Learn about the beta
Office of the Historian

180. Memorandum From the Executive Secretary of the Department of State (Eliot) to the President's Assistant for National Security Affairs12

[Page 1]



  • India-Pakistan: Refugee Problems

The attached contingency study on refugee and resettlement problems in the context of the Indo-Pakistan fighting is forwarded. in response to the request made at the December 6 WASAG meeting. Except where indicated otherwise, the planning assumption is complete Indian/ Bangla Desh control of East Pakistan,

Theodore L. Eliot, Jr. Executive Secretary


[Page 2]


The actual and potential refugee problem in South Asia must be viewed from various points —. humanitarian, political, economic and managerial. This paper assumes the overriding importance of the humanitarian aspect; it does not need to be greatly elaborated. The economic and managerial aspects are technical questions that we raise but cannot answer fully; we lack too much information and our view of the future is too cloudy. Some of these are being addressed in more detail by other papers under preparation in response to the WASAG request. We attempt, however, to flag the most important kinds of problems that we will face.

The political aspect presents problems about which we can make better guesses and which are perhaps more amenable to immediate action. Thus the principal thrust of the paper is in the political area.

There are four classes of refugees/ displaced persons who will probably require assistance on an international basis following the termination of hostilities.

[Page 3]

1. Up to 10 million refugees from East Pakistan who are now in India.

2. Over 600,000 non-Bengali Muslims in East Pakistan who may want to move to West Pakistan, as well as West Pakistani government officials and military in the East.

3. Some 50-100,000 Bengalis living in West Pakistan, most of whom will almost certainly wish to return to East Pakistan.

4. An undetermined number of persons who are being displaced by the fighting within East Pakistan.

The Refugees in India

Nearly ten million* refugees fled from East Pakistan to India in the period since last March. Of these, nearly 3 million are outside refugee camps; an undetermined number of them have merged into the Indian economy and society and may never be heard of again. Of the remainder, who are in camps, about two million are Muslims, the rest Hindus.

The Bengali refugees are both an effect and a cause of the present situation. India found the burden of refugees economically and politically intolerable and has made *These are Indian figures. Pakistan claims only two million. [Page 4]their return to Pakistan a sine qua non of a settlement. Yet even with a settlement, the problem will persist on two levels:

— repatriating and rehabilitating the returnees, and

— dealing with refugees who do not want to or are not allowed to return.

Some of the refugees have reportedly already begun to return. We assume that nearly all the approximately two million Muslims will return to Bangla Desh. Resettlement will cause some problems but should be of manageable proportions.

A number of the Hindu refugees will also want to return — we cannot estimate the proportion. Even assuming that the Bangla Desh government is willing to accept them, they will pose more difficult resettlement problems than will the Muslims. They are less likely to have friends -or family to receive them (whole villages have tended to flee en masse).

[Page 5]

Many Hindus, however, will not want to return — this will be particularly true of those with skills and education but also for many at the bottom of the social scale. This will pose a major problem for the Indian government. New Delhi wants desperately to get rid of these refugees and has built much of its international position on this. At the same time, however, the GOI may be loath to repatriate them forcefully for both domestic and international political reasons. The Bangla Desh government will presumably welcome returning Muslims. They may be less enthusiastic about the Hindus. Should the Bangla Desh government resist full repatriation, New Delhi will doubtless feel compelled to bring strong pressure to bear.

The problem may not be as bleak as suggested above. Much will depend on the circumstances in East Pakistan when fighting stops and the incentives to return that can be offered. A Bangla Desh spokesman has said that his government will welcome all returnees. Should Bangla Desh prove reluctant, India has various levers it can use ranging from economic negotiations to foot—dragging in [Page 6]withdrawing its troops. Still, we must bear in mind that the return of the refugees could create major problems of international concern.

Dealing with those Hindus who remain in India and are accepted by the Indian government will be primarily an economic problem but also one with political import. . Much depends on the numbers involved. If only a few hundred thousand have to be absorbed, the problem will not be great. Should, however, India be faced with many millions of permanent refugees the adjustment will be hard. Keeping these people in camps is distasteful to all concerned but integration into the Indian economy and society will be most difficult. Many refugees have no marketable skill, land is not available, settlement. outside of Bengali-speaking areas of India will be undesirable on cultural grounds, and there will be strong resistance from the Indian public.

[Page 7]

Lacking any idea of the dimensions of the problem we are not able to foresee how it can be dealt with. If the numbers involved are large, India will convincingly argue that it should not hear the entire cost of resettlement. The international community will want to avoid another Palestine-type situation. The role that we will want or be able to play in helping will be affected by the state of our relations with India if and when the problem becomes acute. Sympathetic concern with India's problems could, however, be of some help in rebuilding our relationship with India.

In summary, then, problems can arise:

(1) Resettling those refugees who return and are accepted; this is manageable but will obviously increase in difficulty and cost as numbers increase.

(2) Persuading unwilling Hindus to return to East. Bengal. This will cause problems for India and could be of international concern, since the Indians may be fairly forceful in "persuading" the refugees to return.

[Page 8]

(3) Persuading the Bangla Desh government to accept the Hindu refugees. Although this will be primarily a problem between Dacca and New Delhi, it, too, may have international ramifications.

(4) Resettling such refugees as remain in India. This will become primarily an Indian responsibility, but the international community will be expected to help. The Biharis

The Bihari community has never been completely at ease in East Pakistan and has tended to identify with the West Pakistani government. In the recent unrest, a number of Biharis have sided with Islamabad and participated in actions against the Bengali population. Even if this participation was in fact small, many Bengalis are no doubt thirsting for revenge (and have often loosed violence against Biharis when the opportunity arose.)

Much of the entrepreneurial skill in East Pakistan is in the non-Bengali community. Bangla Desh will need these people's skills but many of them will not want to *Technically, "Bihar's" are non-Bengali Muslims who fled to East Pakistan from present—day India (especially the state of Bihar) following partition in 1947. There are an estimated 600,000 of these in East Bengal now. The term is also used in a broader sense to include all non-Bengali Muslims in East Bengal —- an additional 1-1.5 million. Of these a fairly small number has close ties to West Pakistan. The remainder are reasonablv well intetergrated into East Bengal. [Page 9]stay. We cannot at this time estimate with confidence whether an anti-Bihari program will be launched in West Bengal, nor how many Biharis and other non-Bengalis would want to leave. Thus far, Biharis in areas conquered by Indian/ Muslim forces have reportedly fared well. In prudence, however, we must plan for a situation in which bloodshed could be widespread and several hundred thousand people would have to be evacuated quickly and under pressure.

Aside from the difficulty of finding adequate shipping, it may be extremely difficult to get potential evacuees to ports and into ships. Both transportation and law and order in the countryside may be chaotic and make it difficult to get evacuees to assembly points (although we understand that the Biharis and others likely to leave are generally located in the towns and cities). The withdrawing units of the Pakistani army may not be able to protect them enroute or even once they are gathered at embarkation points. Even if the Indian Army elements in East Pakistan cooperate in protecting the Biharis, as they have said they will, their capability to do so will be limited. Furthermore, the port facilities may be so badly damaged that [Page 10]expeditious loading will be impossible. A final problem would arise if the evacuation were to take place while the Indians still maintained their blockade of West Pakistan and refused to allow the refugee ships to enter Karachi.

It is not at all certain where the 600,000 Biharis can go. They would not be welcome in India (where they lived before 1947) and their welcome in West Pakistan would be questionable. Given these facts in addition to the tremendous logistic problem involved in moving the Biharis, it is clearly desirable that they remain in East Bengal if that is at all possible. Calming anti-Bihari sentiment in Bengal and providing assurances that will be credible to the Biharis will be extremely difficult but must be attempted. In this regard, even while planning for the worst contingencies and keeping-the paramount humani- tarian concern at the fore, we will want to avoid creating either in South Asia or at the UN an atmosphere that will stimulate Bihari expectations of evacuation or international expectations that this is inevitable.

[Page 11]

The problems arising are these:

(1) Reducing the chances of a mass slaughter of Biharis, thus reducing the need for immediate evacuation.

(2) If hasty evacuation is required, protecting the lives of those awaiting evacuation and moving them to West Pakistan as quickly as possible. In practical terms this will mostly have to be done by sea; the numbers will be too large for air evacuation and movement by land across India is problematical. Special problems may arise because of damage to or inaccessibility of port facilities.

(A Related Problem: The Pakistani Army and Officials in East Bengal)

Some 70,000 West Pakistani military were in East Pakistan as of early December. In addition there were some thousands of West Pakistani officials. Many of both categories had families with them in the East. These will all clearly have to be evacuated and will present somewhat similar problems to that posed by a Bihari evacuation. These would be better organized, however, and more [Page 12]centrally located, assuming that the army is able to fall back in an orderly manner to a port area. Also, the Pakistan government will exert greater effort to extricate them. Above all, the army will be better able to establish a perimeter and defend. itself while awaiting withdrawal, if not protected by a formal cease-fire. Evacuation will be a problem, however, and. the Pakistani Government might make a special effort to have the US assist.

Bengalis in West Pakistan

The bulk of the 60,000 Bengalis in West Pakistan will probably want to return to Bang la Desh. Most of them have roots in the East; will have families there to receive them; and for the most part have needed skills. Most will probably not be welcome in West Pakistan. Significant difficulties may arise, however, in the event of anti-Bengali bitterness among the West Pakistan population that could lead to killings of Bengalis. Also, [Page 13]the Islamabad government might see the Bengalis as hostages for the safety of its forces and supporters in the East.

The problems are thus:

(1) Ensuring the safety of the Bengalis in West

(2) Transporting them back to Bangla Desh; this could be done in part by sea and in part by land across India, which will regard the Bengalis benevolently (in contrast to the Biharis).

Displaced Persons in the East We have no way of knowing how many people will be displaced from their homes by the fighting and will need to be resettled. In general, though, we expect that these '. battle refugees will not have fled far from their homes and will return on their own when the fighting passes them by. There will certainly be major requirements for rehabilitation because of destruction caused by the fighting. There is no significant reabsorption or [Page 14]political problem for this group of people, but the economic costs could be substantial.

The problem therefore is

(1) Getting the people back to their homes where necessary and rehabilitating them —- which is part of the large rehabilitation problem that the new government will face throughout its country.

The Humanitarian Problem During the Fighting

There is an additional related problem — the protection of civilian lives and prisoners of war during the fighting.

Innocent people inevitably get killed during the fighting; there are poorly. disciplined armed bands, on both sides of the conflict that have been guilty of atrocities; and soldiers on both sides have real fears of reprisal if they are taken prisoner.

There is thus a need to

(1) Limit: unnecessary killing by all parties concerned by dampening the atmosphere of hatred and revenge even while fighting is going on, and

[Page 15]

(2) Encourage adherence by all parties to the Geneva Convention.


The United States has several interests in the refugee problem:

— Prevention of loss of life and facilitation of repatriation and resettlement.

— Using its assistance in dealing with the refugee question to further its basic goals of (a) reestablishing peace and security in South Asia and (b) building good relations with the governments involved — India, Pakistan and Bangla Desh.

— Preventing the institutionalization of a South Asia refugee problem on the Palestinian model.

— Not becoming saddled with a disproportionate share of the expense and effort involved in moving and resettling refugees.

[Page 16]


The refugee problem will have to be dealt with at three levels:

— by the governments directly involved — India, Pakistan and Bangla Desh;

— by the international community; primarily the. UN but also perhaps the Red Cross and international voluntary agencies;

— by the United States and other countries unilaterally.

We regard this list: as one of descending responsibility; i.e., the governments involved should bear the bulk of the responsibility; only the residual should be handled by the internationil community; and a still much smaller residual unilaterally, with special concern for delimiting our own participation.

Nonetheless, the US has certain unique capabilities that will have to be brought into play directly; there may be specific political gains that we can make through [Page 17]direct involvement; and our history of entanglement in South Asia means that we cannot readily opt out of direct action.

What Needs to be Done

(1) Getting Information

— We lack a clear picture of the magnitude and nature of the problem with regard to the Biharis in East Bengal and Bengalis in West Pakistan. We do not know for certain how many may have to. be moved, where they are, or how great the danger to them is. We have been in contact with the field on these points.

— We do not know how many of the refugees in India will want to return. We are also not certain what the attitudes of the Indian government will be towards unwilling returnees and of the Bangla Desh government towards Hindu returnees. We will probably only be able to gain such information with the passage of time.

— We have no good picture of how great the rehabilitation needs will be of all displaced persons [Page 18]and refugees. This is part of the overall rehabilitation of East Pakistan and cannot be determined until the shooting stops.

(2) Political/ Psychological Actions During the Fighting.

— All parties concerned should be encouraged to ensure adherence to the Geneva Conventions. India has already pledged this.

— Restraint needs to be placed on paramilitary forces operating on both sides in East Pakistan. We should urge New Delhi, Islamabad, and the Bangla Desh government to do this: we must recognize, however, the limitations in this regard of the Pakistan and Bangla Desh governments which lack full control of their partisans in East Pakistan.

— The climate of hatred and revenge must be defused. The GOP should be urged to eschew inti-Bengali propaganda to lessen the threat to Bengalis in the West. The Bangla Desh government should follow a propaganda line that stresses the brotherhood of all living in East Pakistan, whether Bengali or Bihari, Hindu or Muslim. The Indian government should support this line.

[Page 19]

— Our ability to bring these objectives to reality is quite limited. International opinion should be mobilized to lend what support it can. We should encourage special bilateral approaches to India and Pakistan by France, China, the UK, the USSR as appropriate. Prominent world leaders with humanitarian credentials (the Pope, the head of the Red Cross) as well as the Secretary General and UNHCR should be urged to speak out on these matters. Where necessary, we could brief the individuals concerned on the dimensions of the problem. (We have already been in contact with the UNHCR who shares our concerns over the post-hostilities situation. The UNHCR discussed this matter with the SYG, who made a humanitarian appeal for the safety of civilian population on December 7.)

(3) Contingency Planning

— We should inventory the availability of food grains should there be a need for famine relief, although at present it appears that neither availability nor [Page 20]distribution will be a critical problem. On a short term basis, food is available in East Bengal, as well as India and other nearby states. If required, we could undertake to replace stocks sent by these countries to Bangla Desh. Over the longer term we will probably be called upon to supply food direct. High-protein food and medicine could be a more critical short-term requirement. We will need to keep our estimates of requirements and availability in all categories especially current during the critical months that will follow the fighting.

— We should plan for a massive sealift of population between East and West Pakistan, possibly on a crash basis, under international auspices. Sea transport will be the crucial element; US assets, military and civilian, would almost certainly be required, preferably under an international umbrella.

— We should urge the UNHCR to develop and publicize resettlement plans for the refugees in India that would encourage the return of a maximum number of [Page 21]refugees, lest India and the world be confronted with the problem of caring for a permanently unassimilated group of refugees.

— We should encourage appropriate UN organs to make a study of the extent of assistance that will be required for refugees once the fighting stops. We can be of assistance to the UN in the preparation of such a study.

Other Actions

— We should point out to the Indians, Pakistanis and Bengalis that our ability to assist them in the after-math of the war will be conditioned. in part by their per-formance in averting and dealing effectively with humanitarian problems. We, for our part, should look for ways of using assistance to further humanitarian ends.

— We should discuss problems of refugee movement and evacuation with the Indians and urge them to exercise restraint in-destroying port facilities and other infra-structure that may be needed for rapid movement. We should also urge them to plan for and be cooperative in facilitating evacuation of Pakistani forces and, if necessary, the Biharis.

[Page 22]

— We could point out to the GOP that there is only one Bengali leader who might be able to stem an outbreak of violent anti-Bihari sentiment, and that is Mujib. The Pakistanis should take this into consideration in deter-mining when or whether to allow Mujib to return to the East.

— We should informally raise with the Indians and, if possible, the Bangla Deshis our hope that they will hold to a minimum or preferably forego "war crimes" proceedings. Crimes were committed on both sides and a healthy tempering of justice with mercy will be the most states-manlike approach to bringing stability to South Asia,

[Page 23]

— Our international posture on the entire South. Asia problem should focus on humanitarian rather than political concerns. We are stressing this point in our pronouncements at the UN.

— Stress broad international responsibility for the post-hostilities refugee rehabilitation problem: begin to organize post-hostilities planning on an international basis. The India and Pakistan consortia should. be called together for a planning session.

UN Problems

— We will want to consider very carefully the extent and manner that the UN should become involved in the refugee problem. Clearly we want an international umbrella and the UN is the logical choice. On the other hand, a massive repatriation and rehabilitation program could involve the UN in something at least as large as the Congo operation and the lion's share of the financial burden could fall on us.

— The UN should make it clear to the governments of India, Pakistan and Bangla Desh that UN resources —financial and otherwise — are limited. It should [Page 24]attempt to work out with the parties concerned arrangements that would restrict the UN financial role and direct it towards supervisory, technical and. observer responsibilities. The matter of cost would for the most part have to be settled between the three governments on the one hand and individual or multilateral donors on the other.

— At present the role of the UNHCR is limited to that of the "focal point" for assistance to the refugees in India from and through the UN system. The UN East Pakistan Relief Operation (UNEPRO) is in East Pakistan at the request of the GOP to render emergency assistance to the victims of last year's cyclone and this year's civil strife. It is virtually unable to function at present. UNHCR and UNEPRO could perhaps cope with refugee matters if all went well. Any massive programs, however, would require a vote of the General Assembly to enlarge the organizations' mandates as well as the consent of the governing authorities in control of East Bengal. UNEPRO's capabilities in a truce observer or similar mission would be limited unless qualified personnel were assigned. Any such mission would require a mandate from the Security Council.

  1. Source: National Archives, RG 59, Central Files 1970–73, POL INDIA–PAK. Secret; Nodis. Drafted on December 10 by Thomas P. Thornton (S/PC). Thornton also apparently drafted the attached study. Cleared in S/P by Deputy Director Joseph Neubert and Cargo; in NEA by VAN HOLLEN, Quainton, and Constable; and in S/R, IO/UNP, INR/RNA, and AID/NESA.
  2. The contingency paper, prepared for the Washington Special Actions Group, examined the refugee problem created by the conflict in East Pakistan from humanitarian, political, economic and managerial perspectives, but assumed the "overriding importance" of humanitarian considerations.