For the past quarter-century the issue of state-sponsorship of terrorism has been a thorny diplomatic problem and a challenge for the scholarly study of terrorism. The debate over the subject, particularly over the alleged Soviet role in sponsoring terrorism, became mired in strident rhetoric colored by ideological differences and exacerbated by the absence of hard evidence. What terrorism scholars failed to provide to the debate at the time was a comprehensive and dispassionate analysis of the issue. With his book, Deadly Connections: States That Sponsor Terrorism, Georgetown University professor Daniel Byman has rectified that failure and in doing so has performed a great service to the academic study of terrorism. He brings to the subject impressive academic and professional qualifications, including staff work on the 9/11 Commission. The result is a clear-eyed analysis that will be of benefit to every scholar who studies terrorism. It also should be taken seriously by those in government service who have to wrestle with this complex problem.

Byman opens with the obligatory nod to the debate over definitions of terrorism. To the relief of this reader he does not dwell on it at length. He effectively adopts Bruce Hoffman's definition, adding only a caveat limiting the term to the deliberate targeting of non-combatants. Thus, in his view, al-Qaeda's attack on the USS Cole was not terrorism. This standpoint might be challenged by some terrorism scholars, but it does not detract from the central argument of the book. Indeed, Byman provides the clearest explanation this reviewer has seen on what state sponsorship of terrorism means. He defines it as "a government's intentional assistance to a terrorist group to help it use violence, bolster its political activities, or sustain the organization." (p. 10) The forms of help may include providing weapons, training, funding, sanctuary, strategic direction, and political support. His 'model' distinguishes between those states that sponsor terrorism wittingly and those that do so indirectly or simply because they are too weak to oppose the terrorists. He goes on to explain the different reasons why states offer sponsorship, the impact it has on the terrorist groups and their sponsors, and the constraints it imposes on counter-terrorism.

The author uses seven case studies to illustrate the effects — both positive and negative — state support has on terrorist groups and the sponsors themselves. These include four active sponsors: Iran, Syria, Pakistan, and Afghanistan; and three passive: Saudi Arabia, Greece, and the United States (the latter in respect of the Provisional IRA). The Soviet role in supporting terrorists, which is now a matter of documented record, is mentioned only in passing. (p. 1) That subject would benefit from the application of Byman's approach, if only to finally clear the air after the rancorous debates of the 1980s. But, apparently, that will have to wait for another scholar.

The cases show that training assistance is the most common form of support. They also indicate that funding is the most important. But, Byman argues, it is by providing sanctuary that state sponsors permit groups to become more powerful, since it allows them to recruit, organize, train, and plan safely beyond the reach of their enemies. Nevertheless, he makes it clear that sponsorship is a symbiotic 'two-way street' for terrorists and their sponsors, and not simply a unidirectional benefit for the terrorists. Sponsors support terrorism to advance their own goals and usually impose constraints on the terrorists' activities, if only because those activities can rebound against the sponsor.

This is where Byman's study makes a truly original contribution: examining critically the benefits and costs to sponsors. Iran and Syria were able to use their sponsorship of terrorism to further their strategic or ideological goals with some success. But this practice was not risk-free. It led Syria into disastrous confrontations with Israel, and Syria was blamed for Abu Nidal's attacks in 1985-86 simply for hosting the group on its territory. Likewise, Iran's support for Hezbollah while scoring successes against the United States, but at the price of diplomatic isolation and economic hardship. But nothing illustrates the 'down market' side of the terrorist/sponsor equation better than the case of the Taliban regime's support for al-Qaeda. That relationship was so unbalanced in terms of relative power and resources that the roles became reversed. Sanctuary in Afghanistan gave al-Qaeda the secure base it needed to prepare for its global jihad, a clear benefit to the group. But the Taliban were so dependent on al-Qaeda for funding and fighting capabilities that they could not afford to shed that alliance even when al-Qaeda escalated its attacks against the United States. Of course, it is not clear that they wished to do so anyway, and after 9/11 that sealed their fate.

Byman also offers a penetrating analysis of the problems that state sponsorship poses for those countries that wish to defeat or deter sponsored terrorists. A state's protection, support, and sanctuary make it difficult for the terrorists' enemies to deliver a 'knockout' blow against the group, bring them to justice, or de-legitimize their cause. Furthermore, although the Libyan case suggests that states can be persuaded or coerced into abandoning their terrorist clients, it also shows that the process is by no means easy. The instruments available to governments and the international community — political or economic pressure, engagement, using force, or supporting opposition groups — have limited utility; indeed, they can prove to be 'two-edged swords.' States which support terrorism primarily for strategic reasons may change their policies when the costs of sponsoring terrorism exceed the benefits. But radical regimes, such as Iran and the Taliban, have been particularly difficult to dissuade because much of their domestic legitimacy is invested in and rests on supporting like-minded revolutionary movements. Byman feels that such regimes will abandon that support only when their ideological fervor is superceded by pragmatic concerns, such as economics.

Byman closes the book by suggesting some guidelines for governments wishing to persuade state sponsors to abandon their clients. First, he emphasizes that they need to understand what the state gets from sponsorship, as this will determine the most effective tools of dissuasion. Second, they must offer sponsors a way out that provides incentives for change without appearing to reward past support for terrorism. Third, governments must set priorities and stick to them. Byman favors a multilateral approach and creating a strong norm against state sponsorship. Finally, he urges governments to adopt realistic expectations about what can be achieved. They must be willing to settle for degrees of progress over time rather than expecting and requiring complete success.

The book may be criticized on several fronts. As suggested earlier, Byman's choice of definition is likely to leave him open to attack, but that would be the case no matter which definition he chose. It is not an argument he could ever win. Critics might be on stronger ground questioning his heavy reliance on official and secondary sources and the lack of original documents. The author has had to infer motives and other assumptions from less than authoritative sources, which leave his conclusions open to challenge. Yet, given the secretive nature of terrorist groups and their sponsors, and of the security organizations that work against them, it is hard to suggest what other sources he might have used. He has made good use of material in the public domain to develop solid analysis and sound judgements that do not go beyond the evidence. Nor has he suggested that his is the last word on the subject. Rather, Byman has provided an analytical model for understanding the phenomenon of state sponsorship of terrorism that should encourage other scholars to revisit the cases he discussed as well as those he did not explore. That is the major contribution of this work, and it is no small achievement.

David A. Charters is a Professor in the Department of History at the University of New Brunswick.