Jump to ContentJump to Main Navigation
The Hidden Histories of War Crimes Trials$

Kevin Heller and Gerry Simpson

Print publication date: 2013

Print ISBN-13: 9780199671144

Published to Oxford Scholarship Online: January 2014

DOI: 10.1093/acprof:oso/9780199671144.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 26 February 2020

Justice for ‘Asian’ Victims

Justice for ‘Asian’ Victims

The Australian War Crimes Trials of the Japanese, 1945–51

Chapter:
(p.348) 17 Justice for ‘Asian’ Victims
Source:
The Hidden Histories of War Crimes Trials
Author(s):

Narrelle Morris

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780199671144.003.0017

Abstract and Keywords

This chapter examines the Australian war crimes trials from 1945 to 1951. Specifically, it considers the odd Australian statutory jurisdiction based on classes of victims that enabled some Asian victims of Japanese war crimes to be dealt with in the trials; assesses their presence in the Rabaul trial series; and focuses on representations of Asians as witnesses in several notable trials held at Rabaul. Unlike the Tokyo Tribunal, the Rabaul trials cannot be criticized for seeking justice only for Caucasian victims of Japanese war crimes and for deliberately avoiding or otherwise neglecting the ‘hidden history’ of Asian victims.

Keywords:   war crimes trials, Rabaul trials, war criminals, justice, Asian victims, witnesses

One of the common critiques of the International Military Tribunal for the Far East (the Tokyo Tribunal), held after World War II, was that it failed to systematically and comprehensively examine and address crimes committed by the Japanese against the peoples in the Asia-Pacific region, particularly those arising out of Japanese colonialism and wartime occupations, and including sexual crimes against Asian women. Representative of this criticism, for example, is Korean activist Won Soon Park’s conclusion in 1997 that the Tokyo Tribunal was ‘little more than a feast whose guest list included the Western Allies but blatantly omitted the Asian victims themselves’.1

In short, the Tokyo Tribunal has been indicted for being, amongst other things, an example of partial and selective justice along racial, colonial and gendered lines. In fact, the range and complexity of reasons behind the apparent absence of and silence regarding Asia in the Tokyo Tribunal have been articulated by many critics since the end of World War II, and contested again by others.2 Interestingly, recent legal scholarship by historian Yuma Totani has suggested that the Asian presence—for want of a better word—in the Tokyo Tribunal was undoubtedly more vivid and influential than has often been concluded. She draws attention in particular to the efforts of several prosecution teams, including the British Commonwealth team that introduced to the Tribunal evidence of Japanese war crimes committed in a broad geographical area across South East Asia and the Pacific, many of which were committed against Asian and Pacific nationals.3 In her view, such evidence has not (p.349) taken its proper place in studies of the Tribunal due to the manner in which it was tendered and because of the dispersal of related documents. It is apparent from the scholarship of Totani and others that, even after more than sixty years of analysis, the Tokyo Tribunal is still revealing ‘hidden histories’ and the process of understanding the Tribunal remains ongoing.

While the proceedings, judgments and other documents of the Tokyo Tribunal have been interpreted and re-interpreted since it concluded, the Australia’s subsidiary B and C class war crimes trials—300 of which were held pursuant to the War Crimes Act 1945 (Commonwealth) in Morotai, Wewak, Labuan, Rabaul, Darwin, Singapore, Hong Kong and Manus Island in the period 1945–51—have since that time been unreported and virtually unassessed. In part, this was a deliberate creation of ‘hidden history’ by the Australian Government as, for many years after the trials were completed, requests to access the trial proceedings were refused. The Japanese Government, for example, made unsuccessful requests in 1955 and 1959 to be granted copies of the trial proceedings. A third Japanese request came in 1965 and was considered at some length over the next few years, primarily in respect of the concern that releasing the trial proceedings might precipitate criticism of Australia. As a 1967 report into the issue of whether to grant access observed, while the trials were ‘generally satisfactory’ and did not cause ‘any substantial miscarriage of justice’:

Since war crimes trials are a controversial issue in general, they provide material for a troublemaker to use against the country which conducted them…Almost all of the trials of ‘B’ and ‘C’ class criminals have elements appearing on the face of the records which would provide a hostile reader with anti-Australian ammunition.4

One of the report’s conclusions, however, was that a refusal to grant access to the trials ‘might imply that we have something to hide’.5 After considerable consultation, Australian approval was finally granted in 1968 for partial copies of trial proceedings6 to be made available to the Japanese Government and for ‘bona fide Australian Scholars’ to be able to review the trial proceedings.7 There was no suggestion made as to how applicant scholars might be determined to be bona fide, perhaps because, as of 1967, the Australian Government had ‘no record of any interest ever being expressed by scholars’ in the trials.8 It was not until 1975 that the Australian public was granted open access to the trial proceedings, which had, by then, been shifted from the Attorney-General’s Department to the Australian Archives. In announcing his decision to grant public access, the Attorney-General, Mr K. Enderby QC, (p.350) remarked: ‘For too long Australian scholars have been hampered in their attempts to interpret Australia’s history. Restrictions like this one [on access to the trial proceedings] no longer serve a useful purpose…The past should be everyone’s property.’9

While the trials were now open, few scholars and probably even fewer members of the public took up this call to serve as interpreters of history. It is likely that they were daunted by the prospect of having to examine and analyse tens of thousands of pages of transcript, evidence and reports with little overarching or explanatory material for guidance. There may also have been some doubt that such an effort would be worth it. The 1967 report on access had critically (and wrongly) observed that the fact that the trial proceedings lacked written judgments meant they were of ‘little value for research’.10

While more than thirty-five years has now passed since public access was granted, the process of revealing the ‘hidden histories’ of the Australian war crimes trials is no less daunting. In fact, the passage of time and the passing of nearly all those Australian personnel who participated in the trials have now imposed more constraints upon our understanding of them. While the process of understanding the trials has barely begun, one of the ‘hidden histories’ that is becoming clear was that there was a considerable emphasis on Asian victims of Japanese war crimes, far beyond what might be expected of notionally ‘Australian’ trials. To be sure, a few of the Australian trial series, such as those held at Labuan, Morotai and Darwin, almost exclusively heard trials regarding identifiably Australian, British or American victims. However, in the Rabaul trial series—which encompassed more than two-thirds of all trials held, 190 trials—the overwhelming majority of the victims were non-Caucasian, being Indian or Chinese prisoners-of-war or Asian or Pacific civilians who were resident in or transported to territories occupied by Japan. However, as the trials reveal, many of these victims themselves are and will remain ‘hidden’, as their identities and numbers were often unknown.

This chapter examines the odd Australian statutory jurisdiction based on classes of victims that enabled some Asian victims of Japanese war crimes to be dealt with in the Australian war crimes trials, assesses their presence in the Rabaul trial series, and focuses on representations of Asians as witnesses in several notable trials held at Rabaul. Unlike the criticisms that have been made about the Tokyo Tribunal, the Rabaul trials cannot be criticized for seeking justice only for Caucasian victims of Japanese war crimes and for deliberately avoiding or otherwise neglecting the ‘hidden history’ of Asian victims. The question as to whether Asian victims received justice at the Australian trials, as perhaps implied by the title of this paper, however, is very complex and remains to be considered in depth.

(p.351) (I) Jurisdiction over War Criminals

By the close of World War II, it was well established in international law that a belligerent was entitled to prosecute for war crimes those members of the armed forces of the enemy who fell into its hands, irrespective of the place where the crime was committed or of the victim’s nationality, unless the belligerent was somehow prohibited from doing so by international law. There was not always broad agreement as to what wartime conduct was criminal under international law but, as Dr Willard B. Cowles, then a member of the United States’ Judge Advocate General’s Department, concluded in June 1945: ‘[I]‌t is clear that, under international law, every independent State has jurisdiction to punish war criminals in its custody regardless of the nationality of the victim, the time it entered the war, or the place where the offence was committed.’11

The Charter of the International Military Tribunal (for the Nuremberg Tribunal) and the Charter for the International Military Tribunal for the Far East not surprisingly thus defined their jurisdictions by reference to ‘major war criminals of the European Axis’12 and to ‘Far Eastern war criminals’13 respectively. The phraseology deliberately formalized some selectivity; since the war criminals had to be ‘of the European Axis’ and ‘Far Eastern’,14 this effectively excluded the two Charters from being applied to war criminals within the Allied Forces themselves. Neither of the two Charters limited their jurisdiction by reference to the victims of war crimes, although victims impliedly had to have been within the European and the ‘Far Eastern’ war theatres.

Although the principle of universal jurisdiction of belligerents over war criminals was seemingly then recognized in Australia,15 the legislation for Australia’s own war crimes trials, the War Crimes Act 1945, deliberately circumscribed this jurisdiction by explicitly referencing the victims of war crimes. The bill for the Act was drafted with lightning speed by the Commonwealth Attorney-General’s Department without (p.352) much consultation16 in late September 1945, was introduced into Parliament on 4 October 1945, passed both Houses without amendment within a few hours17 and received royal assent on 11 October 1945.18 The Act’s jurisdictional provisions, sections 7 and 12, when read together, established three classes of victims of war crimes into which a victim of an alleged war crime must have fallen in order for an Australian Military Court to be convened:

Section 7—A military court shall have power to try persons charged with war crimes committed, at any place whatsoever, whether within or beyond Australia, against any person who was at any time resident in Australia, and for that purpose, subject to any direction by the Governor-General, to sit at any place whatsoever, whether within or beyond Australia.

Section 12—The provisions of this Act shall apply in relation to war crimes committed, in any place whatsoever, whether within or beyond Australia against British subjects or citizens of any powers allied or associated with His Majesty in any war, in like manner as they apply in relation to war crimes committed against persons who were at any time resident in Australia.19

In short, to be dealt with by an Australian Military Court, while it was irrelevant where the alleged war crime had been committed, the war crime must have been committed against: (1) any person who was at any time resident in Australia; or (2) a British subject; or (3) a citizen of any power allied or associated with His Majesty in any war. These classes appear to demonstrate a claim to jurisdiction over certain war crimes based on an extension to the passive personality principle, as the jurisdiction accrued from the victims essentially being nationals (although that word was not expressly used in either quoted sections of the Act) of the co-belligerent Allied states, including Australia.20

The underlying criteria for the classes of victims noticeably, and rather awkwardly, changed across sections 7 and 12 of the Act from a victim’s residency to allegiance to citizenship. The inelegance of this drafting was quickly acknowledged in practice; Maj K.R. Townley, acting as the Judge-Advocate in the Morotai M8 trial in December 1945, advised the Court that he thought that the categorization of classes (p.353) of victims had been ‘conceived in haste and born in confusion’.21 The Court in that trial agreed, commenting that the draftsmanship on the issue seemed ‘most inept’.22 In fact, the overall drafting of the Act was viewed with some bemusement during the first trials convened at Morotai in late 1945. Lt Col L.J. Byrne, the prosecuting officer in the first ever trial, which began at Morotai on 29 November 1945, described the Act as an example of ‘very curious and quaint methods of drafting’. In a rare moment of concurrence, the defending officer, Capt J.C. Brown, politely agreed, observing that the drafting was ‘to say the least…most unusual’.23

The decision to predicate the jurisdiction of the War Crimes Act 1945 upon classes of victims was certainly deliberate, as the British Royal Warrant of 18 June 1945, upon which the Act was closely modelled, did not define its jurisdiction in relation to victims.24 Regulation 4 of the Regulations for the Trial of War Criminals (UK) attached to the Royal Warrant simply provided that a person had to be located within the limits of command by an officer authorized to convene a Military Court and that it had to appear to the convening officer that the person had committed a war crime at any place within or without the limits of the officer’s command.25 Given the modelling of the War Crimes Bill, as it then was, on the Royal Warrant, it is not unsurprising, therefore, that one of the earliest drafts of section 7 of the Bill similarly said nothing at all about classes of victims but was worded simply as: ‘A military court shall have power, whether within or outside Australia, to try persons charged with war crimes.’26

The insertion of classes of victims into the War Crimes Act 1945 was, however, not without precedent in the Australian war crimes context. Clear parallels exist between the jurisdictional provisions of the Act and the developing phraseology of the Instruments of Appointment of the various inquiries into war crimes commissioned in Australia in 1943, 1944 and 1945. Sir William Webb, upon his appointment as War Crimes Commissioner, was instructed on 23 June 1943, for (p.354) example, to enquire into ‘[w]‌hether there have been any atrocities or breaches of the rules of warfare on the part of members of the Japanese Armed Forces in or in the neighbourhood of the Territory of New Guinea or of the Territory of Papua’.27

These instructions were clearly targeted at those (expressly limited) persons who might have committed war crimes, not the victims of such crimes. The instructions for the second and third war crimes inquiries, however, demonstrated an increasing focus upon the victims of war crimes. Webb’s remit was broadened for his second inquiry, when he was instructed on 8 June 1944 to enquire into ‘whether there have been any war crimes on the part of individual members of the Armed Forces of the enemy against any persons who were resident in Australia prior to the present war, whether members of the Forces or not’.28

The instructions given to the Board of Inquiry into war crimes, also headed by Webb, on 3 September 1945, about a month prior to the passage of the War Crimes Bill, expanded the remit even further to embrace both British subjects and citizens of allied nations. The Board of Inquiry was instructed to enquire into:

[w]‌hether any war crimes have been committed by any subjects of any State with which His Majesty has been engaged in war since the second day of September, [o]ne thousand nine hundred and thirty-nine, against any persons who were resident in Australia prior to the commencement of any such war whether members of the Defence Force or not, or against any British subject or against any citizen of an allied nation.29

The jurisdictional provisions of the War Crimes Act 1945 greatly resemble this final instruction, although the reference to ‘whether members of the Defence Force or not’ was dropped in section 7, while the third class of victims in section 12 was broadened to include any citizens of associated, not just allied, powers.

Given the lack of extant explanation for the wording of the jurisdictional provisions of the War Crimes Act 1945—which can probably be attributed to its swift drafting and the lack of significant consultation on the draft with other government departments or the Australian Army—it is difficult to determine precisely why the drafters deliberately chose to impose limits on the universal jurisdiction over war criminals and, having done so, then selected ‘resident in Australia’ as the criteria for section 7 of the Act, rather than the more normative birth, nationality or, perhaps, membership of the armed forces, as the wording of the second and third Webb appointments might have suggested. Caroline Pappas, one of the few researchers to have examined the processes and practices of the Australian war crimes trials, has suggested that the reason that classes of victims were explicitly referred to in the Act was to ensure that the expansive list of war crimes encompassed by the Act—which went beyond conventional war crimes—was not construed to cover victims who were enemy nationals, for example the Japanese themselves.30 There is certainly some support for the conclusion that the Act’s jurisdiction was designed (p.355) to preclude Australians from being charged under it with war crimes against the Japanese or anyone else. A domestic analysis of the British Royal Warrant undertaken to consider the legislative ‘action necessary’ to institute war crimes trials by Australia had pointed out with some concern, for example, that the terms of the Royal Warrant and its regulations were ‘completely general in their operation’ and that nothing in them prevented proceedings against British subjects accused of war crimes.31

At the same time, however, the drafters of the War Crimes Act 1945 appeared to be faced with the problem of ensuring that however jurisdiction based on classes of victims was established, it was not too limited. If section 7 of the Act had been based on Australian birth, nationality or membership of the armed forces, many victims of war crimes committed in New Guinea or on Nauru Island, for example, might have had their cases slip through the prosecutorial gap. It appears as though residency in Australia was thus chosen as the criteria for section 7 of the Act specifically to ensure that trials could be convened in relation to certain Asian victims of war crimes, as they would be encompassed by being ‘resident in Australia’, although perhaps not by the other possible criteria. A number of Asian victims at the Rabaul trials in fact fell into the class of ‘resident in Australia’, as section 3 of the War Crimes Act 1945 defined ‘Australia’ as including the ‘Territories of the Commonwealth’ and section 4 extended the operation of the Act to ‘every Territory of the Commonwealth’. When read together with the definition of territory in the Acts Interpretation Act 1901 (Commonwealth), this meant that inhabitants of New Guinea and Nauru, for example (the former a mandated territory of Australia and the latter a mandated territory of the British Empire but, under agreement between the United Kingdom, Australia and New Zealand, administered by Australia), were deemed to be resident in Australia for the purposes of the Act.

The use of the unusual phrase ‘resident in Australia’ in the Act proved somewhat confusing to Australian Army officers in the field faced with investigating war crimes and convening trials. The Australian First Army Headquarters, for example, posed several questions to Australian Headquarters (AHQ) in Melbourne throughout November and December 1945 in relation to whether certain persons—namely natives of mandated territories and, more specifically, Chinese civilians born in Rabaul before and after the mandate over New Guinea was granted in 1919—were to be considered as ‘Australian nationals’ for the purposes of the trials. The First Army was advised in each case that, for the purposes of the trials, natives of the mandated territories of New Guinea and Nauru were to be considered as ‘Australian nationals’, although the Chinese civilians in Rabaul must have ‘resided there permanently’ to be so considered.32 The continued use by all correspondents of the phrase ‘Australian nationals’ when considering the Act appears to demonstrate how difficult (p.356) it was to come to grips with the concept of ‘resident in Australia’. The AHQ’s advice to the First Army was, however, partially incorrect: there was no requirement of duration of residency in Australia in the War Crimes Act 1945 nor in the Act’s subsidiary Regulations for the Trial of War Criminals (Commonwealth)33 and, indeed, it could be construed from the wording of sections 7 and 12 of the Act, both of which included the phrase ‘at any time resident in Australia’, that such residency did not have to be uninterrupted or even ongoing at the time when the war crime was committed.

In addition to being difficult to interpret, the criteria of ‘resident in Australia’ soon proved to be impracticable, especially during the Morotai trials in late 1945 to early 1946, during which the victims were predominantly Australian prisoners-of-war. Defending officer Capt Brown argued throughout the first Morotai trial, for example, that the prosecution had failed to establish that the unidentified prisoner-of-war described as the victim in the charge had been ‘resident in Australia’. While a Japanese witness had given evidence that he had been told that one, possibly two, of the several airmen taken prisoner-of-war and executed together had been born in Melbourne and that all three had been stationed in Townsville, Capt Brown submitted to the Court that evidence of birth in Australia was not sufficient proof. Residency, in his mind, did not need to go so far as ‘domicile’; that is, a permanent intention to take up accommodation at a certain place but it must be ‘voluntary’. He argued that the airmen’s stationing in Townsville, if that was indeed the case, did not prove their intention to ‘reside’ there. Maj Townley, acting as the Judge-Advocate, dealt with these submissions in his summing up by suggesting a fairly simple test for ‘residence’. He considered that ‘residence’ did not imply ‘any great degree of permanence’ but it did imply ‘something more than a mere transitory passage’. He advised that:

[I]‌f a person were born in any particular country, you [the Court] may draw the inference from that fact alone, if there is no other evidence, that he comes within the class of person described in Section 7, i.e. a person at any time resident in that country.

He suggested, however, that if residency in Australia pursuant to section 7 could not be proven, then the Court then should proceed to consider whether the victim fell within the ambit of section 12; that is, whether the victim was a British subject or a citizen of any allied or associated power.

Eventually, the process of proving residency in Australia was made more efficient for prosecuting officers in mid-1946 by the addition of regulation 11A to the Regulations for the Trial of War Criminals.34 The new regulation, the subject of which was described as ‘urgent’ by the Secretary of the Army when urging its approval,35 read:

In any proceedings of a military court, a document purporting to be a certificate under the hand of the prosecutor or prosecuting officer that a person referred to in the charge was (p.357) at some time resident in Australia shall be prima facie evidence of the matter so certified without proof of the handwriting of the prosecutor or prosecuting officer.36

As only one such certificate in the hand of the prosecuting officer was ever tendered,37 it appears that the existence of the regulation in of itself effectively thwarted the defence in subsequent trials from pressing the issue of whether it had been proven beyond reasonable doubt that an (obviously Australian) victim was at any time ‘resident in Australia’. The failure to tender such documents in other trials is very curious, especially as there were ongoing but unsuccessful defence submissions throughout the trials that, due to the more complicated international mandate over Nauru, residents of Nauru were not residents of a territory of Australia.38

If the broad intention behind the drafting of the jurisdiction provisions of the War Crimes Act 1945 based on classes of victims was to ensure that no trials could be convened in relation to war crimes committed against enemy nationals, however, it was ineffectual. A victim’s residency in Australia by reason of being an inhabitant of a mandated territory seemed to override in a few cases other identification which would appear far more pertinent. Several trials regarding war crimes committed by Japanese against German and Austrian missionaries resident in New Guinea, for example, were held at Rabaul in early 1946. The defence in those cases submitted that the alleged victims were actually subjects or citizens of an Axis power allied with Japan and therefore did not fall within any class of victim established by the Act; however, these submissions were not accepted by the Australian Military Courts.

The victim in the Rabaul R6 trial, for example, was Father Henry Berger, a German Roman Catholic missionary who was assigned to Ramale Mission in New Britain.39 Berger complained that he had been struck several times and threatened with death by Lt Abe Akihisa, an officer of the 17th Transport Regiment. At trial the prosecuting officer, Maj F.D. Green, submitted that these acts amounted to unlawful assault. At the end of the prosecution case, the defending officer, Maj I.A.H. Spain, submitted that Abe had no case to answer on the ground that the evidence did not support the charge of unlawful assault since the laws and usages of war only operated as between opposing belligerents or belligerents and neutrals. He pointed out that the complainant in this case was a German subject and Japan and Germany were allies at war. Opposing the application to dismiss the charge, Maj Green referred the Court to section 7 of the Act and advised that, by virtue of this section, the Act would apply (p.358) even to Japanese who had been resident in Australia. The Judge-Advocate, Capt F. Ackland, advised the Court that section 7 of the Act referred to ‘any person who was at any time resident in Australia’ and that the term ‘Australia’ included the territories. Capt Ackland pointed out that there was evidence, which had not been challenged, that the German missionary was resident in Australia. He further advised that a person resident in the King’s dominions was entitled to protection of the King’s peace, notwithstanding their nationality and, moreover, that temporary occupation by a hostile power did not alter this entitlement. On the rights of inhabitants, he referred the Court to para 383 of the Chapter XIV of the Australian edition of the Manual of Military Law 1941, which set out that:

It is the duty of the occupant to see that the lives of inhabitants are respected, that their domestic peace and honour are not disturbed, that their religious convictions are not interfered with, and generally that duress, unlawful and criminal attacks on their persons, and felonious actions as regards their property, are just as punishable as in times of peace.

Unfortunately, the Judge-Advocate’s description of New Britain as part of the King’s dominions was, in this case, probably wrong. In the case of Frost v Stevenson in 1937, the High Court of Australia found that New Guinea, of which New Britain was a part, was not part of the dominions of the Crown. As Chief Justice Latham observed, ‘[t]‌he conclusion which I have thus reached is that New Guinea is a place out of His Majesty’s dominions in which His Majesty has jurisdiction’.40

By late 1947, however, there seemed to have been a rethink as to whether it was proper to try Japanese for war crimes committed against subjects or citizens of its own allies. In the Hong Kong HK1 trial, for example, the prosecuting officer, Maj A.D. Mackay, advised the Court that while it would become apparent from the evidence that ‘a number of other internees’ had been massacred alongside twenty-three Australian civilian internees on New Ireland in 1944, the killing of those other internees formed ‘no part of the case against the accused’ and the Court was to entirely disregard that evidence. He found it ‘ironical to note that such other persons were all of German nationality and supposedly the Allies of the Japanese’.41

That the Australian trials should encompass victims of war crimes who were not, by birth or nationality or even residency, ‘Australian’, however, was not natural and inevitable to some critics. The Minister for Post-War Reconstruction, John Dedman, questioned (somewhat outside the remit of his portfolio) in August 1946, for example, as to why Australia had been ‘placed in the position’ of having to try Japanese for war crimes committed against Indian and Chinese nationals,42 though he did not explain why he apparently considered it an unnecessary burden. Brig A.W. Wardell, then acting Adjutant-General of the Army, which was running the trials, drafted the response that:

A war crime is an offence against the law of nations and it behoves any nation to punish proved breaches against such law, whether committed against its own nationals or those of (p.359) another country. This is particularly so where the nationals of another country are resident in the territory of the Commonwealth, or if they are in that territory when the crime was committed.43

In short, Wardell was expressing his conviction that the identity of a victim of a war crime was, in fact, largely irrelevant, as it was the duty of every nation to punish war criminals in order to uphold the law of nations. In addition to apparently asserting the universality principle, however, Wardell’s response also suggested a reliance—not on the passive personality principle as might have been expected from the Act’s use of classes of victims—but on the territorial principle; that is, jurisdiction necessarily arose because the victims of the war crimes were resident in the territory of the Commonwealth of Australia (not just in Australia itself) or were within that territory when the war crime was committed. That a significant proportion of those victims whose cases were dealt with at Rabaul happened to be Asian was perhaps inevitable; it was a simple consequence of the Japanese armed forces’ transportation of enormous numbers of Indian and Chinese civilians and prisoners-of-war to New Guinea and other places in South East Asia and the commission of acts amounting to war crimes against them and the inhabitants of those places. As these places were either a part of Australian territory or fell within Australia’s zone of control in the immediate post-war period, many suspected war criminals fell into Australian custody at the end of the war which ‘behoved’, as Wardell had put it, Australia to convene war crimes trials against them given sufficient evidence to do so.

(II) Numbers and Nature of Asian Victims at the Rabaul Trials

It is a sad measure of the nature and extent of Japanese war crimes committed in the Pacific theatre of World War II that the total number of Asian victims whose cases were eventually dealt with at the Rabaul trials cannot be quantified. Due to a lack of evidence, many of the charges at the trials could only describe the victims as ‘unknown’ and their quantity as ‘a number’. In this manner, for example, the massacres of the population of Ocean Island were described in the charges in the Rabaul R51, R52, R53, R68 and R70 trials as the murder of ‘persons unknown’, a linguistic shorthand that completely understates their extent and significance. Regrettably, the identification and quantification of Asian victims is further hindered by the fact that there was little attempt (even at the Rabaul trials where the greatest number of trials were held) to regularize the particulars of charges, as can be seen by the various examples below:

  • Murder, in that they on Ocean Island on or about 20 August 1945 murdered persons unknown (Rabaul R53 trial);

  • (p.360) Torturing civilians, in that he at Ramale on or about 10 February 1945 tortured Sister Mektil, Sister Cecilia (both civilians) and other civilians (Rabaul R7 trial);

  • Murder, in that they at Teninbaubau, Bougainville on or about 23 January 1945 murdered twelve Indians (Rabaul R90 trial);

  • Murder, in that he at New Guinea in or about January 1945 murdered Hav Mehr Din (Rabaul R96 trial);

  • Ill-treatment of a prisoner-of-war in that he at Mango about 3 March May 1945 ill-treated WO Hor Chin Chun, a prisoner-of-war (Rabaul R118 trial);

  • Murder in that they at New Ireland in or about 1944–5 murdered a number of Chinese civilians, half-caste civilians and natives (Rabaul R127 trial).

Moreover, the issue of jurisdiction was only rarely directly addressed in the charges; that is, the particulars of charges never included a reference into which class a victim fell in the sense of ‘resident in Australia’, ‘British subject’ or ‘citizen of an allied or associated power’. Rather, those who drafted the charges usually concerned themselves only with whether the victim was a prisoner-of-war or a civilian, although if the nationality of the victim was known, it was often appended, as shown in the examples above. A defence objection to the drafting of a charge in the basis that it lacked particularity by not alleging that an unidentified victim fell into one of the classes of victims established by sections 7 and 12 of the War Crimes Act 1945 was overruled in the very first trial at Morotai, mentioned above, and was never, as far as this author knows, raised again.

In addition to distinguishing between prisoners-of-war and civilians, the trials also explicitly distinguished between ‘civilians’ and ‘natives’, a false distinction but one characteristic of this period. The trials consistently used the term ‘native’ both as a noun and an adjective and, of course, this has long since been unpacked as imperialistic, colonialist and racist.44 Most of those involved with the trials only perceived ‘natives’ as a kind of amorphous group, an inevitable (to them) demarcation based on physical appearance, living standards, language and education. As the Australian Military Courts were far less inquisitive about factual details than modern courts today, unless a witness, affiant or declarant in evidence happened, as an aside, to state where a ‘native’ victim was from, contemporary readers of the trials have no idea of the actual ethnic background, family or tribal affiliation or claim to nationality of many victims. The location of the war crime cannot be used as an accurate guide to the origins of victims, as the Japanese armed forces routinely shipped civilian inhabitants of occupied territories all over the Asia Pacific, either as labourers or simply to transfer population groups. The sole survivor to give evidence at the trials relating to the Ocean Island massacres, for example, was one Kabunare, a ‘native’ from Nikunau Island, one of the Gilbert Islands, then a British colony.

(p.361) While in many cases the specific nationality or ethnicity of native victims were not particularized in the charges, the general identity of most other victims can usually be surmised from a reading of the evidence, which begs the question as to why those who drafted the charges were not more specific and consistent in their approach. From the extent to which victim identification is possible from the records of the 190 trials held at Rabaul, Figure 17.1 shows a rough statistical overview of the types of victims. As charges often encompassed multiple victims or trials were convened upon multiple charges, the same trial could include several types of victims. In the Rabaul R173 trial of Lt Gen Adachi Hatazō, for example, the single charge against him, based on the principle of command responsibility, alleged the commission of ‘brutal atrocities and other high crimes against the people of the Commonwealth of Australia and its allies’ in New Guinea.45 The prosecution evidence, however, described war crimes committed by Adachi’s subordinates

Justice for ‘Asian’ VictimsThe Australian War Crimes Trials of the Japanese, 1945–51

Figure 17.1 Categories of victims represented in the Australian Military Court war crimes trials at Rabaul.

(p.362) against Chinese civilians, ‘natives’, Indian and Australian prisoners-of-war and Australian and American soldiers. The Rabaul R173 trial, therefore, is represented in each of the Chinese, ‘native’, Indian, Australian, and American categories.

This statistical breakdown overwhelmingly demonstrates that the majority of victims whose cases were dealt with during the Rabaul trials were of Asian or Pacific origin. For example, Indian, Chinese and ‘natives’ were represented amongst the victims in 176 of the 190 trials overall at Rabaul. By contrast, Australian victims were represented in only seventeen of the trials and British and American victims in only five and two trials respectively. Whatever criticisms might be made about the Tokyo Tribunal or the Australian war crimes trials held in other locations, the Rabaul trial series was not, by far, a story about justice for only Australian or Caucasian victims of war crimes.

(III) Attitudes towards Asian Witnesses at the Rabaul trials

That there was an Asian presence, indeed a certain Asian precedence, in the Rabaul trials does not mean that Asian witnesses, affiants or declarants who gave evidence of war crimes against their comrades, neighbours, family or themselves were viewed in an exemplary light, though to what extent attacks on their competency, credibility and veracity arose from prevalent colonial or racist attitudes of the time or were simply a defence tactic to weaken and destabilize prosecution cases is difficult to gauge. It was certainly a standard defence practice at many trials to submit that the prosecution evidence had been exaggerated or fabricated by victims and prosecution witnesses for a variety of reasons, including to take revenge upon the Japanese or to ingratiate themselves with Australian authorities in hope of personal benefit or to avoid being tried for a crime themselves. Lt Gen Adachi defended himself, and by implication his subordinates, in his trial by asserting, for example, that it was ‘not infrequent that the Indians after their recovery tried to conceal their treason against England and to falsify their position by reporting against the Japanese, maliciously fabricating or exaggerating incidents’.46

According to Phipson’s The Law of Evidence, the eighth edition of 1942, which was the principal text on evidence that was cited in the Australian Military Courts, all witnesses were, in general, to be considered as competent, including ‘believers of all creeds’.47 Witnesses were, of course, permitted to be questioned as to their (p.363) credibility and their general reputation for veracity.48 It is noticeable, however, that Phipson did not extend the warnings about, for example, the danger of convicting on uncorroborated testimony of accomplices49 or the requirement that the testimony of ‘infants, lunatics and drunkards’ be ‘received with caution’50 to witnesses of a certain nationality or ethnic type, such as Chinese, Indian or ‘native’ witnesses. In fact, the only occasions on which Phipson concerned itself regarding the evidence of such witnesses was in regard to various oaths, affirmations and declarations which could be used in relation to them.51 Nevertheless, there appeared to be a minor undercurrent of feeling in the trials that the evidence of Asian witnesses, including Japanese witnesses, although admissible, ought to be considered and weighed differently to the evidence of non-Asian witnesses. Some defence submissions, which appear from their literacy to have been made by Australian defending officers and not Japanese defending officers or civilian defence counsel, openly attributed negative tendencies to Asian witnesses along racial and ethnic lines. Moreover, these defence attacks were usually made in closing addresses to which no response by the witness was possible, even if he or she was present.52

As an example of the defence attacks on Asian witnesses, Chinese witnesses were said to have the ‘habit’ of revenge and the ‘disposition’ to exaggerate even ‘matter[s]‌ of no account’,53 which they did out of malice or because simply exaggeration was ‘common’ to those ‘unlearned coolies’.54 Indian witnesses were similarly alleged to have exaggerated out of their desire for revenge against the Japanese. Capt J.H. Watson, who appeared as the defending officer in many trials at Rabaul, suggested in no less than six separate trials that the prosecution evidence had been produced out of the ‘notorious’ ‘characteristic imagination’ or ‘characteristic exaggeration’ of Indian witnesses.55 In the Rabaul R71 trial Capt Watson submitted, for instance, that it was ‘possible for the Indians, who certainly would not lack motive, to distort simple facts having regard to their notorious capacity for exaggeration’.56

Similar critical comments were made regarding the veracity and credibility of the evidence of ‘native’ witnesses, particularly that the ‘shortness of memory’ of ‘natives’ was ‘well known’57 or that they were ‘prone to exaggeration’.58 That the Japanese accused occasionally held similar beliefs about the ‘native’ witnesses is clear. The two Japanese convicted in the Rabaul R21 trial submitted in their petition, for example, that (p.364) it was a ‘well-known fact’ that natives would ‘unashamedly’ tell lies, disregarding ‘morality and humanity’, as they were ‘slaves of going by nature’.59 In defending his witnesses in this case, Capt J.D. Steed, the prosecuting officer, admitted that ‘natives’ tended to ‘confuse fact with hearsay’. He suggested, however, that if the ‘frills’ of their testimony were removed, a ‘solid core’ of reliable evidence still remained.

Unlike such critical views of the evidence of Chinese and Indian witnesses, the views regarding the evidence of ‘native’ witnesses seemed to receive a bit of an official imprimatur in a few trials at Rabaul. Capt F.D. Green, acting as the Judge-Advocate in the Rabaul R9 trial, instructed the Court, for example, that:

You have formed your opinion of the native Witnesses, and I think you can take notice of the fact that it is very difficult indeed for the Prosecution to get a consecutive story from native Witnesses. There have been many discrepancies. Such discrepancies in fact that if they appeared in the case for the Prosecution with white Witnesses would undoubtedly rule the Prosecution completely out of having made a [p]‌rima facie case. I suggest to you that the fact that there are discrepancies in the evidence of the native Witnesses does not lead the Court to an inescapable conclusion that these Witnesses or any of them are lying. In this respect natives are very like children. They have vivid imaginations, and their evidence must be very carefully scrutinised.60

Similarly, Maj I.A.H. Spain, acting as Judge-Advocate in the Rabaul R10 trial, observed to the Court that the prosecution evidence had been wholly heard from ‘natives’. As such, he warned the Court that it was dangerous to convict upon the evidence of ‘natives’ who ‘like young children are an unreliable class of witness without some corroboration’.61

It is difficult to ascertain, at this late stage, whether the views expressed, in particular about the evidence of native witnesses, stemmed from the speaker or writer’s own beliefs or were simply representative of a more widespread viewpoint. Certainly, these sorts of observations about ‘native’ evidence were not too dissimilar from how Aboriginal witnesses were then viewed in mainstream Australian courts. As one anthropologist, A.P. Elkin, observed in 1947 when considering the ‘unsatisfactory features of native evidence’, Aboriginal witnesses were ‘apt to be labelled “liars”’ due to contradictions in their statements made at different times or when they were questioned in cross-examination.62 In fact, he thought that Aboriginal witnesses appeared to say ‘whatever will get them out of the strange magic-ridden room as quickly as possible’.63 Elkin concluded that the evidence of Aborigines in court was, for a complexity of reasons, ‘unreliable’, and ‘almost inevitably so’, and that the ‘actual court scene’ was ‘apt to be ludicrous and futile’.64 He stressed, however, that it was ‘not lack of intelligence on the part of the Aborigines’ which made them unreliable witnesses, it was the application of:

selected (that is, legal) mechanisms of one culture to persons ‘schooled’ in another culture which provides very different mechanisms for dealing with similar situations…[and that (p.365) this was done] without the medium of a thoroughly understood common language, and without expert assistance in understanding the native cultural background.65

The same observations could be drawn about the Australian Military Courts. Many of the Asian witnesses who appeared before the Courts were unlikely to have been familiar with an adversarial legal system, let alone with the adapted structure and procedures of the field general court-martial system generally used in the trials. Moreover, there was no common language—while English was the principal language of the Courts, Japanese, Mandarin, Cantonese and pidgin (Tok Pisin) were also used.66 There is no evidence whatsoever that the members of the Courts, most of whom were not legally trained, received any guidance on understanding the cultural background of the witnesses whose demeanour they were asked to assess. As Nancy Armoury Coombs points out in her 2010 study of the evidentiary foundations of international criminal convictions, factors such as these can function as impediments to the accurate provision of evidence by witnesses and, consequently, due fact-finding by courts.67

The attacks on Asian witnesses during the Rabaul trials were very significant because, in most cases involving Asian victims, the oral or written record of oral evidence of Asian eye-witnesses to the war crimes formed the bulk, if not all, of the prosecution cases. In contrast to the position at the International Military Tribunal at Nuremberg, there were few, if any Japanese documentary records of war crimes committed in occupied areas that could be tendered in evidence.68 Such documents, if they existed at all (which was increasingly unlikely by 1944–5, given the wide dispersal of units from various command headquarters in South East Asia and the tendency to give oral orders which were carried out immediately), were usually destroyed in response to official orders to dispose of records regarding military matters at the time of the surrender.69 For example, the Japanese armies in locations including Borneo, Malaya and Java were instructed on 20 August 1945 that not only were personnel who had mistreated prisoners-of-war or civilian internees permitted to immediately flee the area but that ‘documents which would be unfavourable for us in the hands of the enemy are to be treated in the same way as secret documents and destroyed when finished with’.70 If the members of the Australian (p.366) Military Courts were thus convinced by the defence to place a lesser weight on or to disregard evidence from those Asian eye-witnesses, then the accused was more likely to be found not guilty.

Given the lack of written judgments by the Australian Military Courts and thus remarks on the evidence or demeanour of specific witnesses or facts found to be proven, it is impossible to determine to what extent negative submissions about Asian witnesses had an impact on the Courts and affected their consideration of the weight to be given to their evidence. The rate of conviction at Rabaul, however, perhaps suggests that the Courts found such arguments generally unconvincing. Of 390 accused tried at Rabaul (some of these being the same persons appearing in multiple trials), 266 were convicted and 124 were acquitted, amounting to a conviction rate of 68.2 per cent.71 This figure was, in fact, slightly higher than the overall conviction rate from the Australian war crimes trials, which was 67.64 per cent.72

Although there was an overwhelming Asian presence at the Rabaul series of the Australian war crimes trials, it is certainly probable that the number of trials merely scratched the surface of the war crimes that were committed. That justice was selective in that sense is regrettable; however, as the distinguished British jurist Lord Wright of Durley, the Chairman of the United Nations War Crimes Commission, observed in 1948: ‘The majority of war criminals will find safety in their numbers. It is physically impossible to punish more than a fraction. All that can be done is to make examples’.73

That some trial personnel felt it necessary to attempt to discredit Asian witnesses on the basis of their race or ethnicity, well beyond what was usually permitted in attacking veracity, is disappointing, though perhaps not surprising for the period. No doubt, though, if there was a wider knowledge of the Australian war crimes trials, some of the most prevalent criticisms of the various international and domestic war crimes trials of the Japanese—that they did not adequately deal with war crimes committed against Asian victims—would be far less sustainable.

Notes:

(1) Won Soon Park, ‘Japanese Reparations Policies and the “Comfort Women” Question’, Positions: East Asia Cultures Critique, 5 (1997), 119.

(2) See, for example, Richard Minear, Victors’ Justice: the Tokyo War Crimes Trial (Princeton, NJ: Princeton University Press, 1971) and Yuki Tanaka et al. (eds), Beyond Victor’s Justice? The Tokyo War Crimes Trial Revisited (Boston, MA and Leiden: Martinus Nijhoff, 2011).

(3) Yuma Totani, The Tokyo War Crimes Trial: The Pursuit of Justice in the Wake of World War II (Cambridge, Mass. and London: Harvard University Asia Centre, 2008), 162–3.

(4) See Lyndel V. Prott, report entitled ‘Release of Records of Japanese War Crimes Trials’ 5 April 1967 (National Archives of Australia (hereafter NAA): A432, 1967/2152), 1, 2, 5.

(5) Prott, ‘Release of Records of Japanese War Crimes Trials’ (NAA: A432, 1967/2152), 11.

(6) The copies were not to include the Judge-Advocate General’s reports, pursuant to the practice not to provide such confidential and privileged reports when transcripts of Australian court-martial proceedings were ordinarily made available.

(7) See the correspondence on this issue in NAA: A432, 1967/2152.

(8) See Department of External Affairs cablegram 11 April 1967 (NAA: A432, 1967/2152).

(9) See Attorney-General’s Department, ‘Access to Historical Records’ Press release (NAA: A1838 3103/10/13/12 Part 16A, 2 June 1975).

(10) Prott, ‘Release of Records of Japanese War Crimes Trials’ (NAA: A432, 1967/2152), 11.

(11) Willard B. Cowles, ‘Universality of Jurisdiction over War Crimes’, California Law Review, 33 (1945), 218. Professor Hersch Lauterpacht described this as a rule which was ‘well established as a matter of practice and which has behind it sound legal principle’: Hersch Lauterpacht, ‘The Law of Nations and Punishment of War Crimes’, British Yearbook of International Law, 21 (1944), 68. See also Kenneth C. Randall, ‘Universal Jurisdiction Under International Law’, Texas Law Review, 66 (1988), 785–841.

(12) See Article 1 of the Charter of the International Military Tribunal, annexed to Agreement for the Prosecution and Punishment of the Major War Criminals of the European Axis, signed in London on 8 August 1945, 82 UNTS 279.

(13) See Article 1 of the Charter of the International Military Tribunal for the Far East, signed in Tokyo on 19 January 1946, amended 26 April 1946, TIAS 1589, 4 Bevans 21.

(14) As Rashid Khalidi points out, the ‘Far East’ is ‘one of many relics of an earlier, Eurocentric era, when things were “near” or “far” or in the “middle” in relation to the privileged vantage point of Europe’: Rashid I. Khalidi, ‘The Middle East as an Area in an Era of Globalisation’, in Ali Mirsepassi, Amrita Basu, Frederick Weaver (eds), Localising Knowledge in a Globalising World: Recasting the Area Studies Debate (Syracuse, NY: Syracuse University Press, 2003), 171.

(15) See, for example, Fry’s assertion that ‘a belligerent’s international competence to try and punish aliens charged with war crimes is unquestioned’, in Thomas Penberthy Fry, ‘The International and National Competence of Australian Parliaments to Legislate in Respect of Extra-Territorial Crime (Including War Crimes)’, University of Queensland, Faculty of Law Papers, 1 (1947), 33.

(16) Gen Thomas A. Blamey, the Commander-in-Chief of the Australian Army, for example, was quite unaware of the drafting of the War Crimes Bill and its passage, as he had drafted his own regulations to permit the trial of war criminals and presented it to his Minister on 3 October 1945: see the correspondence from Gen Blamey to Mr Frank M. Forde, Minister for the Army dated 3 October 1945; and between the Minister for the Army and Mr John A. Beasley, the Acting Attorney-General, dated 5 October 1945, in NAA: A472, W28681. Perhaps in return for the non-consultation, the Department of the Army drafted and presented regulations for the War Crimes Act 1945 to the Executive Council for approval without first submitting them to the Attorney-General’s Department: see Department of the Army, ‘Memorandum for the Secretary’ October 1945 (NAA: A472, W28681).

(17) The sitting session of Parliament was forecast to end on 5 October 1945: see Department of External Affairs, ‘Cablegram to Dr H.V. Evatt’ 26 September 1945 (NAA: A472, W28681).

(18) Act No. 48 of 1945.

(19) Italics added for emphasis.

(20) Another feature that highlighted co-belligerent jurisdiction was section 5(4) of the War Crimes Act 1945 that empowered the Governor-General to appoint as a member (not as the president) of an Australian Military Court one or more officers of the armed forces of any allied or associated power. Such appointments occurred on several occasions during the Rabaul trial series but the appointments of Chinese National Army officers in cases involving Chinese prisoners-of-war were controversial, as sometimes those officers had been prisoners-of-war together with the prosecution witnesses, leading to defence claims that the Courts were not impartial. See, for example, the Rabaul R55 trial (NAA: A471, 80915).

(21) See the Morotai M8 trial (NAA: A471, 80769). Townley later became the President of the Australian Military Courts at Manus Island, 1950–1.

(22) Morotai M8 trial, above n 21.

(23) Morotai M9 trial (NAA: A471, 80718).

(24) Royal Warrant, Army Order No. 81, 18 June 1945. The War Crimes Regulations (Canada) (PC 5831/45), made by Order in Council on 30 August 1945, were also based on the Royal Warrant.

(25) For an analysis of the British Royal Warrant, see United Nations War Crimes Commission, Law Reports of Trials of War Criminals, 1 (1947), 105–10. For an analysis of regulation 4, see Dr Egon Schwelb, legal officer to the United Nations War Crimes Commission, ‘British Regulations for the Trial of War Criminals By Military Courts’ 27 June 1945 (NAA: A472, W28681). For analysis of the war crimes trials held under the Royal Warrant, see A.P.V. Rogers, ‘War Crimes Trials Under the Royal Warrant: British Practice 1945–1949’, International and Comparative Law Quarterly, 39 (1990), 780–800.

(26) See the Attorney-General’s Department files developing the War Crimes Bill, with unidentifiable handwritten annotations (NAA: A2863, 1945/48). A similar draft of section 7 can be seen in an entirely handwritten draft of the War Crimes Bill (NAA: A472, W28681).

(27) NAA: AWM226, 5.

(28) NAA: AWM226, 7.

(29) NAA: AWM226, 8.

(30) Caroline Pappas, ‘Law and Politics: Australia’s War Crimes Trials in the Pacific, 1943–1961’ (unpublished PhD thesis, University of New South Wales, 1998), 94.

(31) See the undated, unattributed report entitled ‘Royal Warrant and Regulations for the Trial of War Criminals: Action Necessary to Constitute Australian War Crimes Courts’ (NAA: A472, W28681).

(32) See various messages from First Army to Landforces, 28 November 1945, 1 December 1945 and 20 December 1945; and responses from Landforces to Landops, First Army, 29 November 1945, 4 December 1945 and 21 December: (NAA: MP742/1, 336/1/382).

(33) Statutory Rule 164/1945, made under the War Crimes Act 1945 on 25 October 1945 and notified in the Commonwealth of Australia Gazette on 26 October 1945.

(34) Statutory Rule 56/1946.

(35) See Attorney-General’s Department, ‘Memorandum for the Secretary’ undated but despatched 12 February 1946 (NAA: A472, W28681).

(36) Statutory Rule 56/1946 (20 March 1946).

(37) The certificate was tendered as an exhibit in relation to an unnamed Australian prisoner-of-war who was murdered in that, after being injured during an air raid, he was shot and his body thrown into the sea by the accused, who claimed that it was a mercy killing, see Hong Kong HK9 trial (NAA: A471, 81656). It is unclear why the certificate was tendered in this case, as there are many trials in which the name of the victim was unknown.

(38) Such submissions only arose, of course, when the particulars of the charge were worded somehow so as to imply that the Nauruans were residents of Australia. In the Rabaul R180 trial, the Nauruan victims were described as ‘natives of Nauru displaced to Truk’ and the prosecuting officer submitted that by virtue of section 12 of the War Crimes Act 1945, the Act applied to war crimes committed against British subjects in any place whatsoever and that the Court might take judicial notice of the fact that natives of Nauru were British subjects: see the Rabaul R180 trial (NAA: A471, 81208).

(39) Rabaul R6 trial (NAA: A471, 80744).

(40) [1937] HCA 41; (1937) 58 CLR 528.

(41) Hong Kong HK1 trial (NAA: A471, 81645).

(42) Dedman presumably did not know that some Japanese had also been tried for offences against Axis nationals.

(43) Brig A.W. Wardell, Acting Adjutant-General, ‘Minute Paper on War Crimes Trials for Secretary, Department of the Army’ 20 October 1946 (NAA: MP742/1, 336/1/980).

(44) For a discussion of the implications of the term ‘native’, see, for example, Bill Ashcroft, Gareth Griffiths and Helen Tiffin (eds), Post-Colonial Studies: The Key Concepts (London: Routledge, 2000), 158. See also Arjun Appadurai, ‘Putting Hierarchy in Its Place’, Cultural Anthropology, 3 (1988), 36–49.

(45) NAA: A471, 81652.

(46) NAA: A471, 81652, Part 1. In trials at Rabaul concerning Indian prisoners-of-war, the principle defence was that the Indians were not prisoners-of-war subject to the protection of international law but that, in seeking the independence of India from Britain, they had been released on parole from the status of prisoners-of-war to become of their own free will either members of the Indian National Army, which was collaborating with the Japanese Forces, or volunteer labourers for the Japanese Forces. The Japanese claimed that it was only after the cessation of hostilities that the Indians asserted that they were prisoners-of-war and had been forcibly made to join the Indian National Army or to serve as labourers.

(47) Sidney Lovell Phipson, The Law of Evidence, (London: Sweet & Maxwell, 8th edn, 1942), 443.

(48) Lovell Phipson, above n 46, 180, 469–74.

(49) Lovell Phipson, above n 46, 477–9.

(50) Lovell Phipson, above n 46, 445.

(51) Lovell Phipson, above n 46, 452–4.

(52) Most witnesses were not present, since the permissive rule of evidence set down in section 9(1) of the War Crimes Act 1945 meant that the Court might consider ‘any document appearing on the face of it to be authentic, provided the…document appears to the court to be of assistance in proving or disproving the charge, notwithstanding that the…document would not be admissible in evidence before a field general court martial’. That this section had the effect in practice of depriving the accused of the right to cross-examine the witnesses against him was a common defence complaint.

(53) See the petition in Rabaul R81 trial (NAA: A471, 80987).

(54) See the defending officer’s closing address in the Rabaul R114 trial (NAA: A471, 81020).

(55) See the defending officer’s closing addresses in the Rabaul R20, R23, R29, R71, R117 and R130 trials (NAA: A471, 80729, 80725, 80736, 80984, 81023 and 81057).

(56) See the defending officer’s closing address in the Rabaul R71 trial (NAA: A471, 80984).

(57) See the defending officer’s closing address in the Rabaul R13 trial (NAA: A471, 80737).

(58) See the defending officer’s closing address in the Rabaul R21 trial (NAA: A471, 80730).

(59) See the defending officer’s closing address in the Rabaul R21 trial (NAA: A471, 80730).

(60) See the Judge-Advocate’s summing up in the Rabaul R9 trial (NAA: A471, 80742).

(61) See the Judge-Advocate’s summing up in the Rabaul R10 trial (NAA: A471, 80740).

(62) A.P. Elkin, ‘Aboriginal Evidence and Justice in North Australia’, Oceania, 17 (1947), 182.

(63) Elkin, above n 61, 183.

(64) Elkin, above n 61,184.

(65) Elkin, above n 61, 187.

(66) The few Indian witnesses who appeared in person tended to use English on the stand. The written statements of Indian witnesses that were tendered in evidence were, however, translated from languages such as Urdu.

(67) Nancy Armoury Coombs, Fact-Finding Without Facts: The Uncertain Evidentiary Foundations of International Criminal Convictions (New York: Cambridge University Press, 2010).

(68) As Nancy Armoury Coombs observes, the ‘high-level Nazi officials who were prosecuted at the Nuremberg Tribunal were convicted on the strength of their own documentation’: Coombs, 11.

(69) This instruction was itself, however, tendered at the Tokyo Tribunal: Totani, above n 3, 105–7. See also Edward Drea, ‘Introduction’ in Edward Drea et al, Researching Japanese War Crimes: Introductory Essays, National Archives and Records Administration for the Nazi War Crimes and Japanese Imperial Government Records Interagency Working Group, 2006, <http://www.archives.gov/iwg/japanese-war-crimes/introductory-essays.pdf> (accessed 3 March 2013), 9–10.

(70) Quoted in Totani, above n 3, 106.

(71) See the chart of statistics in David C.S. Sissons, ‘The Australian War Crimes Trials and Investigations (1942–51)’, <http://www.ocf.berkeley.edu/~changmin/documents/Sissons%20Final%20War%20Crimes%20Text%2018-3-06.pdf>. According to a similar chart compiled by Michael Carrel, the number of accused tried at Rabaul was 392: Michael Carrel, ‘Australia’s Prosecution of Japanese War Criminals: Stimuli and Constraints’, unpublished PhD thesis (The University of Melbourne, 2005), 100. My own count is that 392 accused were tried at Rabaul but, if the multiple trials of the same persons are discounted, the total number of individual persons tried was 330.

(72) Carrel, above n 70, 102.

(73) United Nations War Crimes Commission, History of the United Nations War Crimes Commission and the Development of the Laws of War (London: H.M. Stationery Office, 1948), 22.