Guilty by Association: “Small World” Implications

for Business Strategies in Post 9/11 America

[Working Paper]

By

Carol M. Bast [*][†] and Cynthia A. Brown [**][†]

I. Introduction

Presented with questions concerning the Federal Intelligence Surveillance Act (FISA),[1] the majority of the country’s business leaders and their legal counsel would acknowledge a lack of familiarity with this particular chapter in the annals of federal legislation. Historically, in fact, most business owners possessed little reason for even the most meager exposure to this body of law. The enactment is traditionally associated with foreign intelligence surveillance and the prevention of terrorist activity, and consequently, it is doubtful that FISA would alert further inquiry by the commercial sector, even had it garnered the attention of a business owner or her counsel. As is true of most aspects of America, September 11, 2001 changed things. FISA’s application since 9/11, for instance, and Congress’s responsive amendments to the FISA statute suggest an amplified need for businesses to engage in a focused examination of the law and its prescription for, not proscription of, surreptitious surveillance of domestic commercial activities. The suggestion of need escalates to something much more akin to demand when today’s business interactions are viewed through the lens of network theory’s “small world” problem and the heightened potential for unintended consequences wrought by a business’s lack of knowledge of its own social networks.

This article examines the implications of Milgram’s small world problem theory and the associated concept of “six degrees of separation” for American business strategies subsequent to the September 11, 2001 terrorist attacks. The focus is in many respects one of risk management, and the authors present potential new hazards posed by contemporary business interactions given the latitude FISA confers to American law enforcement agencies. Milgram and those that conducted subsequent studies suggest that the average network of acquaintances that connect any of us numbers no more than six and as few as three. Further, very few of us are cognizant of these networks we create, and more particularly, who is included in these networks. We may be unaware of our social networks, but the networks we create through our acquaintance links have received a great deal of scholarly attention. And, scholars are not alone. Our networks have also garnered the attention of the Executive Branch of the United States, and since 9/11 that attention has only intensified. The authors hypothesize that the invisible social networks created by employees and business owners may unwittingly position a business and its principals as federal surveillance targets and the subject of surreptitious wiretapping, eavesdropping and “sneak and peek” searches. Moreover, the business may never learn of the government monitoring, and under current legislation government disclosure is not mandated. Post 9/11 national security efforts may be contributing to the creation of a desert of fear influencing zealous prosecutors and law enforcement officers to see mirages of treasonous conduct by American businesses. Routine business networking and business development efforts could be leading to commercial relationships that place some American businesses on the government’s list of terrorism suspects and within law enforcements’ surreptitious surveillance efforts.

Part II of this paper briefly reviews the small world experiment and the popularization of the concept of six degrees of separation. The discussion includes later network theory studies and their findings relevant to this paper. Part III presents the FISA legislation and its thirty-year evolution. The article incorporates a comparison of other federal legislation outlining domestic surveillance procedures. Part IV urges businesses to recognize the acute need to consider strategies to forestall the unintended consequences of FISA. Several suggested strategies are included.

II. Milgram’s “Small World” Problem and the “Six Degrees of Separation” Phenomenon

In 1929, Hungarian author Frigyes Karinthy penned a short story entitled Chains[2] wherein we find the earliest published suggestion of the phenomenon more recently dubbed “six degrees of separation.”[3] Karinthy writes, “[t]o demonstrate that people on Earth today are much closer than ever, a member of the group suggested a test. He offered a bet that we could name any person among earth’s one and a half billion inhabitants and through at most five acquaintances, one of which he knew personally, he could link to the chosen one.”[4] Through his fictional character Karinthy hypothesizes that everybody on this planet is connected to everyone else through a chain of fewer than five mutual acquaintances. Karinthy’s character links a “Nobel prizewinner to himself, noting that the Nobelist must know King Gustav, the Swedish monarch who hands out the Nobel prize, who in turn is a consummate tennis player and plays occasionally with a tennis champion who happens to be a good friend of Karinthy’s character.”[5]

As original as the thought was in 1929, the concept that complete strangers could be linked so closely appeared more improbable three decades later as the world’s population seemed to explode.[6] Nevertheless, in the 1950s MIT political scientist Ifthiel de Sola Pool and IBM mathematician Manfred Kochen began questioning what they referred to as “human contact nets” and asking how many acquaintances connected any two random people selected from completely different stratum or population.[7] They hypothesized that increased social stratification reduces the number of intermediaries or links between persons in the same stratum but increases the number of “degrees” between persons in different stratum.[8]

These researchers set about creating a theoretical mathematical model of their version of what Milgram would later denominate the small world problem. In order to do so, they had to first determine the average person’s number of acquaintances; a reliable fact was indeterminable from the social science research available to them. Their research results suggested that the Americans in their sample had acquaintances numbering as high as 2,500 peopled with an average number of approximately 500 people. Possessing this information, Pool and Kochen were ready to tackle the larger question – what is the likelihood that any two people actually know each other. Their mathematical experiment suggests that there is only about one chance in 200,000 that any two Americans selected at random will, in fact, know each other. The odds dropped dramatically when considering whether the two people share a mutual acquaintance. Surprisingly, however, the research indicated there was even a greater probability, better than a 50-50 chance, that the two people will be linked by two intermediaries.[9]

It was, however, a Harvard social psychologist who shared Hungarian roots with Karinthy that pioneered this connection theory and conducted the inaugural social science study of human connectivity. Stanley Milgram enticed by these social networking questions designed the classic experiment that would demonstrate how closely linked earth’s inhabitants might be.[10] He named his study the “small world” problem after the cliché response of strangers who unexpectedly discover they share an acquaintance.[11] Milgram’s work provided the initial empirical evidence that any individual may be connected to any other through a short chain of social ties.[12] The “small world” experiment received a great deal of attention among academic circles across multiple disciplines, but it would take nearly a quarter of a century and John Guare’s play entitled Six Degrees of Separation (1990) before Milgram’s concept escaped the bounds of academia and achieved the popular prominence it now holds.[13] Milgram’s classic work would become part of popular culture and weave itself into the fabric of urban folklore.[14]

A. Stanley Milgram and the “Small-World” Problem

In 1964, Wichita, Kansas provided the setting for Milgram’s “small world” problem inaugural study (the “Kansas study”). He recruited Kansas residents as study participants by placing advertisements in The Wichita Beacon and The Wichita Eagle.[15] According to Milgram, Wichita, Kansas seemed sufficiently distant from Cambridge, Massachusetts to satisfy the “somewhere distant” component of the random sampling of participants who would initiate the acquaintance chain.[16] The objective was to supply the Kansas participants or “starters” with a very official appearing packet and the identity of a person who would serve as the “target” recipient of the packet. The study’s design sought to diminish the possibility that any starter would know the intended target. Along with the identity and address of the target, Milgram provided participants with information about the target’s present and past occupations, educational status and place of childhood.[17]

The participants received the charge to begin a chain by mailing the packet towards the target, but they were restricted to sending the packet only to individuals they knew on a first-name basis. Milgram asked the participants to select a friend as a recipient, but the friend was to be someone who in the participant’s opinion would be more likely to be able to succeed in reaching the target or who would be more likely to know someone else who could more probably reach the target. That friend after receiving the packet would then repeat the process. Each recipient of the packet would continue the chain by serving as a “degree” or intermediary along the packet’s path. The total number of intermediaries required to complete the chain and to successfully deliver the packet to the target represented the number of degrees that separated the Kansas starter from the Massachusetts target.[18]

In order to maintain a record of the intermediaries in each chain, Milgram instructed each person who received the packet and who then forwarded it to a friend to also mail a business reply or “tracer” card back to Milgram at Harvard University. In his pilot study, 60 Kansas residents responded to Milgram’s advertisement, and a total of 40 chains were actually started. Only three of the 40 chains or approximately 8 percent, however, were completed. From this information and the 145 postcards received, Milgram’s pilot study of social connections revealed that the participants were separated, on average, by six degrees.[19]

Milgram would conduct his subsequent study recruiting “starters” from Omaha, Nebraska (the “Nebraska study”). The “target” recipient was a stockbroker who worked in Boston, Massachusetts, and lived in Sharon, Massachusetts. For the second time Milgram would strive to determine how many people would be required to secure the target’s receipt of each packet. In other words, how many people would exist in the chain of mutual acquaintances linking a participating Nebraska resident with the presumably unknown Boston stockbroker. If a participant happened to know the stockbroker on a first-name basis she was to mail the envelope directly to him. Otherwise, participants were restricted to addressing the envelope to someone they knew on a first-name basis but who was someone more likely to know the stockbroker. The Nebraska study revealed, again, five or fewer mutual acquaintances were in the chain linking the original Nebraska participant and the Boston stockbroker.[20]

In May of 1967, Milgram first published the results from his human connectivity experiments in Kansas and Nebraska, studies he referred to as acquaintance chains[21] in the premiere issue of Psychology Today. Somewhat unusual for an academician, Milgram first published his study results in a lay publication rather than an academic journal. His efforts were initially directed at the masses rather than targeting his colleagues and other scholars. He never intended for this early publication to serve as a technical report.

Following the procedures employed in the Kansas and Nebraska studies Milgram and a colleague replicated the “small world” problem experiment. In this replicated study, Milgram and Travers again selected a Boston stockbroker but introduced an experimental variation in the procedure. The researchers varied the starting populations to create three distinct subpopulations from residents of both Nebraska and Massachusetts. Milgram and Travers selected a total of 296 starters from the three groups. They selected two subpopulations from Nebraska by soliciting respondents from mailing lists. These two groups furnished participants who were geographically removed from the designated target. One Nebraska subpopulation consisted of 100 mid-westerners systematically selected because they owned blue-chip stocks and had access to the investment business sector. The second distant subpopulation consisted of 96 Nebraska residents chosen from the population at large. The remaining 100 volunteers were formed the third subpopulation, Massachusetts residents solicited through a Boston newspaper advertisement.[22]

As in the Kansas and Nebraska studies, the 296 initial volunteers received a document which served as the principal tool of the experiment. The researchers supplied the volunteers with information similar to that provided to previous participants.[23] Of the nearly 300 starters selected to participate 217 people actually began chains by forwarding a document to a friend. There were a total of 453 intermediaries who received the folder and were included among the chains. Sixty-four folders or 29 percent successfully reached their targeted destination, the Boston stockbroker. Thus, the study results indicated that the mean number of intermediaries required to connect a starter with the intended target was 5.2 links. [24] The replicated study results, therefore, were consistent with the results from both the Kansas and Nebraska studies.

The Milgram and Travers paper constituted the first technical report employing the “small world” method designed by Milgram. The authors’ primary objective was to once more investigate the minimum number of acquaintances necessary to link two individuals selected randomly from the population.[25] The authors noted, however, that their experiment expanded the Kansas and Nebraska studies’ areas of inquiry to include the ancillary aims of resolving the type of information people would use to strategically move the folder towards the target[26]; the internal structures of the distribution of the chain lengths[27]; the degree of homogeneity in age, sex, occupation, and other characteristics of participants within the chains[28]; and the results of comparisons of complete and incomplete chains[29].

One of the significant conclusions of their study and one of the findings more relevant to this paper is Milgram and Travers’ observation that “people can rarely see beyond their own acquaintance chains; it is hard to guess the circles in which friends of friends – not to mention people even more remotely connected to oneself – may move.”[30] The authors suggest that this point of the small world problem becomes particularly poignant for business strategies in post 9/11 America.

Ultimately, Milgram and Travers concluded that the tangible contribution of their study lies in the investigation “of acquaintance chains to extend an individual’s contacts to a geographically and socially remote target, and in the sheer size of the population from which members of the chains were drawn.”[31] Further, their experiment “demonstrated the feasibility of the ‘small world’ technique, and took a step toward demonstrating, defining and measuring inter-connectedness in a large society.”[32]