Picture credit: Eric Harlow/Daily Mirror/Mirrorpix via Getty Images

Kenya’s history rewritten

Inflated Mau Mau death figures based on incorrect statistical projections have now become accepted as “fact”

This article is taken from the December-January 2024 issue of The Critic. To get the full magazine why not subscribe? Right now we’re offering five issues for just £10.

The alarm bells should have rung as soon as the Buckingham Palace press releases landed. The King’s first state visit to a Commonwealth country, to the former British colony of Kenya, would be an opportunity for him “to express regret over abuses during the Mau Mau Emergency of the 1950s”. The giveaway was the citation by the Palace of an estimate from the Kenyan Human Rights Commission that 90,000 Kenyans had been “killed, tortured or maimed during the British counter-insurgency”. 

There is no evidential basis for such a claim. 1,090 men had been executed after being convicted of capital crimes such as murder or supplying arms to the Mau Mau rebels. 10,540 Mau Mau had been killed in action — mostly by fellow Kikuyu rebelling against the reign of terror Mau Mau had inflicted on those resisting its attempt to win domination over Kenya’s largest tribe. But the notion that some 80,000 had been “maimed or tortured” was a complete invention. 

That the British Foreign Office had allowed the Palace to endorse such nonsense was no surprise — it had a track record in swallowing such claims. No less surprising was the re-emergence, as an “expert” witness in media coverage of the state visit, of Professor Caroline Elkins, the originator of much of the absurdity written about the Mau Mau uprising.

Census calculations

What first appeared as a belated footnote on page 429 of a 475-page tome, based on the PhD thesis of a junior Harvard historian, proved to be a bombshell. Caroline Elkins, the author of Britain’s Gulag, published in 2005, went on to become a leading light amongst liberal US commentators. Despite many hostile reviews, and detailed exposure of the glaring errors in her scholarship, Britain’s Gulag became — and remains — hugely influentual. 

What was the bombshell? Elkins had consulted the Kenya census tables of 1948 and 1962, which covered the eight-year period when the British authorities, having declared a State of Emergency in the face of a rising tide of violence, had suppressed a rebellion from within Kenya’s largest tribe, the Kikuyu. 

What came to be called “Mau Mau” (a phrase of unknown meaning and origin — the rebels called themselves “the Land and Freedom Army”) proved to be the most serious challenge British authority faced in any colony. It was suppressed with considerable force, and ruthless deployment of emergency legislation. The official history of the origins of Mau Mau listed 11,500 Mau Mau killed in action (a figure subsequently corrected to 10,540) and 1,090 hanged for murder or other offences (such as administering the notorious Mau Mau oaths).

But Elkins thought these figures much too low. She calculated that the population growth of three Kikuyu-speaking tribes between the two censuses had been far slower than that of three other tribes. From this she projected that up to 300,000 Kikuyu were “unaccounted for”. 

She posited three ways in which this happened: they were either killed by British forces and their auxiliaries during the shooting war against the Mau Mau; or they became casualties of torture, brutality or outright murder in the detention camps in which tens of thousands of Mau Mau suspects were held during the Emergency; or they fell victims of violence, disease and malnutrition in the 800 “protected” villages that the Nairobi government created to house the vast majority of Kikuyu civilians. 

These guarded villages were designed to cut the Mau Mau guerrillas off from supplies and food, and also keep those who had refused to take the Mau Mau loyalty oaths safe from reprisals. These reprisals had been on such a scale that one District Commissioner estimated “villagisation” had saved 20,000 civilian Kikuyu lives. Elkins saw the villages as no better than Nazi concentration camps, yet provided nothing other than anecdotal evidence to support her claims as to numbers of deaths. But then she came up with her census analysis.

That there was convenient selectivity in the presenting of Elkins’s demographic evidence was immediately apparent; at least, to me. She had set out the separate population sizes for each of the three comparator tribes (Kamba, Luo and Lahya) for both censuses. But for the three Kikuyu-speaking tribes (Kikuyu, Embu and Meru), she only offered a combined total for all three. 

A check with the published census documents revealed the truth: the population of actual Kikuyu had grown no more slowly than the comparator tribes. But the Meru and Embu tribes, who spoke Kikuyu, but were only marginally involved in the uprising, saw their populations increase slowly, or actually decline, in this period. 

So, the Meru and Embu comprised 80 per cent of Elkins’s “unaccounted for”, despite the fact that neither of these two tribes were greatly targeted by the British. Thus the whole claim was based on a statistical illusion through Elkins’s withholding of crucial information.


Elkins’s Britain’s Gulag was published in the same week as another major work by David Anderson. His Histories of the Hanged dealt primarily with the 1,090 Mau Mau executed during the Emergency, but also addressed such issues as the number of detainees. The two books were jointly reviewed in scores of newspapers and journals, rarely to Elkins’s advantage. 

In her book, Elkins thanked the producers of Kenya: White Terror, a BBC documentary broadcast in 2002 for bringing “her research to the screen”. Indeed, her researcher, using a different name, was credited as the programme researcher. By the time Britain’s Gulag was published, Ofcom had ruled that the documentary had unfairly treated the former district commissioner in the colony, Terence Gavaghan, whose responses had been carefully edited to make him appear evasive — rather than reflective — about the treatment in detention camps. 

In her book’s acknowledgements, Elkins also thanked John Lonsdale, describing him as “the most gifted scholar I know”. She separately cited Professor Lonsdale (who is widely regarded as the pre-eminent British historian of East Africa and is a former Director of Studies in History at Trinity College, Cambridge) as having helped with her bogus demographic analysis. 

In fact, whilst Lonsdale did initially draw from Elkins’s statistics the supposition that a population growth shortfall might have been caused by declining fertility during the Emergency, he regarded the idea of 300,000 or even 400,000 Mau Mau deaths floated by Elkins in the BBC documentary as incredible. 

Elkins ignored his advice not to rely on anecdotal testimony unless it was strongly corroborated. She included the testimony from a supposed witness to the Hola massacre, whose account is so laughably false — with at least a dozen “facts” invented — that one can only conclude he was nowhere near the events he so dramatically described. That any serious historian could take this nonsense at face value is very surprising: but Elkins did.

A series of reviewers listed the elementary errors she committed. Lonsdale further distanced himself from Elkins, describing her — in a published review and in private letters — as a “radical supporter of Mau Mau” who understood neither the movement’s origins and actions, nor the relationship between the British and the Kikuyu.

The review in the Journal of African History by Bethwell Ogot — doyen of Kenya’s historians — criticised Elkins’s reliance on oral history: was she certain these were not bogus claims, designed for commercial gain? Elsewhere, he called her “dishonest” and said she had “simply imbibed Mau Mau propaganda rather uncritically … despite massive evidence to the contrary”. As for the Kikuyu who chose to oppose Mau Mau, “Elkins dismisses them simply as imperialist lackeys”, whilst “seeing no atrocities on the part of Mau Mau” (Ogot helpfully listed some of the more gruesome practices).

Susan Carruthers, in Twentieth Century British History, derided Elkins for reducing loyalist Home Guards to “co-opted venal stooges of white power”, while turning settlers and administrators into “cartoonish grotesques”. Her invoking genocide was when “she proves the least reliable guide to history: it is unfortunate, to say the least, that Elkins should confuse Hutu and Tutsi in this formulation”. What happened in Kenya “was not genocide: history is not well served by its sloppy invocation”.

Pascal Imperato, in the African Studies Review, criticised Elkins for “obvious subterfuge”: “draping herself in an academic mantle” when, long before finishing her book, she had “for some time been campaigning for monetary compensation for Mau Mau ‘victims’”. He scorned the “fancy statistical turns on thin data” that resulted in “flawed deductions from the 1948 and 1962 censuses that would astonish demographers”.


It was Lonsdale and Anderson who put Imperato’s prediction to the test. They encouraged an expert demographer, Dr John Blacker, to write an article for the Journal of African Affairs, published in April 2007, entitled “The Demography of Mau Mau”. 

Blacker had worked on the 1962 census in Kenya, and was invited back by post-independence governments to work on subsequent censuses. His verdict on Elkins was devastating: there was “no evidence” to support her claims. They were “based on a misunderstanding of the data”, which she had not properly adjusted for “changes in the tribal classifications used in the 1948 and 1962 censuses”. Most importantly, the age/sex pyramids (the standard tools of demographers) for the Kikuyu showed none of the deep indentations that the loss of hundreds of thousands of adult males would have caused (there were only 400,000 to begin with). 

Calculating back from later censuses, Blacker estimated that the average annual growth of the entire African population between 1948 and 1962 had been 2.7 per cent per annum, with the Kikuyu-speaking tribes growing at 2.65 per cent per annum. He concluded from this and other data sets that the number of “excess” Kikuyu deaths in the period was in the range of 30,000 to 60,000, with the likeliest number being 50,000.

A surprising detail emerged. After examining shifts in mortality rates, he concluded that more than half of the 50,000 (26,000) were children under the age of 10, and the remainder adult males (17,000) and adult females (7,000). He was also confident that the discrepancies he found were excess deaths, not a shortfall in births as a result of a drop in fertility.

On one point he was willing to give Elkins modest support: insofar as children and adult females were concerned, there might have been deaths, as she had argued, as a result of food shortages and increased incidence of disease when the bulk of the Kikuyu population was gathered in the guarded villages. Elkins, of course, had also claimed that fatalities amongst the adult women had resulted from British ill-treatment.

Complaining to the Guardian 

Elkins was undeterred, insisting that Blacker had not addressed her “methodology”. Perhaps Blacker’s reticence on this point did Elkins a favour. Her “methodology” was to take three (why three and why these three?) of the 12 non-Kikuyu tribes, average their very different reported population growth rates, and then compare this average to that for the Kikuyu-speaking tribes (without adjusting for non-comparable components as between 1948 and 1962, or revealing that the three had wildly different growth rates). 

But Elkins still has plenty of support. The Guardian gives her a platform for her views, and the newspaper has regularly used her statistical analysis in its reportage. In one article, it attributed to Elkins the claim (actually made in error by the reviewer of Britain’s Gulag for the London Review of Books) that 100,000 Mau Mau had been killed in the detention camps and guarded villages. 

I wrote to the Guardian to protest. I actually wrote eleven letters, all ignored, including directly to the editor and to the chair of the Scott Trust (which owns the Guardian). I eventually complained to the newspaper’s ombudsman and after a lengthy inquiry he advised that a letter from me should be published. I then had to negotiate this with the managing editor, who insisted that the Guardian’s lawyers read my draft before it could be accepted.

After the agreed text was published, I was subjected to personal abuse in a letter from another Guardian contributor, Richard Drayton, an Elkins supporter at King’s College London, who dismissed Blacker as a colonial stooge (to the amazement and dismay of Lonsdale and Anderson) and insisted (in a signal display of ignorance of both statistics and demography) that the margin of error in Blacker’s calculations could allow for Elkins’s vast estimates. (He was wrong: it could not.) In Drayton’s view, I, as a “TV businessman”, had no place in this argument, anyway. 

In private, Elkins, too, has shrugged me off as “not an academic”, a “Murdoch man” and “a Conservative”. As it happens, I was the youngest student ever to be awarded a double first in history at Cambridge was responsible — as writer, director, producer or executive producer — for dozens of historical documentaries, including three episodes of The World At War. I have served as a visiting professor at three universities, including Oxford. Although I worked at Sky for four years (during which time I spoke to Rupert Murdoch on the phone for a total of one minute), I have also served on the boards of companies hostile to Sky for 20 years. I have never been a member of a political party, and have only once voted Conservative in 11 elections. 

£20m payout

Hostile reviews from experts did not dent the success of Elkins’s Britain’s Gulag. It sold well and won a Pulitzer Prize: which would have done no harm to her successful bid for a tenured professorship at Harvard. Her greatest triumph, though, came in 2011, when a class action claim for compensation, mounted by 5,228 Mau Mau veterans who alleged torture by colonial officials during their time in detention, succeeded: a claim she had supported for many years.

Five test cases had been chosen: two had been withdrawn or dismissed when the UK’s defence was torpedoed by the belated discovery of a trove of 8,800 files from various colonies — including 1,500 from Kenya. The Kenyan documents had not been released to the plaintiffs, despite repeated requests, on the grounds that they did not exist: that is, until they were unearthed by a mid-level official. They had been tucked away in a wing of a country estate owned by the Foreign Office, called Hanslope Park.

This gift to the claimants was rendered even more potent when the Guardian pointed out that Hanslope Park was in part protected by barbed wire fencing. A colleague of Elkins at Harvard, Maya Jasanoff, was inspired to write a piece for the New Yorker, in which she claimed that what came “pouring out” from the “secret stash” held “behind razor wire” at a “facility dedicated to keeping secrets” were “records of systematic, wide-ranging, stomach-churning abuse”.

David Anderson felt particularly vindicated by the revelation, as he had been the historian pressing hardest for files he was sure had been withheld. He later published examples of documents that might have supported the torture litigation if it had not been suddenly settled by the UK government, paying out £20 million to the claimants, as well as funding the erection of a statue in Nairobi in honour of one of Mau Mau’s leaders, Dedan Kimathi. To seal the deal, Foreign Secretary William Hague issued a fulsome apology.

Forgotten files

But once the dust had settled, the truth about the Hanslope disclosure emerged. The files had actually been stored for decades in an anonymous warehouse in Hayes, in West London (no barbed wire there). Whatever their markings, the files were not regarded as secret, and historians who knew what they were looking for were welcome to delve. 

However, the junior staff assigned to Hayes never catalogued the collection, and repeatedly asked more senior specialists at The National Archives (TNA) at Kew to take them over. This request was repeatedly rejected, with TNA assuring the Foreign Office that the files — notably, those from Kenya — simply duplicated those which they already had at Kew, and which were nearly all available to the public. 

When the Hayes lease expired after more than 30 years, the Foreign Office simply transferred all the boxes to one of their own establishments. By this time, the staff who knew at least a little about the Hayes archive had mostly retired. The entire collection was simply forgotten about.

Anderson’s claim that the files were of high significance was at odds with the view of TNA. Having read scores of them, I share TNA’s scepticism. One of the documents flourished by Anderson as proof of his contention was a report by Sir Vincent Glenday on conditions at the controversial Home Guard posts, where Mau Mau suspects had been extorted, beaten, tortured and even killed. In fact, the Glenday Report had never been secret. Indeed, it had been lodged in the House of Commons library in July 1955.

 “Moral balance sheet” 

Whatever the reality, William Hague’s apology to the claimants gave official imprimatur to the Elkins’s approach to Kenyan colonial history. Inexplicably, Foreign Office officials decided to adopt the formulation with which she had introduced her book, using a “moral balance sheet”: “the murders perpetrated by Mau Mau adherents were quite small in number when compared to those committed by the forces of British colonial rule. Officially, fewer than 100 Europeans, including settlers, were killed, and some 1,800 loyalists died at the hands of Mau Mau. By contrast, the British reported more than 11,000 Mau Mau were killed in action.”

This is a most disingenuous presentation of the numbers in the report on the origins of Mau Mau, written by a senior colonial official called Frank Corfield. Tacked on to the last page (it was not even included in the first draft) was a brief table listing Mau Mau and European fatalities, as well as European and African (ie Kikuyu) civilian deaths: but these almost all related to the period after the Emergency was declared in October 1952.

As it happens, the archives at Kew reveal the 1959 revised estimates from the Ministry of Defence of 10,540 Mau Mau killed and 1,832 African (ie Kikuyu) civilians killed. Most of these civilians would have been Home Guards killed — along with their families — when their guard posts were overrun. (Mau Mau barely targeted European settlers: more were killed in road accidents during the Emergency than by Mau Mau.)

But (with a few dozen known exceptions) this is all after October 1952. What of the killing balance sheet in the period leading up to the imposition of Emergency powers?


According to Corfield, what took place before the Emergency was declared was “a holocaust”. For two years, Mau Mau oath administrators operated with near impunity. The Kikuyu heartland was very lightly policed: just 200 colonial district officers, their assistants, and some regular and tribal police oversaw a population of 750,000. 

Loyal chiefs and headmen had minimal ability to resist Mau Mau. Many of them were forced publicly to swear allegiance or be put to death: the standard punishment for refusal was shooting, strangling or hanging. Mass oathing meetings took place in full public view. Anyone brave enough to report any executions faced the same fate. One senior chief, encountering a crowd of hundreds being oathed by Dedan Kimathi, tried to intervene, and was hacked to death, along with his two police bodyguards.

Corfield concluded that “no-one will ever know how many thousands of ‘missing’ Kikuyu died in this way” (Anderson, in offering a précis of Corfield, substituted “hundreds” for “thousands”, without telling his readers he had done so). Corfield referred to confession evidence from later in the Emergency, saying that “the number of Kikuyu who were murdered will never be known, but in Kiambu district alone over 500 bodies were eventually found”. Sir Philip Mitchell, Kenya’s Governor until 1952, also referred to thousands of deaths in his memoirs. 

It was his successor, Sir Evelyn Baring, who had called for a State of Emergency almost as soon as he took office. Three years later, in 1955, he told the Colonial Office: “from the number of corpses produced during the last few months, we now know that the number of murders far exceeded anything imagined in 1952”.

The number of confessions increased greatly in later years. The Special Branch inspector who supervised nearly 70,000 interrogations, Ken Lees, estimated that 20 per cent of those who confessed to him or his team admitted to one or more murders. He, too, carried out exhumations to check the truth of these admissions, and found each to be accurate.

Yet burial was a rare outcome of Mau Mau executions. Most bodies were left for wild animals to eat, or thrown in rivers or ravines, or dropped in wells or deep latrines. That so many bodies were recovered suggests very widespread killings. One colonial official reported finding “large numbers of corpses with rope around their necks” — the fate of those “convicted” by Mau Mau “courts”. Remarkably, Nairobi’s chief pathologist, Dr Morris Rogoff, retained the disinterred remains of 475 victims of Mau Mau violence, never claimed by their families. 

One clue to the number murdered is the number of murderers. By 1957, the authorities had identified nearly 1,700 self- confessed murderers, nearly all living openly, neither detained nor imprisoned. A total of 346 Mau Mau were hanged for murder during the Emergency. 

So we have nearly 2,000 known murderers, but have barely touched on those amongst the detainees (by Ken Lees’s calculation, at least 10,000), let alone the Mau Mau who were killed in action (presumably the most committed of the rebel cadres).

Deaths of children

It was only after Lees first contacted me, less than a decade ago, that I went back to Blacker’s paper, to see how Lees’s estimate of civilian deaths could be reconciled with the figures Blacker had calculated. Sadly, by then, Blacker had died, and I could not put to him what now struck me about the many detailed mortality tables he had included in his essay. It seems that even he had not realised the significance of one of them, which tracked the increase in mortality of children aged 0-5 between 1949 and 1959.

This showed a sharp difference between the first half of the decade and the second, with a rise of 22 per cent in mortality rates in the first five years, but only 17 per cent in the second five years. 

As this 0-5 cohort formed by far the majority of the 0-10 age group (amongst whom Blacker had reckoned there were 26,000 “excess” deaths in the decade as a whole) it would appear that at least 15,000 of these deaths occurred before August 1954 — at which time there were almost no “protected” villages. So the factors cited by Elkins, and repeated by Blacker himself, could not have been relevant. It seems more than likely that these 15,000 — and perhaps many more — were the victims of violence, not disease or malnutrition.

Who would have killed them? The British armed forces? But they were not even deployed until late 1952. Lees’s explanation makes the most sense: the under-10s, unable to run as fast as their older siblings, would have been killed along with parents who refused to take the Mau Mau oaths. It was also the case that Mau Mau recruited children over the age of 10, which might account for a part of the discrepancy in death rates between the under-10s and over-10s.

It seems historians simply cannot get their heads around the reality of what happened. Indeed, in terms of “moral balance”, given that more than half the Mau Mau deaths were inflicted by Home Guards, nearly 90 per cent of all Kikuyu deaths in the period would have been at the hands of other Kikuyu.

There was no genocide in colonial Kenya. Insofar as there was what Corfield called a “holocaust”, Mau Mau were not its victims but its perpetrators. Perhaps Elkins’s confusion between the Hutu and the Tutsi in Rwanda was a Freudian error, rather than the product of sloppy research.

No apologies

The iniquities of the colonial system, based on force, and lacking any meaningful consent, should not be downplayed. For all the improvements in health, education, agriculture, life expectancy and infrastructure, colonial Kenya was a deeply racist and unjust society. Even before the Mau Mau Emergency, there were 20,000 Africans detained without trial each year.

Those Mau Mau who could prove they were brutalised or tortured were entitled to make their claims, even if there is something deeply unedifying in Mau Mau survivors using the UK courts to seek compensation, when their own victims — any that survived the slaughter — have nowhere to look for restitution and have been effectively airbrushed from the history books.

No current historian seems prepared to embrace the implications of Blacker’s paper or listen to Lees’s testimony. Perhaps the tide of revisionism is too strong for any academic to swim against it. Neither Elkins, nor her publishers, nor Harvard, nor the Pulitzer Committee (“meticulously researched history of the highest order”) has ever acknowledged her methodological failings, let alone apologised for them. Instead, they have become the commonly accepted narrative, effectively acknowledged even by the British government.

Today, Elkins holds three professorial posts at Harvard — at the university, where she is director of the Center for African Studies, at the Law School and at the Business School. She writes for the New York Times, the Guardian, The Atlantic and The New Republic. Her latest book, Legacy of Violence, expanding her critique of the British Empire, was listed as a book of the year by the BBC’s History magazine. Her influence continues to grow.

Demand for reparations

King Charles’s state visit to Kenya offered Elkins another opportunity. She used an article in The Observer to demand not only that the King apologise to Mau Mau (“spit it out: say you are sorry”) but that reparations should be paid, if need be from his own pocket (which she estimated, improbably, to contain some £20 billion). 

Old claims from her book were recycled. In the Sunday Telegraph, Elkins was cited as an expert, and her estimate that 160,000 Kikuyu had been detained during the Emergency was solemnly quoted. A BBC online essay used another of her detainee estimates: 320,000 (or four times the official total). She had indeed used both figures — neither with any supporting evidence — as the upper and lower limit of the “actual” number of detainees. 

No other historian has endorsed either number, not least because the higher figure represents nearly the entire male Kikuyu population at the time. In The Times Ben Macintyre wrote an article containing ten mis-statements of fact, including the bogus assertions that there were “160,000 detainees” and “90,000 killed or maimed”. 

Some media coverage even implied that the proscribing of Mau Mau after Kenya became independent was somehow an absent-minded failure to rescind colonial legislation. In fact, Jomo Kenyatta, who had seen Mau Mau in close-up before the Emergency, imposed the proscription when he became Kenya’s first president because he had no doubt how much harm the “hooligans” might do if left unrestricted. The ban lasted 40 years. It is only recently that Mau Mau has re-emerged as the heroic deliverer of independence and supposed victim of British violence. 

When King Charles in his speech at a state dinner expressed “regret” for the “unjustifiable violence” inflicted on Mau Mau detainees, no thought of the African victims of Mau Mau seemed to enter his head, even though it was the British colonial government that failed to halt the Mau Mau slaughter before October 1952. 

Anticipating the age of “truthiness”, the Mau Mau tale-spinners have seized the narrative, and Caroline Elkins, their true believer, will blight the discipline of history until the academy itself chooses to rebel.

Enjoying The Critic online? It's even better in print

Try five issues of Britain’s newest magazine for £10

Critic magazine cover