By Dr. Habib Siddiqui 13 February 2019
Abstract
Genocidal crimes don’t happen in vacuum and require groundwork from the racist and bigoted elements to state and non-state actors to prepare the support base and mobilize the dominant group to perpetrate such heinous crimes. For decades, the Buddhist majority in Myanmar has denied the very existence of the minority Rohingya people who are mostly Muslims and falsely portrayed them as intruders from the district of Chittagong of Bangladesh. As a result, the Rohingyas are victims of genocide. Since 2017 nearly 24,000 of them have been slaughtered, as part of a national project, in Buddhist Myanmar simply because of their distinct ethnicity and religion. Nearly a million of them have been forced to seek refuge inside Bangladesh.
In this paper, the British -era demographic data on the Rohingya people who are mostly settled in their ancestral land of Arakan (Rakhine), bordering today’s Bangladesh are analyzed statistically. The results of the analysis debunk the popular Buddhist myth of Bengali or Chittagonian mass migration from Chittagong in East Bengal (today’s Bangladesh) to Arakan during the British rule of India.
Introduction
Genocidal crimes don’t occur by chance and require years of preparation. The rationale for justifying such horrendous crimes is often provided by the evil geniuses or the ideologues within the dominant group. This ominous task is sometimes shared by other powerful groups within the perpetrating community, e.g., religious, business, social and political leaders, let alone a chauvinist government, stimulating the general population to participate in the elimination campaign against the targeted group, which is viewed as ‘different’. As a matter of fact, ideologues like Julius Streicher acted as the catalysts to expedite the ‘Jewish solution’ in Nazi Germany. They popularized Hitler’s fascist agenda and made his mission a national project inside Germany.
In the context of Burma (now called Myanmar), we see similar state and non-state actors at play to eliminate the indigenous Rohingyas of Arakan (renamed the Rakhine state). Evil, ultra-racist intellectuals like Aye Chan, (late) Aye Kyaw and Khin Maung Saw played the role of Julius Streicher portraying the Rohingya as the infiltrators or intruders and the ‘enemies within’ from the next-door Bangladesh who needed to be killed or wiped out as a ‘virus’, planting unfathomed intolerance against them while the bigoted monks like Wirathu provided the religious rationale for ‘purifying’ the Buddhist motherland from the minority Muslims who are falsely depicted as a ‘threat’ to Buddhism and its way of life. The rest is history! With the direct involvement of all the organs of the government – from top to bottom, from the central to the local level – the genocide of the Rohingya became a national project inside Myanmar, which witnessed the massacre of some 24,000 and mass rape of tens of thousands in the last couple of years alone, plus the forced exodus of nearly a million of them to Bangladesh. More Rohingyas now live outside their ancestral land than inside the Rakhine state of Myanmar.
In spite of the decades of persecution against them since the Japanese occupation of British Burma during the Second World War, the Rohingya Muslims and Hindus have maintained their continuous existence in Arakan that is older than the ethnic Rakhine Buddhist majority who also share the same landmass. [The interested readers may like to read this author’s book – The Forgotten Rohingya: Their Struggle for Human Rights in Burma, available in the Amazon.com.]
In my book – Muslim Identity and Demography in the Arakan State of Burma (Myanmar) – I have analyzed the British era data and put to rest the xenophobic claims of Bengali (or more particularly Chittagonian) infiltration to Arakan by showing that there is no credible evidence supporting that myth, propagated by the criminal Burmese government and its brainwashed supporters within the Buddhist community. And yet, seemingly, there is no shortage of such Buddhist enthusiasts who continue to justify the genocidal crimes of their criminal government against the Rohingya with their half-baked theories and faulty analyses.
Consider, e.g., the work of U Ne Oo (dated 18th August 2014) who claims to be the Coordinator for an organization called the Network for International Protection of Refugees (www.netipr.org/policy/node/39). The pdf file had a catchy title – Rohingya/Bengali: Migration After First Anglo-Burman War – a subject of much interest to me for nearly two decades.
Oo claims to examine competing claims from the pro-Rohingya and Burmese government sources and concludes “that the majority of present day Rohingya/Bengali group had migrated after 1824...”
Analysis
A reading of Ne Oo’s article shows that it was purely a propaganda attempt aimed at justifying the xenophobic treatment of the Rohingyas of Myanmar. It was not surprising that Oo uses Myanmar government source, quoting U Myint Thien’s 2009 paper. The latter is shown as the Director of Historical Research Department, Burma, which had the task of justifying government position on contentious issues like the Rohingyas.
As a highly biased author, Oo has the habit of cherry-picking texts to suit his purpose. He does not surprise us, therefore, when he quotes Rev. Comstock who allegedly was on a mission to Arakan between 1834 and 1844. The Christian missionary or priest is presented as a ‘reliable source’ and quoted by Oo about demography inside Arakan (now called the Rakhine state). Oo writes, “On Page 224, describing the numbers of inhabitants:
“The population of Arakan at the present time (1842) is estimated at about 250,000. Of these, about 167,000 are Mugs, 40,000 are Burmese, 20,000 are Mussulmans, 10,000 are Kyens, 5,000 are Bengalese, 3,000 are Toungmroos, 2,000 are Kemees, 1,250 are Karens, and the remainder are of various races, in smaller number.”
Any unbiased researcher would have problem taking such figures seriously from a priest whose task surely was neither civil administration nor population studies. The figures he cites cannot be reconciled with the population figures reported in 1826 by Mr Paton who was the first administrator of Arakan soon after the East India Company annexed the territory in the first Anglo-Burma War. There, Mr Paton reported that the total population of Arakan did not exceed 100,000 of which 60,000 were Maghs (Arakanese Buddhists) and 30,000 (Rohingya) Muslims.” [The rest being Burmese settlers from outside Arakan.]
Even if one were to assume Comstock’s figures, it is unlikely that the total population had grown to 2.5 times within a mere 16-years period without migration from other territories (since it would require an absurd annual growth rate of approx. 6% for population to grow organically by 150%). How could a Mug (Magh) population grow to 2.78 its size from 60,000 to 167,000 within the same period, if it was not for infiltration from elsewhere in British Burma, when all those territories from Bengal to Burma were under the British rule?
However, Oo is evasive on such anomalies. He does not question why the population of Muslims in Arakan reduced to 20,000 within the same period, a 33% reduction, while the Buddhist Mug population rose by 178% and the Burmese Buddhist population quadrupled. His despicable bias could not be hidden under the rug!
If Oo was sincere and unbiased in his study, he could not have ignored such anomalies with the Buddhist population inside the Arakan state in those early years of British annexation.
Oo cites a table (courtesy of Burmese propagandist Myint Thien) of Maung Daw area (which is close to today’s Bangladesh border) to show the impact of Burma Act 1935. I reproduce the table below before analyzing the data statistically (note the arithmetic error in the net migration column for 1929).
| Year | Immigration | Emigration | Net Migration |
| 1928 | 324000 | 278000 | 46000 |
| 1929 | 307000 | 398000 | -92000 |
| 1930 | 273000 | 365000 | -92000 |
| 1931 | 252000 | 205000 | 47000 |
| 1932 | 240000 | 225000 | 15000 |
| 1933 | 220000 | 200000 | 20000 |
| 1934 | 299000 | 185000 | 114000 |
| 1935 | 216000 | 188000 | 28000 |
| 1936 | 219000 | 184000 | 35000 |
| 1937 | 199000 | 195000 | 4000 |
Based on the data above (supposedly for the Muslim population alone), Oo concludes, “On examining the pattern of Indian immigrant/emigrant from above Table in 1921 to 1937,
there is a surge in 1934 by Indian immigrants who had opted to remain in Burma.”
Such a conclusion regrettably betrays the underlying facts of seasonal workers and lacks common-sense logic in the British-ruled India and Burma. While on the surface, the above table may give the impression that there was a net gain in population in Maung Daw since 1934 and that there was a net gain of 12,600 people in Maung Daw over a period of ten years, one cannot ignore the fact that this net gain boils down to a very small number of only 1,260 people per year compared to the average annual immigration of 254,900 people and emigration of 242,300 (see the graphical summary reports below). That is, the net gain in population was less than 0.5 percent. The data above also fails to explain the net loss of 137,000 Muslims in the first three years. How could there be more people opting for emigration than immigration?
As I have noted in my earlier work on Muslim demography in Arakan, the so-called immigrants were all seasonal workers who returned to their ancestral land inside Bangladesh once their tenure or temporary job was completed, and/or no new employment availed them. Treating population data in an isolated or static way as if they are immobile like building structures is ludicrous. Sadly, Oo not only makes such attempts but also makes conclusions that are symptomatic of cherry-picking with data without ever asking the question – why? But more problematic is his utter lack of statistical knowledge with demographic data while claiming to draw conclusions from the data to support the criminal xenophobia against the minority Rohingya people, which has resulted into genocidal crimes against the Rohingya by his fellow Buddhist marauders inside Myanmar.
In what follows, I share statistical analysis of the data for both immigration and emigration within the said period. (Note: we are limited here by small data sets, i.e., 10 yearly data.)
The graphical summary of the immigration data shows that the data follow normal Gaussian distribution function while the emigration data fails to exhibit normal behavior with a probability (P) value far below 0.05.
The Individual and Moving Range (I-MR) charts for the immigration data below do not exhibit any out of control points while again the emigration data, as pointed out earlier, do show such unusual trends for the 2nd and 3rd year (i.e., 1929 and 1930).
In order to compare the two sets of data, immigration vs. emigration, we shall employ three statistical tests. The firs of these is the 2-sample t-test (which assumes normality). The p-value of 0.665 below show that there is not enough statistical evidence to reject the null hypothesis that the two samples are same.
Two-Sample T-Test and CI: Imm, Emi
Method
| μ₁: mean of Imm |
| µ₂: mean of Emi |
| Difference: μ₁ – µ₂ |
Equal variances are not assumed for this analysis.
Descriptive Statistics
| Sample | N | Mean | StDev | SE Mean |
| Imm | 10 | 254900 | 43636 | 13799 |
| Emi | 10 | 242300 | 78825 | 24927 |
Estimation for Difference
| Difference |
95%
CI for Difference |
| 12600 | (-48507, 73707) |
Test
| Null hypothesis | H₀: μ₁ – µ₂ = 0 |
| Alternative hypothesis | H₁: μ₁ – µ₂ ≠ 0 |
| T-Value | DF | P-Value |
| 0.44 | 14 | 0.665 |
Next, we employ the Mann-Whitney non-parametric test, which with a p-value of 0.212 show there is not enough statistical evidence to reject the null hypothesis that the two samples are same.
Mann-Whitney: Imm, Emi
Method
| η₁: median of Imm |
| η₂: median of Emi |
| Difference: η₁ – η₂ |
Descriptive Statistics
| Sample | N | Median |
| Imm | 10 | 246000 |
| Emi | 10 | 202500 |
Estimation for Difference
| Difference |
CI
for Difference |
Achieved Confidence |
| 28500 | (-41000, 78000) | 95.48% |
Test
| Null hypothesis | H₀: η₁ – η₂ = 0 |
| Alternative hypothesis | H₁: η₁ – η₂ ≠ 0 |
| W-Value | P-Value |
| 122.00 | 0.212 |
Next, we employ pairwise comparison to test if the two data sets are statistically different. Again, the p-value of 0.54 shows there is not enough statistical evidence to reject the null hypothesis that the two samples are same.
Paired T-Test and CI: Imm, Emi
Descriptive Statistics
| Sample | N | Mean | StDev | SE Mean |
| Imm | 10 | 254900 | 43636 | 13799 |
| Emi | 10 | 242300 | 78825 | 24927 |
Estimation for Paired Difference
| Mean | StDev | SE Mean |
95%
CI for μ_difference |
| 12600 | 62475 | 19756 | (-32092, 57292) |
µ_difference: mean of (Imm – Emi)
Test
| Null hypothesis | H₀: μ_difference = 0 |
| Alternative hypothesis | H₁: μ_difference ≠ 0 |
| T-Value | P-Value |
| 0.64 | 0.540 |
Conclusion: There is not enough evidence to reject the null hypothesis that the two sets of data are the same. That is, the immigration and emigration data are statistically similar.
Concluding Remarks
This study fails to find statistical evidence in support of Oo’s claim that “the majority of present day Rohingya/Bengali group had migrated after 1824...” The tabulated data shared by Oo failed to prove that during the British-era Muslims had moved into Maung Daw (and for that matter into other parts of Arakan). On the other hand, something that I have noted in my book – Muslim Identity and Demography in the Arakan State of Burma (Myanmar) – there is credible evidence to suggest that there was unnatural growth of Buddhist population inside Arakan during the early decades of the British occupation. To put succinctly: the so-called influx to Arakan was caused by the Rakhines (and other Buddhists) and not the Rohingyas (or the so-called Chittagonians from Bangladesh or British-ruled East Bengal).
Far from the mission of protecting the refugees, as his organization’s name suggests, Oo’s flawed work appears to be a smokescreen to justifying Myanmar government’s on-going criminal activities that have resulted into forced exodus of the Rohingya people than anything else. Such attempts by chauvinist individuals are simply inexcusable.