Hanging for murder in late colonial Burma

Abstract

The first decades of the twentieth century saw a marked rise in the number of recorded murders in British-ruled Burma. The sentence for murder prescribed by the Indian Penal Code—in force in Burma as a province of British India—was death or transportation for life. However, while Burma's murder count was rising sharply, the number of executions taking place in the province's jails remained broadly stable, with perhaps a tendency to fall, and indeed it fell decisively at the end of the 1930s. By that point, a high standard of evidence was being increasingly demanded for conviction for murder and increasingly the death penalty was being imposed only when murder had clearly been premeditated. In other words, in the administration of justice in murder cases, the rule of law had been markedly tightened. The focus in the article will be on the local pressures, on the practices and perspectives of Burma's colonial administration and of the Burmese themselves that led in time to a decisive fall in the number of convicted murderers who were hanged for their crime.

The first decades of the twentieth century saw a marked rise in the number of recorded murders in British-ruled Burma. The rise was sustained, with few significant reverses and indeed three major surges, in the first half of the 1920s, at the beginning of the 1930s, and in the final years of that decade. In the five years 1905–09 there had been on average 345 true cases of murder recorded by the police each year: in 1935–39, the figure was 876.¹ The sentence for murder prescribed by the Indian Penal Code, 1860, section 302—in force in Burma as a province of British India—was death or transportation for life. Yet, it is interesting to note, during those first decades of the twentieth century when Burma's murder count was rising sharply, the number of executions taking place in the province's jails remained broadly stable with perhaps a tendency to fall, although there were also occasional pronounced spikes. In the five years 1905–09, there had been on average 72 executions each year: in 1935–39, the figure was 70.² As a rough calculation, a murderer in 1905–09 had a one in five chance of being hanged: in 1935–39, the odds had lengthened to one in twelve.

For some British officials, the falling execution rate explained Burma's rising murder count, at least in part. In his 1936 annual report, Burma's Inspector-General of Police commented that ‘[t]he fear of hanging need not … deter anyone from murder as the odds against it are [now] more than 8 to 1’.³ The position underpinning the Inspector-General's comment—that a clear prospect of being hanged is a strong deterrent to murder—is contentious to say the least.⁴ In fact the issue here is not whether hanging is a deterrent to murder (it is not), but whether it has a significantly or even marginally greater deterrent effect than an alternative penalty. In the context of late colonial Burma, this may open up an interesting line of inquiry. As noted above, in colonial Burma the alternative to hanging was transportation for life, transportation by ship to the penal colony on the Andaman Islands in the Bay of Bengal. Interestingly, however, it would appear that the empire's officials, notably in British India, often held that, for cultural reasons, transportation was more feared than the death penalty.⁵ In other words, hanging was the lesser not the greater deterrent. But then in cases of unpremeditated murder, murder committed without prior reflection, perhaps neither hanging or transportation would deter. To pursue the Inspector-General's comment above and the positions that underpinned it would require a detailed, nuanced examination of each of these complex and methodologically fractious issues. Sadly, the material in Burma's colonial archives is insufficiently rich to sustain that examination.

This article therefore takes a quite different direction, to seek to understand why in the first four decades of the twentieth century, when colonial Burma's murder count was rising sharply, the number of executions taking place in the province's jails remained broadly stable with perhaps a tendency to fall, and indeed fell decisively at the end of the period. Increasing public and political opposition in Britain to the death penalty for murder in the 1920s and 1930s, feeding into imperial opinion, may well have been a significant influence. But the focus here will be on the local pressures that led over time to a substantial fall in the proportion of convicted murderers who were hanged for their crime. Some of those pressures were quite specific. For example, the following section will explore at length the argument frequently made by British officials at the time that the prospect that an accused murderer if convicted would hang was significantly reducing guilty verdicts in murder cases and allowing the almost-certainly guilty to walk free. But the fall in the proportion of convicted murderers being hanged in this period may also have reflected, in ways that are often difficult to plot precisely, major shifts then taking place in Burma's politics, economy, and society.

To illustrate that argument: the inter-war decades saw substantial advances in colonial Burma's formal political structures, advances that put Burmese politicians into positions of power. In 1923, a dyarchy government structure was established, in which Burmese ministers for the transferred functions (agriculture, excise, health, public works, forestry, education) were responsible to a largely-elected legislative council. That structure was discarded under the 1935 Government of Burma Act, which came into effect in 1937, and replaced with a cabinet of nine Burmese ministers, led by a prime minister and responsible to an elected House of Representatives.⁶ Burmese ministers and their senior officials no doubt had their own views on hanging convicted murderers, that is its impact in deterring murder but also its moral justification. To further illustrate the argument above: through those same decades, colonial Burma experienced considerable political, economic, and social turmoil, crises which seriously challenged the colonial state and often (the point here) its administration of criminal justice. Burmese nationalist politics, emerging at the end of the First World War and thereafter near-relentless in its agitation against the British presence, was riven by intense internal rivalries that, fuelled by an unruly vernacular press, often spilled over into disorder and violence. ‘[T]he absence of any generally accepted standards of permissible public conduct or tactics gave to the political arena something of the atmosphere of a jungle. No holds were barred’.⁷ Then, in the final days of December 1930, a major rebellion erupted in Tharrawaddy District, north of Rangoon, rapidly sweeping through much of the Lower Burma delta, in time reaching as far as the Shan States in the northeast.⁸ Although the rebellion was broken by mid-1931—the alleged leader, Saya San, was captured in the August and executed the following November—serious disorder continued well into 1932. Local military forces were augmented by seven battalions from India, comprising over three-and-a-half thousand troops. The final months of 1930 also saw a dramatic collapse in rice prices—the onset in Burma of the world depression—that forced vast numbers of Burmese cultivators to default on their agricultural loans and lose their land.⁹ By the mid-1930s, agriculturalists owned less than half the land in the Pegu and Irrawaddy Divisions, comprising ten of the principal rice-surplus districts of Lower Burma. And a final crisis: in May 1930, rioting broke out in Rangoon between Burmese and Indian labourers that left, according to an official report, around one hundred dead and about a thousand injured.¹⁰ There were further anti-Indian clashes in 1938, again with substantial casualties. As noted earlier, these crises often created serious challenges for Burma's criminal justice administration, a point to which this article will return, specifically with respect to capital punishment, that is in determining the circumstances in which convicted murderers would hang.

I

Through the 1920s and 1930s, the annual report on police administration repeatedly noted the reluctance of Burmese witnesses to murder, as Buddhists required to refrain from taking life, to give evidence in murder trials ‘in case such evidence should result in the man [the accused] being hanged’.¹¹ Or again, ‘even eye witnesses … are in many cases unwilling to give evidence [that could result in the] death penalty, feeling that one man has already met his death, why be a party to causing another death, a Buddhist religious scruple,’ the report argued, ‘which it is difficult to combat’.¹² As a result, murder cases collapsed in court or even before reaching court, and the accused walked free.¹³

The Buddhist precept to refrain from taking life also strongly influenced the contribution to murder trials of the Burmese assessors who, qualified by their education and character, were appointed at sessions courts to assist the judge in determining the facts in a case and to advise or instruct him on local custom or practice. Throughout this period, the annual report on the administration of criminal justice complained that ‘assessors are continually on the look-out for some reason to advance in order to find an accused not guilty … a large number of findings [guilty verdicts] in murder cases are in the teeth of the assessors’ opinion’.¹⁴ In 1913 the sessions judge at Prome reported that he had found summing up in murder trials ‘practically useless’, for often, as soon as he started, the assessors would declare that they were satisfied of the innocence of the accused.¹⁵ But judges too could be driven by their principles or beliefs to acquit in murder cases, even where the evidence pointed to guilt. In his 1939 report, the Inspector-General of Police, R.C. Morris, pointed to those sessions judges—he did not identify whether Burmese or British—who ‘are reluctant to pass death sentences and sometimes take the easier line of acquittal where on the evidence they might well convict’.¹⁶

The Inspector-General then proposed that the Burma administration abolish the death penalty, for a limited period, as an experiment.¹⁷ If it were abolished, he argued, ‘a much larger percentage of convictions would be obtained as witnesses would be more readily forthcoming if they knew they were not assisting in getting a fellow creature hanged’—and, he might have added, as the grounds on which assessors insisted on acquittal were removed and as judges were released from the tension between their beliefs and the demands of the law.¹⁸ The Inspector-General stated that many senior officials agreed with this position, and indeed through the 1920s and 1930s, there were repeated calls in the pages of the annual report on police administration for the abolition of the death penalty for murder. For example, in 1926 the Deputy Commissioner, Meiktila, declared:

The reluctance among Buddhists to give evidence against a man in case such evidence should result in the man being hanged is so strong that it makes one think capital punishment is unsuited to a Buddhist country. I think it possible that there would be a very large increase in convictions for murder if transportation for life were substituted for hanging … it would be well worth while to abolish capital punishment, provided that the transportation was [to] outside the country and really a life sentence, not twenty years or so.¹⁹

But there were clearly many senior officials who were opposed to abolition, seeing capital punishment as a deterrent to murder. Abolition might well increase the conviction rate but it would also increase the number of murders being committed. Indeed, even as he proposed the abolition of the death penalty for a limited period, as an experiment, the Inspector-General of Police had concern: ‘[t]here is perhaps a danger that the abolition … will encourage dacoits, robbers and burglars to kill in order to reduce the chances of arrest and conviction’.²⁰

At the end of the 1920s, an earlier Inspector-General of Police, R.W. MacDonald, had proposed an alternative measure to mitigate the impact of the death penalty in reducing the conviction rate, that is the classification of murder into two degrees, as in the United States.²¹ Broadly, premeditated murder would be first degree murder and unpremeditated murder, second degree. The death penalty would be retained only for first degree murder. Three-quarters or more of murders that came to trial, it was said, were unpremeditated, second degree murder, and in those cases, in court, witnesses, assessors, and judges would no longer be deflected by the thought that a conviction would send a man to the gallows. The obvious objection to the retention of the death penalty for premeditated murders only was that in many cases it would be difficult to decide whether the killing was premeditated or not.²² Take the case of the feuding villager who set out, he swore, merely to threaten a rival with a knife: but the confrontation had run out of control and the rival had been killed, struck with a blow from that knife.²³ It would then be for the police and courts to decide whether the killing was premeditated, that the accused had in fact set out determined not merely to threaten but at the least to seriously wound. If that were their conclusion, the charge would be first degree murder and, if convicted, the murderer would be hanged. The decision was clearly a difficult one for the police and courts but with the most profound consequence for the accused.

This leads to a broader issue that clearly caused considerable unease among senior officials in late British Burma when they reflected on the retention of capital punishment for murder. It was occasionally remarked in the pages of the annual report on police administration that for reasons in part peculiar to Burma, there were many individuals charged and convicted of murder and now being led to the gallows who, except for a small twist of fate, would have faced a far less serious charge, perhaps simple hurt, carrying a brief prison sentence, or perhaps no charge at all. They had been unlucky.²⁴

The reasoning here was that a substantial proportion of the murders committed in Burma in the 1920s and 1930s arose, apparently, from sudden quarrels, often on social occasions and commonly fuelled by drink, which had then erupted into violence. The crucial factor in these violent clashes was that those involved almost invariably resorted to the dah, a Burmese knife, carried as a matter of course by adult males, for it was a basic agricultural and household implement in everyday use. Of 885 murders (509 classified as premeditated and 376 as unpremeditated) analysed in the police administration report for 1930, no less than 562, almost two-thirds, had been committed with a dah.²⁵

While a dah was undoubtedly wielded with clear murderous intent on many occasions, when it was drawn as a quarrel suddenly erupted into violence, the intention, it was said, was rarely to kill but merely to intimidate, at worst to inflict a minor wound. As the Inspector-General of Police, R.W. MacDonald, noted in his 1924 annual report, the unpremeditated murder ‘is usually committed in a fit of passion or following the effects of drink and it is improbable that the accused really meant to commit murder … cases of unpremeditated murder … in all but degree are hurt cases’.²⁶ Nevertheless a blow from a dah, whatever the assailant's intention, could so easily end in death, depending on a number of unforeseen factors—the violence of the blow, the part of the victim's body struck, and indeed the speed with which medical help could be found. The crucial point, then, was that the dah was lethal but unpredictably and often unintentionally so. In contrast, argued a British official, Maurice Collis, in 1926, the Indian carried not a lethal knife but the lathi, a long iron-bound stick, often of bamboo, a blow from which, although often severely painful, was rarely fatal.²⁷ In other societies, an insult spat out in a drunken brawl would be met by a fist, again severely painful but rarely causing death.²⁸

In late colonial Burma, then, an argument or brawl could so easily end with a charge of murder—or, with good fortune, a charge of simple hurt. As the Inspector-General of Police noted in his 1934 report, ‘the question whether an assault becomes a murder … in Burma is merely a matter of luck!’²⁹ That turn of fate could of course have disastrous consequences, death from a stab wound for the victim, death by hanging for the individual whose thrust with a dah had proved to be unintentionally fatal. For some officials, the fact that men were being sent to the gallows on a mere twist of fate undermined the retention of the death penalty for murder, specifically unpremeditated murder. ‘It is with real regret,’ recorded a Deputy Inspector-General of Police in 1929, ‘that I read of the death penalty being enforced on those persons—the majority of those convicted—whose luck has changed … a hurt case into a murder case.’³⁰ It was a regret reinforced by an awareness that in cases of premeditated murder, which were often carried out by hit-men careful to leave no incriminating evidence, only rarely was an accused brought to trial.³¹ The hired killer rarely faced justice. It was the wretched labourer, unexpectedly caught up in the village brawl and with no murderous intent, who went to the gallows.

In short, the abolition of the death penalty was frequently argued on practical grounds: that the prospect of sending a man to his death was discouraging witnesses from giving evidence in court and thereby allowing the guilty to walk free; that as the law made no distinction between premeditated and unpremeditated murder, simply prescribing death or transportation for life for murder, it was too often only a small turn of fortune that was sending a man to be hanged. But there were also officials who opposed the death penalty on moral grounds, who saw judicial execution as unacceptable in a civilised order. The well-known voice here is George Orwell, who served in the Indian Imperial Police in Burma between 1922 and 1927. In 1931 he published a brief essay, ‘A Hanging’, in the English literary journal, The Adelphi, an account of an execution in a Burmese jail. As the condemned man was marched towards the gallows, he skipped to one side to avoid a puddle.

It is curious, but till that moment I had never realised what it means to destroy a healthy, conscious man. When I saw the prisoner step aside to avoid the puddle, I saw the mystery, the unspeakable wrongness, of cutting a life short when it is in full tide … He and we were a party of men walking together, seeing, hearing, feeling, understanding the same world; and in two minutes, with a sudden snap, one of us would be gone—one mind less, one world less.³²

‘I watched a man hanged once,’ declared Orwell in The Road to Wigan Pier, published in 1937: ‘it seemed to me worse than a thousand murders.’³³

But many senior officials in late colonial Burma were opposed to the abolition of capital punishment for murder. Indeed, with the murder count rising sharply, some sought to strengthen the deterrent (as they saw it). At the end of the 1920s there were calls for execution by guillotine, firing squad, or the electric chair, ‘as hanging is a popular method among Burmans of committing suicide and is not feared’.³⁴ The Deputy Commissioner at Amherst, a Burman, proposed public executions, as the ‘present humane method of hanging privately in the jail has no terror at all’.³⁵ And in the mid-1930s, it was suggested that the convicted murderer's last days be filmed and the film be widely shown in public.³⁶ The proposals were not taken up. Nevertheless, the fact that they were advanced at all, and in particular that they were reported in the annual reports on police administration, suggests that resistance within the service to the Inspector-General's call for the abolition of the death penalty for a limited period, simply as an experiment, would have been fierce.

II

Under section 374 of the Code of Criminal Procedure, 1898, when a sessions court passed a death sentence for murder, the trial proceedings were submitted to the High Court for confirmation of the punishment. The figures for capital sentences referred to the High Court for confirmation may therefore be taken as the figure for death sentences passed in sessions courts in each year. Setting that figure against murder trial convictions would then provide an indication of the proportion of convicted murderers sentenced by sessions courts to death. The figures for 1905 to 1940 are given in table 1.

Table 1. Murder trial convictions and capital sentences referred to the High Court, 1905–40

Sources and notes:

[a] Report on the Police Administration of Burma. The figures here are for cases in which a conviction was secured. In some murder trials, more than one accused would have been before the court and have been convicted.

[b] Report on the Administration of Criminal Justice in Burma (RACJB). Prior to 1922, death sentences passed in Lower Burma's courts were referred to the Chief Court for confirmation, those passed in Upper Burma's courts were referred to the Judicial Commissioner. From 1922, all death sentences were referred to the newly constituted High Court in Rangoon. From 1905–21, the RACJB gave separate figures for Lower Burma and Upper Burma. In all but two of those years, the Upper Burma figure was explicitly for persons. The Lower Burma figure was occasionally for cases, in two years for persons, but most frequently as ‘sentences of death’, an ambiguous term that nevertheless internal evidence suggests can be interpreted as cases. As the Lower Burma figure was invariably much higher than the Upper Burma figure, the figures in column [b] prior to 1922 are regarded here as being for cases. From 1926, the RACJB recorded the number of death-sentence cases referred to the High Court for confirmation and the number of persons involved in those referrals. In 1930, for example, 163 cases in which sentences of death had been passed by Courts of Sessions were referred to the High Court, involving 197 persons. The figures in the table from 1926 are for the number of cases referred to the High Court, to maintain consistency with the pre-1921 figures and to allow for a comparison with the figures for convictions in column [a] which, again, are for cases.

In reading table 1, two of its features must be kept in mind. First, as explained in the notes, while the number of murder trial convictions reported each year in the RPAB was given in terms of cases (and in many cases more than one person stood accused and had been convicted), prior to 1922 the number of capital sentences referred to the High Court, as reported each year in the RACJB, was most often given in terms of persons or ‘sentences of death’, and only occasionally explicitly in terms of cases. That disparity in measuring convictions and referrals clearly undermines the calculation of the proportion of convicted murderers being sentenced by sessions courts to death in the years prior to 1922. Second, referrals to the High Court in a particular year did not derive exclusively from the murder convictions in that year (the 123 capital sentences referred to the High Court for confirmation in 1905 did not derive solely from the murder trial convictions in 1905). This was simply because, following conviction and sentencing by the court, it would clearly take some time, occasionally beyond the end of the year, for the court to prepare a statement of the trial proceedings for submission to the High Court for confirmation. Obviously that temporal misalignment between convictions and referrals further undermines the calculation of the proportion of convicted murderers being sentenced to death.³⁷

The undermining of that calculation can be mitigated to a degree by taking a broad view of the figures, for then a long-term shift away from hanging is clearly evident. From 1905 through to the late 1910s, a high proportion of those convicted of murder, perhaps nine in every ten, were sentenced by the trial judge to death.³⁸ The proportion then fell substantially, rose moderately in the early 1930s, and then fell again, to six or less in every ten, through to the end of that decade. The fall in capital sentence referrals to the High Court from the mid-1930s, while murder convictions remained roughly constant, is striking. As noted in the introduction, the Indian Penal Code prescribed two sentences for murder, death or transportation for life. But it did not specify which of the two punishments must be imposed in which circumstances. In other words, the question of when or why a convicted murderer would hang, rather than be transported, was left to judicial discretion in every case.³⁹ The shift in sentencing away from the death penalty—although even at the end of the 1930s around half of convicted murderers were still being sentenced to hang—may well have been the result of trial judges now paying closer attention to the precise circumstances of the murders being recounted before them in court. They were now increasingly distinguishing where possible between premeditated and unpremeditated murder, between murder committed in the course of violent robbery and murder arising from an argument that unintentionally turned violent, and increasingly handing down a death sentence only in cases of the former. If this were indeed so it could be seen as the judiciary's response to the views, repeatedly made in the pages of the police administration report through the 1920s and 1930s, first that the prospect of sending a man to his death was substantially reducing the conviction rate in murder trials, and second that in cases of unpremeditated murder, hanging was clearly an unjustly severe sentence.

III

In order to review the death sentences passed in sessions courts, the High Court had before it, as noted earlier, a record of the trial proceedings in each case. In addition, under the Code of Criminal Procedure, the High Court could make further inquiry or seek further evidence (if required in the presence of the condemned prisoner) or instruct the relevant sessions court to do so. That said, since between 150 and 200 capital sentences were normally referred to the High Court for confirmation each year from the mid-1920s, clearly only a limited time could be spent on reviewing each case. Table 2 shows the decisions of the High Court—to confirm the death sentence, to reduce the sentence to transportation or imprisonment, to acquit, or to order a retrial—for the years 1926 to 1940.

Table 2. Referral of death sentences to the High Court, 1926–40

Source: Report on the Administration of Criminal Justice in Burma

Notes:

[a] In Table 1, column [b] recorded the number of cases referred to the High Court for confirmation of the death sentence. Column [a] in this table records the number of cases for disposal by the High Court in that year, and therefore includes any cases left over from the previous year. In the four years 1937–40, the RACJB gave no figure for cases for disposal but a figure for cases disposed of during the year, and that is the figure inserted here. However, it is strongly likely that all cases for disposal in each of those years were indeed processed during the year.

[b] The RACJB refers to the number of persons ‘involved in these references’. Internal evidence makes it clear that this is the number of persons involved in the cases for disposal by the High Court in that year. The figures for the three years 1931, 1932, and 1933 include the appeals against death sentences handed down by the Special Tribunals and Special Judges, newly established in order to provide for the speedy trial of individuals accused of offences arising from the rebellion.

[c] The figures for the three years 1931, 1932, and 1933 include the confirmations of the death sentences imposed by the Special Tribunals and Special Judges.

It is clear from table 2 that in those years, the mid-1920s through to the end of the 1930s, the High Court intervened substantially with respect to both convictions and sentencing in murder trials held in sessions courts. For example, in all except four of those years, at least one in ten of the individuals who had been convicted of murder and sentenced to hang in a sessions court was now, following a review of the trial proceedings, acquitted and walked free. In 1929, that was no less than 32 of the 192 (one in six) of those whose sentence was before the High Court for confirmation. Or again, roughly two in ten (but in a few years closer to three in ten) had their sentence reduced to transportation or imprisonment. In 1929, that was 55 of the 192 death sentences (almost three in every ten) under review. The High Court confirmed that the 55 were guilty of murder. But it rejected, again after reviewing the trial proceedings, the sentence that had been handed down by the sessions court, that they would hang. Seen from a different perspective, in 1929 the High Court confirmed less than half the death sentences referred to it by sessions courts, 89 of 192. In three further years in this period it confirmed barely half the death sentences, and in the remaining years, only roughly six in ten.

As noted earlier, in these decades there had already been a substantial shift away from the death penalty in the sentencing of convicted murderers in sessions courts, strikingly so from the mid-1930s, perhaps as trial judges paid closer attention to the precise circumstances of the murders being recounted before them. The High Court, reducing the sentences of roughly two in every ten condemned murderers to transportation or imprisonment, secured a further shift away from the death penalty. But in this it seems unlikely that the High Court was paying still closer attention to the circumstances of each murder, simply because, reviewing 150 to 200 capital sentences in a year, this would have been impracticable. Rather, it is suggested, the High Court was now increasingly taking the view that in a modern criminal justice administration, built on the rule of law, capital punishment should be used sparingly, a punishment only for the most brutal murders. In other words, for the High Court it was an issue of colonial modernity. That argument is seen most clearly in the High Court's high rate of acquittal—in 1929 one in six of those sentenced to hang for murder were on review acquitted and walked free—for in those cases it was determining that sessions judges were convicting (and sentencing to death) ‘on insufficient evidence’.⁴⁰ The High Court was making it clear that a high standard of evidence was required before it would confirm a conviction for murder. In fact the police repeatedly railed against ‘the very high standard of evidence required [even] by Sessions Judges’.⁴¹ In his 1929 report, the Inspector-General of Police, R.W. MacDonald, was strikingly outspoken:

the dice are heavily loaded in favour of the murderer. Even after the police have with great difficulty arrested the murderer and secured sufficient evidence the case breaks down in the Court of the committing magistrate or, if committed, in the Sessions or if it has been safely steered through both these Courts a learned Judge of the High Court discovers a legal flaw or a reasonable doubt.⁴²

Nevertheless the High Court stood firm. Indeed one India Office official in London declared at this time that the judges of the High Court in Rangoon were ‘fanatical in their attempts to maintain a [high] standard of justice’.⁴³ In other words, the High Court saw that to demand a high standard of evidence to convict, above all in murder trials where conviction could well lead to the gallows, must be central to colonial modernity.

In addition to reviewing all death sentences imposed in courts of session, each year the High Court itself tried a small number of murder cases. Table 3 shows the decisions of the High Court in those cases for the years 1926 to 1940.

Table 3. Murder trials held at the High Court, 1926–40

Source: Report on the Administration of Criminal Justice in Burma

Note: In some years, for example 1935, the combined figure for convictions and acquittals was less than the number of persons standing trial. The explanation is either that one or more verdicts were still pending at the end of the year or that the High Court had in one or more cases ordered a retrial.

In a number of these years, the conviction rate for individuals facing trial in the High Court for murder was strikingly low, 7 convictions from 22 accused in 1926, 4 from 15 in 1928, 5 from 14 in 1935. This was a conviction rate substantially below that in murder trials held in sessions courts.⁴⁴ Without knowing why particular murder cases were tried in the High Court rather than in sessions courts—were they allocated to the High Court because they were notably complex or in some way high profile: or were they simply local, arising from murders committed and investigated in Rangoon itself?—it is difficult to explain the low conviction rate. The only point to be made here (and it is an important one) is that the low conviction rate clearly indicates that the High Court was again demanding a high standard of evidence before it would convict in murder cases. Moreover, the fact that it more frequently imposed the death penalty than transportation or imprisonment would seem to confirm that it would convict (and send to the gallows) only on a secure verdict. In 1926 no less than 14 of the 22 before the High Court charged with murder were acquitted. These were the standards sought by a modern criminal justice administration built on the rule of law.

IV

The High Court, either reviewing the murder verdicts and sentences referred to it from sessions courts or trying murder cases itself, marked the end of the judicial process for those charged with murder in late colonial Burma. The first column in table 4 records the number of death sentences imposed by Burma's courts—that is by the conclusion of the judicial process—in each year from 1905 to 1940.

Table 4. Death sentences imposed by Burma's courts and judicial executions carried out in Burma's prisons, 1905–40

Sources and notes:

[a] Report on the Administration of Criminal Justice in Burma.

[b] Report on the Prison Administration of Burma. The Indian Penal Code prescribed the death penalty for a number of offences in addition to murder, for waging war against the King-Emperor, for abetment of mutiny, for abetment of the suicide of a child or an insane person. It would be safe to assume, however, that conviction for murder accounted for the vast majority of the judicial executions that took place in Burma's prisons in this period.

In each year from 1905, almost invariably the number of judicial executions taking place in Burma's prisons ran below the number of death sentences imposed by its courts. In other words, each year a number of death sentences, often a substantial number, were not carried out. There are several possible explanations. There were cases no doubt of a lawless individual—a dacoit—being convicted of two or more separate murders and being given two or more death sentences, but of course he could be hanged just once.⁴⁵ A condemned man could die in prison while awaiting execution, of natural causes or by his own hand. But most frequently, condemned men had appealed to the political administration for clemency. Several hundred rebels, convicted and sentenced to death by special tribunals for murder committed during the Saya San rising at the beginning of the 1930s, verdicts and sentences confirmed by the High Court, petitioned the Burma Government—the Governor in Council—for clemency.⁴⁶ When the Burma administration rejected a petition, the condemned man could then petition the Government of India—the Governor-General in Council—as Burma at this time was a province of British India. In fact a formal petition was not actually required for the India administration to review a death sentence, for in June 1931, as the first special tribunal hearings were being concluded, Delhi asked Rangoon to refer to it all those rebel cases in which it had rejected an appeal for clemency and the sentence of death therefore still stood. In order to enable the Governor-General in Council to reach a view on a referred case, Rangoon sent to Delhi a copy of the special tribunal's judgment, a copy of the High Court's order, and a statement by the Burma administration of the grounds on which it had rejected the appeal. Where the Governor-General in Council, having reviewed the papers, proposed not to exercise clemency, the case, together with the growing volume of papers, was referred to the Secretary of State for India in London. He and his India Office officials did not make a final judgment. But where they had concerns as to the soundness of the Government of India's proposed rejection of clemency, they would ask Delhi, and by implication Rangoon, to think again.

That many-staged process, frequently extending through to the Secretary of State in London, resulted in the sentences of a substantial number of condemned rebels being commuted. According to table 4, from 1931 to 1934, 694 convicted murderers (of course not all rebels) were sentenced to death by Burma's courts but just 499 (again not all rebels) were executed in Burma's prisons. The process was extended because the execution of a potentially large number of rebels would be certain to cause considerable unease within the home government and among the informed home population. It was therefore politically important to show that each case was being fully considered at the highest levels before a death sentence was finally confirmed. But that political concern did not arise in non-rebellion cases, and here, it would seem, the condemned murderer's petition for clemency was heard only by the Governor in Council and if rejected went no further. That said, in April 1941, the Rangoon Gazette reported a recent dramatic increase in the number of death sentences commuted by the Governor in Council, from 11 in 1937 and 9 in 1938 to 46 in 1939 and 48 in 1940.⁴⁷ In those last two years then, more than half the death sentences being imposed by Burma's courts, 73 in 1939 and 85 in 1940 according to the table, were commuted to transportation or imprisonment, not by judicial process but by political intervention. And that intervention explains the dramatic fall in executions in 1939 and 1940, to 27 and then 31, down from 83 and 74 in the two preceding years (table 4). The dramatic fall in executions, it should be added, occurred under a Burmese minister.

The commutation of death sentences on that scale may well have been eased by the creation of procedures and perspectives in the first half of the 1930s that enabled the administration to process the large volume of appeals for clemency on the part of the Burma rebels. Commutation of death sentences had now become embedded in the government's thinking, the result, it must be repeated, of the challenge to the criminal justice administration arising from the Saya San rebellion. But there was a further factor. In an interview with the Rangoon Gazette in April 1941, the Home Minister, U Aye, noted that as it stood the law relating to murder made no distinction between premeditated murder and murder committed on the spur of the moment, a flaw, he argued, that forced judges to impose a death sentence on conviction, whatever the circumstances.⁴⁸ Of course the Indian Penal Code prescribed an alternative sentence for murder, transportation for life. But Burma's judges would impose that sentence only when there were extenuating circumstances relating to the convicted individual, for example, his young age. Prevented by a flaw in the law from using their discretion and imposing the lesser sentence, the Minister continued, judges were now increasingly recommending that the death sentence they had reluctantly imposed be commuted by the administration. Indeed the recent greatly increased exercise of clemency by the Governor in Council had in almost every case followed that recommendation from the sentencing court.

That the right of the condemned murderer to appeal for clemency was now being used by the administration to distinguish between premeditated and unpremeditated murder, thus correcting a perceived flaw in the Indian Penal Code, is confirmed in a passage in a memoir by a senior British official, F.S.V. Donnison, from the late 1930s the Secretary to the Government of Burma, Judicial Department, advising a Burmese minister.

[The] most vexatious and troubling task [for the Judicial Department] was to advise the Minister and Governor in the matter of exercising their prerogative to commute sentences of death passed by the High Court. As soon as a death sentence was passed the records of the case would be sent to my office. They would appear at once before me, distinguished by a red label. They would come straight to me without any recommendation from the office. All other papers were examined in the office first and submitted to me with recommendations, wherever possible, by the Branch Superintendent concerned, and by the Under-Secretary. But when a red label appeared I would take the case record home and put aside all other business in order to soak myself in the facts and background of the murder. In the past, commutation of sentences had, I think, been rare. In my time, taking into account the changing climate of opinion, the fact that I was advising a Burmese Minister to whom, as a Buddhist, all taking of life was wrong, and my own convictions concerning the death penalty, I sought to distinguish between premeditated murders, and murders committed in the heat of passion, recommending that the death penalty be enforced for the former only, and commuted for the latter. These recommendations were accepted in virtually every case, though I think that on a few occasions the Minister recommended, and the Governor exercised, leniency despite my recommendation to the contrary.⁴⁹

It is important to add that at around that time the Burma government was using the appeal for clemency to mitigate a further perceived flaw in the law. Under section 398 of the Indian Penal Code, if at the time of attempting to commit robbery or dacoity the offender was armed with a deadly weapon, the mandatory sentence was a term of imprisonment of not less than seven years. But that sentence was seen as being too severe if either the weapon was not used in that offence or the offender was a minor, and therefore in those cases judges and magistrates were instructed by the High Court to submit a recommendation to the Governor in Council for remission in part of the sentence they had been forced to impose.⁵⁰

According to Donnison, the judges on the High Court were ‘strongly critical’ of political interference in sentencing for murder, ‘for under the Penal Code the punishment for murder was either death or transportation for life, and they, understandably, contended that they had already, and with better facilities than were at our disposal, made a choice of the appropriate penalty. In their position I would have taken the same view.’⁵¹ But in that case: why not simply amend the law? In fact in his interview with the Rangoon Gazette in April 1941, the Home Minister reported that the government would shortly be introducing an amending bill that would differentiate between premeditated murder and murder committed on the spur of the moment, in effect allowing judges to use their discretion and impose a less severe sentence for a less serious crime, unpremeditated murder. But legislation had risks, as became clear when the administration considered amending the Indian Penal Code with respect to the mandatory sentencing of robbers and dacoits armed with a deadly weapon to a minimum of seven years, section 398. The Penal Code and Code of Criminal Procedure (Amendment) Bill, 1939 made provision for the repeal of that section, thereby giving judges and magistrates discretion in sentencing for robbery and dacoity.⁵² But crucially the administration had long hesitated to introduce that amending legislation, fearing that Burma's legislators would take the opportunity to table further amendments to the Indian Penal Code, in the words of the Governor, ‘of a far more sweeping and undesirable character’.⁵³ Inevitably that concern would also discourage amending the Indian Penal Code section 302 relating to murder. In addition, as argued earlier, to create robust definitions of premeditated and unpremeditated murder was far from simple. In the circumstances, it was clearly preferable to continue to use the right of appeal for clemency to overcome that perceived flaw in the law.

V

The fall in the number of executions being carried out in Burma's prisons from the middle of the 1930s, and in particular the sharp decline in 1939, a decline sustained in 1940, undoubtedly caused unease for a substantial group of senior officials, convinced that the death penalty was an effective deterrent to murder, an unease surely reinforced by the fact that Burma's murder count was soaring through these same years. The calls at the end of the 1920s (noted earlier) for execution by guillotine, firing squad, or the electric chair, ‘as hanging … is not feared’, and for public executions, as ‘hanging privately in the jail has no terror at all’, speak of an iron faith in capital punishment (even if in a yet more brutal form) as a deterrent. Even the Inspector-General of Police who proposed the abolition of the death penalty for a limited period, as an experiment to assess the impact on conviction rates, feared that it might ‘encourage dacoits, robbers and burglars to kill in order to reduce the chances of arrest and conviction’ (again as noted earlier). Nevertheless, as this article has sought to explore, in the 1920s and 1930s, Burma's judicial and then political administrations substantially reduced the number of convicted murderers who were hanged.

This article has focused on the local pressures that brought about that reduction but, as was noted in an opening paragraph, imperial opinion and practice also had an influence. From the 1920s a substantial strand of political opinion in Britain argued for the abolition of capital punishment for murder, principally on the grounds that judicial execution could have no place in a civilised society but also that its alleged deterrent effect was highly questionable. Notable here were the establishment in 1925 of the National Council for the Abolition of the Death Penalty, the recommendation by a parliamentary select committee in 1930 that capital punishment be suspended for a trial period of five years, and in 1938 a Commons vote on a Private Member's motion in favour of experimental abolition, again for five years.⁵⁴ Senior officials in Burma, both British and Burmese, were clearly familiar with the arguments. Indeed in his interview with the Rangoon Gazette in April 1941, the Home Minister, U Aye, disclosed that he had been studying the ‘abolition literature’, including the pamphlets produced by the Howard League for Penal Reform, and he referred to the select committee's recommendation to the Commons that capital punishment be abolished for five years as a trial.⁵⁵ In Belgium, he added, although the death penalty remained in law, there had been no executions since 1867. However, there was no suggestion that the Home Minister himself supported abolition, an important point to be taken up shortly.

Burma's absorption of imperial practice in the sentencing of convicted murderers was clearly evident in the sharp increase in the use of the appeal for clemency at the end of the 1930s. The practice was firmly established in Britain itself. The punishment prescribed by English law in the first half of the twentieth century was death, and the presiding judge could not impose a lesser sentence. However, that rigidity was mitigated by the exercise of the royal prerogative, which rested in the hands of the Home Secretary. In the fifty years from 1900 to 1949, 1,210 individuals were sentenced to death in England and Wales but of those almost half, 553, had their sentence commuted.⁵⁶ And the exercise of the prerogative was common in other parts of the Empire too:

it was not unusual in Kenya or Nyasaland, for example, for half of the death sentences passed in a year to be commuted, primarily because the majority of murders involving Africans were regarded as unpremeditated and resultant from quarrels between friends and family, types of murder which were regarded as less threatening to law and order and which consequently did not warrant the ‘extreme penalty of the law’.⁵⁷

As noted above, abolition was not proposed by Burma's Home Minister in his 1941 interview (and of course neither was it secured in Britain in this period). It is interesting to note that in 1929 and again in 1931, a member of the Indian Legislative Assembly, Gaya Prasad Singh, sought to introduce a bill to abolish capital punishment but he made no progress.⁵⁸ In other words, in sentencing for murder, Burma's judicial and political administrations were seeking in this period only to restrict the imposition of the death penalty in essence to conviction for premeditated murder. Punishment for premeditated murder was hanging: and both Donnison, despite his ‘own convictions concerning the death penalty’, and the minister he advised, a Buddhist, sanctioned the execution of convicted murderers. In other words, at issue here was the rule of law. It was clearly unjust that the ‘majority of those convicted—[those] whose luck has changed … a hurt case into a murder case’, the wretched labourer, unexpectedly caught up in the village brawl and with no murderous intent, should go to the gallows, and at the end of the 1930s, the political administration used the appeal for clemency to commute that sentence to imprisonment and thereby establish a measure of justice. A reinforcement of the rule of law was earlier evident in the High Court's high rate of acquittal—in 1929 one in six of those sentenced to hang for murder were on review acquitted and walked free—a declaration that sessions judges were convicting (and sentencing to death) ‘on insufficient evidence’. The rule of law demanded a high standard of prosecution evidence if a murder conviction was to be confirmed by the highest judicial authority.

‘The single most important exemplar of the claimed beneficence of the Empire,’ argued Martin Wiener in 2009, ‘was its system of laws, and by the nineteenth century there was great pride in spreading the benefits of English law around the world.’⁵⁹ For those in imperial service, from officials in London down to district officers, the imposition of the rule of law, structured and impersonal, was a powerful argument for the colonial presence. It secured the authority of and justification for colonial rule in the eyes of those who ruled.⁶⁰ But as Wiener then rapidly added, the claims for the rule of law in the colonial context hit huge contradictions. Perhaps most importantly, while

[p]erhaps the best-known principle of [English] law was the equality of individuals—that all were equally subject to its strictures, and that all could equally claim its protection … [yet] the most basic working principle of empire … was inequality—a necessary inequality of power between the British ruling class and the subject populations.⁶¹

This article has attempted to show, however, that in late colonial Burma, the rule of law was markedly tightened in one important respect, in the administration of justice in murder cases. By the late 1930s, a high standard of evidence was being increasingly demanded for conviction for murder, and increasingly the death penalty was being imposed only when murder had clearly been premeditated.