Pre-Emptive Action Including Air Strikes or Sabotage Operations are not Effective Non-Proliferation Tools

Assessing pre-emptive action as a non-proliferation tool through the contrasting cases of  Iraq’s Osirak airstrike and Iran’s advanced nuclear programme.

This article will explain that pre-emptive actions are not effective tools of non-proliferation, for which evidence is drawn from two cases, those being Iran and Iraq. It will be shown that the effectiveness of pre-emptive actions is strongly influenced by contextual factors like the level of development of the targeted nuclear programme. The first section will outline how the question is understood, and that effectiveness is defined as being a long-term impact, not a delay of nuclear proliferation. The case of Iraq will focus on the pre-emptive action taken until 1981,  covering the Osirak airstrike and the overestimated effectiveness attributed to the airstrike.

The section will prove that pre-emptive actions had only been effective as long as Iraq had been dependent on foreign support but failed to produce conditions for non-proliferation, as it not just drove the program underground but even stimulated a more sophisticated attempt to construct a nuclear bomb. Worse, a wrongly assumed effectiveness of the airstrike on Osirak led to a neglect of Iraq’s ability to rebuild.

The case study of Iran will create a contrast to a more advanced nuclear programme and thereby show that the pre-emptive actions used so far have not been effective. At first it will be explained that a theoretical airstrike on Iran would not just be ineffective due to the lack of intelligence on the clandestine programme but also potentially lead to war.

The pre-emptive actions used, those being assassinations and sabotage, will be shown to have been ineffective tools of non-proliferation because Iran’s programme is already too advanced to be seriously hindered by them. The potential of cyber-attacks is at the end also explored through the case of Stuxnet, which had also been much less effective then claimed in 2010.

Scope and Definition

The primary interest of this essay is to explore the effectiveness of pre-emptive actions as tools of non-proliferation which means that all actions taken must be judged against their impact upon the targeted country’s ability to potentially produce nuclear weapons. Actions can be considered effective if they have either frustrated the targeted country’s efforts to the point that the country itself relinquishes its nuclear programme or forced by the damaged cause to submit to outside pressure. This is done by either breaking the targeted countries political will^[1] to continue the nuclear weapons programme or through inflicting unbearably high economic costs^[2].

The pre-emptive actions have therefore to create long-term policy changes to be considered effective not merely cause delays which are not understood to be a satisfactory outcome. The reason being that postponing nuclear proliferation is simply gambling on future conditions in the hope that they are better, which was the rationale behind Osirak^[3].

The greatest limitation of this essay which must be kept in mind is that because of the pre-dominant focus on pre-emptive actions are other tools of coercive diplomacy ignored like sanctions on Iran’s nuclear programme. Therefore, must further studies be conducted to explore if pre-emptive actions become more effective when utilised alongside other tools of coercive diplomacy.

Iraq

Israel’s pre-emptive campaign started in the early 1970s in response to the discovery that France was assisting in Iraq’s nuclear programme^[4]. While Israel would keep plausible deniability were the acts of sabotage and assassination as outlined by Uri Sadot closely linked to Mossad^[5]. Sadot makes a compelling argument for the effectiveness of the pre-emptive actions but does acknowledge that the goal was to delay Iraq^[6]. As already stated, is that not an effective approach to non-proliferation because it did neither force Iraq to abandon the programme nor stopped their progress.

So, did the assassinations of Iraqi nuclear scientists caused some loss of expertise^[7] but failed to stop the nuclear programme from advancing. Rather than pre-emptive actions, was it Saddam himself who delayed the nuclear programme by arresting key scientists like Jafar and Shahristani as Shiite conspirators^[8]. The assassinations were therefore ineffective tools of non-proliferation whereas the sabotage campaign was effective in cutting Iraq off from its foreign supplier. This was done by targeting the facilities of the

Italian company SNIA with bombings^[9] while subjecting French scientists to a prolonged harassment campaign^[10]. Fearing for their families lives an increasing number of scientists requested to be transferred to other programs cutting Iraq off from valuable sources of expertise^[11]. The mounting media pressure on the French government to stop supporting Saddam’s nuclear programme¹² was only intensified following the bombing of containers holding nuclear reactor parts intended for shipping to Iraq, in Toulouse^[12].

Caving to public pressure the French government stated it would only supply LEU to Iraq’s reactor in the future^[13], thereby preventing Iraq’s from acquiring more HEU for a nuclear weapons programme. While not a permanent success were the sabotage actions effective and as such it can be stated that pre-emptive actions can be effective tools of non-proliferation should as in this context the targeted country be dependent on outside support. An alternative to those covert pre-emptive actions are direct kinetic actions like the airstrike on the Osirak reactor which shall be looked at next.

The significance of the 1981 airstrike can only be understood if it is appreciated that the facility constituted what Kreps and Fuhrman have called chokepoint facilities. Those are facilities which are irreplaceable for the continuation of the nuclear programme^[14]. Osirak was such a facility because its destruction and the refusal of France to rebuild it^[15] meant that Iraq could no longer acquire nuclear weapons through the plutonium reprocessing path.  It can therefore be stated that depending on the context pre-emptive actions can be effective as tools of non-proliferation because it prevented Iraq from pursuing the plutonium reprocessing path.

This rests on the assumption that the reactor could have been used as part of a nuclear weapons programme which has been called into question for example by Dan Reiter^[16], which would make the strike otherwise ineffective. But at the time it was assumed that the strike had destroyed Iraq’s nuclear capabilities^[17] which was an overestimation as it never destroyed the whole nuclear programme. Worse, it installed a false sense of security in the international community which regarded the programme as crippled and therefore underestimated Iraq’s ability to rebuild^[18].

Under this cover Saddam rebuilt his now clandestine programme to focus on uranium enrichment^[19] and as was discovered after the Gulf War had massively expanded it in size and sophistication^[20]. Because this oversight had been caused by an overestimation of the airstrike’s effectiveness, the prior statement must be revised. While the airstrike was successful in preventing further plutonium enrichment, failed it as an effective tool of non-proliferation because it never fully stopped the programme. Iraq was in 1991 as outlined by Hegghammer indeed closer to the nuclear threshold as they had been in 1981^[21], meaning that pre-emptive actions had failed to create conditions for non-proliferation but merely bought time.

As such must the impact of pre-emptive actions be closely monitored to ensure that they have been truly effective as otherwise they may create a false sense of security. The Bush administration fell victim to this overestimation in 1991 and attributed much of the delay in Iraq’s nuclear programme to the airstrike^[22] which as will be shown next was not correct.

While proponents of the Osirak case like Kreps make the important argument that the strike delayed the nuclear programme thereby preventing Iraq from acquiring nuclear weapons until the Gulf War^[23], was that as will be shown not the result of the airstrike. The impact of the airstrike has not just been exaggerated but worse misunderstood as rather than delaying the programme, it served as a stimulus which accelerated the nuclear programme^[24] which had, as Hegghammer calls it, been at drift^[25] between nuclear energy and weapons.

Saddam allocated following the strike new funds to the program and gave them a prior missing clear political direction this being the creation of a nuclear bomb²⁷. With quickly expanding infrastructure and increasing staff capacity²⁸, the programme was making enormous progress between 1981 and 1991 in establishing uranium enrichment capacities and an indigenous knowledge base^[26].

What the proponents of the pre-emptive action got wrong in retrospect was that not the airstrike had caused massive delays to the program but its dysfunctional management style^[27]. Both under the leadership of Jafar and Hussein had time been wasted on pet scientific projects^[28] and compartmentalisation³² , so did teams even compete under Hassan in a tournament style over the same projects^[29].

This undermined Iraq’s ability to achieve nuclear proliferation before the Gulf War and therefore it must be concluded that pre-emptive actions had proven ineffective as tools of non-proliferation. While Iraq’s initial reliance on foreign support and limited capacities had made them vulnerable to pre-emptive actions, they never managed to stop Iraq’s nuclear programme and as such showing that against determined countries pre-emptive actions are not effective. War appeared to have been suited better to end Iraq’s nuclear ambitions because it was the invasion of Kuwait and the following Gulf War that caused the destruction of Iraq’s nuclear capabilities which estimated by Jafar could have had otherwise enough plutonium for a working bomb ready in 1993 or 1994.^[30].

Iran

The case of Iran is fundamentally different from the one present before because Iran has both a broad indigenous knowledge base^[31] and sophisticated nuclear infrastructure^[32] therefore underlining the importance of the context on the effectiveness of pre-emptive actions. Potential airstrikes are much less likely to be effective as it would be necessary to hit all facilities at ones^[33]. This would be difficult as it is to be expected that Iran has many still unknown clandestine facilities^[34] which is often overlooked by proponents of the option like Koenig^[35] meaning that detailed intelligence would be required first^[36].

Further, would such a strike undoubtedly have a negative impact on regional stability^[37]. Acts of revenge carried out by Iran and its proxies would have to be expected and could range from bombings attacks on Western targets^[38] to escalations of secretariat violence in the Middle East⁴³. Because airstrikes would not stop Iran from rebuilding its nuclear programme^[39], would that make the grim scenario described by Betts of an invasion the only permanent alternative^[40] and therefore war.

A war with Iran would plunge the Middle East into chaos^[41] and it is therefore considered in this essay that airstrikes would not be an effective tool of non-proliferation if the target is an advanced and sophisticated nuclear programme like Iran’s. As already outline in the case of Iraq are pre-emptive actions like airstrikes or assassinations only effective if the targeted country’s nuclear programme has no sophisticated indigenous base but is relying on outside support and chokepoint facilities.

Following the assassination of scientist^[42] have proponent of such methods like Hayden claimed that it not just caused “white defection” but also seriously delayed Iran’s programme^[43]. This again echoes the sentiments of prior usage of pre-emptive actions as a tool to bide for time which is not an effective approach to non-proliferation.

While the programme had indeed been delayed, was the actual impact much smaller because as Richard Mayer rightly points out is Iran’s nuclear programme too advanced to be seriously delayed or hindered by the loss of single individuals^[44]. The same is true for the various acts of sabotage directed against Iran’s nuclear supply chain including the delivery of faulty parts^[45].

Those acts have caused delays but have not effectively undermined Iran’s nuclear programme, despite Iran’s reliance on imports of certain material like maraging steel^[46]. The pre-emptive actions had the opposite effect by giving Iran an incentive to further sophisticate its domestic production capabilities^[47] while also expanding its smuggling network for nuclear equipment^[48].

The outlined pre-emptive actions have in the case of Iran proven therefore to be ineffective as tools of non-proliferation, and it must be assumed that against a sophisticated nuclear programme, they at best can achieve short, limited successes. Further, it must be considered that the targeted country will adapt over time to the pre-emptive actions taken, limiting their effectiveness in the long term.

A pre-emptive action which has so far not been addressed are cyber-attacks which provide a new non-kinetic approach. The in 2010 discovered Stuxnet virus^[49] which had infected Iranian computer networks was long believed to have been effective against the Iranian nuclear programme causing massive damage^[50]. Ralph Langer went as far as claiming that Stuxnet was in the end even more effective than a military strike regarding the destruction it caused without having started a conflict^[51]. This statement as will be shown is not just incorrect, but it also led to a similar overestimation of effectiveness as with the Osirak strike.

Barzashka has made an important contribution in that regard as she gave a detailed overview of the available data regarding the damage that was caused by Stuxnet^[52]. While Iran ran an estimated 23% fewer centrifuges from mid-2009 to mid-2010 which was presumably the damage caused by Stuxnet^[53]  the total number of operational centrifuges merely decreased by more than 300^[54] with many new machines being installed^[55]. This means that while Stuxnet caused damage to the Iranian nuclear programme, was it not that significant because Iran was able to compensate for the losses.

The cyber weapon therefore failed as a tool of non-proliferation as it could not harm Iran nuclear efforts as results of the country’s sophisticated industrial support base which proves the contextual constraints on the effectiveness of pre-emptive actions. What is even worse is that like the Osirak reactor strike it appears that Stuxnet served as a catalyst for Iran which had been made aware of shortcomings in its security system which it then improved.

Maher has therefore rightfully claimed that the timing for Stuxnet was badly chosen as it did not hit Iran at critical moment but served as an incentive for improvements⁶²; after all, despite the supposedly crippled programme was Iran able to achieve 20% enrichment in 2010^[56], proving the viruses lack of effectiveness. As such, have cyber weapons so far proven to be ineffective tools of non-proliferation.

In this article it has been argued based on the presented case studies, that pre-emptive actions like airstrikes or sabotage were not effective tools of proliferation because they failed to cause enough damage to the targeted nuclear programmes to lead to their termination. While it must be acknowledged that pre-emptive actions were effective in delaying the nuclear programmes of both Iraq and Iran can that not be considered to be advancing the goal of non-proliferation as it delayed the problem and did not solve it.

The effectiveness of pre-emptive actions motivate is heavily dependent on contextual factors like the level of sophistication of the targeted nuclear programme or a reliance on foreign supplied material and technology. While Iraq’s programme had been initially vulnerable to such pre-emptive actions is Iran’s programme supported by a sophisticated industrial base too advanced to be effectively damaged by acts of sabotage or assassinations of individual scientists.

Airstrikes can be effective tools of non-proliferation if the targeted country is reliant on a few critical facilities like Iraq in 1981 on the Osirak reactor. But should the programme be too advanced or even partially clandestine as in Iran would airstrikes fail as tools of non-proliferation and may trigger regional conflicts.

Stuxnet, while showing the vulnerability of nuclear infrastructure to cyber-attacks^[57] has ultimately proven ineffective as it failed to inflict serious damage on Iran’s nuclear programme. A bigger problem which has been revealed concerning the effectiveness of pre-emptive actions is that by exploiting the weaknesses of the targeted countries programme it incentivises the country to address the weaknesses making repeated usage of pre-emptive actions less effective.

Furthermore,  pre-emptive actions may motivate the targeted country to pursue proliferation much more vigorously like in Iraq. While pre-emptive actions have therefore proven mostly ineffective as tools of non-proliferation can they under certain conditions be considered, but then must their results be closely monitored to avoid overestimation of their impact.

Bibliography

Barzashka, Ivanka. “Are Cyber-Weapons Effective? Assessing Stuxnet’s Impact on the Iranian Enrichment Programme.” The RUSI journal 158, no. 2 (2013): 48–56.

Betts, Richard K. “The Osirak Fallacy.” The National interest 83, no. 83 (2006): 22–25.

Braut-Hegghammer, Målfrid. “Revisiting Osirak: Preventive Attacks and Nuclear Proliferation Risks.” International security 36, no. 1 (2011): 101–132.

Clawson, Patrick, and Michael Eisenstadt. “Halting Iran’s Nuclear Programme: The Military Option.” Survival (London) 50, no. 5 (2008): 13–19.

Colin H. Kahl. “Not Time to Attack Iran.” Foreign Affairs, 2012.

Dobbins, James, Alireza. Nader, Dalia Dassa. Kaye, Frederic M. Wehrey, and Brian Matthews. “Coping with a Nuclearizing Iran.” Santa Monica, CA: Rand Corporation, 2011.

Farwell, James P., and Rafal Rohozinski. “Stuxnet and the Future of Cyber War.” Survival (London) 53, no. 1 (2011): 23–40.

Fitzpatrick, Mark. “Can Iran’s Nuclear Capability Be Kept Latent?” Survival (London) 49, no. 1 (2007): 33–58.

Hymans, Jacques E. C. “Achieving Nuclear Ambitions: Scientists, Politicians, and Proliferation.” Cambridge: Cambridge University Press, 2012.

Kreps, Sarah E., and Matthew Fuhrmann. “Attacking the Atom: Does Bombing Nuclear Facilities Affect Proliferation?” Journal of strategic studies 34, no. 2 (2011): 161–187.

Kroenig, Matthew. “Time to Attack Iran: Why a Strike Is the Least Bad Option.” Foreign Affairs 91, no. 1 (2012): 76–86.

Maher, Richard. “The Covert Campaign Against Iran’s Nuclear Program: Implications for the Theory and Practice of Counterproliferation.” Journal of strategic studies 44, no. 7 (2021): 1014–1040.

Reiter, Dan. “PREVENTIVE ATTACKS AGAINST NUCLEAR PROGRAMS AND THE ‘SUCCESS’ AT OSIRAQ.” The Non-proliferation review 12, no. 2 (2005): 355–371.

Sadot, Uri. “Osirak and the Counter-Proliferation Puzzle.” Security studies 25, no. 4 (2016): 646–676.

Footnotes

[1] Kreps, Sarah E., and Matthew Fuhrmann. “Attacking the Atom: Does Bombing Nuclear Facilities Affect Proliferation?” Journal of strategic studies 34, no. 2 (2011), p. 164.

[2] Sadot, Uri. “Osirak and the Counter-Proliferation Puzzle.” Security studies 25, no. 4 (2016), p. 657.

[3] Ibid. p. 365.

[4] Sadot. p. 656.

[5] Målfrid Braut-Hegghammer. “Revisiting Osirak: Preventive Attacks and Nuclear Proliferation Risks.” International security 36, no. 1 (2011), p. 108.

[6] Sadot. p. 656.

[7] Hegghammer. p. 108.

[8] Ibid. p. 108-109.

[9] Sadot. p. 657-658.

[10] Ibid. p. 658.

[11] Ibid. p. 658. ¹² Ibid. p. 657.

[12] Reiter, Dan. “PREVENTIVE ATTACKS AGAINST NUCLEAR PROGRAMS AND THE ‘SUCCESS’ AT OSIRAQ.” The Non-proliferation review 12, no. 2 (2005), p. 357.

[13] Ibid. p. 667-668.

[14] Kreps. p. 162.

[15] Ibid. p. 172.

[16] Reiter. p. 358.

[17] Ibid. p. 356.

[18] Hegghammer p. 129.

[19] Betts, Richard K. “The Osirak Fallacy.” The National interest 83, no. 83 (2006), p. 23.

[20] Ibid. p. 22.

[21] Hegghammer p. 125.

[22] Reiter. p. 356.

[23] Kreps. p. 170.

[24] Ibid.  p. 171.

[25] Hegghammer p. 109. ²⁷ Kreps. p. 171. ²⁸ Reiter. p. 362

[26] Reiter. p. 362-363.

[27] Hymans, Jacques E. C. “Achieving Nuclear Ambitions: Scientists, Politicians, and Proliferation.” Cambridge: Cambridge University Press, 2012, p. 122.

[28] Hymans. p. 100.  ³² Ibid. p. 99.

[29] Ibid. p. 104.

[30] Reiter. p. 363.

[31] Kreps. p. 181.

[32] Betts. p. 23.

[33] Kreps. p. 182-183.

[34] Ibid. p. 182.

[35] Kroenig, Matthew. “Time to Attack Iran: Why a Strike Is the Least Bad Option.” Foreign Affairs 91, no. 1 (2012), p. 79-80.

[36] Clawson, Patrick, and Michael Eisenstadt. “Halting Iran’s Nuclear Programme: The Military Option.” Survival (London) 50, no. 5 (2008), p. 14-15.

[37] Fitzpatrick, Mark. “Can Iran’s Nuclear Capability Be Kept Latent?” Survival (London) 49, no. 1 (2007), p. 48.

[38] Ibid. p. 48. ⁴³ Ibid. p. 48.

[39] Colin H. Kahl. “Not Time to Attack Iran.” Foreign Affairs, 2012, p. 170.

[40] Betts. p. 24.

[41] Colin. p. 169-170.

[42] Dobbins, James, Alireza. Nader, Dalia Dassa. Kaye, Frederic M. Wehrey, and Brian Matthews. “Coping with a Nuclearizing Iran.” Santa Monica, CA: Rand Corporation, 2011.  p. 86.

[43] Maher, Richard. “The Covert Campaign Against Iran’s Nuclear Program: Implications for the Theory and Practice of Counterproliferation.” Journal of strategic studies 44, no. 7 (2021), p. 1018.

[44] Ibid. p. 1018.

[45] Ibid. p. 1021.

[46] Fitzpatrick. p. 39.

[47] Ibid. p. 39.

[48] Maher. p. 1025.

[49] Farwell, James P., and Rafal Rohozinski. “Stuxnet and the Future of Cyber War.” Survival (London) 53, no. 1 (2011), p. 23.

[50] Ibid. p. 23.

[51] Barzashka, Ivanka. “Are Cyber-Weapons Effective? Assessing Stuxnet’s Impact on the Iranian Enrichment Programme.” The RUSI journal 158, no. 2 (2013), p. 49.

[52] Ibid. p. 52.

[53] Farwell. p. 29.

[54] Barzashka. p. 52.

[55] Ibid. p. 53. ⁶¹ Maher. p. 1024  ⁶² Ibid. p. 1024.

[56] Barzashka. p. 54.

[57] Farwell. p. 25.