The temptation of centralism is great, and so, too, is the half-life of its decay long. "National unification" is one of the major themes of recent world history, and it is closely linked to the idea of centralization. To centralize appears to be a primal human drive, tied to the need for transparency, clarity, and hierarchies. Perhaps there is even an aesthetic and psychological need behind it, with deep-seated anthropological roots. There are also good reasons to regard centralization as a crucial point in raising the efficiency of political systems and a main factor in the success of powerful nation-states. But how lasting is that success? Are not the increase of power and the unquestionable benefits of centralization sooner or later outstripped by its costs – which are also linked to power? For power has the dangerous tendency to make those who hold it "stupid and happy," because they enjoy at least a partial dispensation from the constant need to put their views across in competition with the non-powerful, and from the need to learn from others. Power is not only "evil in itself' (as Jakob Burckhardt remarked, endorsing Julius Schlosser); it also bears the seeds of its own destruction and is thus something dangerous not only for the subjugated but also for those who wield the power.

     The example of France led to other countries throughout Europe – and indeed throughout the world – copying the same model. This model initially combined absolute centralism with mercantilistic national- economic centralism and then with Jacobin egalitarian centralism, and thus enabled the country to achieve notable successes in culture, commerce, and in its machtpolitik. The struggle to gain political power, both outwards and inwards, finds its best allies in centralism. Against this background, it is obvious that, on the other side of the coin, skepticism about power and skepticism about centralism have to combine and form an alliance. In terms of the history of political ideas, liberalism and centralism did not manifest themselves as a thesis and an antithesis – even in Switzerland. To the contrary, advocates of freedom placed all their hope in what was at least a federal state central organ, which was, for its part, more disposed towards freedom than the most backward regional authorities of the confederation of the day; these being hostile towards the idea of freedom. The most striking example of this can be found in the federal state centralization introduced in the USA in the light of the victory in the American War of Secession (1861-65) by the Northern States over the Confederated Southern States that had left the Union. The slavery in the South provided entirely liberalist reasons to justify this centralization.

     The "Devil" of absolutist, interventionist anti- liberalism was driven out by the "Beelzebub" of a relatively liberal centralism. What was ignored in the process was the fact that such a central government itself inevitably tends towards increasingly anti-liberal interventionism – unless the necessary countervailing forces are mobilized.

     Political theory has long suppressed the realization that, given conditions of free or relatively free communication, open competition among systems quickly results in a high overall level of liberalism, even without the "detour " of compulsory liberalization via central governmental control (and this is true even among the most backward societies, which in the long run cannot afford expensive, uneconomical authoritarian systems when faced with the competitive pressure of more liberalized neighbors). Moreover, this has been corroborated in dazzling fashion by the more modem political economy (that of public choice). This realization is based on facts known in part even to Edward Gibbon, then recognized by the American anti-federalists, and found explicitly in the thinking of Senator John C. Calhoun (1782-1850) – facts which are still doubted in many quarters to this day. In the European Union, for example, reliance is placed on centrally controlled harmonization instead of on non-central deregulation. Many liberals have themselves long fallen victim to the discreet charms of centralism – and not to their advantage. Power always tends towards centralization, and centralized power tends towards absolutism. Lord Acton's warning against the absolutely corrupting effect of absolute power is at the same time a warning against centralism. Those who nourish the hope that it will be possible to keep central government free of the corrupting tendencies of power and to staff it with a freedom-loving elite, overestimate the virtues of both the electorate and the elected, and underestimate the normative power of structural processes even over well- intended functionaries.

     Centralism as a national political strategy may very well enjoy success for one or even two centuries. But today, anyone examining the various studies on the global ranking of nation- states can hardly avoid the observation that a broad range of indicators provide empirical evidence for the success of smaller, non-central systems. Centralism and non-centralism have different rhythms of generating costs and benefits for the general public. The collective benefits of centralization generally come in an early phase, which may last quite long: decades and even centuries in large systems such as France, for example. The general benefits of non-centralism, which are linked to learning ability, flexibility, robustness, and the avoidance of large-scale collective mistakes, often need a longer time to become visible and palpable. Wars, which may be seen as the pathological excesses of centralized power, create benefits for centralized systems that interrupt and mask the very gradual process of their decay. There is a "centralization dividend" and a "non-centralization dividend." The former increases in times of war, the latter in times of peace. But in peacetime, as soon as centralism's costs become greater than its benefits, centralized nations try to hide their lack of success by implementing the principle at its next-higher level in order to participate in the benefits of the competition of systems within a larger national or supranational structure. In this way, the "moment of truth" about the failure of centralism may be postponed again for decades.

     But do we really need to go through the collective experience of this programmed failure yet again?

     These remarks are intended as a warning against "flight to a larger structure." They urge that solutions to collective problems be found at the smallest possible level, and they propose structures in which only a minimum of problems requiring collective solutions will arise because responsibility is perceived and accepted not as "decentralized," but rather as "non-central," dealt with in small, peacefully competing units.

     The somewhat uncommon use of the term "non- central" is more than merely a personal quirk of this author. The point is to find a descriptive term that avoids the aggressive use of "anti" and does not resort to the dangerous image of decentralization. Decentralization presupposes the centre, and thus remains attached to the very ailment for which it claims to be the cure. Naturally, given the prevalence of already centralized institutions, decentralization seems an appropriate countermeasure, a kind of "orderly retreat" from misconceived structures. But there is an undeniable danger that this delicate operation would leave the mental ties to the centre untouched, so that the result would be merely a more refined form of centralized administration in which the political and economic costs are shoved off onto the periphery and the benefits remain with the centre. Anti-centralism, on the other hand, tends towards secession, which can inflict very high conflict costs. Non-centralism, however, leads to a kind of peaceful new ordering of responsibilities and delegated authority, which by its very nature must be structured from the bottom up, rather than from the top down, if constructive competition is to result from it.

     World history offers few encouraging examples of this approach, because large empires have collapsed mostly as a result of wars and crises. Switzerland, with its confederated structure, cannot provide a model. The technique of political "cell division" – that is, the way to get from misconceived centralized structures to a more progressive non-centralism – has yet to be discovered. The thoroughly unpeaceful and not yet completed de-centralization of Yugoslavia is not a particularly encouraging example. It may have generated a number of experiences which could be evaluated positively. But who is interested in these? Great powers and global powers prefer a transparent world with a few big (though comparatively less powerful) partners. Kissinger's wish for a central telephone number at which he could "call Europe" is evidence of this, and the sympathy felt in America for what remains of the Russian empire (which is still far too centralized) speaks volumes.

     As to the annoying and economically successful small players, they are there to be dominated at least politically (and fiscally!). That is what all the fuss about the primacy of politics is about – for ultimately that adds up to the primacy of power and the state, with the subtle inclusion of military security for the big powers, for which the smaller must generally pay too great a price in losses of sovereignty.

     In the course of my international lecturing on the subject of "Swiss Federalism and Non-Centralism" I am sometimes asked: "When did Switzerland begin its process of decentralization?" Here is my response: It never began that process; all it has managed to do is to successfully avoid – or at least put the brakes on – the process of centralization for an uncommonly long time. The country's only centralist experiment, Helvetia under French dominance, lasted barely five years, from 1798 to 1803, or less than half of one percent of its 700-year history. Switzerland is moving very slowly in the direction of "more centralism," and perhaps the first step towards non- centralism consists in active skepticism towards centralism, in the refusal to push the process forward and accelerate it. The best political path to less centralism and more non-centralism has yet to be found, or invented. And it is to be characterized not as a "return" but rather as a new start, as perhaps the greatest and most exciting political challenge to liberalism in our new century. It is quite evident that the easily misunderstood term "decentralization" is inappropriate for it, and that it would also be wise to avoid the idea of "secession," which is equally susceptible to misunderstanding.

     A more positively toned term would have to connote the level at which a large proportion of political problems can be resolved jointly and flexibly: the level of what is known here in Switzerland as the Kommune or the Gemeinde – that is, the township or municipality. In Switzerland, an autonomous township (with the power to levy taxes) is not simply an administrative unit with certain subsidiary powers left to it, and public funds assigned to it, by "higher" levels of government. With something of nostalgic emotion, it is often referred to here as the "primal unit of government." Communalism, autonomism, localism, regionalism – there are reasons for and against the use of each of these terms. Negatively defined, the secret of Switzerland's success resides in the persistent resistance to centralism; positively defined, it may be seen in the functioning of competitive municipal bodies, in the combination of personal autonomy and municipal autonomy.

     There are various commonly used labels for these political principles. Most usual are "federalism" and "subsidiarity." As I shall illustrate in a later chapter on terminology, both terms carry a high risk of misunderstanding and intentional or unintentional misinterpretation. The politically crucial question: “How does this affect centralism?" (or “How does it affect power?") can be answered with respect to both those terms if the unequivocal concept of "non-centralism" is attached to them: Federalism in the sense of non-centralism, subsidiarity in the sense of a distribution of power in favor of personal autonomy on the one hand, and municipal autonomy on the other. All political structures above the level of the local township entity (and even that one as well, in fact) must be justified – and, from the standpoint of liberalism, the reasons for them must be very strong indeed.

     Unfortunately, no nation-state is in the happy position of being able to reconstitute its political structures starting from a tabula rasa and to reorient them along the lines of a few fundamental principles. But at the level of Europe as a supraordinate entity, we are at an early stage of creating political structures, still in a phase at which it would definitely be worthwhile to avoid making certain mistakes. At the national level, however, every proposal for political reform can be measured against the concepts of liberal ideals. The process of gauging its compatibility with liberalism could be supplemented by examining the reform's compatibility with non-centralism. This would help liberals not to be re-infected with the delusions of centralism. Perhaps we also need to have in reserve some proposals for political solutions, to be brought out after the grand collapse of the modem welfare state – a scenario that is not to be wished, but one that cannot be ruled out. Perhaps we should prepare a kind of peaceful and bloodless bankruptcy proceeding for central governments, one in which non-centralism could play an important role in any new beginning.

     Switzerland today is the result of various constellations of rather unique circumstances. There are good reasons to view it as a special case in European history. Special cases cannot serve as models, because they cannot be very successfully copied if certain peripheral conditions are different. But although Switzerland is no model, it can be seen as a quite successful experiment, and something can be learned from experiments even when conditions vary greatly.

     The Swiss political system rests on two main points: first, a skepticism about power and rejection of domination (which is always "foreign" domination, to a certain extent), and second, the realization that there are collective problems which must be resolved in common, flexibly, as cooperatively as possible, and without outside intervention. Both these points are aptly presented in Friedrich Schiller's play "William Tell": the anarchistic "Tell principle" and the communitarian "Rütli alliance principle." A political system cannot survive on the basis of the negation of power alone. Neither the "anti" nor the "non" is sufficient as a long-term principle for political survival.

     This presentation, "In Praise of Non-Centralism," is also a small gesture of gratitude for a valuable gift which the Swiss Confederation was privileged to receive from a German citizen, and which will soon celebrate its 200th anniversary: Schiller's William Tell.

Robert Nef studied law in Zürich and Vienna. From 1961 to 1991 he was Research Associate in the Law Department at the Swiss Federal Institute of Technology in Zürich. He has been Director of the Liberales Institut since 1979. Since 1994 he has been editor and co-publisher of the journal "Schweizer Monatshefte". Hi is author of Which Constitution for Which EU? and In Praise of Non-Centralism .