Beyond the Horseshoe - A New Model to Separate Dissent from Extremism

Left vs right is the wrong map. Beyond the Horseshoe introduces the Pluralistic Gradient Model of Radicalisation (PGMR), a research-backed way to separate legitimate dissent from anti-pluralist extremism. Drawing on McCauley & Moskalenko, Moghaddam’s staircase, Kruglanski & Bélanger’s 3N, Bandura on moral disengagement, Haslam on dehumanisation, Europol’s TE-SAT 2024, and UNDP fieldwork, the essay shows how radicalisation unfolds as a process, and pinpoints the indicators that mark a crossing into danger. The payoff is practical: proportionate, rights-respecting responses that protect democratic resilience without criminalising protest.

A Berlin motorway at rush hour, a line of young people sit down and glue their hands to the tarmac. Days later, police in three German states arrest members of a far-right cell who had stockpiled weapons and fantasised about collapsing the grid. Both scenes were labelled “extremism” in the same week’s commentary. The first was exasperating, perhaps reckless; the second conspiratorial and violent. The conflation is more than a rhetorical shortcut. It is a category error that risks licensing disproportionate state power and, perversely, fuelling the escalation it is meant to prevent. In this essay I propose a different lens: not a single ideological spectrum or a tidy horseshoe, but a process that tracks how radical critique can harden into anti-pluralist extremism and, at the limit, violence. I call this the Pluralistic Gradient Model of Radicalisation (PGMR).

I begin from a premise that is empirical rather than polemical: the decisive fault line for democratic risk is not left versus right but pluralism versus anti-pluralism. Comparative political science has long moved beyond one-dimensional maps. The Chapel Hill Expert Survey and related work by Liesbet Hooghe, Gary Marks and colleagues show European party systems structured along at least two axes—economic left–right and a social–cultural GAL–TAN (Green-Alternative-Libertarian and Traditional-Authoritarian-Nationalist) dimension, with positions on Europe and identity increasingly sorted along the latter. This multidimensional cartography explains why movements that look similar in style (anti-elite, insurgent, distrustful) can differ fundamentally in their ethical commitments to plural competition. In that light, the horseshoe’s tidy convergence at the “extremes” becomes a misleading optical illusion rather than a guide to risk.

“The line between radicalism and extremism is crossed when anti-pluralism becomes doctrinal and extraordinary means are endorsed.”

PGMR is a process model with a clear threshold. Building on Clark McCauley and Sophia Moskalenko’s Friction, which distinguishes the radicalisation of opinion from the radicalisation of action, and on Fathali Moghaddam’s “staircase” in which perceived injustice and narrowing options bring people to floors where violence is contemplated, I trace five phases: grievance awareness after moral shock or identity threat; a cognitive opening that renders unorthodox frames plausible (as Quintan Wiktorowicz documents in his classic study of Al-Muhajiroun); ideological consolidation that simplifies moral universes; an extremist turn in which pluralism itself is rejected and coercion is morally legitimated; and, finally, violent extremism where coercion is operationalised. The line between radicalism and extremism is crossed when anti-pluralism becomes doctrinal and extraordinary means are endorsed.

That threshold has both analytic and normative footing. Cas Mudde’s account of the populist radical right describes a “thin-centred” ideology that claims exclusive representation of a pure people; Jan-Werner Müller’s What Is Populism? shows how this logic delegitimises rivals as enemies rather than opponents, corroding the very rotation on which democracies depend. In policy, the UK’s Commission for Countering Extremism reframed the problem as “hateful extremism”, behaviours that incite or amplify hatred or make the moral case for violence and erode others’ rights, arguing that law and policy should be calibrated to this domain short of terrorism. Taken together, these sources justify the PGMR claim that radicalism - disruptive, even unruly - can be democratically vital, whereas anti-pluralist extremism, even when non-violent, signals a crossing that merits exceptional scrutiny.

Any model that claims to be useful must speak to present risk. Europol’s EU Terrorism Situation and Trend Report 2024 sets a clear baseline for 2023: 120 terrorist attacks in seven EU states (98 completed, 9 failed, 13 foiled). Separatist actors perpetrated 70 attacks (all completed); left-wing and anarchist actors were responsible for 32 (23 completed); there were 14 jihadist attacks (5 completed), and two right-wing attacks were foiled. Across 22 Member States, 426 individuals were arrested for terrorist offences, most of them (334) in jihadist cases; jihadist violence was the most lethal, with six people killed and twelve injured. A plural threat landscape therefore requires a pluralistic analytic lens and a disciplined way to differentiate radical from extremist across ideologies.

“Most who hold radical views never cross the behavioural threshold, and that prevention which treats attitude as destiny will misfire both ethically and empirically.”

Two cognitive pivots mark the extremist turn on the gradient. The first is moral disengagement, the repertoire of justifications that allows people to transgress their own inhibitions: euphemism, diffusion of responsibility, advantageous comparison and, crucially, dehumanisation. The second is dehumanisation itself. Nick Haslam’s synthesis distinguishes “animalistic” and “mechanistic” forms, denying either uniquely human traits or human nature, each of which erodes the moral salience of competing lives. When such cues become recurrent in a movement’s internal discourse and public repertoire, and when they are paired with an explicit rejection of rivals’ legitimate standing, I treat the crossing into extremism as underway.

Drivers and accelerants are distributed across levels. Arie Kruglanski, Jocelyn J. Bélanger and Rohan Gunaratna’s “3N” framework (Needs, Narratives, Networks) connects a search for significance after humiliation or loss to identity-affirming stories and the social scaffolding that makes escalation thinkable. A recent Campbell systematic review by Michael Wolfowicz and colleagues then sharpens a distinction central to PGMR: factors that correlate with radical attitudes and intentions only weakly predict violent behaviour; criminogenic variables and some social-psychological factors matter more for action than for opinion. It follows that most who hold radical views never cross the behavioural threshold, and that prevention which treats attitude as destiny will misfire both ethically and empirically.

Justice capacity is not peripheral to prevention; it is constitutive of it. The UNDP’s Journey to Extremism in Africa reports, based on thousands of interviews with former fighters, emphasise that triggers for joining violent groups were less doctrinal than experiential: state abuse, arbitrary arrest, killings, along with economic precarity and local conflict dynamics. Where citizens experience the state as predatory, the gradient steepens; where institutions are procedurally fair and responsive, the slope can invert. A serious model must therefore bridge security and development rather than treat them as substitutes.

Contemporary dynamics complicate the picture. Affective polarisation (the transformation of opponents into objects of distrust or disgust) has grown in some democracies and remained stable in others, but where it intensifies it raises citizens’ tolerance for exceptional measures against out-groups. The best work here is careful: Shanto Iyengar, Gaurav Sood and Yphtach Lelkes document the rise of partisan animus; subsequent reviews map cross-national variation and warn that affect is not identical to ideological distance. I treat affective polarisation as a background condition that can accelerate movement along the gradient, especially when combined with perceived institutional closure.

“Treating the internet as a single conveyor belt obscures the heterogeneity that policy must accommodate.”

Digital architectures act as multipliers rather than prime movers. Studies of YouTube before the platform’s 2019 changes found pathways linking mainstream, “alt-lite,” and explicitly extremist channels; Horta Ribeiro and colleagues’ audit suggested migration from milder to more extreme content was common in that era. Later work by Homa Hosseinmardi et al. argued recommendations were not the primary driver of traffic to radical content and that consumption concentrated among already engaged users, while other audits claim persistent bias for right-leaning users. Meanwhile, Christopher Bail and colleagues’ field experiment on Twitter showed that forced exposure to opposing views increased ideological rigidity among Republicans and did not depolarise Democrats - an effect the authors emphasise is context-dependent. The reasonable conclusion is measured: online systems can lower friction and normalise transgressive talk for some users some of the time, particularly where social validation and identity threat are present, but treating the internet as a single conveyor belt obscures the heterogeneity that policy must accommodate.

Consider, with that nuance, a current European dilemma. German prosecutors have charged organisers from the climate group Letzte Generation with forming a criminal organisation under §129 - a move that civil-liberties groups view as a chilling escalation against disruptive but non-violent protest. At the same time, courts have convicted members of the “United Patriots” for plotting to overthrow the government and kidnap the former health minister, Karl Lauterbach, with sentences of up to eight years. PGMR compels a disciplined distinction: the first sits on the radical, pluralist side of the gradient. It may infringe laws and frustrate commuters yet does not deny rivals’ legitimate standing or valorise violence; the second explicitly fuses anti-pluralist ideology with justificatory narratives for coercion. To collapse these phenomena into a single “extremism” demeans the law’s claim to procedural fairness and drains public trust.

Scope matters. A fair critique is that defining extremism as anti-pluralism smuggles liberal morality into analysis. I answer with clear scope conditions: PGMR is designed for open and semi-open polities that profess pluralism as a constitutional value. In authoritarian or occupied contexts, anti-pluralist resistance may be framed by protagonists as a tragic necessity; moral evaluation then requires a different baseline. The model does not pretend to be universal. It aims instead to discipline assessment where plural competition is the proclaimed rule of the game.

Nor should a gradient be allowed to explain everything and nothing. I propose evidentiary indicators at the point of crossing: recurrent moral-disengagement cues and dehumanising repertoires; explicit claims to exclusive representation of “the people” or to enemies’ illegitimacy, of the sort Müller diagnoses; and behavioural thresholds such as facilitation, fundraising, training and operational planning as recognised in criminal law and security practice, consistent with how Europol and national services code preparatory acts. Analysts should evidence several such indicators before coding a movement as having crossed into extremism. The point is not to smuggle ideology into risk assessment, but to anchor diagnosis in observable speech and conduct.

“Governments can align prevention with development by prioritising procedural justice, economic inclusion and local legitimacy.”

A further critique comes from security realists who stress that prevention must act early and sometimes harshly. I propose calibration, not leniency. Intelligence and policing are necessary where there is credible preparation for violence. Yet dragging coercive tools down the gradient to cover radical but pluralist dissent corrodes legitimacy and, as the disengagement literature shows from Tore Bjørgo and John Horgan onwards, can foreclose the exit ramps that reduce violence. Open channels to desist and reintegrate, paired with credible alternatives to grievance, are not indulgences but investments in public security.

The practical implications follow even if implementation is hard. Education that teaches the mechanics of moral disengagement and dehumanisation inoculates against manipulative rhetoric; Bandura and Haslam provide the curriculum. Media outlets can retire the horseshoe trope and adopt an anti-pluralism test when deciding whether a movement deserves the “extremism” label. Platforms can add friction at the point of acceleration, interstitials or down-ranking for content that pairs dehumanisation with calls to coercion, while protecting contentious but pluralist speech. Governments can align prevention with development by prioritising procedural justice, economic inclusion and local legitimacy - the very variables UNDP associates with lowered recruitment into violence. Practitioner networks such as the EU’s Radicalisation Awareness Network have long cautioned that online and offline dynamics interact; policy should follow suit.

A final empirical check is salutary. If the TE-SAT numbers show anything, it is that Europe’s threat landscape is plural and fluid. Jihadism remains the most lethal, separatism the most active in number of incidents, and right-wing and left-wing/anarchist actors persist in lower volumes with potentially high-impact lone-actor risk, alongside minors’ involvement, interest in 3-D printed weapons and the opportunistic use of AI-enabled deception. These are not reasons to abandon proportionality; they are reasons to insist on it. PGMR offers a vocabulary for doing so: treat radicalism as a democratic resource and an early-warning system; treat anti-pluralist extremism as a threshold that, once crossed, justifies exceptional vigilance; reserve coercion for where there is evidence of preparation for violence.

“The Pluralistic Gradient Model of Radicalisation clarifies the terrain, differentiates risk, anticipates misuse and gives citizens, journalists and officials a common language.”

The argument is not merely taxonomic; it is an ethic for institutions and a habit of public discourse. The vitality of a pluralist order might be measured not by the absence of radicalism, but by the presence of credible off-ramps that keep radicalism from curdling into extremism. That measure focuses attention on the architecture of participation, the ethics of enforcement and the textures of public debate. It encourages humility about what coercion can achieve and ambition about what responsive institutions can repair. In a decade likely to be defined by geopolitical shocks, climate disruption and information disorder, a clear lens may be the most practical gift theory can offer. I propose that the Pluralistic Gradient Model of Radicalisation is such a lens: it clarifies the terrain, differentiates risk, anticipates misuse and gives citizens, journalists and officials a common language. Most importantly, it anchors the defence of democracy in the very pluralism it seeks to protect.

For editors and readers who will want to verify and link scholarship: I have drawn on McCauley and Moskalenko’s Friction (mechanisms; opinion versus action), Moghaddam’s “staircase,” and Wiktorowicz on cognitive opening; on Bandura’s account of moral disengagement and Haslam’s dual-form dehumanisation; on Kruglanski, Bélanger and Gunaratna’s 3N model; on Wolfowicz et al.’s Campbell review distinguishing attitudes, intentions and behaviours; on CHES/Hooghe/Marks and Kriesi for the GAL–TAN and cleavage transformations; on Europol’s TE-SAT 2024 for the EU threat baseline; on UNDP’s Journey to Extremism in Africa for triggers and disengagement; on RAN practitioner evidence concerning online/offline interaction; and on contemporary work on polarisation and digital architectures including Bail et al., Horta Ribeiro et al. and Hosseinmardi et al. The German examples refer to §129 charges against Letzte Generation and to the convictions of the “United Patriots” linked to Reichsbürger networks.

If democratic societies adopt this differentiation as habit rather than posture, the conversation about extremism becomes less performative and more useful. The question shifts from “How do we stop radicalism?” - a question that has never had a democratic answer - to “How do we keep the gradient within the bounds of contestation?” The answer will never be simple. But a public able to name the crossing, and institutions capable of responding proportionately on either side of it, is the beginning of a politics that can be both resilient and free.

Sources

Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, M. B. F., Lee, J., Mann, M., Merhout, F., & Volfovsky, A. (2018). Exposure to opposing views on social media can increase political polarization. Proceedings of the National Academy of Sciences, 115(37), 9216–9221. https://doi.org/10.1073/pnas.1804840115

Bakker, R., Hooghe, L., Jolly, S., Marks, G., Polk, J., Rovny, J., Steenbergen, M., & Vachudova, M. A. (2020). 2019 Chapel Hill Candidate Survey (Version 2019.1) [Codebook and data set]. University of North Carolina at Chapel Hill. Available at https://chesdata.eu

Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3(3), 193–209. https://doi.org/10.1207/S15327957PSPR0303_3

Benford, R. D., & Snow, D. A. (2000). Framing processes and social movements: An overview and assessment. Annual Review of Sociology, 26, 611–639. https://doi.org/10.1146/annurev.soc.26.1.611

Bjørgo, T., & Horgan, J. (Eds.). (2009). Leaving terrorism behind: Individual and collective disengagement. Routledge. https://doi.org/10.4324/9780203884751

Commission for Countering Extremism. (2019). Challenging hateful extremism. UK Home Office.

Europol. (2024). European Union terrorism situation and trend report (TE-SAT) 2024. Europol. https://www.europol.europa.eu/cms/sites/default/files/documents/tesat_2024.pdf

Haslam, N. (2006). Dehumanization: An integrative review. Personality and Social Psychology Review, 10(3), 252–264. https://doi.org/10.1207/s15327957pspr1003_4

Horta Ribeiro, M., Ottoni, R., West, R., Almeida, V. A. F., & Meira Jr., W. (2020). Auditing radicalization pathways on YouTube. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency (pp. 131–141). Association for Computing Machinery. https://doi.org/10.1145/3351095.3372879

Hosseinmardi, H., Ghasemian, A., Clauset, A., Möbius, M., Munger, K., Rothschild, D., & Watts, D. J. (2021). Examining the consumption of radical content on YouTube. Proceedings of the National Academy of Sciences, 118(32), e2101967118. https://doi.org/10.1073/pnas.2101967118

Iyengar, S., Lelkes, Y., Levendusky, M., Malhotra, N., & Westwood, S. J. (2019). The origins and consequences of affective polarization in the United States. Annual Review of Political Science, 22, 129–146. https://doi.org/10.1146/annurev-polisci-051117-073034

Kriesi, H., Grande, E., Lachat, R., Dolezal, M., Bornschier, S., & Frey, T. (2006). Globalization and the transformation of the national political space: Six European countries compared. European Journal of Political Research, 45(6), 921–956. https://doi.org/10.1111/j.1475-6765.2006.00644.x

Kruglanski, A. W., Bélanger, J. J., & Gunaratna, R. (2019). The three pillars of radicalization: Needs, narratives, and networks. Oxford University Press.

McAdam, D., Tarrow, S., & Tilly, C. (2001). Dynamics of contention. Cambridge University Press.

McCauley, C. (2022). The ABC model: Commentary from the perspective of mechanisms of radicalization. Terrorism and Political Violence, 34(7), 1364–1372. https://doi.org/10.1080/09546553.2020.1763964

McCauley, C., & Moskalenko, S. (2011). Friction: How radicalization happens to them and us. Oxford University Press.

Moghaddam, F. M. (2005). The staircase to terrorism: A psychological exploration. American Psychologist, 60(2), 161–169. https://doi.org/10.1037/0003-066X.60.2.161

Mudde, C. (2004). The populist Zeitgeist. Government and Opposition, 39(4), 541–563. https://doi.org/10.1111/j.1477-7053.2004.00135.x

Müller, J.-W. (2016). What is populism? University of Pennsylvania Press. https://doi.org/10.9783/9780812293784

Orsini, A. (2020). What everybody should know about radicalization and violent extremism. Studies in Conflict & Terrorism. Advance online publication. https://doi.org/10.1080/1057610X.2020.1738669

Radicalisation Awareness Network. (2022). Online radicalisation: What we know. European Commission, Directorate-General for Migration and Home Affairs.

Smith, A. G. (2018). How radicalization to terrorism occurs in the United States: What research sponsored by the National Institute of Justice tells us. National Institute of Justice. https://www.ojp.gov/pdffiles1/nij/250171.pdf

United Nations Development Programme. (2017). Journey to Extremism in Africa: Drivers, incentives and the tipping point for recruitment. UNDP. https://journey-to-extremism.undp.org

United Nations Development Programme. (2023). Journey to Extremism in Africa: Pathways to recruitment and disengagement. UNDP. https://journey-to-extremism.undp.org

Wiktorowicz, Q. (2004). Joining the cause: Al-Muhajiroun and radical Islam (working paper). University of Chicago workshop The Roots of Radical Islam.

Wiktorowicz, Q. (2005). Radical Islam rising: Muslim extremism in the West. Rowman & Littlefield.

Wolfowicz, M., Litmanovitz, Y., Weisburd, D., & Hasisi, B. (2021). Cognitive and behavioral radicalization: A systematic review of the putative risk and protective factors. Campbell Systematic Reviews, 17, e1174. https://doi.org/10.1002/cl2.1174