Humane Studies Review

Volume 7, Number 1 Winter 1991/92

Institutional Bases of the Spontaneous Order: Surety and Assurance

by Albert Loan

Bibliographic Essay

Systematically explaining the emergence of spontaneous orders by "invisible hand" processes has been, since Adam Smith, an integral part of the research program of economics. Cooperation is shown to emerge in a society of self-interested individuals, who spontaneously create a peaceful and productive social order through their repeated interaction with each other, without consciously intending to do so. Since the writing of David Hume's Treatise On Human Nature in 1740, many social theorists have elaborated on the way rules that promote social order emerge without deliberate attempts to create them. In a community of people who interact frequently, the principle of reciprocity serves to reinforce cooperative behavior since the benefits of cheating are often outweighed by the cost of losing the trust and the business of other members of the community. Even in instances where strict reciprocity does not occur, due to the number of individual actors being large or because contacts between any two individuals are typically "one off." mechanisms can emerge to coordinate activity on a voluntary basis. (For recent treatments of spontaneous social order theory see Robert Axelrod, The Evolution Of Cooperation [New York: Basic Books, 1984]; Robert Sugden, The Economics of Rights, Cooperation, and Welfare [Oxford: Basil Blackwell, 1986]; Anthony De Jasay, Social Contract, Free Ride [New York: Oxford University Press, 1989]; and David Schmidtz, The Limits of Government [Boulder, CO: Westview Press, 1991]. For a view of the spontaneous emergence of arrangements for dispute settlements, see Peter Stein, Legal Institutions: The Development of Dispute Settlements [London: Butterworths, 1984], and Legal Evolution [New York: Cambridge University Press, 1980].) These writers have sought to show how the rules on which such an order is founded, such as the observance of property rights and the performance of promises, could result from rational self-interested behavior, despite the apparent incentives to cheat and to "free ride."

This paper examines some historical evidence that supports this recent body of work. It examines how the institutions of suretyship and assurance spontaneously emerged as the unintended result of self- interested action and how they embodied the stabilizing tacit norms of trust, fellowship, and thrift, which served to promote and preserve these institutions.

The Nature of Suretyship and Assurance
Suretyship and assurance are closely related institutions in which each member of a group assumes some form of responsibility for the others in the group, thereby diffusing among the group the burden of risks which might otherwise fall on a single individual. Suretyship involves members of the group assuming some degree of responsibility for any unpaid debts, defaults on agreements, negligence, and so forth of other members of the group. In the case of assurance members pool resources to provide for relief from a hazard they face in common as members of that group. Both, though the precursors of the modern institutions of insurance, are in their purest form reciprocal arrangements, whereas insurance generally involves one party (designated in advance as the insurer) assuming some of the risk faced by another (the insured) in exchange for a monetary premium. All three institutions operate on the principle of diffusing the burden of various kinds of risk, but in the case of insurance it is a specialized intermediary who manages the diffusion of the risks and in the end takes final responsibility.

The earliest well-documented instances of suretyship in the West are traceable to forms of fellowship within the Saxon clans of England and their tribal Germanic forbears, and to tribal Celtic society. The borh system and frankpledge are notable examples of early forms of suretyship in medieval England, while other instances can be found in the experience of Medieval Iceland and Celtic Ireland. There are isolated documented examples of assurance dating back to as early as the 10th century BC in the Eastern Mediterranean. (For a useful account of the assurance of sea voyages during this time and of marine insurance in general see William Winter, Marine Insurance: Its Principles and Practice [New York: McGraw-Hill, 1952].) In many cases there is considerable controversy over the origins of particular forms of suretyship and assurance, partly because many types date back to periods for which records are very scarce.

The rules and norms that characterized these institutions allowed them to function and survive, in one form or another, centuries of political and social change. They owe their origin to a spontaneous informal process reliant on custom and traditions, the functions of which were unknown to those whose welfare they increased. What P.H.J.H. Gosden wrote with regard to assurance within the British friendly societies of the 18th and 19th centuries applies as well to other forms of mutual risk-bearing:

they do not owe their origin to parliamentary influence; nor to private benevolence; nor even to the recommendation of men of acknowledged abilities, or professed politicians. [They] originated among the persons on whom chiefly they were intended to operate: [these men] foresaw how possible, and even probable, it was that they, in their turn, should ere long be overtaken by the general calamity of the times and wisely made provision for it. (P.H.J.H. Gosden, Self-Help: Voluntary Associations in the 19th Century [London: Batsford Press, 1973, p. 91])
The earliest examples of suretyship are connected with the enforcement of customs and laws. In many societies, laws have existed and been enforced in the absence of anything which we would call a state, by purely voluntary action. In Anglo-Saxon England there existed a well developed system of customary law. (For an excellent discussion see Bruce Benson, The Enterprise of Law [San Francisco: Pacific Research Institute for Public Policy, 1990, pp. 21-30].) Similar and even more elaborate systems of surety and customary, non-state law flourished in Iceland and Celtic Ireland. (For these, see Jesse L. Byock, Medieval Iceland [Berkeley, CA: University of California Press, 1988]; William I. Miller, Bloodtaking and Peacemaking: Feud, Law and Society in Saga Iceland [Chicago: University of Chicago Press, 1990]; and D. A. Binchy, ed., Studies in Early Irish Law [Dublin: Irish Universities Press, 1936].) Enforcement in all of these cases was informal and primarily governed by custom, rather than by a state.

The dominant type of "peace-surety" in pre-Norman England was the borh. Borh was an institution for "securing peace observance among the lower classes," and was regarded at the time to be "the base of social order." (William Alfred Morris, The Frankpledge System [New York: Longmans, Green & Co., 1910, p. 25]. This is an extensive and rigorous historical treatment of the borh system and its later development into frankpledge. Other accounts can be found in T.F.T. Plucknett, A Concise History of the Common Law, 5th ed. [Boston, MA: Brown, Little & Co., 1956, pp. 628-32], and Helen Maud Cam, The Hundred And The Hundred Rolls [London: Methuen, 1930].) The borh system was a voluntary form of peace surety and lasted until the Norman conquest of 1066. In its earliest form, it consisted of a group of twelve neighbors, often, though not always, members of the same clan or kin-group. It provided that if an individual committed a crime, disturbed the peace, or was delinquent on a debt, there was a group of individuals (the borh) who would share the burden of tracking, capturing and bringing that delinquent person to justice.

An individual's borh also served as a kind of credit reference and guarantee by vouching for his integrity and agreeing to make recompense in his stead if the situation required. The members who stood as borh were able to bear such risks for two reasons. First, the burden was dispersed among them and second, the frequency of contact and the voluntary nature of the bond helped deter actions in violation of custom. Kinsmen or neighbors were not under obligation to agree to stand as surety for an individual whose reputation they questioned, which made the maintenance of one's reputation of paramount importance and further restrained dishonest or criminal behavior.

It is true that some form of suretyship was required by the crown at various times in medieval English history, but there was usually no stipulation as to the form of organization to be employed for ensuring accountability among the people. More importantly, since "social relations were maintained only with people who shared surety protection," there were strong incentives to engage in these voluntary associations even without the mandate of the crown. (Leonard Liggio, "The Transportation of Criminals: A Brief Political Economic History," in Randy E. Barnett and John Hagel III, eds., Assessing The Criminal: Restitution, Retribution, and The Legal Process [Cambridge, MA: Ballinger Press, 1977, pp. 273-94]) The importance of reputation and good character were essential for the establishment of credit and to secure prospective partnerships for negotiations and exchange.

Because others in the group provided insurance (credit) for all members...they would not accept or keep someone who was not of good character. Consequently, members of a surety organization could disclaim someone who committed an egregious wrong, providing strong incentives to abide by the law...In effect, everyone who wanted to participate in and benefit from the social order was bonded. (Benson, 1990, p. 23)

It is clear from this that the borh system was sustained by self-interested behavior, but to show how it was in an individual's interest to participate in the institution once it was established is not the same thing as providing an invisible-hand type explanation of how the practice first emerged. Such accounts must go beyond a merely functional explanation of the institution to identification of the process by which it became established. The challenge in identifying how a practice such as that of borh was initiated lies in the fact, well known since the time of Hobbes, that in the absence of an established convention of mutual cooperation it appears that cooperative behavior (such as willingness to stand as surety for a family member) is not self-reinforcing and could not gain a foothold in society. This conclusion derives from the observance that, regardless of whether others behave cooperatively or not, an individual may gain an advantage for himself by cheating on a cooperative arrangement or by not participating in one while gaining the benefits. The claimed result is the irony of the well-known "prisoners' dilemma" game in which the players rationally choose not to cooperate with each other even though this outcome is less beneficial than mutual cooperation.

Robert Sugden and Robert Axelrod utilize game theory to point out, in the tradition of David Hume, that the uncooperative equilibrium that would seem to preclude institutions such as suretyship need not obtain in a situation where individuals expect to interact with each other repeatedly. In such a case, the short-run benefits of avoiding the cost of cooperating must be weighed by "the shadow of the future," the expectation that noncooperation may result in retaliation in future interactions, while cooperation may foster similar behavior in others. Sugden reasons that even in a world where most individuals obstinately refuse to cooperate, if the expectation of future interactions is high a growing minority may find that a strategy of "Tit-for-Tat," which bravely initiates cooperation while retaliating against defection, can result in a successful convention of cooperation. Axelrod likewise illustrates how a "cooperative cluster" of as little as two individuals following a Tit-for-Tat strategy may grow in number and establish a general convention of cooperation even in a community initially composed of unconditional non-cooperators. Hayek, in a recent work, argues that communities which develop voluntarist means of coordinating actions and regulating behavior will be more successful than others which, for one reason or another, fail to do so. (F. A. Hayek, The Fatal Conceit [London: Routledge, 1988]) This implies that Axelrod's model of the emergence of cooperative clusters or communities may be tied into an evolutionary model of historical development.

Moreover, work by Brubaker, Goldin and, most recently, Schmidtz suggests ways in which cooperation can arise even when strict interpretation of the prisoners' dilemma would suggest that no individual has any strong incentive to cooperate with others to supply some collective good. (Kenneth S. Goldin, "Equal Access vs Selective Access: A Critique of Public Goods Theory," and Earl R. Brubaker, "Free Ride, Free Revelation, or Golden Rule?", both in Tyler Cowen, ed., The Theory of Market Failure [Fairfax, VA: George Mason University Press, 1988, pp. 69-92 & 93-110]; Schmidtz 1991, passim but particularly pp. 92-111) When individuals know that a good will not be provided unless they cooperate, the incentive to do so may be strong enough to lead to such cooperation, even when not everyone in the group goes along. One crucial factor is the capacity to create a mechanism to exclude non-cooperators and, even more important, punish defectors. (A key argument here is that of Goldin, that most goods can be made exclusionary, i.e. non-public, hence the key decision is whether or not to supply the good on an exclusionary or open basis.)

Civil peace within a community, such as was maintained by the borh system, is the classic example of a collective good being provided through voluntary action. A community would collectively bind themselves, usually by oath, to live together in peace and punish malefactors. In one of the most dramatic examples, the formation of urban communes in medieval France and Italy, this was symbolized by the erection of walls which separated the cooperating community from the outside non- participants. Another example was the movement for the "Peace of God" and "Truce of God" in northern Europe during the twelfth century, where large numbers of lords and cities would collectively bind themselves not to war against each other and to cooperate in punishing anyone who reneged on the agreement. Institutions such as outlawry and excommunication evolved to enforce these "conditional assurance contracts," as Schmidtz terms them. (Schmidtz, 1991, pp. 66-79. For historical examples of this kind of process see Henri Pirenne, Medieval Cities [Princeton, NJ: Princeton University Press, 1925], Charles Petit-Dutaillis, Les Communes Francaises [Paris: Albin Michel, 1947]; W. F. Butler, The Lombard Communes [New York: Haskell House, 1969 repr. of 1906 ed.]; and, for the "Peace of God" movement, Harold Berman, Law and Revolution: The Formation Of The Western Legal Tradition [Cambridge, MA: Harvard University Press, 1983].)

Anthony de Jasay goes further and argues that even when there are free riders on the public good supplied by others' free cooperation, it may still be in the interests of the cooperators to supply the good and carry the free riders -- to be "suckers" in the usual terminology. Again, the key factors are the knowledge of the actors that the good will not be supplied at all in the absence of their cooperating, the existence of a mechanism to ensure that it will be supplied if a given number cooperate, and some means of identifying the cooperators.

These theoretical treatments of the process by which cooperation, conventions or institutions emerge spontaneously are perfectly compatible with the historical treatments of the borh system, which emphasize the role played by reciprocity and the importance of ties such as kinship and geography which would guarantee frequent interaction. The Icelandic example shows how surety and other cooperative relations could evolve and sustain themselves even where the individuals involved were not necessarily connected by kinship ties or geographical contiguity.

After about the year 960 Iceland was divided into thirty-nine chieftaincies, each headed by a chieftain or godi (plural godar). Each godi had authority over a thing. This was an association of individuals or households who voluntarily chose to be bound to a particular godi. Among its many other functions, the thing acted as an instrument of collective surety, in the same way as the borh. It was also the main agency for settling legal disputes, seen purely as private affairs between individuals. The chieftaincies were grouped into thirteen varthings. The meetings of things and varthings were judicial assemblies at which disputes were heard and settled, with serious matters held over to the collective meeting of all the godar at the annual althing. The thing was not a geographical entity; at the local level the thingmen of various godar were intermingled with no clear geographical base for any one thing. Also, things did not necessarily correlate with kin-groups. Although in the beginning things were primarily composed of kin-groups, they later expanded to include members without kin relations. (Byock, 1988, pp. 55-76, 108-36; Miller, 1990, chs 6-8, pp. 179- 300)

The Icelandic case also shows the importance of another factor: the use of representation or fictive collective individuals in reducing the number of those who must decide to collaborate to an optimum level. There being only thirty-nine godar made the decision to outlaw any defector easier to arrive at. Other examples of this were the leagues of cities which were a feature of the middle ages, the most notable being the Hanseatic League. Here the number of individual merchants and operatives bound by the collective agreements of the league (to mutual peace, collective security, and common commercial law) was very large, but the actual entities making up the league were a manageable number of cities. This also demonstrates the point that, contrary to much ignorant or malicious comment, voluntary social organization of this kind is compatible with large, complex societies, and not found only in "face-to- face," "tribal" communities. In fact, as Hayek and others have argued, in the "great" or "extended" society of modernity, free and voluntary cooperation becomes more, not less, important. Voluntary cooperation through these kinds of networks and institutions is what enables use to be made of the great mass of tacit knowledge scattered throughout the population, so making modern society possible and more social than its "primitive" forbears. (Hayek, 1988, passim but especially pp. 15-19)

Moreover, such systems are also adaptable to changing political and demographic conditions. The Anglo-Saxon borh is a good example of this. With time, it came to be based less and less on actual (as opposed to fictive) kin ties, if indeed this had ever been dominant. Benson argues that the borh system came to lose its family character because of the increased mobility of the population, which caused kinship reciprocity to break down while making reciprocity among friends and neighbors of the same church or vil more significant (1990, p. 23). Maitland, by contrast, argued that the system had never been purely, or even primarily, kin based. (Sir Frederick Pollock & Frederic William Maitland, History Of English Law To The Time Of Edward I, 3rd ed. [Cambridge: Cambridge University Press, 1968, pp. 568-71]) Either way, borh was sufficiently robust as an institution to cope with great social change.

The Replacement of Borh by the Frankpledge System
The Anglo-Norman scheme of suretyship called frankpledge is sometimes confused with the borh practices from before the Norman conquest. Like borh, frankpledge involved a group of men pledging or bonding an individual, although the groups established for suretyship under frankpledge were the groups of ten known in medieval society as the tithing, not necessarily friends or relatives. Although similar in purpose and in many particulars, there were two crucial differences between borh and frankpledge: first, frankpledge seems mainly to have been an institution imposed by the crown; secondly, whereas an individual could decline to stand as surety in another's borh on the basis of knowledge of that person's character, those who stood as suretyship under frankpledge had no choice in the matter. Morris notes that:

Between the voluntary pledging of a man by his neighbours in 1030...and the duty, in 1115, of every man in a tithing to serve for every other man in the tithing without right of refusal or withdrawal, no matter what the character of the associates, is a break that can be explained only by governmental action of a deliberate and rigorous nature....(p. 29-30)
This change was significant. The effectiveness of the borh system had been due to everyone's knowing that the likelihood of receiving suretyship from his neighbors depended on his own actions and reputation. The elimination of the element of choice and the replacement of kinship and friendship relations as the basis of suretyship with an arbitrary group decided by the authorities eviscerated suretyship by removing the crucial voluntary element. The borh system had developed spontaneously, largely the result of voluntary intercourse and probably a considerable amount of trial and error over a century and a half of Anglo-Saxon legal history. It was both adapted and adaptable to the changing demography and political environment of medieval England. Frankpledge, on the other hand, was imposed by William I for reasons of expediency and greed. Frankpledge was intended "to invoke a more stringent borh system" (Morris, p. 31), and also served the purpose over the eleventh and twelfth centuries of diverting compensation into fines paid to the crown, so swelling the King's revenue. By supplanting a voluntary system, dependent on a composite of tacit rules of trust which had developed spontaneously over centuries and had proved to be beneficial and progressive, with a compulsory arrangement, essential elements which made borh an effective surety system were lost. Benson recounts how under the Normans voluntary participation in the maintenance of civil peace declined and had to be replaced by the extension of more authoritarian methods of keeping the peace, such as the introduction of local sheriffs with broad powers of law enforcement. The "shadow of the future" which had reinforced voluntary participation in the maintenance of order was diminished under frankpledge. Thus, frankpledge could not function in the way that borh had, since it drastically reduced the role of reputation that had been at the heart of the Saxon legal system.

Suretyship and the Growth of International Trade
The social function that suretyship served is illuminated by a look at the cultural and commercial history of Europe as a whole. By the twelfth century, networks of trade were being extended deeper into England and the continent from commercial centers in Northern Italy and coastal cities on the Mediterranean. The uncertainties associated with international commerce during this time were considerable. Not the least of those were risks relating to the trustworthiness or prudence of prospective transactors. These were often high because of the tenuous and infrequent nature of trade in the more isolated regions of Europe. An examination of the emergence of suretyship within the context of expanding medieval markets is especially important. It is even more interesting when viewed in light of the argument that the expansion of the groups within which people interact and the resulting decrease in the frequency of interaction between individuals endangers social stability. It is a significant point that the problem of increasing anonymity is not new to modern civilization, and that history provides examples of how it has been effectively dealt with by practices such as suretyship.

The extension of markets into previously isolated inland regions of France and Britain during the 11th and 12th centuries often took the form of periodic fairs for which merchants traveled great distances to sell their merchandise. These fairs were invaluable institutions for coordinating trade between people from distant parts of Europe and beyond. The famous fairs of Champagne, which ran almost continually, alternating in one of four towns of that region, attracted merchants from all parts of the Western world during the twelfth and thirteenth centuries. As a result, they became "a sort of clearing house for merchandise and currency exchanges between the Mediterranean and the North Sea basin." (Robert Lopez and Irving Raymond, Medieval Trade in the Mediterranean World [New York: Columbia University Press, 1990 repr of 1955 ed., p. 80]. This is a very thorough documentation of medieval trading practices. Ellen Moore's The Fairs of Medieval England [Toronto: Pontifical Institute of Medieval Studies, 1985] examines these practices as they occurred in England.)

The success of the medieval fairs is not hard to understand, and the function they served is of interest for understanding the further development of mutual risk-sharing practices. To appreciate the role played by such practices, it is useful to examine how the fairs came to be in the first place. In the absence of efficient communication, large towns, and well developed permanent commercial establishments in these regions, there was a tremendous amount of uncertainty associated with trade, especially international trade. The prospects of buyers and sellers meeting were low and difficult to ascertain, credit arrangements were extremely risky, and the problem of differing currencies was serious -- all of this since the scope and frequency of personal interactions between transactors were severely limited. In this context, the actions of the merchants exemplify Sugden's arguments about cooperative clusters. Merchants must have attempted to seek out locations that were central and accessible, and recognized as such by other merchants as well as the local population. Small groups of merchants who converged on a particular place and time in this attempt would have done well to agree to return again at a specified time and to announce this to other merchants and to the locals. Such groupings would consolidate and grow as word of the time and place of their meeting spread both locally and internationally. As the same merchants (or their agents) returned again and again and the amount of trade grew sufficiently to lead to convergence on certain standards of trade and currency exchange, the uncertainty of trade in these regions was greatly reduced, and transactions that were too costly or risky to be undertaken before were now regular occurrences.

The fairs made it possible for commercial relationships between individuals to develop, for reputations to be established and for accepted practices for assuring performance of contract to evolve. Most notably, they led to the emergence of the Law Merchant, a transnational system of customary law enforced by informal courts of Piepowder (from pieds poudres, or dusty feet), which first appeared at the Champagne fairs and soon spread all over Europe, becoming more systematic as they did so. (For accounts of the Law Merchant and its origins, see Wyndham Anstis Bewes, The Romance Of The Law Merchant [London: Sweet & Maxwell, 1986 reprint of 1923 ed.]; William Mitchell, An Essay On The Early History Of The Law Merchant [New York: B. Franklin, 1969 reprint of 1904 ed.]; and L. E. Trakman, The Law Merchant [Littleton, CO: F. B. Rothman, 1983].) The spontaneous appearance of the Law Merchant and its surety mechanisms is particularly interesting as an instance where such an institution evolved despite the lack of repeated contacts between players, showing that anonymity does not necessarily lead to a crippling "prisoners' dilemma." (For a discussion of this point see Paul R. Milgrom, Douglass C. North, and Barry R. Weingast, "The Role of Institutions in the Revival of Trade: The Law Merchant, Private Judges, and the Champagne Fairs," in Economics and Politics, Vol. 2, 1990, pp. 1-25.)

Nor was the Law Merchant the only institution to evolve from the fairs. In the fairs of medieval England, rules were established that further reduced the uncertainty of itinerant trading by extending the liability for repayment of debts to any merchant within the guild of which the original borrower was a member. The guilds were local voluntary associations whose members practiced the same trade. Craft and merchant guilds served to set standards of workmanship, rules of the business, terms for apprenticeship, and conventions for adjudicating disputes. By providing suretyship among members, the guilds converted costly international disputes into more manageable disputes within the local guilds and reduced the scope for defaulting on agreements. The custom of mutual liability for debts among members of the guilds was advantageous because frequent interaction, reciprocity, and bonds of fellowship between members greatly reduced the gains that could be had from defaulting on a trading obligation. Membership in a merchant guild was very valuable since it communicated one's trustworthiness to other merchants. Members were thus discouraged from taking actions that might endanger their continued acceptance within the guild. (See Antony Black, Guilds and Civil Society [Ithaca, NY: Cornell University Press, 1984].)

The importance of mutual accountability for promoting trade in England was demonstrated by the contraction of trade that occurred after 1275, when the First Statute of Westminster legally abolished the practice. By 1283, Edward I was dismayed

...because merchants who in the past have lent their substance to various people are impoverished because there was no speedy law provided by which they could readily recover their debts on the day fixed for payment, and for that reason many merchants are put off from coming to this land with their merchandise to the detriment of the merchants and of the whole kingdom. (Moore, p. 120)
Before continuing, it is important to take notice of the social dynamics underlying this historical account. The search for new profit opportunities led entrepreneurs in the established European centers of commerce to try to expand their trade into the hinterland of the continent. A new market institution evolved -- the fair, which made this expansion possible by establishing a process of expanding trade geographically. Yet, the physical distance and cultural differences that separated merchants gave rise to problems of accountability. The problem was that the expansion of the groups within which people interacted diluted the influence of reciprocity. The response was twofold: first, the extended, expanded trade was preserved by returning the function of deterring dereliction to smaller, localized social units that could bring reciprocity and fellowship to bear on their members. The second response was the emergence of impersonal voluntaristic institutions such as the Law Merchant which, like modern credit bureaus, transmitted and pooled information about individual merchants, making it available to their fellows even when they seldom met face-to-face. Contemporary institutions such as credit card companies and bureaus are in fact excellent examples of private solutions to apparently insuperable public good problems, occurring in a system marked by anonymity and ephemeral contact: again this demonstrates that such solutions are not purely a feature of "primitive" societies. (For an excellent discussion of the theoretical problems involved in credit, and the appearance of the credit bureau as a solution, both historically and in modern society, see Daniel B. Klein, "Cooperation Through Collective Enforcement: A Game-Theoretic Model of Credit Bureaus," forthcoming in the Journal of Economics and Politics.)

Two complementary forms of institutions, then, tend to emerge together and reinforce each other. On the one hand are institutions (like media of exchange, standard forms of contract, and fairs) which are comprised of conventions that guide people's expectations for the coordination of their plans. On the other hand are institutions (like courts, credit bureaus, police, and in the present context, guilds), which reinforce cooperative rules of conduct typically susceptible to breakdown due to shirking or free riding. As development of the former coordinating institutions fosters the extension or dispersion of markets, they create the need and incentive for reinforcing institutions that bring problems of accountability back to a level where reciprocity and fellowship are effective and enable individuals meeting for the first -- and maybe only -- time to discover each other's past records.

One important conclusion to be drawn from the history of the extension of commerce in Europe and of suretyship in England is that voluntary risk-shifting institutions such as suretyship serve the overall order of society in ways that reach beyond what we normally think of when we speak of "the reduction of uncertainty." Most notably, they allow for both the extension of the market order and improved social stability. Contrary to the arguments of some historians, such institutions do not imply a holistic or collectivist view of identity. Rather, they enable individuals to pursue their own ends more effectively by acting cooperatively with others. As one author puts it, "The crucial point about both guilds and communes was that here individuation and association went hand in hand ... One achieved liberty by belonging to this group." (Antony Black, 1984, p. 65, emphasis original)

The Emergence of Assurance: British Friendly Societies
The roots of the friendly-society sort of mutual assurance against the risks of illness, accident or other disability appear to be closely connected with earlier institutions of suretyship, although they served a different purpose. While suretyship arrangements operated primarily by reducing the risks imposed on those who would trade with members of the group, assurance served to cope with the risks faced by the group members themselves. The history of these assurance institutions bears out the fact that spontaneous ordering processes like those identified as the source of voluntary suretyship can explain the existence of assurance institutions as well. By the 16th century, the power of the guilds in England had grown so great that they were seen as a threat to the rule of the crown and the government undertook to suppress them. But as the guilds declined, a new form of fellowship arose to replace them. The "friendly societies," as they were called, were voluntary associations of working men organized "to meet their social and convivial needs as well as to insure against the hazard of sickness and death." (P.H.J.H. Gosden, Self-Help: Voluntary Associations in the 19th Century [London: Batsford Press, 1973, p. vii]. See also Gosden's The Friendly Societies in England 1815-75 [New York: Augustus M. Kelley, 1967]; two works by William H. Beveridge, Voluntary Action [London: George Allen & Unwin, 1948], and, with A. F. Wells, The Evidence for Voluntary Action [London: George Allen & Unwin, 1949]; and two older works: Samuel Smiles, Self Help [London: J. Murray, 1958], and Frederick Eden, The State Of The Poor [London: J. Davis, 1797]. These make up the classic literature on friendly societies. For more recent work, see David G. Green, Working Class Patients And The Medical Establishment: Self-Help in Britain from the Mid-nineteenth Century to 1948 [New York: St. Martins Press, 1985], and the same author's Mutual Aid or Welfare State: Australia's Friendly Societies [Sydney, NSW: Allen & Unwin, 1984]; and Ian R. Christie, Stress and Stability in Late Eighteenth Century Britain [New York: Oxford University Press, 1984, ch. 4]. In the American context, see David Beito, "Mutual Aid for Social Welfare: The Case of American Fraternal Societies," in Critical Review, Vol. 4, No. 4, 1991, pp. 709-36.)

There is disagreement over whether direct historical connections may be drawn between the friendly societies and the merchant and craft guilds that came before them, but, as F. H. Eden observed, there is

an extraordinary coincidence between the rude simplicity which pervades the ordinances of [some of] the Saxon Guilds, and some of the modern friendly Societies. (Eden, 1797, p. 590)
Unlike the merchant and craft guilds, however, their membership was not dominated by any one trade, but was comprised of men of all trades and work backgrounds. The friendly societies charged their members regular subscriptions so that the pooled funds could be used for the relief of members who were unemployed or who, because of illness, injury, old age or other misfortune, could not support themselves or their families. (The wide occupational spread of the membership was of great assistance here: as each occupation tended to have its own trade cycle it was unlikely, barring a major catastrophe, that all or a majority of the societies' members would be unemployed at any one time.) These societies were numerous and large during the 18th and 19th centuries. Eden estimated in 1801 that there were about 7,200 societies with a total of 648,000 members (Eden, 1797, p. 7). By the end of the 19th century, the number enrolled in the great "Affiliated Orders" such as the Oddfellows, Druids, and Forresters ran to several million (Gosden, 1967).

By the late 19th century, most friendly societies supplied not only financial relief to those whose circumstances required it, but offered treatment by a medical practitioner when needed (Gosden, 1973, p. 112). In Britain, the societies were able to provide the full range of medical services, since by pooling their resources they were able to build and run dispensaries and to retain a panel of doctors on a permanent basis. Because the societies exercised customer control over the doctors, they were able to keep down medical prices to an affordable level, as well as control the service provided by the medical practitioners and so check the cartelizing tendencies of the organized medical profession (Green, 1985, passim).

Early self-help groups were not, as one might expect, politically collectivist in their organization or intent. Rather, they were fiercely individualistic and resisted all external attempts to regulate them. These organizations, priding themselves on their autonomy, consciously fostered among their members the ethics of self-reliance, prudence, and frugality. Unlike the compulsory insurance schemes of later years, friendly societies were not established for the purpose of the redistribution of wealth. (Black, 1984, pp. 177-81; Gosden, 1973, p. 10; Hayek, 1960, p. 287-89)

Like the guilds that were their historical predecessors, the friendly societies were viable because their rules and organization assured close and frequent contact between members, offering the maximum scope for reciprocity and the development of reputations. The development of affiliated orders provided a mechanism for transmitting information about a person's character and reputation to those who did not know them directly. In much the same way that reciprocal and reputational dynamics discouraged guild members from defaulting on loans or failing to perform contracts, they counteracted the temptation of individuals in a mutual-aid situation to free-ride on the contributions of others in the group. Ties of fellowship were strengthened by the fact that the friendly societies organized not just for the purpose of mutual indemnity but as centers of conviviality and friendship, just as their name implied. This factor was usually overlooked by Parliament, which on some occasions tried to copy the success formula of the friendly societies when establishing compulsory mutual aid societies for various groups of citizens. In reference to a 1792 Act of Parliament, which attempted to set up a compulsory mutual society for the relief of skippers and keelmen employed in the coal trade on the river Wear, Eden claimed that "one of the most important functions of local friendly societies, for which the Act obviously made no provision...was to bring men together to spend a convivial hour with their neighbours...." (Gosden, 1973, p. 8) Like other legislative forays of this sort, this mandatory society was unsuccessful, apparently because it was seen as a tax on the workers and not as a vehicle for social intercourse (Eden, 1797, pp. 614-615).

Eden was convinced that the state should not attempt to regulate friendly societies, for if it did so 'the inclination of the labouring classes to enter them will be greatly damped, if not entirely repressed." In this he was entirely accurate in his observations, for in the next century even that mild form of state protection offered by registration was to be found objectionable by many members and even in the later 19th century, the Chief Registrar believed that there were probably as many members of unregistered as of registered societies.' (Gosden, p. 10)
This evidence appears to highlight the importance not only of the frequency of association, but its voluntary nature for the nurturing of reciprocity and friendship.

The parallel between the character of the social rules involved in assurance and in suretyship is clear. Each harnesses the influence of reciprocity and reputation to reinforce cooperative behavior. Each strengthens the function that trust serves in social relations by maintaining close interaction among individuals through practices based on fellowship. Not surprisingly, explanations of the emergence of assurance have been made along very similar lines to those discussed for suretyship. Sugden provides an invisible hand explanation of the development of mutual aid societies which runs exactly parallel to the process he identified for the establishment of cooperative conventions in the prisoners' dilemma game. (See Sugden, 1986, ch. 7 for this discussion.)

Suretyship and assurance may seem dry, narrow topics, of interest only to the specialist. In fact, as this paper demonstrates, they are basic institutions for any liberal order, and their study can cast light on many questions of vital import to those interested in understanding and promoting the classical liberal ideal of a free society. The voluntary cooperation of individuals through these institutions has played a crucial role in the historical development of free societies. In the economic arena, they have been essential to the expansion of trade and the economic empowerment of individuals through greater wealth and a more flexible and productive economy. They have often been central to the maintenance of social peace and civil order. They have been the mechanism for protecting individuals and communities against hazards through the provision of a wide range of benefits and services, from unemployment insurance through old age provision to health care.

Moreover, all of this was done on a voluntary basis, without any resort to coercion or forced redistribution of resources. The historical evidence presented here should force us all to reexamine many common presumptions: that such "services" as law and police can only be provided by the state, that social welfare is best provided by some form of forced redistribution, and that voluntary cooperation in civil society labors under the burden of a crippling prisoners' dilemma.

All of this means that much of our currently accepted social and economic theory is contradicted by the facts of history. Consequently, there is a wealth of areas open to the eager iconoclast, for anyone who wishes to demonstrate that the idols of the modern academy have feet of clay. In economics and game theory, there is clearly a need for work which takes into account the actual historical experience, rather than trying to force it into the discredited model of classical public goods theory. We need in particular more work along the lines of that of Schmidtz, Klein, and some of the contributors to the Cowen volume. This work should seek to explain the precise mechanisms and dynamics involved in voluntary cooperation. One important question is that of the role and impact of technology; another, that of whether there may be an optimum number of participants for the formation of such voluntarist institutions as those described here.

History and sociology also have many opportunities for contribution. The former needs to do more work on the many historical instances of voluntary cooperation, most of which are relatively untouched by modern scholars. For example, there are almost no modern studies of friendly societies and mutual aid, in contrast to the enormous number of studies of state welfare and its history. Sociologists can explain how institutions such as suretyship interact with other social institutions and practices, and how the internal dynamics of societies can sustain or undermine voluntarist institutions.

All of this also has practical application to today's situation. Voluntary institutions such as surety and assurance embody norms of reciprocity, trust, honesty, fellowship, and thrift without which no stable social order is possible. The evidence shows that when these norms are articulated and expressed through voluntary action, they are enhanced and strengthened to everyone's benefit. Attempts to mimic the invisible-hand process that has generated them will not only fail; they will actively undermine and destroy these norms. Theory and empirical research combine to suggest four things: first, that such norms and institutions are needed for the successful functioning of any society; second, that the more complex the social order, the greater the need for them; third, that such institutions may appear spontaneously but cannot be deliberately created; finally, that much state action will undermine or destroy these norms and institutions, with potentially catastrophic effect.

Albert Loan is a graduate student in Economics at George Mason University.

Copyright 1992 by the Institute for Humane Studies.

Return to Winter 1991/92 Humane Studies Review menu IHS Publications or IHS homepage