Sagan, Scott D. "Why Do States Build Nuclear Weapons?: Three Models in Search of a Bomb". International Security, Vol.21, No.3 (1996): 54-86.

Sagan, Scott D. "Why Do States Build Nuclear Weapons?: Three Models in Search of a Bomb". International Security, Vol.21, No.3 (1996): 54-86.

• States build nuclear weapons for three reasons: the security model, where nuclear weapons defend the state against foreign threats; the domestic politics model, where nuclear weapons fulfill domestic political goals or the aspirations of state bureaucracies; and the norms model, where nuclear weapons are an important symbolic element of national identity, normally linked to claims about modernity (55).
• The security model is based on neorealist IR theory that states exist in an anarchic self-help system, assuming that strong states build nuclear weapons because they can, while all other states build nuclear weapons as deterrents against invasion or ally to a nuclear power. The most common expression of these dynamics is a domino effect, by which nuclear proliferation to reduce a threat posed by a rival threatens others, thus encouraging further proliferation (57-58).
• This was the case during and following World War II, as the great powers initially rushed to build the bomb because they feared another countries obtaining it first. The American use of nukes on Hiroshima and Nagasaki in August 1945 compelled the USSR to develop nuclear weapons to protect itself against the USA, while Soviet nuclearization pressured Britain and France to develop nukes. Hostilities with both the USA and the USSR led China to develop its own nukes. Chinese nuclearization threatened India and led to Indian nuclearization, which in turn threatened Pakistan and led to Pakistani nuclearization (58-59).
• Most recent examples of denuclearization have only occurred through massive political change that totally rearranged perceptions of threats to national interests. This occurred in South Africa after the end of the Cold War, in South America through the sharp reduction in tensions between Brazil and Argentina, and in Belarus, Ukraine, and Kazakhstan through joint Russian and American security guarantees (60-61).
• The main implication of the security model is that, since states will pursue nuclear weapons if not adequately protected, US nuclear commitment to allies, including a first use policy, are critical to nonproliferation. The general commitment of nonnuclear states to the Non-Proliferation Treaty is also very important, as it forestalls regional nuclear arms races (61-62). These goals have become more difficult to achieve since the end of the Cold War, which has created a multipolar world in which US security guarantees are less trusted (62).
• Much of the scholarship underlying the security model is bad, since these scholars often adopt the theory that national security threats caused nuclearlization, then work back to identify the threat. They do not consider other possibilities and do not conduct rigorous analysis (63).
• The domestic politics model says that the development of nuclear weapons is driven by the interests of the nuclear energy establishment, the military bureaucracy, or popular political movements. Rather than responding to national security interests, bureaucratic actors shape perceptions of national interest to encourage extreme perception of threat and thus encourage the development of nuclear weapons (63-64). This theory dovetails with the security model, but answers the question at what point it is decided that foreign threats require a nuclear deterrent (65). 
• The domestic politics model more accurately explains Indian nuclearization than does the security model. Chinese nuclearization started a conflict between the Indian nuclear energy establishment, which wanted to build a nuke and fraudulently lowballed the cost of a program in its statements to the government, and the Indian government, which believed that a nuclear program would be too expensive (66-67). Indira Gandhi's decision to approve Indian nuclear tests in 1974 appears to have a political motived and used to boost public support for her government during a time of severe economic hardship and widespread unrest (67-68).
• Contrary to the expectations of the security model, India neither started its nuclear weapons program nor sought security guarantees from nuclear powers in response to Chinese acquisition of the bomb in 1964 (65-66).
• The initial nuclearization of South Africa also appears to have been motivated by lobbying by the nuclear energy establishment and mining interests to acquire more government funding, as the program predates the stated reason for nuclearization: Cuban intervention in Angola in October 1975. Moreover, the military was not consulted about many aspects of the nuclear program, resulting in bombs that actually could not be delivered by air and did not meet certain safety requirements (69-70). 
• The denuclearization of South Africa in 1989 was also more motivated by domestic political concerns than by international conditions, as the de Klerk government's initial request to dismantle the nuclear weapons came in September 1989. Furthermore, the destruction of nearly all information from the program indicates that one goal was to prevent the nuclear weapons from falling into the hands of the ANC or White extremists (70-71).
• The end of nuclear programs in Argentina and Brazil in the 1980s is also better explained by the end of military governments and the emergence of liberal civilian governments in both countries. These new liberal government were dominated by banking and business interests who viewed nuclear programs are unnecessarily costly pet projects of the military (71).
• The domestic politics model implies that the decision to nuclearize is internal and cannot be greatly affected by US policy. Tools to curb proliferation would have to be more indirect, such as making aid conditional on reduction in military or nuclear research budgets, supporting anti-nuclear scaremongering propaganda campaigns, supporting civilian control of the military, and giving ambitious nuclear laboratories the opportunity to work with American labs (71-72).
• The Non-Proliferation Treaty is an important document within the domestic politics model, as it provides a nexus for domestic opposition to nuclear weapons, cultivates a foreign policy establishment committed to maintaining the nonproliferation regime, and subjects governments to IAEA pressure. The commitment of great powers, like the USA, to banning nuclear tests and first use doctrines also sets norms that could be followed by political actors in non-nuclear countries (72-73).
• The norms model says that decisions to acquire nuclear weapons are determined by notions about whether this action is internationally acceptable and whether it fits within a constructed national identity. International institutions and norms designate certain behaviors as rational and legitimate, while degrading others; these categories influence state action. Nuclear weapons are pursued because states view them as an important part of fulfill national identities as 'modern' states (73-74).
• Similar to how abolitionist norms only ended slavery because they were adopted by powerful states like Britain, norms of nuclear proliferation only have an impact because major countries have adopted certain norms through institutions like the Non-Proliferation Treaty (75-76).
• Norms regarding proliferation have massively changed over the past several decades, as seen in the difference in normative pressures affecting France in the 1950s and Ukraine in the 1990s (76). In the 1950s, nuclear weapons were a symbol of being a 'modern' and 'important' nation -- which was particularly important to France after the loss of Algeria in 1958 (78) -- whereas in the 1990s, Ukrainian retention of nuclear weapons would have been condemned as 'irrational' and punished by the USA and other major powers (76, 81-82). These new norms have largely been cemented by major countries working to raise the prestige of adhering to the Non-Proliferation Treaty (76, 82).
• Neither the security model nor the domestic politics model can adequately explain Ukrainian denuclearization: it still faced threats from Russia over Crimea and Russian minorities, and there was widespread support for retaining nuclear weapons, including among the military and political establishment (80).
• Ukraine also supported denuclearization as a way of asserting its foreign policy independence from the USSR, as a movement to destroy its nuclear arsenal had started in 1990 and was supported as a way to show the Ukrainian government's distance from Soviet leadership (81).
• The security model explanation for French nuclearization, that the failure of the 1956 Suez War led France to seek nuclear weapons to reduce vulnerability to Soviet pressure and dependence on the USA, does not account for the fact that the Mendès France government started its nuclear program in 1954, nor that nuclear Britain also capitulated to Soviet and American pressure over Suez in 1956. Additionally, no other European states with nuclear capacity pursued nuclear weapons during this same time period, despite a buildup of Soviet forces (77-78), and Charles de Gaulle did not ever believe a Soviet attack in Europe was likely (78-79).
• The connection between French national identity, prestige, and nuclear testing was so ingrained that France was very reluctant to stop nuclear testing even in the 1990s (79-80).
• The Non-Proliferation Treaty has been extremely important in changing international norms and attitudes about nuclear weapons. whereas, nuclear weapons were once a symbol of pride and gave a country the same status as the USA, USSR, or China, unauthorized nuclear arsenals now relegate countries to the category of pariahs like Iraq, North Korea, Pakistan, and India. The Non-Proliferation Treaty also makes denuclearization look legitimate and noble, rather than unpatriotic (82).
• The norms model implies that the main goal of US policy should be to shape international norms on nuclear weapons, while also recognizing that US policy is itself shaped by the norms imbedded in agreements like the Non-Proliferation Treaty. The USA should try to enhance the prestige of non-nuclear nations that may attempt to acquire bombs, possibly by offering them UNSC membership. The most important move is for the USA to support the Non-Proliferation Treaty, although certain provisions, like abandoning its first strike nuclear policy, should be pursued with great caution as they may have other consequences (82-84).
• Most American policymakers and IR scholars assume that states will develop nuclear weapons when they face a military threat that cannot be repealed through conventional means. If this is not the case, then states will remain non-nuclear (54).
• Although it explains the majority of cases of nuclearization (85), this explanation is inadequate, as it does not recognize that nuclear weapons also serve nonmilitary ends in domestic politics and in bureaucratic infighting (55, 85). Any successful counter-proliferation policy will need to recognize that different states pursue nuclear weapons for different reasons at different times (56-57, 85).
• Nuclear proliferation is an important issue, even after the signing of the Non-Proliferation Treaty in 1995, because that treaty still allows states to withdraw given 3 months notice, and because an increasing number of countries have the industrial and scientific base to theoretically create nuclear weapons if they so desired (55-56).
• The increasing number of states with the ability to create nuclear weapons demonstrates that we must pay attention to the reasons why states want to build nukes in the first place, since -- unlike attempted nuclear proliferators like Iran, North Korea, Iraq, and Libya -- it will be extremely difficult to control the supplies needed for countries with a latent nuclear weapons capacity to accomplish full nuclearization (56).
• Most of the policy recommendations based on different models of nuclearization are compatible, but there is tension between the norms model recommendation that the USA commit utterly to nonproliferation, and the security model recommendation that American nuclear guarantees are a critical part of nonproliferation (86).