An International Symposium

Publisher: Croatian Heritage Foundation & Croatian Information Centre
For the Publisher: Ante Beljo
Expert Counsellor: Dr. sc. Dragutin Pavlicevic
Editor: Aleksander Ravlic
Graphic Design: Gorana Benic - Hudin
Printed by: TARGA
Copies Printed: 2000
ISBN 953-6525-05-4

Count Nikolai Tolstoy

historian; research on inforced repatriation of Croats and other nationalities after the World War II
Court Close, Southmoore, nr Abingdon
Berkshire OX 13 5Hs ENGLAND



In 1945, the overwhelming majority of Croatian people returned to Yugoslavia from Austria were not killed at Bleiburg itself, but following their recrossing of the Drava. However, historically the involuntary repatriation of Croats in that year has long borne the name of the Austrian town where their Calvary began. I do not propose on this occasion to attempt any detailed account of the fate of the unfortunate victims after they had been returned to Yugoslavia, nor to attempt any statistical estimate, since these are topics at present undergoing specialised research within Croatia.

I intend here to concentrate attention on one aspect of the greater event, which to this day remains a strange and sinister mystery: the decision of the British military authorities to hand the Croats over to be slaughtered has never received any satisfactory explanation. It is an enigma which I have been researching now for nearly twenty years, to which even now I am unable to provide a coherent account, which is consistent with currently available evidence and historically more satisfactory account contained in my book The Minister and the Massacres (1986), and the curious version of events which appears in the British Government’s authorised report, The Repatriations from Austria in 1945 (1990).

It is an exceptionally difficult history to explore, largely because of the unusual obstacles placed in the path of anyone attempting to investigate it. The English historian Herbert Butterfield once wrote:

‘There are two maxims for historians which so harmonise with what I know of history that I would like to claim them as my own, though they really belong to nineteenth-century historiography: first, that governments try to press upon the historian the key to all the drawers but one, and are anxious to spread the belief that this single one contains no secret of importance; secondly, that if the historian can only find the thing which the government does not want him to know, he will lay his hand upon something that is likely to be significant’.1

In my case the situation has been almost the other way round. The British Government permitted me to inspect a few carefully-selected drawers, while the remainder were kept firmly closed. Before attempting my own explanation, an important matter needs to be emphasised. That is the distinction which should be drawn between the tragedy of the Croats driven back to Tito at Bleiburg on 12 May 1945, and the subsequent fate of the smaller body of Croats who remained in Austria following the Bleiburg tragedy.

The events at Bleiburg are simply described. During the first fortnight of May 1945, as the war drew to a close in Yugoslavia, terrified people of all ethnic categories in Yugoslavia streamed across the Karavanken mountains and the River Drava in a desperate attempt to surrender to the British. What they sought above all was protection from the Communist Partisans. Fearful massacres were being perpetrated behind the Yugoslav lines, and there were few who did not anticipate a ghastly fate in the event of capture, regardless of their actions during the chaotic years of occupation and war.

Shortly after midnight on 13 May the British 5th Corps Headquarters in Austria estimated that ‘approximately 30,000 POWs, surrendered personnel, and refugees in Corps area. A further 60,000 reported moving north to Austria from Yugoslavia. I am taking all possible steps to prevent their movement along roads, but this will NOT completely prevent them as they are short of food and are being harassed. Should this number materialise food and guard situation will become critical’. The 60,000 referred to were Croatian Domobran and Ustache military formations, followed by a vast concourse of civilian refugees.

By 15 May, the head of the advancing Croatian column arrived in the meadows just south of Bleiburg in southern Carinthia. There the Headquarters of the British 38th Infantry Brigade had been established a few days earlier within the massive walls of Bleiburg Castle overlooking the town on the edge of the adjacent forest. The Croatian commander, General Herencic, together with his interpreter Danijel Crljen, drove up to the castle, where they attempted to negotiate a surrender on terms with the British Brigadier Patrick Scott. However they had no sooner made themselves known to Scott, than the Yugoslav General Milan Basta arrived on the scene and insisted on joining the talks. Basta and Scott swiftly decided that they would compel Herencic to surrender all Croats under his command to the Yugoslav forces. Scott made it bluntly clear to the General that he would not under any circumstances permit the Croatian exodus to advance further into British-occupied Austria, and that he would deploy all forces he could muster to assist Basta in compelling submission if required. Eventually, after passionate arguments on both sides, Herencic recognised this aggressive display of force majeur, and reluctantly accepted the surrender terms. General Basta assured Brigadier Scott that everyone returned to Yugoslavia would be treated humanely and decently, and that the Croats consequently had nothing to fear. Scott dutifully reported this pledge to his superiors: whether he believed it is another matter. Meanwhile in the fields to the south, lying just out of sight of the castle of Bleiburg, a vast mass of people was gathered in a state of terror and confusion. They comprised the vanguard of what was effectively a fleeing nation.

A terrible panic began, as Basta’s Partisans opened fire from the woods on both sides upon the largely defenceless crowd collected below in the valley. Many people were wounded and killed. How many died in the fields beside Bleiburg I have been unable as yet to establish with any precision. Over the years I have obtained many accounts by eyewitnesses of what occurred. In addition graves of the fallen have been identified, and it seems that subsequently bodies were removed by the Austrian Black Cross and interred elsewhere. My impression is that the number of fatalities at Bleiburg itself was not great by comparison with what was happening elsewhere at the time, and may not have amounted to more than a few score. This suggestion may be imperfectly understood by many of the large number of Croats and sympathisers of other nationalities who attend each year at the commemorative service on the site. However I believe that as historical awareness advances, it will be increasingly appreciated that the annual obsequies are observed in memory of all those Croatian victims who died at the hands of the British and their Communist allies during the dark days of 1945, and not just those who fell in the immediate vicinity.

The great majority of people herded back to the guns of the Partisans were massacred during ensuing weeks and months, after they had recrossed the Yugoslav frontier. Thereafter they were subjected to the infamous death marches, which still await exhaustive investigation. Fortunately, now that Croatia is once again an independent nation, historians are at long last enabled to examine the issue on a free and scientific basis.

I turn now from the grim but historically relatively straightforward succession of events at Bleiburg to the vexed and convoluted question of British responsibility for crimes against the Croatian people. The Supreme Allied Commander in the Mediterranean was Field-Marshal Sir Harold Alexander, whose authority extended to Southern Austria. His Headquarters had been established at the royal palace of Caserta, outside Naples. The chain of command passed down through 15 Army Group (General Mark Clark) at Florence, to the British 8 Army (General Sir Richard McCreery), whose headquarters was in north-east Italy near Udine. 8 Army comprised two corps: 13 Corps, which faced Tito’s forces in Trieste and along the Isonzo, and 5 Corps (Lieutenant-General Sir Charles Keightley), which as has been seen occupied Southern Austria across the Yugoslav frontier to the north. On 15 May Alexander reported to the Combined Chiefs of Staff: ‘Approximately 600,00 German and Croat Troops of Army Group E moving into Klagenfurt area’. For some twenty-four hours it was wrongly believed at Caserta that a huge body of Croats had actually surrendered to 5 Corps in Austria, and Allied Force Headquarters (AFHQ) was obliged to decide what should be done with them. This error appears to have arisen from a genuine misapprehension during successive transmissions of the report from Austria.

Clearly Alexander felt that this influx was more than the British occupying force in Austria, which consisted of a Corps comprising some 25,000 men, could be expected to look after. On 16 May he issued this instruction to Air Vice-Marshal Lee, his military emissary at Tito’s headquarters:

"Commander of Allied troops in Austria reports that approximately 200,000 Yugoslav Nationals who were serving in German armed forces have surrendered to him. We should like to turn over immediately to Marshal Tito’s forces and would be grateful if Marshal Tito would agree to instruct his commanders to accept them and to arrange with GOC Five Corps the rate at which they can be received, and handing-over point on Austrian frontier south of Klagenfurt for return to Yugoslavia".

A few days later Tito replied, thanking the Field-Marshal. By now however events had overtaken these exchanges, and the Croats were already within the Yugoslav dictator’s grasp. Hindsight and moral judgements should be employed by historians with caution. My own belief, for what it be worth, is that General Herencic committed a grave error when he agreed to surrender to Basta. He was fully aware of the inevitable fate of the thousands of unfortunate people for whom he was responsible. The alternative course would have been to advance further into Austria, provoking Partisan attacks on their flanks and British military resistance ahead. While the Domobran forces were surely capable of fending off the Titoist irregulars, British artillery, armour, and air power presented a formidable obstacle. However Scott himself conceded that the forces at his disposal were insufficient to obstruct the passage of the Croatian exodus for long. Scott’s decision to compel the Croatian withdrawal appears to have been reached unilaterally, and at this early phase of the British occupation I suggest that he had little choice but to react to events as best he could with the scanty forces at his disposal. Had Herencic ordered a peaceful advance and dispersal into the British zone, it is certain that British troops would have opened fire, inflicting casualties on the dense crowd of Croats whose likely extent is impossible to estimate. At the same time it may be questioned whether British troops would have continued for long shooting at a mass of panic-stricken and largely defenceless fugitives. Evidence of the likely British response is available in the contemporary logbook of Captain Nigel Nicolson, Intelligence Officer to 1 Guards Brigade. Early on the evening of 19 May, 3 Grenadier Guards reported: "10000 Croats just arrived at Ferlach. 3 GG told to tell all Titoist in the neighbourhood and are NOT to let the Croats over bridge whatever happens". However it was not long before the implications of this order registered with 6 Armoured Division Headquarters, which half an hour later issued this qualifying rider:

‘NOT to fire at Croats if they attempt to rush bridge. (If they have women and children)’.

Such were the circumstances of the Communist capture of the half-million or more Croats fleeing from slaughter at the hands of the Communists. I now move to a mysterious aspect of this tragedy, understanding of which has yet to be fully achieved. As has been seen, the Croats at Bleiburg did not surrender to the British, who cannot fairly bear more than tangential blame for the dreadful atrocities which ensued. Certainly there exists nothing in international law which requires a belligerent to accept the surrender of units demanding to be taken prisoner. The Croatian surrender at Bleiburg took place on 15 May 1945. As the War Diaries make clear, what daunted the Allied command was the enormous number of fleeing troops and refugees reported to be advancing into Carinthia, at a time when 5 Corps had barely established its presence in the region, and when relations with Tito were dangerously inflammatory. Prior to this, from 12 May onwards, numerous smaller bodies of Croatian soldiers and civilians had succeeded either in arranging a formal surrender to British forces, or in infiltrating undetected into their zone of occupation. Since it was clearly unnecessary to guard people who were desperate to remain in British custody, the fugitives were either directed to large camps improvised for their reception, or simply told to stay put where they found themselves. By 15 May 5 Corps reported to 8 Army that they held some 25,000 Croats.

Prior to the Bleiburg crisis, British forces had made no attempt to halt these lesser incursions, and accepted their surrender without recorded reservation. For the present the internees settled down as best they could in the British zone, safe (so they thought) from the clutches of Tito’s executioners. Marauding bands of Partisans who sought to open fire on the refugees in their camps were deterred by patrolling British guards. Explicitly on occasion, and implicitly throughout, the British command accepted that their 25,000 uninvited "guests" lay under the protection of international law. The British Government was responsible for the protection and humane treatment of prisoners-of-war under the terms of the 1929 Geneva Convention. 2 Initially 5 Corps Headquarters does not appear to have contemplated any other course. Had they chosen otherwise, the refugees’ arrival in the British zone of occupation could readily have been prevented, since across was confined to bridges across the Drava.

Such was the situation up to the middle of May. Yet from the 15th onwards 5 Corps policy towards the captive Croats changed drastically, from one in accord with the laws of war and dictates of humanity to one of ruthless co-operation with the greatest mass purge of the Yugoslav Communist regime. During the third week of May arrangements were made for all Croats in Corps custody to be transported into the hands of Tito, so that he might extend his genocidal policy to those Croats who believed themselves safe from return to Yugoslavia. Given the general awareness of Tito’s attitude towards the wartime state of Croatia, the notoriously brutal nature of his regime, and the atrocious behaviour of his troops within the British zone of Austria, there can be little doubt that those who arranged their repatriation nurtured no illusions about the fate to which their charges were being consigned.

As early as 13 may 1 Guards Brigade War Diary had reported : ‘Slovenes and Serbs mostly concentrated [in] Viktring cage. None of these can be repatriated except to almost certain death at hands of Tito’.

If that was the fate anticipated for the Serbs and Slovenes, how much worse was it likely to be for the Croats! The Partisans made little attempt to disguise their appetite for a bloody retribution. Until forcibly prevented by British troops, they repeatedly attempted to murder inmates of Viktring camp, south of Klagenfurt. On 25 May Captain Nicolson’s logbook recorded:

‘100 further Croats ... are already on the way to Yugoslavia by train - en route for the slaughter-house ... Information came from Tito officer who was in charge of loading of dump at Maria Elend’.

What was it that caused this dramatic and dishonourable change in policy? The pattern of events shows clearly that the decisive intervention occurred on 13 May, when Harold Macmillan unexpectedly arrived at Corps Headquarters. Macmillan was at the time Minister Resident in the Mediterranean, a post which was effectively that of political adviser to Field-Marshal Alexander. In this capacity he possessed authority to communicate directly with the Foreign Office and the Prime Minister. On 12 May Macmillan had arranged with Alexander to fly to 8 Army in north-east Italy, where he intended to consult with General McCreery over the Allies’ deeply worsening relations with Tito. As he reported to the Foreign Office on the eve of his departure, he intended to advise McCreery on the political situation, and receive in return a military assessment from those on the spot. Macmillan spent the evening of the 12th visiting McCreery and Lieutenant-General Harding, whose 13 Corps faced the Yugoslavs along the line of the Isonzo.

At this point there occurred a dramatic change to Macmillan’s schedule. Instead of flying back to Naples as originally intended, he unexpectedly flew north over the mountains into Austria. There he spent two hours in discussion with Keightley and his staff. What happened at their conference can only be inferred from the evidence, since Macmillan never disclosed the motive for his altered itinerary and the nature of the discussion at 5 Corps Headquarters.

In his diary, which was probably compiled the next day, Macmillan expatiated at some length on what was evidently one of the more important issues laid before him by Keightley:

‘To add to the confusion, thousands of so-called Ustashi or Chetniks, mostly with wives and children, are fleeing in panic into this area in front of the advancing Yugoslavs. These expressions, Ustashi and Chetnik, cover anything from guerrilla forces raised by the Germans from Slovenes and Croats and Serbs to fight Tito, and armed and maintained by the Germans - to people who, either because they are Roman Catholics or Conservative in politics, or for whatever cause are out of sympathy with revolutionary Communism and therefore labelled as Fascists or Nazis. (This is a very simple formula, which in a modified form is being tried, I observe, in English politics.).

Macmillan’s diary was compiled with a view to eventual publication, and is consequently not always as candid or complete as it might otherwise have been. The passage quoted invites some obvious questions, and cannot be naively taken au pied de la lettre, as it has been by Macmillan’s sycophantic biographer Horne and the authors of the Government-sponsored "Cowgill Report".

Plausible inferences may be drawn, categorised as follows:

1. It is clear that the whole of this information derived from General Keightley.

2. Given the brief time available for their meeting, and the pressing urgency which led Macmillan to alter his original travel plans so dramatically, the topic is unlikely to have represented mere small talk.

3. Though the passage recounting the visit to Klagenfurt is written in a style appropriate to a personal journal, it conveys the impression of reflecting the formal agenda which must have governed such a discussion. The topics appear to be listed in order of importance.

(I) The Yugoslavs had openly declared their intention of annexing Southern Carinthia, where their troops were behaving with increasing truculence.
(II) Among great numbers of surrendered enemy forces, 5 Corps held 40,000 surrendered Cossacks and White Russians, whose return was claimed by the Soviets. Marshal Tolbukhin’s army, which had halted within the bounds of the allotted British zone, was likewise known to hold a number of liberated British prisoners of war.
(III) Various categories of "Yugoslavs" had arrived in panic-stricken flight before the advance of Tito’s armies, as described in the passage above.

Macmillan concluded his account of the meeting by explaining: "We had a conference with the general and his [staff] officers covering much the same ground as those with Generals McCreery and Harding yesterday. He gave us his story and we gave him ours. I feel sure it was useful and helpful all round".

Thus, as might be expected, Keightley tabulated the major problems facing him in Austria, to which Macmillan responded with appropriate advice or directions. It is surely significant that each of the issues raised was governed by political factors, which Macmillan was pre-eminently qualified to address. Macmillan paraphrases the responses he provided for the first two issues.

(I) The Yugoslav aggression: "We have to look on, more or less hopelessly, since our present plan is not to use force and not to promote [provoke?] an incident".
(II) The Cossacks and White Russians: "We decided to hand them over ... I suggested that the Russians should at the same time give us any British or wounded who may be in his area".

However no indication appears in Macmillan’s diary as to what if any advice he proffered on problem (iii). This omission appears the more curious the closer it is considered. The first point to note is that, if we discount Macmillan’s characteristically florid language, his account implies that Keightley’s report on the Yugoslav refugees was explicit and detailed. It covers the whole language of anti-Communist Yugoslavs held at Viktring and elsewhere by 5 Corps at the time of Macmillan’s arrival:

1. ‘Chetnik’ Slovenian troops, being ‘guerrilla forces armed and maintained by the Germans ... to fight Tito’: i.e. Slovenian Domobranci.

2. Croatian forces, falsely categorised en blocas ‘so-called Ustashi’, in reality largely comprising ‘guerrilla forces armed and maintained by the Germans ... to fight Tito’: i.e. Croatian Domobrani

3. ‘Chetnik’ Serbs: i.e. anti-Tito Serbian formations acting under the authority of Generals Mihailovic or Nedic. The first three groupings were accompanied ‘mostly with wives and children’.

4. Roman Catholic and Conservative elements ‘out of sympathy with revolutionary Communism’: i.e. civilians from varied ethnic groups in Yugoslavia who had reason to fear a Communist take-over.

Macmillan’s listing is confirmed by the War Diary of 6 Armoured Division for the same day, which reported: ‘Position with regard to surrendered personnel in the Divisional area was now very roughly as follows:- ...

'Mercenary Tps.

(a) In battle Group Seeler 21,000 Slovenes, Serbs and [White]
(b) Croats. Area Eisenkappel, military strength 7,000 plus 3,000

Macmillan’s account of his conference with Keightley remains the only full version available, since both of Keightley’s senior staff officers, Brigadiers Low (Aldington) and Tryon-Wilson, deny having been present. (The absence of the Corps Commander’s two senior staff officers on such an occasion is remarkable). 5 Corps must have presumably have recorded a summary for its own reference, but if so it has disappeared from the War Diary along with so much else that the British Government subsequently deemed compromising. Accordingly we are obliged to rely on Macmillan’s version, which at least has the advantage of being written at the time. However analysis reveals some curious anomalies.

Macmillan records the advice he gave in respect of the first two of Keightley’s points, but does not reveal his response to the question of the Yugoslav refugees. The omission is curious, in that so far as the Cossacks were concerned Keightley had already received precise instructions how to treat captured Russians, in the form of a carefully-worded directive issued by 8 Army on 13 March. In the case of the Yugoslavs, however, the position was unclear. On 3 May 8 Army had issued a ruling that "Chetnicks, troops of Mihailovitch, and other dissident Yugoslavs ... will be regarded as surrendered personnel and will be treated accordingly. The ultimate disposal of these personnel will be decided on Government levels". The context of the order, however, was the surrender of all German forces in Italy. Furthermore it contained no explicit allusion to Croats, thousands of whom had now surrendered to 5 Corps.

It seems inconceivable that Keightley did not seek Macmillan’s advice on this essentially political question, and that Macmillan did not provide him with some guidance. The picture becomes the more puzzling when Macmillan’s motive for unexpectedly flying to Klagenfurt is taken into account.

The authors of the "Cowgill Report" assert that Macmillan flew to meet Keightley in order to explain to him the need for extreme tact in dealing with the Yugoslavs, since three days earlier Keightley had requested permission from McCreery to be permitted to shoot at Yugoslavs who disobeyed British commanders.3 Though advanced as settled fact, this suggestion represents pure speculation, and is clearly designed to substantiate the Report’s thesis that Macmillan only encountered the refugee problem in Austria en passant, and played no material part in the decision to have them despatched to the Communists.

Had Macmillan thought it necessary to advise Keightley in person there can be little doubt that he would have planned to fly to Austria at the outset of his expedition. Moreover this does not explain why he subsequently concealed the decision to repatriate the Cossacks and Yugoslavs from the Foreign Office.

Recently a senior staff officer at 5 Corps Headquarters confirmed the accuracy of my suggestion that Keightley contacted Macmillan while he was with McCreery, and requested him to fly north and advise him how he should treat the Cossack. In a recorded interview held at the Imperial War Museum, Brigadier C.E. Tryon-Wilson recalled in 1990 that during the Italian campaign,

"I think when the history of that campaign is dealt with you will find that in many cases 5 Corps were in a position sometimes to go to Harold Macmillan, sometimes direct to Alex".

Going on to describe the problems facing 5 Corps in Austria, the Brigadier recalled a visit he made on 10 or 11 May to red Army Headquarters at Voitsberg.

"Now soon after we arrived in there [Austria] - and it was within a few days of General Keightley going up - I was asked to go up for two reasons, really: one was to have a look at the area through which we might have to operate; and secondly to make a contact with the other side. I didn’t at that particular time go up with any instructions, or intention to talk about the handing over of the Cossack. Because - again, I refer to the 78 Div ones - they were moved a long way, in the hopes that the Russians didn’t know too much about them, because they didn’t want them to be handed back. 46 Div, which was much closer, had quite a lot. General Keightley had prior to that (I think) - my journey - he had (I think, rightly) he had already had contact with General - with Harold Macmillan. And he’d told him what the problem was, and he had mentioned - or perhaps he’d mentioned that we had some White Russians. But we certainly at that particular stage, until the 15th I know (and it’s a thing which sticks in my mind) we really didn’t know the numbers or the names of anybody, because we had to tell the Divisions they had to feed the chaps out of the reserves that they’d got there. And Harold Macmillan had said: "Well, look, if you’re going to hand these chaps back, and you want to hand them back, the only thing I can do is tell you that you’ll get a better deal if you go directly to the top - not through an intermediate. And we suggest that you go straight to General Tolbukhin and sort the thing out".4 Brigadier Anthony Cowgill, Lord Brimelow, and Christopher Booker, The Repatriations from Austria in 1945: The Report of an Inquiry (London, 1990), pp. 4 The advice Tryon-Wilson ascribed to Macmillan appears a little confused, but its principal point is plain. Keightley requested Macmillan’s attendance specifically for the purpose of advising him on the policy he should pursue with regard to the Cossacks.

Furthermore this first-hand account confirms that prior to Macmillan’s visit 5 Corps had been anxious to protect the Cossacks from betrayal to the Soviets. Precisely what the Minister said to Keightley is not recorded, but the gist of it is indisputable. Ten days later Keightley informed 8 Army commander General McCreery: "As a result of verbal directive from Macmillan to Corps Comd at recent meeting we have undertaken to return all Soviet Nationals in Corps area to Soviet forces". Hitherto both Keightley and Macmillan had withheld all reference to this "verbal directive" from their colleagues, and the only reason that the 5 Corps Commander chose to reveal it was in the context of an attempt to reverse a newly-received order from Alexander forbidding him to use force to compel Cossacks to return "home".

At present I a, concerned with the fate of the Croats rather than that of the Cossacks. However there exists abundant reason to believe that the repatriation of both peoples represented the outcome of an identical policy decision. Before Macmillan’s arrival the evidence indicates that 5 Corps had neither the intention nor the desire to hand anyone over to be maltreated or killed. Thereafter a radical shift in policy occurred, which required extensive deception of the Allied command, to say nothing of the unfortunate prisoners. The 5 Corps war diary and other military records have been substantially doctored, a procedure which would scarcely have been necessary had all proceedings been above board. The "Cowgill Committee" was at pains to scout the idea of any conspiratorial activity on the part of Macmillan or Keightley, on general grounds of implausibility. However it is possible to provide a telling example of the ingenuity with which Macmillan succeeded in duping his "friend" Alexander. On this return to Naples on 14 May Macmillan succeeded in persuading Alexander’s Chief Administrative Officer, General Robertson, to issue an order requiring the handover of Cossacks and Yugoslavs to the Soviet and Yugoslav Communists. Macmillan’s diary is silent sent late that night by Alexander Kirk, Macmillan’s American counterpart as political adviser to Alexander, to the State Department in Washington. ‘This afternoon General Robertson, Chief Administrative Officer AFHQ requested us to concur in a draft telegram to CG British Eight Army authorising him to turn over 28,000 Cossacks (see our 797 of October 16, 1944, Midnight), including women and children to Marshal Tolbukhin, and further instructing him to turn over to Yugoslav Partisans a large number of dissident Yugoslav troops with exception of Chetniks. ‘General Robertson stated that Macmillan, who talked with CG Eight Army yesterday, had recommended this course of action. We asked whether the Russians had requested that these Cossacks be turned over to them, and Robertson replied in the negative and added "But they probably will soon". We also asked General Robertson what definition he proposed to give to "Chetnicks" and he was very vague on this point. We then stated we could not concur without referring the matter to our Government. CAO expressed disappointment that we did not seem to agree with him on this point but added that he was faced with a grave administrative problem with hundreds of thousands of German POW’s on his hands and could not bother at this time about who might or might not be turned over to the Russians and Partisans to be shot. He would have to send his telegram in spite of our non-concurrence.

‘Department’s views would be appreciated urgently’.

Given this conflict of opinion between the US and British Political Advisers, it is likely that Robertson would have reverted to Macmillan for confirmation of the course he now adopted. At 4.36 that afternoon (14th) Robertson despatched an order to 8 Army for onward transmission to Knightly, which required the prompt hangover of ‘Russians’ (i.e. the Cossacks), and concluded with this instruction: ‘All surrendered personnel of established Yugoslav nationality who were serving in German Forces should be disarmed and handed over to Yugoslav forces’. Copies were sent to Alexander’s Chief of Staff, General Morgan, who was on the point of departing on an extended mission to North Italy and Austria, and Macmillan. Significantly none was sent to Kirk, who would have observed that even the tentative saving clause regarding ‘Chetniks’ was dropped from the final version. The omission suggests that it was included in the draft in what proved to be the vain hope of gaining Kirk’s. Having decided to proceed without his approval, Macmillan and Robertson seized the opportunity of extending the order. It was this order which those responsible at 5 Corps employed as justification for the repatriation operations which continued throughout the second half of May. It is a strange but seemingly indisputable fact that Alexander remained wholly unaware of the existence of this order until 21 May. Precisely how it was kept from him is uncertain, but the events which followed establish the omission beyond reasonable doubt. From 16 May onwards he was engaged in elaborate discussions with Eisenhower, whose purpose was the evacuation of the Cossacks to SHAEF custody. At the same time it was his declared intention to transport the Yugoslav prisoners and refugees in Austria to camps in Italy. It was not until 21 May that General McCreery came to query the discrepancy between this policy and that prescribed in the ‘Robertson order’, in response to which Alexander issued fresh clarificatory orders. It would be absurd to suppose that the two Supreme Allied Commanders went to all this trouble in the full knowledge that a diametrically different policy was already being put into effect. Fortunately it is unnecessary to rely on inference and general grounds of plausibility, since evidence of extensive deception is further to be detected in the contemporary records. On 17 May Alexander issued this emotive appeal for direction to the Combined Chiefs of Staff: ‘To assist us in clearing congestion in Southern Austria we urgently require direction regarding final disposal following three classes:

(a) Approximately 50,000 Cossacks including 11,000 women, children and old men. These have been part of German armed forces and fighting against Allies.
(b) Chetniks whose numbers are constantly increasing. Present estimate of total 35,000 of which we have already evacuated 11,000 to Italy.
(c) German Croat troops total 25,000.

In each of above cases to return them to their country of origin immediately might be fatal to their health. Request decision as early as possible as to final disposal’.

The wording indicates plainly enough the extent of Alexander’s humanitarian concern for the helpless fugitives, and his objection to delivering them to their enemies. In the present context, however, the content of the signal is of secondary concern to the manner of its transmission. Though the format establishes that the message emanated from the Supreme Allied Commander in person, it was actually despatched from the office of his Chief Administrative Officer, General Robertson. This represented regular procedure, though in this instance it raises a significant query concerning Robertson’s role in the policy of forced repatriation.

Three days earlier, at Macmillan’s instigation, he had issued the infamous ‘Robertson order’ cited supra, which ordered 8 Army to hand over ‘all Russians’ to the Soviets and ‘all Yugoslavs serving in German forces’ to Tito.

When he received his copy of the Field-Marshal’s signal of 17 May, Robertson must have recognised that Alexander was unaware of the existence of the prior order, which conflicted with his concern for the prisoners’ welfare and made his appeal to Eisenhower superfluous. Why in that case did he not alert Alexander to the discrepancy? 5

It appears inescapable that Robertson deliberately withheld reference to his order of 14 May, whose callous provisions he well knew flouted the humanitarian intentions of the Field-Marshal. It may perhaps be questioned whether a such a deception was possible within the tightly-knit framework of a military headquarters. Alexander was notoriously a ‘hands off’ commander, who was inclined to leave much of the routine work to his capable subordinates. However this may be, fortunately there exists confirmatory evidence of the extent of the deception and indicates the skilful manner in which it was effected.

It will be recalled that late on 14 May Alexander Kirk, the American political adviser at AFHQ had reported to the State Department his dissent to Robertson’s proposal to hand over Yugoslav prisoners to Tito. The proposed move was in direct violation of agreed Allied policy, and on 16 May Assistant Secretary of State Grew instructed Kirk to lodge a formal protest with AFHQ on behalf of the US Government. The same day (17 May) that Alexander issued his appeal to the Combined Chiefs of Staff, Kirk’s deputy Carmel Offie registered formal objection with the Deputy Chief of Staff (General Lemnitzer), General Robertson, and Harold Macmillan:

‘I wish to refer to my non-concurrence in the telegram which the CAO despatched to MACMIS with regard to disposition of certain Yugoslav nationals who have surrendered to the Allies.

‘The Department of State has informed me urgently that in its opinion no distinction should be made between dissident Yugoslav troops and Chetniks and that the American position, with which the Foreign Office has agreed, with respect to dissident Yugoslav troops or anti-Partisans, has clearly been established.

‘You will recall that the British Ambassador in Belgrade proposed some two weeks ago that there were three alternatives available in connection with handling of these Yugoslavs:

(a) that they should be used as auxiliary troops;
(b) that they should be handed over to the Yugoslav Army; and
(c) that they should be disarmed and placed in refugee camps.

At that time the Department of State and the British Foreign Office agreed that alternative (c) was the only possible solution.

‘In summary, therefore, we believe that the troops in question who wish to surrender to American or British commanders in Northeast Italy should be disarmed and placed in base camps for investigation; that those wishing to return to Yugoslavia as individuals should be permitted to do so; and that all others should be removed to refugee camps and those against whom there is evidence of war crimes should be handled as such’. Alexander needed no persuading in this respect, and it was on the same day (17 May) that he issued an order providing for the evacuation of Chetnik and other ‘dissident’ Yugoslav prisoners in Austria to camps in the rear area of Northern Italy known as District One (‘Distone’). Next day a gratified Kirk reported back to the State Department: ‘S[upreme] A[llied] C[ommander] has informed Eight Army and ... Fifteenth Army Group that chetniks and dissident Yugoslavs infiltrations into areas occupied by allied troops should be treated as disarmed enemy troops and evacuated to BRIT concentration area. Total number believed about 35,000 AFHQ taking up question of final disposition’.

The order (known as ‘Distone’) to which Kirk referred required the evacuation of all surrendered Serbian, Slovene, and Montenegrin troops in Austria to camps in Italy, where thousands of their compatriots who had earlier surrendered in Italy were held. No reference was made in the order to the estimated 25,000 Croat troops held by 5 Corps in Austria, since the ruling had been issued in response to a specific enquiry from the 5 Corps Chief of Staff (Brigadier Low) regarding the disposition of ‘Jugoslav Royal Army? units. The Croats could not be quartered alongside their inveterate enemies, and so it was necessary to retain them for the time being in Austria. Any decision as to the ultimate fate of all these captured troops of Yugoslav nationality now rested with the Combined Chiefs of Staff, to whom Alexander had referred the question on 16 May.

So far as Kirk was concerned all appeared to be well. AFHQ policy was now in alignment with that agreed by the United States and Great Britain, and there appeared no longer to be any question of repatriating fugitives to be butchered by a vengeful Tito. More than two months were to pass before Kirk discovered that both he and Alexander had been victims of an elaborate deception practised by their own colleagues.

It was on 14 May that General Robertson showed Kirk the draft of the order approved by Macmillan, which provided for the handover of all Yugoslav prisoners held in Austria. Robertson had ignored Kirk’s protest, and shortly afterwards issued his notorious order FX 75383 providing for what he cynically anticipated would be the likely slaughter of Cossacks and Yugoslavs. However the text of the formal US protest of 17 May lodged at AFHQ reveals that by then Kirk’s office had come into possession of what they presumed to be a copy of Robertson’s order. Kirk’s deputy Carmel Offie referred to it as ‘The telegram which the CAO [Robertson] despatched to MACMIS with regard to disposition of certain Yugoslav nationals who have surrendered to the Allies’.

The reference betrays the manner in which the American Political Adviser was duped. ‘MACMIS’ was the abbreviation for the Maclean Mission to Tito, a section of which was based at Trieste to report on Yugoslav moves in the disputed frontier zone. In reality, however, the sole order sent to Macmis ‘with regard to disposition of certain Yugoslav nationals who have surrendered to the Allies’ was not Robertson’s order FX 75383 of 14 May, but Alexander’s signal FX 75902 of the 15th, which arranged for the return to Tito of the 200,000 Croats. Since the belief that the 200,000 had surrendered to the British was swiftly discovered to be mistaken, the second signal effectively became dead letter as soon as it was issued.

It was in this ingenious manner that Offie was gulled into believing that the signal effectively set aside Alexander’s order of 17th May was that which Robertson had Shawn Kirk on 14 May. On 18 May Kirk accordingly reported to the State Department that Alexander had issued orders superseding Robertson’s signal, which consequently no longer posed a threat to Yugoslav prisoners. All must now have appeared well both to Kirk and Alexander, who were however unaware that the real Robertson order had not been explicitly superseded. It was retained by 5 Corps, who were to use it as justification for their subsequent handover of tens of thousands of Yugoslav nationals. It was this Machiavellian procedure which enabled the Robertson order to remain dormant, awaiting reactivation when required.

The conspiracy involved a high degree of skilful duplicity, with the consequence that its unravelling is also a fairly complex process. A brief summary of successive events will however serve to clarify events.

13 May
Following Macmillan’s visit to 5 Corps, both he and Keightley omit all reference in their otherwise detailed reports to the presence of tens of thousands of Yugoslavs in the Corps area, and to their decision to deliver them to the Communists.

14 May
At Macmillan’s instigation Robertson issues his order for the Yugoslavs to be handed back to Tito. Kirk is carefully omitted from the circulation list.

16 May
Alexander’s Chief of Staff, General Morgan, visits 5 Corps. Reporting to Alexander, he explains that the presence of ‘25000 Croats [and] 25000 Slovenes’ is imposing a severe strain on 5 Corps resources. He clearly cannot have been informed by Keightley of the Robertson order, which provided a remedy for the problem.

Alexander accordingly requests directions from the Combined Chiefs of Staff for the disposal of Yugoslavs in Austria. Again, it is inconceivable that he would have done this had he been aware of the existence of the Robertson order, which already provided for them.

17 May
The US Political Adviser is misled into believing that the Roberson order has been superseded.

Aldington at 5 Corps issues the following order, extending the category of those required to be repatriated, and taking care not to transmit a copy to higher command:

‘all Yugoslav nationals at present in the Corps area will be handed over to Tito forces as soon as possible. These forces will be disarmed immediately but will NOT be told of their destination. Arrangements for the handover will be co-ordinated by HQ in conjunction with Yugoslav forces. Handover will last over a period owing to difficulties of Yugoslav acceptance. Fmns will be responsible for escorting personnel to a selected point notified by this HQ where they will be taken over by Tito forces’.

General McCreery instructs Keightley:
‘Pending outcome of present Governmental negotiations with Yugoslavs you will avoid entering into any agreements with Yugoslav Commanders’.

18 May
Aldington receives Alexander’s order of the previous day: ‘Chetniks and dissident Yugoslavs infiltrating into areas occupied by Allied troops should be treated as disarmed enemy troops and evacuated to British concentration area in Distone [Italy]. Total numbers including eleven thousand already in Distone believed about thirty five thousand’.

19 May
Despite the clear terms of the last two orders Aldington enters into a written agreement with Yugoslav Colonel Ivanovich, committing 5 Corps to hand over all Yugoslavs in the area specifically including ‘Chetniks and dissident Yugoslavs’, and relying on the Robertson order for his authority.

21 May
Alexander learns for the first time of the existence of the Robertson order in consequence of General McCreery’s enquiry as to whether is still expresses AFHQ policy.

It must be apparent by now who was masterminding this elaborate train of deception. Macmillan enjoyed a particularly close working relationship with Robertson. On 8 January 1945, for example, he noted in his diary: ‘I like doing business with General Robertson, for he is a very clever man’. It was probably on the morning of 14 May that he approached the General, explaining the problem (as he saw it) of the Russian and Yugoslav prisoners whose surrender has been accepted by 5 Corps. Between them they devised the order despatched that day to 5 Corps, which flouted Allied policy by requiring the handover of Yugoslav prisoners to Tito. This move was kept a closely-guarded secret from Alexander, whose ignorance is proved inter alia>by his elaborate arrangements in keeping with Allied policy.

It was Alexander’s ‘Distone order’ of 17 May that endangered the whole conspiracy, which depended on the Robertson order’s surviving unrepealed at 5 Corps. It can be seen why Lord Aldington was so concerned at the 1989 libel hearing to pretend that the ‘Distone’ order for some mysterious reason never reached his Headquarters (to which it was directed), and why his neighbour Judge Davies was at equal pains to withhold from the jury the evidence which proved the contrary.

It was on 15 may that Assistant Secretary of State Joseph Grew required Kirk to protest against the issuing of the ‘Robertson order’, and on 17 May that Kirk’s deputy office reported that he had registered his ‘non-concurrence in the telegram which the CAO despatched to MACMIS’. The likely date on which Offie was accordingly 16 May. On that day Macmillan spent some time with Offie, advising him on signals to be sent to the State Department, after which: ‘As part of regular routine, I had a conference with General Robertson on various Italian questions ...’

Historians have increasingly come to recognise the extent to which deviousness and duplicity ranked among Macmillan’s prevailing characteristics, along with a cynical contempt for humanity. He possessed both motive and opportunity for misleading the Americans, and the substitution of Alexander’s outdated signal FX 75902 of 15 May for Robertson’s FX 75383 of the 14th was precisely the sort of deception was the condemnation to almost certain death of some 50,000 people merely inflated the sense of power which Macmillan’s deeply-rooted sense of inferiority ceaselessly craved.

I have not space here to analyse the complex machinations which followed over the next week, which confirmed the fate of the unsuspecting Croatian prisoners-of-war. Suffice it to say that between 19 and 22 May thousands of Croats were transported to the hands of Tito’s executioners by means of further lying and deception.

It was not until August that Kirk came to learn of the deception which had been practised on him. On 14 August he reported bleakly to the State Department: ‘On receipt to your telegram 719, August 6 we addressed memorandum to Supreme Allied Commander in accordance with Department’s instructions. We have today been informed by Deputy Chief of Staff on behalf of Supreme Allied Commander that decision to turn over to Tito Yugoslav nationals under reference was made on grounds of military necessity in view of conditions existing at that time. It was stated that Supreme Allied Commander took note of our non-concurrence and pointed out that British Resident Minister had concurred in proposed action but that in any event Supreme Allied Commander took his decision because of conditions existing of which he was better aware than Dept. The communication from Deputy Chief of Staff added that in view of divergent political views expressed to him on subject, by Resident Minister and ourselves, Supreme Allied Commander suspended transfer of dissident troops as soon as emergency conditions ceased to exist. It was set forth that while Supreme Allied Commander of course seeks the advice of his political advisers on all occasions he must reserve unto himself right to decide matters of an urgent military nature as he sees fit. ‘In conversation with Alexander this morning he stated to us that he was obligated to receive surrender of almost 1,000,00 Germans in mid-May and could not deal with anti-Tito Yugoslavs as he would have liked. We stated we had nothing to add to our memorandum under reference except to point out to him again that Resident Minister acted contrary to policy agreed upon after consultation by Department and Foreign Office.

British apologists for mass murder gleefully seized upon this signal to ascribe responsibility to Alexander for the repatriation operations, and so to absolve the Conservative prime Minister Macmillan. Such an interpretation is not only diametrically at variance with the evidence, but is implicitly refuted by the very explanation reported by Kirk. Alexander’s explanation that ‘in view of divergent political views expressed to him on subject, by Resident Minister and ourselves, Supreme Allied Commander suspended transfer of dissident troops as soon as emergency conditions ceased to exist’ can only refer to the Bleiburg crisis on 15 May and the ‘Distone order? of 17 May, which required the evacuation of ‘dissident’ Chetniks to Italy.

It was characteristic of Alexander that he should accept blame for the misdeeds of his colleagues and subordinates. As one of his ablest generals recalled: ‘Anyhow you had a great feeling of trust in him [Alex] as you knew that he would back you whatever happened, and that if things went wrong, he would accept full responsibility for far more than his own share of the blame’.6

In any case Kirk must by this time have acquired a fairly full appreciation of what had occurred in reality, and he made it plain whom he believed to be ultimately responsible for the treachery and slaughter:

‘We stated we had nothing to add to our memorandum under reference except to point out to him again that Resident Minister acted contrary to policy agreed upon after consultation by Department and Foreign Office’.

General Robertson’s order upon which Lord Aldington relies for justification of his part in arranging the repatriation of Croats and others in May 1945, read as follows: ‘All surrendered personnel of established Yugoslav nationality who were serving in German Forces should be disarmed and handed over to Yugoslav forces’. The accepted interpretation of the Geneva Convention is that uniform determines citizenship. If the Croats were regarded as part of the German armed forces, they should have been treated as such and held as prisoners-of-war of the power to which they surrendered, i.e. the British. In fact Aldington made no attempt to determine the citizenship or status of any of the Russian and Yugoslav prisoners in 5 Corps hands, and sent them to be killed indiscriminately. During the 1989 libel trial in London, Lord Aldington and his fellow Chief of Staff at 5 Corps in 1945, Brigadier Tryon-Wilson, defended the former’s classification of civilians accompanying surrendering Croat forces as ‘camp followers’. The claim was designed to legitimise the inclusion of civilians among Croats surrendered to Tito, who would not otherwise have been covered by the orders 5 Corps claimed to have fulfilled. Though this ploy served Aldington’s purpose at the time, in reality it served to aggravate the cynical violation of international law. Article 81 of the 1929 Geneva Convention provides that civilians engaged in this type of relationship with the military ‘have the right to treatment as prisoners-of-war’.7



1 Herbert Butterfield, History and Human Relations (London, 1951), p. 186

2 Cf Appendix

3 Brigadier Anthony Cowgill, Lord Brimelow, and Christopher Booker, The Repatriations from Austria in 1945: The Report of an Inquiry (London, 1990), pp.

4 The authors of the "Cowgill Report" refer to my suggestion 'that Macmillan flew to Klegenfurt 'expressly' to discuss the problem of the Cossacks' as one of a succession of 'unfounded assumptions'. Since Brigadier Tryon-Wilson was himself a member of the 'Cowgill Committee', the authors must have been aware of the validity of my conjecture.

5 The 'Cowgill Committee' (of which Aldington was an unacknowledged member) was clearly embarrassed by this inconsistency, to which its authors provide this curious attempt at an answer: "We do not know how consciously he [Robertson] was seeking political cover for the order he had given in the heat of a grave emergency three days before' (The Repatriations from Austria in 1945, i. p. 75). To which it is sufficient to respond (i) that the text emanated from Alexander, and was merely transmitted by Robertson; (ii) by no possible interpretation can it be interpreted as 'seeking political cover' for an order to which it makes no reference, whose provisions were in direct conflict with those indicated in Alexander's signal.

6 Unpublished memoir of General Sir Oliver Leese.

7 Gustav Rasmussen (ed.), Code des prisonniers de Guerre: Commentaire de la convention du 27 juillet 1929 relative au traitement des prisonniers de guerre (Copenhagen, 1931), p. 130. This clause reflected a provision enshrined in Article 13 of the Hague Convention (ibid., pp. 26-27).