).
Frank
...and of course his Q Vic report is written sequentially in terms of events at RD.

For those following this... here is the footnote from
Dr Adrian Greaves in Journal 32 of the AZWHS.

[You must be registered and logged in to see this image.]

I see that Dr. Greaves has stated in his end note that Chard's first report is an "accurate and mistake-free account."

Acceptance at last that Adendorff was present for the defence?

Neil

I see that Dr. Greaves has stated in his end note that Chard's first report is an "accurate and mistake-free account."

Acceptance at last that Adendorff was present for the defence?

Neil.....

Hmm, not sure there is consensus on that!.. i have long believed that to be the case, that he did
indeed stay and assist in the defence, despite the report of his later arrest..

Got to say Neil loving the detail in your book.. how much input did Jim have? was it the bio's.

Hi Les,

I am of the opinion that Adendorff was present. I just wondered if Dr. Greaves had changed his stance, since he admits that Chard's report (which cites Adendorff as a defender) contains no errors.

Thanks for the comments about my book. Jim's contributed with the foreword and in giving permission to use a few of his images. I have to say that he's a thoroughly decent bloke and I was very glad when he agreed to write my foreword.

I'm glad you're enjoying it. Have you finished it yet Les?


Neil

Xhosa
I had rather hoped that my Brave Fugitive essay (pp. 65-66 in particular) had nailed the Adendorff question beyond doubt that Adendorff's presence was proven.
Apart from Chard, Hook saw him. With Chelmsford’s column when it returned, Norris-Newman, Stafford, Trpr. Symons, Harford all saw him (in fact Harford sketched him!)

Yes Julian you did nail it!. but there will always be naysayer's.. the question posed
did not directly ask for my opinion, and it must the case that some might of not
even seen your essay.. no doubts from me..

Indeedy! Then those that say 'nay' will also have to state 'why'.

"You say yes
I say no
You say why
And I say
I don't knowwww."
Would be interesting to get more from AG but don't hold your breath.

With all due apologies to those lads from Liverpool.

All you need is love.