Given the fog of mutual suspicion that surrounds the issue, perhaps the best starting point in the search for a solution to the Kashmir dispute is, as President Musharraf said before and after the Agra summit, to rule out the solutions that both countries cannot live with. The first two solutions that will have to be rejected are the original demands of the two countries, which still remain their official positions. Since the early 1950s, Pakistan has been calling upon India to honour its commitment to hold a plebiscite in the State of Jammu and Kashmir to decide whether it should belong to Pakistan or India. But a moment's reflection should show not only that this is no longer feasible, but that were it so, it would almost certainly result in Pakistan losing Azad Kashmir.

The main problem with a plebiscite today remains what it was in 1948, namely, determining precisely who would be qualified to vote in it. In 1948, as correspondence to be found in the British Library shows, Pakistan was worried that it might lose the plebiscite if a large number of Kashmiri Muslims (especially from Jammu, Poonch and Rajouri) who had crossed, or been driven over, the border into Pakistan were not allowed to vote. By the same token it wanted all the Hindus and Sikhs who, it claimed, had settled in Jammu after partition, to be excluded from the vote. These demands constituted a tall order, but Nehru accepted them, at least initially. However, Karachi's fear was revealing. If in a state with 77 percent Muslims, it was afraid that the absence of a relatively small proportion would make all the difference, then it already knew that a large proportion of the Muslims, i.e. the followers of Sheikh Abdullah, who made up a majority in the valley, would not vote to join Pakistan.

The problem of deciding who could vote has become all but insurmountable today. Three-quarters to four-fifths of the 1947 inhabitants of Jammu and Kashmir are dead. Their progeny have the right to vote but who precisely are they and how will they prove their right to vote if they do not reside in one or other part of the old princely state? How will they do so if they reside abroad? Most have taken citizenship there. Should they even be considered for inclusion in the voters' list? One would think not.

The truth, as the British conceded in private memoranda between the Commonwealth Relations Office in London and the UK delegation to the UN in 1948, is that ascertaining who should be allowed to vote on the basis of origin was difficult even then. It would be almost impossible today. In practice one would have to confine the plebiscite to the present population of the Jammu and Kashmir princely state. That would go even more heavily against Pakistan today than in 1947, because while Jammu has filled up with Hindus and Sikhs, Azad Kashmir has lost much of its original Muslim population.

Some idea of what could happen in Indian Kashmir if there were a plebiscite today was obtained by a MORI poll commissioned by the UK based 'Friends of Kashmir' which is headed by the veteran champion of Kashmiri accession to Pakistan, Lord Avebury. Held in April 2002, it showed that given two choices, India or Pakistan, 61 percent favoured staying in India and only 6 percent favoured joining Pakistan.

The Indian government is, needless to say, aware of implications of the changing population composition as indeed of the MORI poll. But the reason why it will not even consider a plebiscite today is that in a plebiscite Pakistan would have absolutely no option but to appeal to the Kashmiris on the basis of religion. That would inflame communal passions and just probably set off a Hindu backlash in India. No one who has the good of the subcontinent at heart could possibly want to do that.

If a plebiscite is out, then so is the Indian claim that the whole of Jammu and Kashmir should belong to India. It is true that Maharaja Hari Singh acceded to India under the Indian Independence Act. It is also true that he did not really sign the Instrument of Accession under duress, but had been trying to accede to India without much success since the middle of September 1947, i.e. five weeks before the raiders from the NWFP arrived in Kashmir. But Nehru's government knew, although it never conceded it in public, that there had been a fairly spontaneous revolt in the Jhelum valley and other parts of what is now Azad Kashmir, against the Maharajah's decision to accede to India.

This was because the Muslims of Jhelum, Poonch and Rajouri were ethnically entirely different from those of the vale of Kashmir. They were, for the most part, Mirpuris, who spoke Punjabi and had little in common with the Kashmiri speaking people of the Vale. That is one of the reasons why he accepted a cease fire without trying to recover the Jhelum valley or all of Poonch. That is also one of the reasons why Mrs. Indira Gandhi signed the Simla Agreement which conceded Azad Kashmir to Pakistan so long as Pakistan did the same for Jammu and Kashmir, till a final agreement could be hammered out.

Legal and ethnic issues apart, there is another powerful reason why, despite all its advocacy of a plebiscite, Pakistan cannot even contemplate a mode of settlement that could bring the whole of Maharaja Hari Singh's Kashmir into India. For that would make India contiguous with the NWFP. In 1947 this was unthinkable, for Pakistan had won a plebiscite in the NWFP under a property qualification that allowed only 7 percent of the population to vote, by a single percentage point. A common border with India would have made the infant state of Pakistan indefensible from the start.

That was the reason Lt. Gen. Akbar Khan gave in his book, 'Raiders in Kashmir', for sending the Pakhtun tribesmen to Kashmir. Today the Pakhtuns have been more or less fully assimilated into the Pakistani state, but can anyone lay bets that this will never come undone? In particular can anyone be sure of what would happen to Pakhtun sentiment if the U.S. were to ask the Pakistan army to punish the Pakhtun tribes that are sheltering the Taliban in the Tribal Areas, and Gen. Musharraf or a future leader were to comply?

Now that we have eliminated both the official positions, are there any intermediate, more acceptable solutions available? Over the years, both India and Pakistan have mooted one alternative each; India has let it be known in private that it will accept the Line of Control as the international border, with or without some rectifications. Pakistan has often mentioned, although never formally, the possibility of implementing what is known as the Dixon Plan (after Sir Owen Dixon), although it was, in fact, first mooted during the first days of the tribal invasion by V.P. Menon. This is to leave Jammu and Ladakh with India and transfer the valley to Pakistan. A variant of this plan would make the valley a UN trust territory for a specified number of years, to be followed by a plebiscite in the valley alone. The plebiscite would give Kashmiris only two choices -- India or Pakistan.

The LoC as International Border
The Indian proposal has a good deal of international support. Not only has the LoC endured for 55 years, but there are cease fire lines all over the world that have, through the passage of time, become accepted as international boundaries. Other countries have become tired of the Kashmir dispute and grown increasingly apprehensive of its potential to trigger a war on the subcontinent that could turn nuclear. The Simla agreement of 1972, moreover, strongly pointed towards the LoC becoming the basis for a final settlement of the Kashmir 'situation', after the resumption of normal relations between the two countries. If all else failed, India could insist on this particular meaning of the agreement and would probably get abundant support abroad for its interpretation.

In any case, India is much larger than Pakistan, and has much more staying power that it is capable of forcing this 'solution' to the Kashmir dispute down Pakistan's throat. Its army is in control of the state; it seems capable of accepting the level of attrition that the jihadis from Pakistan are capable of inflicting almost indefinitely; the genuine Kashmiri insurgency has been over since 1994 and violence in the state is nourished almost entirely by 'guest militants'. Thermal imaging and other new equipment have sharply raised the kill rate of jihadis crossing from the other side of the Line of Control. Kashmiris are tired of the insurgency and pine for peace. They may, therefore, eventually settle for fair elections and their own governments in Kashmir.

But for India to opt for this 'solution' would be extremely short-sighted. For it would do nothing to resolve the tension between India and Pakistan. As long as this continues, all Pakistan governments will be tempted to keep nourishing jihadis for Kashmir on one border while kowtowing to the U.S. and relentlessly pursuing another kind of jihadi on the other border. It does not need much intuition to see that this policy is not sustainable in the long run. Sooner or later, therefore, Pakistan will get dragged into the vortex of violence that is being created in Afghanistan and the Middle East. Chaos will then be right at India's doorstep.

The Dixon Plan
If the LoC is not an optimal solution even from India's point of view, the Dixon Plan is even more impracticable. Whatever may have been possible in 1947, when V.P.Menon first mooted the idea to the British Acting High commissioner in Delhi, such a solution would be totally unacceptable to India for the same reasons as the original UN proposal. No plebiscite can be held without Pakistan appealing to communal sentiment in Kashmir. The risk that communal passions will not remain confined to Kashmir alone, and could trigger a powerful backlash against Muslims in India is one that no responsible Indian government can take. The fact that India is a democracy, in which the opposition would pounce upon any government that showed an inclination to do so, hardly needs to be underlined.

There are, however, weightier reasons for not conceding such a demand. The original Dixon Plan has been bypassed by time. In 1947 the Kashmiris, along with the subjects of other princely states, were treated as chattel that could be handed over from one ruler to another at will. Today, after eleven elections, (including three that all concede were fair elections, in 1977, 1983 and 2002) and a huge spread of literacy and higher education in the valley, this is no longer possible. Kashmir has a militantly nationalistic intelligentsia, extremely conscious of 'Kashmiriyat' -- its Kashmiri identity -- which will not allow anyone else to decide its fate.

A Plebiscite with Three Options
What is theoretically possible is a plebiscite with three options -- Pakistan, India and independence. We have no clear indication of how people in the valley would vote if they were given three choices, because the only scientific opinion poll, the MORI poll, was so obviously designed to give Pakistan ammunition for use against India in international fora that its questionnaire did not even contain the third option of independence. But if the 29 percent who claimed they were 'undecided' between Pakistan and India can be assumed to have been in favour of independence, and if all of them can be assumed to have been from Kashmir valley -- both tall but not totally unwarranted assumptions -- then it would seem that somewhere around 60-65 percent of Kashmir valley dwellers would opt for independence.

A three-option plebiscite would greatly reduce the risk of a communal backlash in India against Muslims. But the question that will immediately arise is: why should Kashmir valley be singled out for such a plebiscite? Why should the three options not be extended to the whole of the original state of Kashmir. At the very least, why should it not include Azad Kashmir? There is, after all a substantial movement for independence in Azad Kashmir as well.

A long moment's reflection shows that neither Pakistan nor India would be receptive to the idea of a three option plebiscite. In India, no ruling coalition, acting on its own, would be able to endorse such a proposal because it would immediately be accused by the opposition of betraying the nation. Any major constitutional change in Kashmir or any step that could require such a change would have to be endorsed by all the major parties in the country. Such an endorsement would be exceedingly hard to obtain. Not only will the opposition accuse the government of giving away a part of the country, but also endanger the federal structure of the country. Kashmir would become a precedent for every aspiring state leader trying to whip up public support to win a state election. That is a risk that no central government in Delhi will ever take.

Pakistan has demonstrated its aversion to the idea by never once endorsing it during five decades of incessant campaigning to secure Kashmir. Even when it was inciting the leaders of the JKLF to take up arms against India, during talks held in Muzaffarabad, in the late 1980's, it never offered the 'third option'. In February 1992, when the Pakistan security forces fired upon JKLF cadres trying to cross the LoC into Indian Kashmir at Chinari, Prime Minister Nawaz Sharif was sufficiently rattled by the reaction to endorse the third option of independence for Kashmir. The storm that broke out in Pakistan forced him to issue retractions every few hours for the next two days. The plain truth is that for all the strategic and military reasons, Pakistan cannot endorse any solution that makes it run the risk of losing Azad Kashmir.

Is a Condominium Feasible?
But what if Indian and Pakistan agreed to join the valley to all or the larger part of Azad Kashmir and turn it into a kind of condominium. This would involve giving the new state of Kashmir sovereignty in all internal affairs and jointly guaranteeing its defence against external threats. Proposals for limited or constrained sovereignty somewhat along these lines have been proposed by the New York-based Kashmir Study Group headed by Farouq Kathwari. This proposal has the merit of meeting the aspirations of the majority of the Kashmiris while largely avoiding the risk of a communal backlash in India. However, at the present state of India-Pakistan relations it suffers from one fatal flaw.

Any joint guarantee of security or, for that matter, the implementation of any agreement between the two countries to respect Kashmir's sovereignty presupposes a high degree of trust between the two countries. That trust is precisely what does not exist today. In its absence Pakistan will constantly wonder whether every agreement Srinagar signs with New Delhi does not signal a closer integration with India. It will view every seat offered in an Indian medical or engineering college to a Kashmiri, every new road or rail link between India and Kashmir, and every addition to the Indian armed forces in Ladakh or Jammu a part of a larger design to reabsorb Kashmir into India.

India will entertain similar fears about Pakistan. In addition, after 50 years and three attempts by Pakistan to annex Kashmir, a condominium arrangement of the above kind will not really end the suspicion in New Delhi that Pakistan's main purpose is to get the Indian army out of Kashmir so that it can make a fourth attempt on terms far more favourable than those that exist today. Nor will a condominium arrangement necessarily bind organisations like the Lashkar-i-Tayyiba or the Jaish-i-Muhammad. New Delhi will, moreover, view the arrival in Kashmir of every Maulvi preaching orthodox Deobandi, Wahhaby or Ahl-e-Hadith variants of Islam, as preparation for fomenting a rebellion on religious grounds by discrediting Rishi Islam and thereby undermining one of the main pillars of Kashmiriyat. Indeed many of the key leaders of Kashmiri nationalism today regard the 1972 decision by a Congress government in Kashmir to lift the ban on the Jamaat-i-Islami as the first and most grievous blow against Kashmiriyat, and it was dealt not by Pakistan but by India.

Such suspicions will make it difficult for India not to keep troops in a constant state of readiness for re-entering Kashmir. There will, in short, be no peace and no let-up of tension. In many ways, therefore, we will all find ourselves back where we were in 1947.

Towards a Lasting Solution
The above discussion brings us at last to what, in this writer's opinion, is the key requirement of a lasting solution to the Kashmir problem. This is the building up of trust between the two countries. Without trust no solution is likely to endure. But trust too cannot be built up in a day. We, therefore, need to establish a process - a phased succession of actions by both countries - whose successful implementation will build up trust step by step, even as it leads towards a final solution. What is more, this final solution cannot be one that is acceptable only to India and Pakistan. It must be acceptable to the people of Kashmir too. It must, in short, be one in which all parties give up some of their aspirations in exchange for the benefits of lasting peace, but from which none emerges as a loser.

Both Delhi and Islamabad, in their saner moments acknowledge that what is needed is to start the process and go step by step, without jumping to its end point. But even this movement gets bogged down before it starts because each fears that the other will try to push the process itself in a direction the other does not want to go -- India towards acceptance of the LoC, and Pakistan towards some version of the Dixon Plan. That is why, watching the numerous abortive attempts to start the dialogue since the mid- 1990s, I have come to the conclusion that the process cannot be divorced completely from the end.

Both countries need to have an idea of the kind of solution that they could accept and sell to their own people, even before the process of détente begins. This need not be a precise idea. Indeed there would be virtue in imprecision because it would enlarge the room for negotiation. But if there is not some narrowing of the gap that exists at present between even the 'informal' positions held by Delhi and Islamabad, it will be difficult even for discussions on process not to get bogged down.

Let me, therefore, take the bull by the horns and suggest the broad outlines of what a final settlement of the Kashmir dispute could look like, and the process by which India and Pakistan could arrive at it. To do so we do not have to invent something new. History has enough examples of how similar problems have been resolved elsewhere.

Learning from Tyrol
The model I have in mind is that of Tyrol, a German speaking trans-Alpine region, which falls partly in Austria and partly in Italy. Southern Tyrol came into Italy after the fall of the Austro-Hungarian Empire in 1918, and enjoyed a troubled relationship with it for the major part of the twentieth century. All that is a part of the past. The reason is that, well before the formation of the European Union, Italy and Austria had learned how to turn the existence of Tyrol between them for a big minus into a big plus. Today Southern Tyrol enjoys a considerable measure of autonomy within Italy, with its own elections and its own laws over a wide range of subjects.

It enjoys immense tax advantages in comparison to the rest of Italy, which make it a magnet for industry. It has complete freedom of communications with northern Tyrol, Austria and the rest of Europe. People from the rest of Europe are free to visit, to trade with, invest and live in the Tyrol. Tyroleans, needless to say, enjoy unhindered access to both Italy and Austria and now, of course, the whole of Europe. What Italy and Austria have done is to reduce the significance of the international border between the two parts to the point where it no longer impinges upon the lives of the people of Tyrol, Austria and Italy.

Yet, Southern Tyrol remains a part of Italy and northern Tyrol of Austria. The foreign relations of the two parts remain securely in the hands of the two countries. They vote separately for the European parliament. Presumably their defence, insofar as it is not subsumed in the defence arrangements of the European Union and NATO, would also be the separate responsibility of the two countries. Tyrol is, therefore, a sort of condominium, in which the areas of responsibility for defence and foreign affairs of the region have been demarcated by history instead of by explicit agreement.

The key element in the success of the arrangement between Austria and Italy is trust. Admittedly, it has taken the better part of a half century to build. But India and Pakistan can do it in a much shorter period of time. Today the defining event that created the two Kashmirs is already half a century behind us. In the intervening years both countries have learned, with the Kashmiris, that there is no military solution to the problem.

On paper at least, both the Kashmirs enjoy immense autonomy. Indian Kashmir has a separate constitution and the power to redefine its relations with India. It has used this power to increase integration in the past but can use it to reduce it as well if it wants to. Pakistan has throughout maintained the fiction that Azad Kashmir is an independent nation, although this has been even more of a fiction than Kashmir's autonomy. An essential stage in the step by step resolution of the Kashmir problem would be to clothe this skeletal autonomy with real flesh.

India and Pakistan can then allow the elected representatives of the two Kashmirs to decide whether or not they want to set up a common consultative council and the subjects on which they will coordinate their policies. High on their list, of course, will come the opening of the LoC between the two parts at various crossing points for travel and trade between the two Kashmirs. It will then be up to India and Pakistan to work out how this will be confined only to Kashmiris.

But why should it remain confined to the Kashmiris? Why should Pakistanis not be able to visit Indian Kashmir and Indians not be allowed to visit Azad Kashmir and the northern areas. The people of these regions can only gain from the increase in tourism that will result. Greater contact will also break down stereotypes and build friendships.

The next obvious step would be to extend freedom of trade and transit between the two Kashmirs to freedom of trade and transit for Kashmiris from each part of Kashmir with the whole of the other country. In practice this would be difficult to keep separate from a more general move towards full trading and consular relations between India and Pakistan. But that is something that civil society in Pakistan is already asking for with increasing assertiveness, for the simple reason that its benefits for Pakistan far outweigh its benefits for India.

The process outlined above will take several years to work out because it is immensely complex. But it is worth remembering that the European Union also came into being a full thirty years after the creation of the European Common Market. What we need to remember is that in the building up of trust, the first steps are the most difficult. If India and Pakistan agree, however tacitly, that the above road is broadly the only one that they can follow, then each step that they go down it will make the next step easier, and the process itself faster. In the end the benefits of a simultaneous resolution of the Kashmir dispute and the coming into being of a SAARC free market, will so far outweigh the costs that people will be left wondering why it all took so long.

(Prem Shankar Jha has written extensively on the Kashmir issue. He was the editor of the Hindustan Times and author of Kashmir 1947- The Origins of a Dispute.)

References

• Prem Shankar Jha, Kashmir 1947 - The Origins of a Dispute, (Delhi: Oxford University Press, London: Pluto Press, 2003).