Jeroen Gunning, King’s College London
The unprecedented brutal attacks carried out by Hamas on 7 October 2023 came as a shock not just to Israel’s intelligence community but to International Relations scholars and social scientists more broadly. Social science has a dismal record at predicting events such as the Iranian revolution, the fall of the Berlin Wall, 9/11, the Arab Uprisings, Assad’s fall. Like IR, social movement theory (SMT) is ill-equipped for predictions. However, when it comes to explanatory power and descriptive accuracy, SMT does rather better. The same reasons why SMT is bad at predicting – recognition of the fundamental contingency of events and an abundance of possible causal factors – are why it is good at explaining events after they happened, especially compared to more parsimonious IR theories which bracket domestic dynamics from their analysis. This raises questions about the value of predictiveness and parsimony – whether in terms of number of variables (“parsimony by components”) or number of outcomes and thus predictive accuracy (“parsimony by constraints”) (Dubova et al. 2025) – as a basis for judging a theory’s worth. This essay argues that the shortcomings of other theoretical approaches in both predicting and explaining 7 October might open the door to more consideration for SMT by IR scholars. Though international dynamics are core to any explanation, social movement and other domestic dynamics offer critical context for making sense of an important case. IR theories’ search for parsimony hinders consideration of this complexity.
What is Social Movement Theory?
SMT is not a “theory” in the strict sense of the word, but a dynamic (and often contradictory) collection of models rooted in different research traditions, ontologies and epistemologies which explain social movement dynamics focusing (to different degrees) on mechanisms, repertoires, relations, interactions, frames, beliefs, identities, cultures, emotions and structural conditions. While SMT was predominantly structuralist in the second half of the 20th century, the field has become more focused on the micro-practices of agents, their life experiences, emotions, beliefs and events (Goodwin and Jasper 1999; Jasper 2010; Gillan 2020). SMT has also become more modular, both in the sense of emphasising how movements combine different broadly identifiable building blocks (causal/functional mechanisms) to mobilise,[1] and in the sense of scholars drawing on different theories in the tradition of “modular theory-building”: “link[ing] compatible elements across otherwise distinct, and often interdisciplinary, theoretical realms… to build broader, flexible, multi-scalar theoretical frameworks that can be used … [to explain] high complexity, dynamism and variation” (Ponte and Sturgeon 2014, 196). In this sense, it manifests all three aspects of what Sil and Katzenstein (2010, 411) call “analytic eclecticism,” by “downplaying unresolvable metaphysical divides and presumptions of incommensurability” between theories, focusing on “the messiness and complexity of concrete dilemmas facing “real world” actors,” and “offer[ing] complex causal stories that… selectively recombine analytic components—most notably, causal mechanisms—from… competing research traditions.”
This trend has gone hand in hand with greater acknowledgement of the fundamental historical, situational and conjunctural contingency of movement outcomes (Goodwin and Jasper 1999; Morris 2000; Gunning and Baron 2014). Practices are caused by a highly contingent interplay of causal mechanisms, including broader structural and mobilising structures as well as agential strategy, emotions and culture (Platt 2004). Political opportunity and threat structures (POTS) do not simply exist, they are interpreted by agents in complex organisational, ideational and relational webs whose practices in turn affect those structures. Activists interpret/use the building blocks differently depending on how other actors act; how they interpret those actions, threats and opportunities; how emotions, habitual practices and culture shape those interpretations, etc. – i.e. similar structures and collections of agents can produce very different outcomes. Yet combining complex structural and agential analysis remains a challenge for SMT both in terms of linking the different types of explanation (e.g. interpreting “mechanisms” contingently rather than deterministically) and in terms of deciding which factors/dynamics are most important (Gunning and Baron 2014; Chalcraft 2016).
Because of the plurality of theoretical explanations and the move away from law-like generalisations, different SMT scholars emphasise different building blocks to explain movements, making explanations even more ambiguous and fragmented (though richer). The variety of SMT approaches ensures that there will never be a single SMT prediction. This makes SMT unfit to predict any particular outcome confidently but simultaneously good at explaining unexpected outcomes. For, following Tetlock (summarised in Sil and Katzenstein 2010, 414), “grossly inaccurate forecasts are more likely… when experts… [rely] on a single parsimonious approach to explain many things and [depend] excessively upon “powerful abstractions to organize messy facts….”.”
SMT and 7 October
SMT can identify a number of contingent dynamics that explain both Hamas’s turn to shocking violence and the timing of the violence. Three sets of dynamics stand out – all an interplay between structural and strategic changes, organisational dynamics, and actors’ lived experience, emotions, calculations and interactions.[2]
Gaza’s blockade: As Sara Roy (2023) has shown, the de-development of Gaza has been ongoing for decades, with Israel gradually destroying its economic base through expropriation, dispossession, destruction and deinstitutionalisation. Its once flourishing orange and flower industries were destroyed and, from the 1990s, most Gazans were forbidden from crossing the border. After the blockade’s formalisation in 2007 following Hamas’s consolidation of power, Israel reportedly limited what went into Gaza to sustain bare life while preventing a humanitarian disaster. After each ‘mowing the lawn’ war, as Israeli generals called the regular wars to cut down Hamas’s capabilities, more of Gaza remained destroyed as rebuilding materials could not enter. By the late 2010s, Gaza had become virtually unliveable, with over 80% of the population living under the poverty line and dependent on humanitarian aid (UNRWA n.d.). Further, between 2008-2023, over 20 Palestinian civilians (the vast majority in Gaza) were killed for every Israeli civilian killed (OCHA n.d.). This lived experience, produced by the structural and strategic context, arguably created an “emotion culture” (Gould 2004) that can help to explain the brutality of the violence on 7 October (the blockade and civilian deaths feature centrally in Hamas’s official explanation of Operation al-Aqsa Flood; Hamas 2024).
The previous rounds of armed conflict had resulted in limited temporary easing of the blockade (Skare 2025), but none of them resulted in its ending. None of Hamas’s political overtures had been taken seriously either. Hamas engaged in numerous national unity governments negotiations and, despite having won the last national elections (2006), conceded to not being directly part government and granting Fatah control of Gaza’s Egyptian border. Its 2017 Manifesto publicly accepted a Palestinian state on the 1967 borders and no longer contained the antisemitic tropes that had permeated its 1988 Charter, distinguishing between Jews and Zionists. It refused to recognise Israel or give up the right of return of Palestinian refugees – but, that is not uncommon for anticolonial armed groups, especially in settler colonial contexts (Gunning 2025b, para 28).
The fact that Hamas was willing to accept the 1967 borders as a matter of national consensus in its Manifesto was significant, particularly as hardliners had become stronger by 2017 (among them, Mohammed Deif, head of the Qassam Brigades, and Yahya Sinwar, head of Hamas in Gaza, who both would go on to plan the 7 October attacks). The publication of the Manifesto appears to have been an attempt by Hamas’s pragmatists to bind the hardliners to a consensually produced document (Skare 2025). Hamas’s organisational culture values consensus and loyalty to the movement’s formal decisions, shaping how its leaders responded to the structural conditions of blockade and marginalisation. Following the Manifesto’s publication, Hamas in Gaza, under Sinwar’s leadership, made the most extensive concessions to Fatah in the 2017 National Unity Government negotiations, although the negotiations ultimately failed (Hroub 2025).
Hamas endorsed the Great March of Return in 2018, a broadly non-violent grassroots protest of the blockade, which resulted in over 200 killed and 36,000 injured. This, combined with the high death rate of civilians since 2008 and the lack of international support, was a key turning point in how Hamas leaders, and particularly the hardliners, viewed political options (Hubbard and Abi-Habib 2023; Skare 2025). The final straw seems to have been the cancellation of Palestinian elections in 2021 by President Abbas, allegedly for fear of losing to Hamas. Elections had been seen as central to reuniting Gaza and the West Bank and ending the blockade. Their cancellation, coupled with the intensification of violence in May 2021 over Israeli evictions in Jerusalem’s Sheikh Radwan area, strengthened hardliners within Hamas and the detailed planning for 7 October appears to have begun shortly after (building on earlier planning that appears to have intensified after the failure of the Great March of Return, with the first announced joint “war games-style exercise” led by Hamas in 2020) (BBC 2023).
Illustrating the contingency of Hamas’s actions, hardliners, while complying with the movement’s decision to attempt pragmatism in 2017-2019, began preparing for more extensive military action for if/when pragmatism failed. They had been strengthened by the lack of results of political strategies, the tunnel economy, and Iranian support (Skare 2025, 185–87, 191). The close personal ties between Deif and Sinwar, Sinwar’s closeness to the Brigades and his election to head of Hamas’s Gaza Political Bureau, and their shared hardline views, shaped by their personal histories, facilitated this strengthening. In 2018, at the start of the Great March of Return (and partly to facilitate it), a joint operations room was set up with a dozen Palestinian armed groups which reportedly began assisting military attacks in November 2018 (alongside the Great March) and would play an important role in the May 2021 Sword of al-Aqsa war (Musa 2019; Abu Amer 2021). Hamas’s rapprochement with Iran and Hizbullah, after falling out over the Syrian war in 2012 under Khaled Mishal’s leadership, accelerated following the election of pro-Iranians Sinwar and Saleh al-Arouri (Deputy Head of Political Bureau) in 2017 and Mishal’s stepping down as Head of Political Bureau (Jawhar 2022) – showing the importance of agency and personal life trajectories in shaping the structural and strategic context. By 2021, a “joint security room” had reportedly been established in Lebanon which similarly helped Hamas in the May 2021 war (Majid 2022). Illustrating the interplay between regional dynamics, specific Hamas leaders and their beliefs, and the movement’s organisational evolution, Arouri and others had settled in Lebanon after having been forced out of Turkey following the latter’s rapprochement with Israel (Jawhar 2022).
A speech Sinwar gave after the May 2021 war underlines the multiple scenarios Hamas had been preparing for, ranging from non-violence to extreme violence, contingent on how Israel and the world would respond to the Palestinians’ plight: “We hoped that the free and civilized world and international organizations would respond [to our non-violent March of Return]… [but] the conscience of the free world didn’t wake up… if the world does not respond, and if this occupation does not respond or commits serious crimes, then our resistance must act to deter them. … [We have] an existing and known scenario, and the rehearsal has been made for it [i.e. the Sword of Al-Aqsa confrontation]” (Hroub 2025; italics in original).
The brutality of the violence seems to have been calculated to change the equation of occupation and blockade. The timing of the attacks was further driven by what Hamas leaders saw as two existential threats (structural and strategic context interpreted through Hamas leaders’ beliefs, personal histories, etc.):
West Bank annexation: In 2023, the West Bank had seen a sharp escalation of violence – with over a 1,000 attacks by Israeli settlers and over 200 Palestinians killed, the highest since the al-Aqsa Intifada – and a record in new approved settlements (Gunning 2025a). Israeli ministers openly called for the West Bank’s annexation and threats against the al-Aqsa Mosque, Islam’s third holiest, had intensified (both threats feature in Hamas 2024). Hamas had initiated violence in response to threats to the al-Aqsa Mosque before – partly because it saw this as a red line and a sign of Israel making Palestinian statehood impossible, partly to show solidarity with Palestinians in Jerusalem and the West Bank, but also to reclaim leadership of Palestinian politics, against Fatah’s ineffectiveness and its being seen as Israel’s policeman (Hubbard and Abi-Habib 2023).
In 2023, the intent was also to trigger an uprising. Sinwar had already alluded to this in his 2021 speech if Israel and the world would not respond: “the West Bank will explode with all its power. Our people will attack all the settlements at once” (Hroub 2025). Public support for a return to armed struggle in the West Bank had jumped from 35% to 54% in 2023 (PCPSR n.d.) and, after 7 October, Hamas and the Brigades have called repeatedly on West Bankers to rise up while its operatives there have increased attacks (alqassam.ps). In the Brigades’ eyes, triggering an uprising required shocking violence and a shocking response from Israel (e.g., Khalid al-Hayya, currently head of Hamas’s negotiations team and one of the five-men leadership of the Political Bureau, telling the New York Times: “What could change the equation was a great act, and without a doubt, it was known that the reaction to this great act would be big”) (Hubbard and Abi-Habib 2023). While this can be explained strategically, the emotion culture caused by the effects of the blockade, occupation and high death count helps to contextualise the decision to use extreme violence to cause an extreme reaction.
Regional normalisation: normalisation with Israel, without a Palestinian state, was another strategic factor that would change Hamas’s POTS. Four Arab states had signed normalisation agreements in 2020. Rumours about Saudi Arabia, historically a stalwart supporter of a Palestinian state, being next intensified in September 2023, when Netanyahu announced at the UN that Israel was “at the cusp” of a deal with Saudi Arabia, followed by unprecedented high-level visits to each other’s states (Knell 2023). For Hamas, this meant the end of the possibility of a Palestinian state as it would only leave Gaza, just 1% of historic British Mandate Palestine – ostensibly contained, with Hamas as prison warden. This would be both a profound structural change and one that would lock Hamas as an organisation into a position where the internal tensions between those who saw themselves as primarily a resistance movement and those who wanted to govern would continue, without end. As al-Hayya said after 7 October:
Hamas’s goal is not to run Gaza and to bring it water and electricity and such… Hamas, the Qassam and the resistance woke the world up from its deep sleep and showed that this issue must remain on the table. …This battle was not because we wanted fuel or laborers… It did not seek to improve the situation in Gaza. This battle is to completely overthrow the situation.
To derail regional normalisation, the rationale again seems to have been that shocking violence was required – to “change the entire equation and not just have a clash”, in al-Hayya’s words, “putting the Palestinian issue back on the table, and now no one in the region is experiencing calm” (Hubbard and Abi-Habib 2023). Sinwar is reportedly to have told colleagues on October 2 that “an extraordinary act” was required to derail Saudi normalisation and “that it was time to activate the plan for an attack that [Hamas] had been planning for some two years “to bring about… a strategic shift in the paths and balances of the region with regard to the Palestinian cause.”” (ToI 2025)
Reflections on SMT’s explanatory power
The interplay between key structural and strategic changes in Hamas’ POTS outlined above and how a small subset of Hamas leaders interpreted and responded to them goes a long way in explaining, retrospectively, why Hamas acted the way it did. Each of these dynamics was clearly visible before 7 October (including Hamas’s military training sessions) and, in theory, an SMT scholar could have predicted a scenario in which an attack of unprecedented reach and shock would take place. SMT’s modularity and its recognition of contingency enable it to, at least in principle, forecast multiple scenarios. But precisely because SMT foregrounds contingency, such a scenario would have been one among many and difficult to predict with any certainty – partly because it lies so far outside of what Hamas had done before, but mainly because the effect of each of these dynamics is contingent on the sequential interplay between key actors’ agency, interpretations, relations and interactions.
Hamas had further pursued an intentionally ambiguous/multi-pronged strategy over the years, distinguishing between medium- and long-term goals (a Palestinian state on the 1967 borders vs liberation of all of historic Palestine) and developing capabilities along multiple axes (diplomatic, political, social, military) (see also Skare 2025 on the intentional ambiguity of Hamas’s New Manifesto). This means that it could be flexible in what it did in response to different pressures without undermining its narrative fidelity with past statements. That it opted for 7 October rather than political engagement does not mean that the 2017 Manifesto did not authentically reflect one possible version of Hamas; it just means that the interaction between POTS, internal organisational dynamics and particular leader trajectories created conditions favouring this strategy. For SMT, this means even more contingency, thus making it more difficult to predict.
The inability to make narrowly accurate predictions is also partly due to the sheer number of (f)actors considered by SMT which has been accused of “becoming a sponge that soaks up virtually every aspect of the social movement environment… Used to explain so much, it may ultimately explain nothing at all” (Gamson and Meyer 1996, 275). This critique was levelled at SMT’s structuralists. But scholarship on agency, culture and emotions has tended to eschew structuralist insights and grand theory and become fragmented into “subspecialities” with often highly situated and similarly multi-causal accounts of contingent mechanisms (Jasper 2010, 973–74). The more structuralist and the more contingent/agential SMT approaches need to be combined to explain situated movement dynamics. But that further increases the number of possible (and non-deterministically contingent) mechanisms.
Here, SMT can benefit from critical realism: its distinction between the “real” (the underlying causal mechanisms that make up our world), the “actual” (the mechanisms that have been actually activated in a particular set of events), and the “empirical” (what has been experienced/observed); but also its notions of “emergent properties”, “abduction” and “retroduction”. Each of SMT’s modular elements describes real causal mechanisms that could play a role. Which of these mechanisms is activated depends on the actual constellation of (f)actors that have produced observed events (e.g. 7 October versus political engagement). To arrive at an explanation, one has to first abductively hypothesise which mechanisms might explain the events (“theoretical redescription”), then retroductively work back from the events to reconstruct the most plausible combination of “real” mechanisms that appear to have been actualised (Fletcher 2017, 188–89). Thinking back, I arrived at the above explanation two days after 7 October 2023 in preparation for a radio interview – I did not yet have access to Hamas statements explaining their reasoning, yet through retroduction, starting with the events, then reflecting on the real causal mechanisms that could be at play (drawing on SMT, conflict studies, etc.) and what actual structural, organisational and ideational changes could be observed over the years (drawing on media, reports and secondary literature on Hamas/Gaza/Palestine/Israel, as well as my own, long-term analysis of the movement), I settled on what appeared to be the most critical mechanisms that had been actualised.
Crucially, these mechanisms coming together produce “emergent properties that are more than, and different from, the sum of their constituent parts” (Easton 2010, 121). The causal properties of each of the (f)actors described above alone or even interacting as independent building blocks cannot explain 7 October – treating each “module” as semi-independent is one of the dangers of a modular approach. The causal properties of each (f)actor shaped others, sequentially and concurrently, thus creating emergent properties that become more than their parts. A good modular theory “illuminates not only how specific variables influence [practice] in isolation, but also how key variables… themselves co-evolve in larger systems” (Ponte and Sturgeon 2014, 217). 7 October thus pushes SMT to reflect more on how the different (f)actors interact and shape each other and to distinguish between “real” and “actual” mechanisms retroductively to arrive at an explanation that does not include “everything under the sun” (Goodwin and Jasper 1999, 31) but singles out the most important actual mechanisms from the plethora of real ones.
SMT (and IR and social science generally) can also be critiqued for not thinking boldly enough outside of what has occurred in the past – unexpected events are unexpected because they are rare/unprecedented, but they are still products of visible structural changes and agential actions that, though varied, nevertheless follow a broadly known logic.
Finally, SMT has historically focused on mobilisation within states. The rise of globalisation theories and increasingly visible forms of cross-national mobilisation have “forced social movement theorists to rethink their obsession with the nation state and to recognize the importance of different arenas” (Jasper 2010, 974). However, much of SMT is still methodologically nationalist/statist – and without understanding the regional and international dynamics that shaped Hamas’s leadership, their lived experience and calculations, it would be impossible to arrive at a compelling explanation of 7 October. Crucially, though, as Jasper (2010, 974) warns, “we cannot understand the special nature of the global without understanding the microfoundations that make up global, national, and local politics”. To be compelling, SMT thus has to combine an analysis of micro-practices and broad regional and global dynamics. If it does, it is much more likely to be successful as an explanatory framework than the more parsimonious IR theories which operate on a macro level.
References
Abu Amer, Adnan. 2021. “غرفة العمليات المشتركة للمقاومة في غزة نواة لـ”جيش التحرير” | أخبار سياسة | الجزيرة نت.” Aljazeera/الجزيرة نت. May 28, 2021.
BBC. 2023. “How Hamas Built a Force to Attack Israel on 7 October,” November 27, 2023.
Chalcraft, John. 2016. Popular Politics in the Making of the Modern Middle East. Cambridge: Cambridge University Press.
Dubova, Marina, Suyog Chandramouli, Gerd Gigerenzer, Peter Grünwald, William Holmes, Tania Lombrozo, Marco Marelli, et al. 2025. “Is Ockham’s Razor Losing Its Edge? New Perspectives on the Principle of Model Parsimony.” Proceedings of the National Academy of Sciences 122(5).
Easton, Geoff. 2010. “Critical Realism in Case Study Research.” Industrial Marketing Management 39(1): 118–28.
Egeland, Jonathan. 2024. “Making Sense of the Modularity Debate.” New Ideas in Psychology 75(December): 101108.
Fletcher, Amber. 2017. “Applying Critical Realism in Qualitative Research: Methodology Meets Method.” International Journal of Social Research Methodology 20(2): 181–94.
Gamson, William, and David Meyer. 1996. “Framing Political Opportunity.” In Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings, edited by Doug McAdam, John McCarthy, and Mayer Zald, 275–90. Cambridge: Cambridge University Press.
Gillan, Kevin. 2020. “Temporality in Social Movement Theory: Vectors and Events in the Neoliberal Timescape.” Social Movement Studies 19(5–6): 516–36.
Goodwin, Jeff, and James Jasper. 1999. “Caught in a Winding, Snarling Vine: The Structural Bias of Political Process Theory.” Sociological Forum 14(1): 27–54.
Gould, Deborah. 2004. “Passionate Political Processes: Bringing Emotions Back Into the Study of Social Movements.” In Rethinking Social Movements: Structure, Meaning, and Emotion, edited by Jeff Goodwin and James Jasper, 155–76. People, Passions, and Power. Lanham, MD: Rowman & Littlefield.
Gunning, Jeroen. 2025a. “The Politics and Effects of Proscription – Written Evidence for ‘The Hamas Case.’” Riverway Law.
———. 2025b. “Written Evidence Submitted to the Parliamentary Committee on Foreign Affairs’ for ‘The Israeli-Palestinian Conflict’ Inquiry (IPC0109).” The Israeli-Palestinian Conflict.
Gunning, Jeroen, and Ilan Zvi Baron. 2014. Why Occupy a Square?: People, Protests and Movements in the Egyptian Revolution. New York, NY: Oxford University Press.
Hamas. 2024. “Our Narrative… Operation Al-Aqsa Flood.” Hamas Media Office. January 21, 2024.
Hroub, Khaled. 2025. “Postscript: What Did Sinwar Want? — In His Own Words.” Jadaliyya – جدلية. 2025.
Hubbard, Ben, and Maria Abi-Habib. 2023. “Behind Hamas’s Bloody Gambit to Create a ‘Permanent’ State of War.” New York Times, November 8, 2023.
Jasper, James. 2010. “Social Movement Theory Today: Toward a Theory of Action?” Sociology Compass 4(11): 965–76.
Jawhar, Souhayb. 2022. “Lebanon: New Strategic Base for Hamas.” Carnegie Endowment for International Peace. October 17, 2022.
Knell, Yolande. 2023. “Landmark Saudi and Israeli Trips as Normalisation Talks Progress.” BBC News. September 26, 2023.
Majid, Mohammed. 2022. “‘الجزيرة’ تكشف معلومات عن الحرب الأخيرة في غزة.” Anadolu Ajansı. 5 2022.
Morris, Aldon. 2000. “Reflections on Social Movement Theory: Criticisms and Proposals.” Contemporary Sociology 29(3): 445–54.
Musa, Raed. 2019. “غرفة العمليات المشتركة.. ‘قيادة أركان المقاومة’ في غزة.” Aljazeera/الجزيرة نت. April 11, 2019.
OCHA. n.d. “Data on Casualties.” United Nations Office for the Coordination of Humanitarian Affairs – Occupied Palestinian Territory. n.d.
PCPSR. n.d. “Index PSR Polls.” Ramallah: People’s Company for Palestinian Survey Research. n.d.
Platt, Gerald. 2004. “Unifying Social Movement Theories.” Qualitative Sociology 27(1): 107–16.
Ponte, Stefano, and Timothy Sturgeon. 2014. “Explaining Governance in Global Value Chains: A Modular Theory-Building Effort.” Review of International Political Economy 21(1): 195–223.
Roy, Sara. 2023. “The Long War on Gaza.” The New York Review of Books, December 19, 2023.
Sil, Rudra, and Peter Katzenstein. 2010. “Analytic Eclecticism in the Study of World Politics: Reconfiguring Problems and Mechanisms across Research Traditions.” Perspectives on Politics 8(2): 411–31.
Skare, Erik. 2025. Road to October 7: A Brief History of Palestinian Islamism. London: Verso.
Tarrow, Sidney. 1993. “Modular Collective Action and the Rise of the Social Movement: Why the French Revolution Was Not Enough.” Politics & Society 21(1): 69–90.
ToI. 2025. “Report: In Days Prior to Oct. 7, Yahya Sinwar Said ‘Extraordinary Act’ Needed to Derail Saudi Normalization.” Times of Israel, May 18, 2025.
UNRWA. n.d. “Gaza_15 Years of Blockade.” n.d.
[1] This goes beyond ‘modular’ in Tarrow’s (1993) sense of repertoires becoming readily diffusible. See Ponte and Sturgeon (2014), Egeland (2024).
[2] For distinction between strategy and structure, see Goodwin and Jasper (1999).