Trading on reputation: stateless justice in the Medieval Mediterranean.

Institutions and the Path to the Modern Economy: Lessons from Medieval Trade, by Avner Greif, New York: Cambridge University Press, 526 pages, $80

IN THE EARLY 11th century a trader from Tunisia, momentarily residing in Sicily, wrote a long, distraught letter to a business associate in Egypt. Sumhun ben Da'ud complained to his correspondent, Joseph ben 'Awkel of Fustat (old Cairo), that "my reputation is being ruined." Sumhun had asked Joseph to pay two of Sumhun's creditors in Fustat, and Joseph hadn't done it. Now Sumhun complained that those unpaid creditors were writing to parties all along the Mediterranean basin, letters "filled with condemnations," which have "reached everyone."

Sumhun and Joseph were both members of a social group, the Maghribi traders, that emerged in the 10th century. As the social and political environment in Baghdad under the Abbasids deteriorated, many Jewish traders left, heading for the western part of the Arab world, the "Maghrib." They were ignored by and in turn largely ignored the Fatimids, the nominal rulers of their new home.

Avner Greif, an economic historian at Stanford and a 1998 recipient of a MacArthur Foundation "genius" grant, diagnosed the social context of this letter and related documents in a brilliant series of articles in the late 1980s and early '90s. His work from those years is summed up in his latest book, Institutions and the Path to the Modern Economy. It tells a story whose implications cut to the heart of both political philosophy and political practice.

If there's one thing about the nature and purpose of government that almost everyone agrees on, it's that a state-supplied judicial system is indispensable. Even those who see the state as a necessary evil make sure to include the judicial system in the "necessary" part. But the story of the Maghribis is the story of a complicated system of international trade that mostly refused to use government courts.

The Fatimid dynasty, founded in 909 by Abdullah al-Mahdi, was on the rise just as the Abbasids to the east were in decline. When the Jewish emigres who became the Maghribi traders began arriving, Fatimid terrain consisted of present-day Morocco, Algeria, Tunisia, and Libya. That realm soon expanded, incorporating Egypt in 972, then Syria, Sicily, and parts of the Italian coast somewhat later.

As Greif tells it, the Fatimids had a weak bureaucracy that saved itself administrative trouble by relying on community associations. If the newcomers wanted to govern their own affairs among themselves, they were welcome so to do. And they did. The emigres developed a new identity. In time, they came to write casually of "our people, the Maghribis, the travelers." These travelers proved very successful, claiming a niche of expertise in long-distance commerce. Greif gives an example of their influence when he describes how, around the year 1050, "the Muslim ruler of Sicily imposed a 10 percent tariff (instead of the 5 percent tariff specified by Islamic law) on goods imported to Sicily by the Maghribi traders. The traders responded by imposing an embargo and sending their goods to the rival trade center, Tunisia. The embargo was effective; after a year the Sicilian ruler removed the extra tariff."

Some of the Maghribi traders' success came from avoiding the judicial system of the Fatimids. The Maghribis had access to two legal systems, the Muslim or the Jewish. As residents of the Islamic world, they could sue within the Muslim legal system. Separately, the Fatimid caliphate recognized the Jewish legal system as having authority over those who chose to use it. But the Maghribi traders saw both as inefficient, uninformed, and time-consuming.

In a paper published in the American Economic Review 13 years ago, Greif elaborated more on this point than he does in his most recent book. That paper cited one of the rare occasions when Maghribis did invoke the courts. The original plaintiff's grandchildren were still seeking redress on their grandfather's behalf 50 years later. No wonder the Maghribis generally avoided such a Jarndycean fate, preferring what Greif calls an "informal code of conduct." They relied, to put it simply, on each other's handshakes.

When the end came, it was because the Maghribis lost the umbrella of Fatimid tolerance and laissez-faire policies. This occurred in two stages. Late in the 11th century, the rising naval power of the Italian city-states limited the ability of the Maghrihis to trade profitably across the Mediterranean. Accordingly, they moved. They remained a cohesive social group but shifted their domicile and activity to the Red Sea and the Indian Ocean, where they now became the intermediaries between Egypt and the Far East. Egypt was by this time the core of what remained of the Fatimid caliphate, as the remainder of its once-broad realm was either under attack or in revolt.

Two-thirds of the way through the 12th century, the final Fatimid redoubt fell. Saladin conquered Egypt on behalf of the Damascus-based Zengid dynasty, and soon thereafter he declared himself sultan. Once in power, he prohibited any non-Muslim from engaging in trade between Egypt and the Far East. As a result of that edict, the Maghribi traders could no longer carry out their unique role and faded into the broader Jewish community.

When the Maghribi traders Sumhun and Joseph were having their differences in the early 11th century, though, they were living through the height of both Fatimid and Maghribi success. The two correspondents had a personal history. Not a "vertical" history, as employer to employee, but a horizontal history, in which either man could act as the principal and the other as his agent for a particular transaction. The horizontal nature of Maghribi relationships was, Greif tells us, crucial to their coalition.

This particular relationship, though, had soured. Joseph must have concluded at some point that Sumhun was cheating. What was he to do? He wouldn't appeal to the Fatimid courts. The Maghribis took care of their own affairs through a sort of shunning--or, in the more formal language Greif prefers, a "multilateral reputational mechanism" for punishment. So Joseph didn't pay Sumhun's creditors, and presumably explained to them why he wasn't doing so. This led to their letters to other Maghribi acquaintances and to Sumhun's plaintive letter quoted above.

The informality of Maghribi relationships allowed great flexibility. Concerning one transaction, a trader in Lebanon wrote to his partner in Fustat: "Do whatever your propitious judgment suggests to you." Behind this simple statement is an understanding that time lags in communication, the vagaries of wind and tide, and the certainty of the unpredictable required a lot of ad hoc, on-the-spot decisions for which detailed formal contracts, drafted with an eye to potential litigation, could be a hindrance. Greif also quotes a pepper-trading merchant who displayed much the same spirit of mutual trust and flexibility. This fellow wrote to his partner after a transaction involving two loads of pepper ended with an unexpected outcome. "Settle my account with yourself and give the balance to my brother-in-law," he said, "for you are a very busy man."

A fascination with the Maghribis has been a constant in Greif's career--they were the subject of his doctoral dissertation--but over time he has seemed increasingly interested in putting some distance between himself and the near-anarchist implications of their relationship to the state. (The Maghribis needed the Fatimids to protect them from other sovereigns but otherwise wanted to be left alone. They certainly didn't want the Fatimids interpreting or enforcing their handshake deals.) So while the theme of Greif's early discussions of the Maghribi traders might well have been that they succeeded without the state, even in its role as interpreter and enforcer of contracts, the theme of his more recent writings has been that they were a historical dead end because they weren't intertwined with the state whereas the merchants of Genoa and Venice, a little later, demonstrated the way forward. In the 12th century, Greif tells us, the Genoese "ceased to use the ancient custom of entering contracts by a handshake and developed an extensive legal system for registering and enforcing contracts." This extensive legal system also included bills of lading and, in time, double-entry bookkeeping. Greif argues that these Genoese legal innovations made the rising city-state integral to commerce.

Maybe. But allow me to make a few points on behalf of the Maghribi way of doing commerce.

To an extent, Greif is contending that the Italian model proved superior because, from the perspective of history, it was the one that worked. The Italian cities set off the great expansion of the West, which in time underwrote colonial empires that carved up the rest of the world. But history is filled with contingencies. Greif himself notes that the Maghribi model disappeared not because of an internal failure but because of changing external circumstances. The Italians didn't outcompete the Maghribis; they merely had the good fortune to be situated in a more hospitable political environment.

And institutions aren't all-or-nothing packages. The Maghribis might have adopted elements of the Genoese system, such as bills of lading or the new system of accounting, had they survived into subsequent centuries. But that doesn't mean the results would have looked like Genoa.

Beyond the evidence of history, Greif offers a game theory argument that the Maghribi model isn't replicable because it only works within a small community, one where everybody can know everyone else and have a sense of who deserves shunning, through a process akin to that of small-town gossip. As long as "the Maghribi traders coalition survived," he writes, "its functioning crucially depended on maintaining an appropriate size.... [A] larger coalition implies a slower circulation of information and hence delayed punishment." Any delay of punishment is a discount in its value and accordingly a decrease in its effectiveness. Even if the small town is spread out through much of the Mediterranean, from Gibraltar to Asia Minor, it still retains its narrow demographic limits. Given those limits, it is inevitable that at some point any actual such coalition will be outflanked and outlasted.

But the considerations that require smallness of such a coalition, in Greif's model, were quite specific to the period. They are technology-dependent. The speed of communication is an obvious example. An increase in the number of people on my email list doesn't slow the speed with which I can communicate with all of them. Similarly, Greif tells us it was difficult for medieval parties to "retaliate collectively against a cheater not personally known to them ... due to the challenge of describing him to those who were not cheated." Does this amount to saying that the courts of a sovereign were necessary in a world without photography?

Furthermore, it obviously isn't the case that Maghribi only did business with Maghribi. They only entered into agency relations (requiring promises and trust) with one another, and they dealt in cash on the barrelhead with the rest of the world. But common experience indicates that dealings of the latter sort can become dealings of the former sort, or at the very least that they could become sufficiently habitual and amicable to constitute the glue of a broader coalition. In some ways--some ways--you could compare the Maghribis with certain Anabaptist communities in the United States in the 20th and 21st centuries. Mennonites and the Amish typically have informal, "horizontal" relations with one another, enforce their internal norms through ostracism, and try to avoid outsiders' court systems. These traits have not prevented them from fruitful interchange with the surrounding society. The Maghribis, of course, were much more outward-looking than the Amish.

Greif is an indefatigable researcher and a wonderfully acute thinker, and I recommend his work highly. But he may have erred in treating the Maghribi way as a road to nowhere. It might be better conceived as a road not taken.

Christopher Faille ([e-mail]) is a correspondent for Lipper HedgeWorld, a Web-based investment news service.