New Times,
New Thinking.

  1. World
  2. Europe
26 January 2016updated 09 Sep 2021 1:57pm

Ending the new Thirty Years war

Why the real history of the Peace of Westphalia in 17th-century Europe offers a model for bringing stability to the Middle East.

By Patrick Milton

A man hangs upside down in a fire. Others are stabbed to death or tortured; their womenfolk offer valuables to save their lives – or try to flee. Elsewhere, women are assaulted and violated. In another image the branches of a tree are weighed down with hanging bodies, and a religious symbol is proffered to a victim as the last thing he will see on Earth. The caption describes the hanged men as “unhappy fruit”.

This could be Syria today: but it is Europe, in the mid-17th century, at the height of the Thirty Years War. The artist who recorded these horrors was Jacques Callot, who saw the French army invade and occupy Lorraine in 1633. He was perhaps the closest thing his time had to a photojournalist.

The Thirty Years War, within which the occupation of Lorraine was just a short episode, has been cited as a parallel in new discussions of the Middle East by a range of foreign policy practitioners, including Henry Kissinger and the president of the US Council on Foreign Relations, Richard Haass, academics such as Martin van Creveld and journalists such as Andreas Whittam Smith. Like the original Thirty Years War, which was in fact a series of separate but interconnected struggles, recent conflict in the Middle East has included fighting in Israel, the occupied territories and Lebanon, the long and bloody Iran-Iraq War, the two Gulf wars, and now civil wars in Iraq and Syria. As with the Thirty Years War, events in Iraq and Syria have been marked by sectarian conflict and intervention by peripheral states (and still more distant countries) fighting proxy wars. Both the Thirty Years War and the present Middle Eastern conflicts have been hugely costly in human life. The Peace of Westphalia that ended the Thirty Years War in 1648 has also featured in comment of late, usually along with the observation that recent events have brought about the collapse, at least in parts of the Middle East, of ideas of state sovereignty that supposedly originated with Westphalia.

Yet that is a myth, a serious and perhaps fatal misunderstanding of the Westphalian treaties. The provisions of the treaties in fact set up a structure for the legal settlement of disputes both within and beyond the German statelets that had been the focus of the conflict, and for the intervention of guarantor powers outside Germany to uphold the peace settlement. And, as we shall see, the real history of Westphalia has much to tell us in the present about the resolution and prevention of complex conflicts.

 

***

 

Give a gift subscription to the New Statesman this Christmas from just £49

Germany is the prosperous heart of the continent today, but in the early 17th century the “Holy Roman Empire of the German Nation” was the disaster zone of Europe. It was politically fragmented, with the various princes, bishops, towns and the emperor himself all vying for influence, greatly complicated by religious differences between Roman Catholics and followers of various forms of Protestantism. The empire lay at the centre of Europe and was thus the point at which the great-power interests of nearly all the main protagonists in the international system intersected: the French, the Habsburgs, the Swedes, the Ottomans and even the English regarded the area as vital to their security. So Germany both invited intervention by its neighbours and spewed out instability into Europe when the empire erupted in a religious war in 1618 that lasted three decades.

Domestically, the root of the Thirty Years War, just as with many Middle Eastern ­conflicts today, lay in religious intolerance. The security of subjects governed by rulers of the opposing religious camp was often at risk of their governments’ attempts to enforce doctrinal uniformity. With the creation of cross-border confessional communities, as well as antagonisms both within and between the territorial states, rulers became increasingly willing to intervene on behalf of co-religionist subjects of other princes – another parallel with the contemporary Middle East.

Initial attempts to solve these problems failed. After a series of wars following the Reformation, a religious peace was ­concluded at the imperial Diet of Augsburg in 1555. This was a milestone in the devel­opment of confessional cohabitation, because it embodied, for the first time, a recognition of the importance of creating a legal-political framework to manage religious coexistence. Although the treaty helped foster peace for many years, it was nevertheless deficient. First, the princes granted each other toleration only between themselves, not among subjects within their territories. The “Right of Reformation”, or ius reformandi, gave princes the power to impose their confession on their subjects: a form of religious compulsion later encapsulated in the phrase cuius regio, eius religio (“the religion of the prince is the religion of the territory”).

Rulers became increasingly willing to intervene on behalf of co-religionist subjects of other princes

This was a state-centred solution; it ignored the concerns of the princes’ subjects apart from guaranteeing their right to emigrate. Partly designed to undercut interventionist impulses by consigning confessional affairs to an inviolable domestic sphere, the treaty text stated: “No Estate [territory] should protect and shield another Estate or its subjects against their government in any way.” Second, the state-centred settlement was increasingly unsatisfactory for most Protestant states, as it had inbuilt structural advantages for the Catholic side. Calvinism was not recognised and remained officially a heresy. Furthermore, the Catholic princes began to rely on majority voting to sideline Protestants at decision-making assemblies such as the Reichstag or Diet, which in effect
was the German parliament. And the Catholic Church embarked on a major evangelising effort to reverse the effects of the Protestant Reformation through popular preaching – the Counter-Reformation, a prime mover for which was the Jesuit order. Taken together, these factors left Protestants feeling increasingly under pressure, and more radical Protestants were constantly trying to revise the settlement. The formation of hostile princely religious alliances – the Protestant Union in 1608 and the Catholic League in 1609 – was symptomatic of the general “war in sight” atmosphere characterising central Europe at the turn of the 17th century.

The resulting war was, just like the current Middle Eastern conflict, a set of interlocking political-religious struggles at local and regional levels. These provoked and enabled extensive external interference, which in turn exacerbated and prolonged the conflict. Non-state and sub-state actors played important roles in that epoch as they do now: corporate groupings of noble subjects (estates) and private military entrepreneurs; terrorist groups and aid organisations. The war began as an insurrection of the Bohemian nobility against their Habsburg rulers, and soon escalated into a much broader confessional conflict within the empire. But it also became a struggle between competing visions of the future political order in central Europe – a centralised imperial monarchy against a more federally organised, princely and estates-based constitution – which in turn folded into the long-standing Habsburg-Bourbon struggle for European supremacy.

The war was immensely destructive: arguably the greatest trauma in German history. It resulted in an overall loss of about 40 per cent of the population, which dropped from roughly 20 million to 12 million. The war was not merely quantitatively, but qualitatively, extreme. Such atrocities as the massacre and burning of Magdeburg in 1631, which killed over 20,000 people, resonate in the German popular imagination to this day. The war also caused its own refugee crisis. Cities such as Ulm hosted huge numbers relative to their pre-war population – 8,000 refugees taken in by 15,000 inhabitants in 1634, a situation comparable to the one faced by Lebanon today, where one in four people is a Syrian refugee. The resulting shifts in the religious balance often sparked unrest in previously quiet areas, a phenomenon we are beginning to see in the Middle East as well. In those days no one had come up with the concept of toxic stress – but the trauma was no less for that.

 

***

 

Eventually, the war between the Holy Roman emperor, the princes, Sweden, France and their respective allies was brought to an end by the now-famous Treaties of Münster and Osnabrück (collectively known as the Peace of Westphalia). In roughly the past century and a half, however, their nature and implications have been completely misunderstood. The misconception – still frequently repeated in many textbooks, in the media, by politicians, and in standard works on international relations – maintains that the Peace, by granting the princes sovereignty, inaugurated a modern “Westphalian system” based on states’ sovereign equality, the balance of power and non-intervention in domestic affairs. This fallacious notion of Westphalia was later picked up uncritically by political scientists, scholars of international law and historians, leading to the remarkably persistent and widespread Westphalian myth.

The real Westphalia was something quite different. Although the Right of Reformation was officially confirmed, it was in effect nullified by the imposition of the “normative year”. This fixed control of the churches, the right of public worship, and the confessional status of each territory to the state it had been in, on 1 January 1624. This was an innovative compromise arrangement that set a mutually acceptable official benchmark for faith at a point in time at which neither side had gained supremacy. By establishing a standard applicable to all, it also represented a convenient means of avoiding the conflicts of honour inherent in early-modern negotiations in which princes were asked to make concessions.

The practical outcome was that a princely conversion could no longer determine the religious affiliation of the subject population in question. The imperial judicial tribunals retained extensive authority to enforce the confessional and property rights of princes’ subjects (many of which were stipulated at Westphalia). The external guarantors, France and Sweden, were granted a right to intervene against either the emperor or the princes, in order to uphold Westphalian rights and terms. So, this “true Westphalia” is better characterised as an order of conditional sovereignty.

Princes were entitled to rule for life, but crucially were required to respect their subjects’ basic rights, such as religious freedom (including that of Calvinists), enjoyment of property and access to judicial recourse, while also respecting the rights of fellow rulers. If they failed in their duties towards their subjects or the empire they could in theory and practice become targets for intervention, which in some cases entailed deposition from power.

That central Europe avoided another religious war after 1648 shows the success of Westphalia’s conflict regulation mechanisms. At a time of renewed religious dispute in the early 18th century, a statement issued by the Protestant party at the imperial Diet commented on the improvements that Westphalia had brought to the imperial constitution, stating: “The refusal of Territorial rulers to accept that other fellow states protect foreign inhabitants and subjects was one of the greatest causes which led to the wretched Thirty Years War. It is precisely this wound which has been healed by the Peace of Westphalia.”

Westphalia was thus seen as a corrective measure, opening up domestic affairs to mutual and reciprocal scrutiny, on the basis of clear principles agreed by all. It provided an effective system for the “juridification” of conflict, whereby confessional strife (which certainly continued) was channelled into a legal-diplomatic framework and defused through litigation and negotiation, if necessary with the threat of external intervention by a guarantor power, rather than being settled by warfare.

 

***

 

Where in 17th-century Europe Protestants were alarmed by the revanchism of the Catholic Counter-Reformation, through which the emperor (with the support of his Spanish Habsburg cousins) sought to restitute property and lands confiscated from the Catholic prince-bishoprics by Protestant princes during the previous century, so in the Middle East today Shia communities feel under pressure from the new wave of aggressive Wahhabi/Salafi jihadism which similarly regards their faith as heresy and abomination. Or, if you choose to accept the Saudi or Wahhabi version, you could regard Iran and the Shias as the threatening hegemon. One way or the other, both Iran and Saudi Arabia feel insecure in the region, menaced by enemies, to a degree paranoid and liable to miscalculate the true nature of the threat to them and their faiths.

Moreover, the position can change. After the Swedish intervention in Germany in 1630, the Catholics, previously triumphant, were thrown on the defensive and their worst nightmares began to come true. For an eventual settlement to become possible, it was necessary for disillusionment with religious aggrandisement to set in. That might still seem to be some way off in Syria and Iraq now; yet perhaps not so far off. At an earlier stage some Sunnis at least, in Iraq and elsewhere, became disillusioned with al-Qaeda when it was seen to be able to offer no more than continuing violence, with no prospect of any kind of victory. It will be necessary first to defeat Da’esh, or Islamic State, but disillusionment with it could set in quite quickly when its millenarian project is seen to suffer severe setbacks. It will nonetheless be necessary to deal with the Wahhabi origins of the jihadi problem, in Saudi Arabia, as Michael Axworthy argued in his New Statesman article of 27 November 2015.

It would be highly desirable as part of a wider Westphalia-style settlement also to make progress towards a solution of the conflict between Israel and the Palestinians. Yet such a settlement should not be seen as necessarily dependent on that. The Israel/Palestinian question is not an important factor in the present situation in Syria or Iraq, nor has it been among the prime concerns of al-Qaeda or Islamic State, which have both been much more focused on toppling Arab states in the Middle East.

Another aspect of the conflict in the Middle East is that both Iran and Saudi Arabia see themselves as the legitimate leader of the community of Islam as a whole. Just as Christendom was pulled apart by religious conflict in the 17th century, yet Catholicism and Protestantism were still horribly bound together, like cats in a sack, by a shared history and shared faith, so too with contemporary Islam. The traditional territory of Islam is still, in some sense, a coherent whole in the minds of Muslims. In a way reminiscent of that in which the Holy Roman emperor’s authority was still recognised by the Protestant states of the empire, albeit reluctantly and with bitter resentment, so Shia Muslims have to accept Saudi Arabia’s de facto guardianship of the holy places of Medina and Mecca. A settlement in the Middle East could take strength from the lingering sense of a common heritage in the region.

 

***

 

The creation of Iraq, Syria and Lebanon as sovereign states after the First World War owes something to the European state model that is linked in the minds of many to the mythical Westphalia. Some would say that the model was artificial and unsuited to the complex political reality of those countries; that the continuing collapse of Iraq and Syria (with Lebanon looking fragile) is at least in part a consequence of the bad match. But it may be less the borders of those states that have been the problem than the internal political nature of the states as they were established.

The new nations’ borders for the most part followed the boundaries of previous Ottoman administrative districts, including those abolished with much fanfare by Islamic State 18 months ago. Such is the ethnic, religious and tribal complexity of the peoples they contain that they are likely to be difficult to divide up in any less artificial or more satisfactory way. Any attempt to redraw borders extensively is likely to deepen and exacerbate the chaos. In the Westphalia settlement, with only a few exceptions, the pre-war borders of the German statelets were retained; it was the way the states related to each other and the confessional diversity of their subjects that changed. There is a lesson here.

Sectarianism, the interference of neighbouring states, the breakdown of earlier state arrangements, the exodus of refugees –all of these are features of a region that has become, as a recent New Statesman leader put it (quoting Karl Kraus), a “laboratory for world destruction”. Some in the contemporary Middle East are aware of past religious extremism and conflict in Europe and ask how we overcame it historically. Therefore, it is in no way patronising to offer the lessons of those past traumas: it is part of our shared human experience, our collective memory. That is what history is – or can be. The Westphalia myth, in supporting a notional model of the modern state which has failed in both Iraq and Syria, may have contributed to the terrible conflicts we have seen unfolding in recent years in those countries. The real Westphalia, by contrast, could contribute to a solution.

It showed ways to turn interference in wars into guarantees of peace

Its application to the Middle East requires an inclusive conference with representatives from all recognised states in the region, plus potential “guarantor” powers. The negotiations would have to start from the assumption that the “truth content” of the various positions has to be set aside for now, and would have to end with a recognition that sovereignty would be conditional and involve the transfer of some prerogatives to common institutions modelled on the old German imperial ­supreme judicial institutions and/or the Reichstag. Populations would not necessarily be guaranteed democratic participation in the first instance, but governments would be obliged to respect certain vital rights, including the free exercise of religion and, in certain circumstances, that of judicial appeal outside their local jurisdictions. Toleration would thus be “graded”, Westphalian-style, with the recognition of a dominant religion or system in each territory, but with safeguards for minorities. As with Westphalia, rulers would be constrained by duties towards their own subjects (for that is what they are, at present), but also towards respecting each other’s integrity as well as that of the whole system. The whole arrangement would then have to be placed under external guarantee of agreed regional and global powers.

All this requires political will and engagement, obviously, but it must begin with some intellectual legwork. To this end, the Forum on Geopolitics at the University of Cambridge has established a “Laboratory for World Construction”, drawing on expertise in both cases, to begin to design a Westphalia for the Middle East.

 

***

 

There will be no a “quick fix”; the Westphalia negotiations took five years and ultimately failed to end the related war between Spain and France (which lasted until 1659). By 1648 the various warring parties in central Europe had reached a state of general exhaustion, and disillusionment with religious extremism.

But the lessons of the real treaties of Westphalia, which provided means for the legal resolution of disputes and showed ways to turn external interference in conflict into external guarantees for peace, could be a significant contribution to eventual settlement of the Middle East’s problems.

Bringing peace to the Middle East will not be easy, and many have failed before. Yet if it could be done in mid-17th-century Germany, a problem no less intractable, then anything is possible.

Brendan Simms is the director of the Forum on Geopolitics at Cambridge

Michael Axworthy is the director of the Centre for Persian and Iranian Studies at the University of Exeter

Patrick Milton is a postdoctoral fellow at the Free University of Berlin (POINT programme) and co-ordinator of the Westphalia for the Middle East “Laboratory for World Construction” at the Forum on Geopolitics

Content from our partners
Building Britain’s water security
How to solve the teaching crisis
Pitching in to support grassroots football