Negative social tipping dynamics resulting from and reinforcing Earth system destabilization

In recent years, research on normatively positive social tipping dynamics in response to the climate crisis has produced invaluable insights. In contrast, relatively little attention has been given to the potentially negative social tipping processes that might unfold due to an increasingly destabilized Earth system and to how they might in turn reinforce social and ecological destabilization dynamics and/or impede positive social change. In this paper, we discuss selected potential negative social tipping processes (anomie, radicalization and polarization, displacement, conflict, and financial destabilization) linked to Earth system destabilization. We draw on related research to understand the drivers and likelihood of these negative tipping dynamics, their potential effects on human societies and the Earth system, and the potential for cascading interactions (e.g. food insecurity and displacement) contributing to systemic risks. This first attempt to provide an explorative conceptualization and empirical account of potential negative social tipping dynamics linked to Earth system destabilization is intended to motivate further research into an under-studied area that is nonetheless crucial for our ability to respond to the climate crisis and for ensuring that positive social tipping dynamics are not averted by negative ones.

1 Introduction

Recent advances in research on Earth system tipping points (ESTPs) (e.g. Armstrong McKay et al., 2022) paint an increasingly alarming picture of the state of our planetary system. Understanding tipping points and other forms of non-linear change is now widely recognized as critical to managing and responding to change in complex systems (Scheffer, 2009). We define social tipping points on the basis of mathematics of dynamical systems (Strogatz, 2000). Specifically, tipping points in dynamical social systems are critical thresholds where a small change in a variable describing the state of the social system or in a parameter capturing external influences leads to an often abrupt qualitative change in the dynamical social system; i.e. the social dynamical system undergoes a phase transition from one state to another (Winkelmann et al., 2022). Tipping occurs because positive feedback mechanisms create self-reinforcing loops, where a small change in one component of the system triggers changes that further reinforce the initial change. Tipping is further enabled by weak negative feedback mechanisms that tend to stabilize a dynamical system. Tipping is usually difficult to reverse due to hysteresis that locks the system within the new state or within the trajectory to a new state, even if the original drivers for the change are removed (Wiedermann et al., 2020; Winkelmann et al., 2022). Normatively speaking, social tipping points can be both positive (predominantly beneficial to humans and the natural systems) or negative if they result in catastrophic consequences for human societies and ecological systems (IPCC, 2022; Lenton et al., 2023).

Increasing attention is also being paid to cascade effects that connect different systems, implying that a change in one system may trigger further change in another system (Liu et al., 2023). Here, we consider a tipping cascade to take place when one tipping point triggers the crossing of another tipping point (Klose et al., 2021). We focus here on negative social tipping processes that have the potential to feed back to the Earth system, further destabilizing it; i.e. we are interested in processes where the Earth system destabilization contributes to social system destabilization, which then further destabilizes the Earth system (e.g. due to lack of cooperation), creating a potential feedback loop. We note that this paper focuses on climate ESTPs, but the same rationale can be broadly generalized to other Earth systems.

Although research on the potential for positive social tipping dynamics in various systems (e.g. food, energy, transportation, financial, and behavioural) has started to emerge (Tàbara et al., 2018; Otto et al., 2020; Lenton, 2020; Lenton et al., 2022; Winkelmann et al., 2022; Milkoreit, 2023), there has been limited research on negative social tipping dynamics that might be triggered by climate change (Laybourn et al., 2023). This is noteworthy, not least because early research on tipping points in the social sciences was mostly concerned with undesirable social processes, such as rapid and non-linear patterns of urban racial segregation in the United States (Schelling, 1978). More recently, researchers have used dynamical systems analyses to empirically study tipping points in school segregation (Spaiser et al., 2018), political instability of countries (Grimm and Schneider, 2011), and rapid proliferation of misinformation (Törnberg, 2018).

We argue that studying negative social tipping points in the context of Earth system destabilization is important because it highlights the risks generated by overshooting temperature thresholds such as 1.5 °C (Bustamante et al., 2024). While indeed every tenth of a degree of temperature increase matters, framing around climate policy is moving in the direction of making overshoot socially acceptable. Overshoots are presented as temporary, with the deployment of carbon dioxide removal (CDR) being able to recover temperatures back into the “safe zone” by the end of the century. The risks of ESTPs, however, make overshooting very dangerous, as overshooting may trigger ESTPs, which then cannot be reversed even if we return to lower global warming after the overshoot period. Triggering ESTPs on the other hand poses the risk of escalating climate change impacts (Wunderling et al., 2024). Moreover, overshooting would lead to further ecological destabilization (Singh et al., 2023), which might be reversible in terms of returning to lower global warming; however, in the meantime, ecological destabilization could trigger negative social tipping points described here, and these negative social tipping points could feedback to the Earth system, further destabilizing it, potentially leading to ESTPs being triggered. We believe that understanding these potential complex interactions is important because humans have agency and can make decisions to try to prevent such escalating processes. None of the scenarios described here is inevitable, and although many dynamics are already unfolding today, we have not reached a point of no return.

Negative and potentially catastrophic consequences are unequally distributed, both internationally and within individual societies (Pereira et al., 2024). Research has emphasized that low-income countries that have often contributed least to the destabilization of the Earth system will bear the brunt of the climate change impacts (IPCC, 2022; Lenton et al., 2023). Moreover, within each society, it is the most vulnerable groups, such as children (Thiery et al., 2021; UNICEF, 2021), women (Denton, 2002), minority groups (Berberian et al., 2022; Donaghy et al., 2023), and generally the less affluent (Thomas et al., 2019), who will be most affected by climate change impacts. Triggering negative social tipping points will have considerable consequences for these vulnerable groups, further amplifying their vulnerability and stressing the need for climate justice (Newell et al., 2021).

From this perspective, we pose the following questions. (1) What are the potential negative social tipping points that the destabilization of the Earth system could trigger? (2) To what extent could the triggering of negative social tipping points further destabilize the Earth system? (3) How do these negative tipping elements interact, and what cascades could these interactions cause? (4) What research and modelling approaches are suitable for studying negative social tipping points and cascades? (5) What intervention options are available to prevent negative social tipping points and cascades?

2 Mapping out negative social tipping

We identify five negative social tipping processes that according to some existing evidence could be triggered by Earth system destabilization (see Fig. 1). The part or subsystem of a larger system that can pass a tipping point is referred to as the tipping element. Drawing on the positive social tipping element framework developed by Otto et al. (2020), we identify four social tipping elements (TEs) that have the potential for negative tipping processes (TPs): socio-psychological systems (TE1), political systems (TE2), human settlements (TE3), and financial markets (TE4). Figure 1 provides an overview of these tipping elements and the tipping that could be triggered within these tipping elements: anomie (TP1.1), radicalization and polarization (TP1.2), displacement (TP2.1), conflict (TP3.1) and financial destabilization (TP4.1). All of these processes can unfold across different levels of social structure on different timescales and spatial scales. Specifically, tipping can occur as rapidly as hours triggered by a major shock event or unfold more slowly (years) over cascading pathways as the effects of ESTPs accumulate. Tipping can also occur only locally, affecting a specific community, or spread across a nation or the globe. Figure 1 also indicates the potential for interactions between various negative tipping elements. The interactions between different TEs indicate different possible destabilization pathways that could lead to the crossing of negative tipping points across scales. This illustrative selection is based on evidence for tipping processes in these subsystems and evidence that Earth system destabilization has a direct effect on these subsystems.

Figure 1 Tipping elements (TEs) and associated negative social tipping processes (TPs) with the potential to further destabilize the Earth system. The identified interactions between the various negative tipping processes mean that they can potentially reinforce one another, making destabilization more likely. Earth image source: https://pngimg.com/image/25350, last access: 6 September 2024 (CC BY-NC 4.0).

2.1 Anomie

The concept of anomie, which was introduced by Durkheim (1893, 1897) to describe the breakdown of norms and social order and its relationship to suicide patterns in societies, has evolved over decades of social research (Abrutyn, 2019; Twyman-Ghoshal, 2021). We define anomie as a state of a society or community that is characterized by a breakdown of social norms, social trust, social ties, and social reality, resulting in social disorder and disorganization, disorientation, and disconnection. These syndromes manifest on the individual level through mental health deterioration, increased suicide rates, and/or increased deviant behaviour (Brown, 2022; Teymoori et al., 2017). Although this is a relatively new area of research, there is increasing evidence to suggest that changes in the Earth system can contribute to anomie. For instance, anomie has been observed in the aftermath of natural disasters, made more likely by climate change (Miller, 2016) and it has been suggested that Earth system destabilization may result in a new form of anomie, called environmental anomie (Brown, 2022), where sudden changes to the physical landscape can upend the established social order and undermine people's ability to comprehend, relate to, and function within their environment. For instance, people from Paradise (California, US) that survived the devastating Camp Fire in 2018 reported how the wildfire event undermined their ability to comprehend the world around them because their familiar environment became unintelligible (e.g. they struggled to determine wind direction), meaning that they were no longer able to relate to and function within their environment. This resulted in a breakdown of self-efficacy, with a sense of unreality taking hold (e.g. burning tree branches falling from the sky). This experience of environmental anomie was further exacerbated when the affected individuals witnessed that traditional authorities were overwhelmed and unable to respond to the physical chaos, which undermined confidence and led to an individuation of suffering and feelings of social isolation, i.e. experience of general anomie. With the breakdown of social order, people were forced to fend for themselves and rules (e.g. regulating traffic) were no longer observed (Brown, 2022).

Beyond anomie resulting from extreme weather events caused by escalating climate change, there is also evidence for a rise in anomic experiences, particularly by young people and children around the world, contributing to a mental health crisis. In a first comprehensive study, surveying 10 000 children and young people (aged 16–25 years) in 10 countries (Australia, Brazil, Finland, France, India, Nigeria, Philippines, Portugal, UK, and USA), Hickman et al. (2021) found that more than 45 % said their feelings about climate change negatively affected their daily life and functioning, 75 % reported that they find the future frightening, and 83 % said they think people (adults) have failed to take care of the planet. However, it is not just the young experiencing the effects of climate change on mental health – it is negatively affecting the mental health and emotional wellbeing of people of all ages globally, but they more profoundly affect those of poor and vulnerable populations (Lawrence et al., 2021; Clayton et al., 2017), women, and Indigenous people (IPCC, 2022; Sultana, 2022). For a summary of other studies, see Fig. 2.

Figure 2 Examples of the impact of extreme weather events on mental health across the world based on Carleton (2017), Clayton et al. (2017), Jermacane et al. (2018), Atwoli et al. (2022), Hamideh et al. (2022), Lawrence et al. (2021), and Ferreira et al. (2023).

The extent of tipping dynamics in anomie have not yet been studied directly, but some studies have demonstrated tipping dynamics in phenomena that can serve as proxies for the anomic state of a society or community. Specifically, (complex) contagion processes (see Table 1) have been observed for mental disorders and distress, including suicide (Scatà et al., 2018; Paz, 2022); for deviant behaviours (Busching and Krahé, 2018); and for distrust (Ross et al., 2022). Societies or communities that are already in a zone of social instability (e.g. high rates of anti-social behaviour, increasing deviant behaviour such as crime or substance abuse, and high rates of mental health problems) due to other factors, such as poverty, rising inequality, and failing institutions (Burns, 2015), or because of a gradual erosion of social norms, which can also affect affluent communities (Piff et al., 2012; Bursztyn et al., 2020), are particular at risk to tip into an anomic state when additionally being faced with ecological destabilization (Douglas et al., 2016). Anomie tipping can also result from a single extreme event; for instance, it can be triggered by an ESTP being breached. Such an event can instantly disintegrate whole communities, scattering members of the community in the aftermath (i.e. interaction with displacement), leaving them with depleted social and mental resources (Miller, 2016) and establishing the perception that society as a whole is failing (Teymoori et al., 2017). Tipping in this case can be described using logistic map models (Bruun et al., 2017), which can model how coupled systems can tumble towards chaotic system behaviour (see Table 1). Natural and human-caused disasters can bring communities together and strengthen cooperation; however, research suggests that when the experience of solidarity and unity in the disaster aftermath starts to wane, communities can experience increasing disillusionment and depression followed by social disintegration (i.e. anomie) if they are left without adequate, long-term support (Townshend et al., 2015).

Table 1 Models and methodological approaches for studying negative social tipping points and cascades.

Anomie can have feedback effects on the Earth system, further destabilizing it through various pathways. When social norms disintegrate, certain pro-social behaviours and collective actions that are necessary to slow down the climate crisis may diminish (Constantino et al., 2022; Schneider and van der Linden, 2023; Lettinga et al., 2020). Without strong social norms and social ties supporting collective action and fostering reciprocity, trust, and cooperation, it becomes increasingly challenging to implement effective measures to address accelerating Earth system destabilization, hence increasing the likelihood for Earth system tipping (Fehr et al., 2002; Thøgersen, 2008; Malerba, 2022). Moreover, mental health problems weaken people's capacity to seek solutions, fostering collective inertia and increasing susceptibility to conspiracy theories, potentially further undermining trust and cooperation to prevent further Earth destabilization (Burden et al., 2017; de la Sablonnière and Taylor, 2020; Green et al., 2023).

2.2 Radicalization and polarization

Radicalization can be a reaction to perceived external threats, including ecological threats. Research suggests that people can respond to climate change and other ecological threats by becoming more authoritarian and derogative against out-groups (Fritsche et al., 2012; Jackson et al., 2019; Taylor, 2019; Russo et al., 2020; Uenal et al., 2021; Spaiser et al., 2024). This effect can be further exacerbated by the well-documented effect of heat on aggressive behaviours, including online hate speech (Stechemesser et al., 2022). Current trends seem to suggest increasing polarization (Dunlap et al., 2016; Vihma et al., 2021; Cole et al., 2023; Smith et al., 2024), i.e. a rise of the political right, which is increasingly attracting the political centre (Levitsky and Ziblatt, 2018; Halikiopoulou, 2018; Layton et al., 2021); obstructing climate action; and increasingly diverging from the political left and centre-left, which is demanding climate action (Aasen, 2017; Lockwood, 2018; Gustafson et al., 2019). This polarization is also driven indirectly by Earth destabilization, as it is at least partly a response to climate mitigation policies that are perceived as a threat to the existing socio-economic system, status, and identity (Dunlap et al., 2016; Hoffarth and Hodson, 2016; Daggett, 2018; Clarke et al., 2019; Benegal and Homan, 2021; Ehret et al., 2022; Brännlund and Peterson, 2024) and can be further exacerbated by inequality and general economic decline (Winkler, 2019; Stewart et al., 2020; Hübscher et al., 2023), which again can be partly linked to Earth destabilization at least in some parts of the world (Méjean et al., 2024; Dietz et al., 2021). However, as climate change progresses and becomes a more concrete existential threat throughout the world (Huggel et al., 2022), we may see even socially liberal individuals developing increasingly authoritarian and reactionary views (Gadarian, 2010; Hetherington and Suhay, 2011; Huddy and Feldmann, 2011; Hirsch, 2022). At that stage we may see radicalization taking a different direction, with currently fringe political ideologies such as ecofascism taking hold. Ecofascism reinterprets white supremacy ideology in the context of the climate and ecological crisis with the goal to defend habitable areas for the white race and decrease world population (Taylor, 2019). A couple of recent right-wing terrorists have already self-identified as ecofascists, such as Brenton Tarrant, who killed 51 people during a terror attack on a mosque in Christchurch, New Zealand, in 2019. A few months later, Patrick Wood Crusius killed 23 people in El Paso, United States, legitimizing his actions again with ecofascist ideologies (Achenbach, 2019). Certain ecofascist themes seem to also appear increasingly in public debates (Thomas and Gosink, 2021).

Radicalization can exhibit tipping dynamics. Research has described radicalization, e.g. the spread of right-wing ideology (Youngblood, 2020), through complex contagion processes (see Table 1). Similarly, the spreading of extremist content on social media has been observed to follow complex contagion processes (Ferrara, 2017). Indeed, polarization and radicalization around climate change have been observed to be on the rise online (Weber et al., 2020; Teen et al., 2020; Falkenberg et al., 2022), at times displaying non-linear, accelerating diffusion dynamics (Centre for Countering Digital Hate, 2023), and these processes are fuelled by corporate funding (Farrell, 2016; Teen et al., 2020). Moreover, processes of “cross-pollination”, the merging or previously separate radical clusters facilitating further contagion, have been documented (Kimmel, 2018; Baele et al., 2023), including for climate denial (Agius et al., 2020). Polarization has also been observed to follow tipping dynamics. Leonard et al. (2021) describe, for instance, how subtle public opinion shifts from left and right can have a differential effect on the self-reinforcing processes of elites in the US, causing Republicans to polarize more quickly than Democrats. As self-reinforcement pushes societies toward the critical threshold, polarization speeds up. Political polarization tipping, often accompanying radicalization of certain segments of the population, has been found to be difficult to reverse due to asymmetric self-perpetuating trajectories (Macy et al., 2021).

Radicalization and polarization can have feedback effects on the Earth's system, destabilizing it further. According to research (Stanley et al., 2017; Stanley and Wilson, 2019; Julhä and Hellmer, 2020), authoritarian and social dominance attitudes are negatively related to environmental attitudes and support for environmental and climate change policies. Indeed, right-wing ideology has been repeatedly correlated with climate change denial (Hornsey et al., 2016; Hoffarth and Hodson, 2016; Czarnek et al., 2020; Julhä and Hellmer, 2020). When climate change is denied, no attempts are made to mitigate climate change. On the contrary, decisions may be taken to further prop up high-emitting industries (Ekberg et al., 2023; Darian-Smith, 2023). There is, however, increasingly a retreat of pure climate denial (primary climate obstruction), instead we see a rise in secondary and tertiary climate obstruction, which can include deliberate polarization of societies on the issue (Kousser and Trantr, 2018; Goldberg and Vandenberg, 2019; Lamb et al., 2020; Mann, 2021; Flores et al., 2022; Ekberg et al., 2023; Burgess et al., 2024). Moreover, research demonstrates that the increasing success of the radical right also influences the policies of mainstream parties (Abou-Chadi and Krause, 2020); i.e. even if radical parties are not in government, they still can undermine climate policies.

2.3 Displacement

Acute and slow-onset environmental pressures, such as heatwaves, long-term temperature and humidity changes, extreme weather events, and sea level rise (e.g. due to the melting of Greenland glaciers, and the West Antarctic Ice Sheet), are likely to impact the migration (voluntary) and displacement (forced, involuntary) circumstances of a large proportion of the global population (Mastrorillo et al., 2016; Berlemann et al., 2020; Hauer et al., 2020; Hoffmann et al., 2020; Lu and Romps, 2023). In the context of ESTPs, sea-level rise is projected to be one of the most costly and irreversible consequences of climate change (Hauer et al., 2020; McLeman, 2018; Kaczan and Orgill-Meyer, 2020; Armstrong McKay et al., 2022). Another rapid-onset hazard is land degradation due to permafrost melt, both in coastal areas and inland (Irrgang et al., 2022; Streletskiy et al., 2023). Accelerated Polar warming or Arctic amplification warms Arctic surface temperatures by a factor 2–4 times faster than the rest of the globe (Rantanen et al., 2022), which – in addition to the direct impact on permafrost thawing – results in the loss of protective sea ice and consequently rapidly increasing coastal erosion (Casas-Prat and Wang 2020; Nielsen et al., 2022; Wunderling et al., 2024). As the proportion of the global population living in coastal regions continues to grow, likely surpassing a billion people this century, this will have profound implications for both individuals and societies (Hauer et al., 2020; McLeman, 2018; Kaczan and Orgill-Meyer, 2020). However, sea level rise is not the only driver of adaptive mobility (Gioli et al., 2016). Even if international efforts towards mitigating climate change are successful (RCP 4.5 – low emissions scenario), models have projected drought-induced international displacement to increase substantially by the end of the 21st century. High emissions scenarios (e.g. RCP 8.5) would push the number of displaced due to droughts up even further (Smirnov et al., 2023).

Displacement can happen suddenly, and amplifying or positive feedbacks can increase or maintain the dislocation of populations even after the extreme weather event or initial shock has passed. This can create a cycle that reinforces, extends, or renders the displacement permanent. Displaced populations must grapple with the loss of their livelihoods, often by identifying new temporary sources of income that can become permanent due to the challenges of returning to origin communities (Young and Jacobsen, 2013; Wilson, 2020). The displacement is often linked with turning away from traditional ways of life and economical support, e.g. in the cases of Arctic Inuit population fishing, hunting, and trapping (Ford et al., 2023; Streletskiy et al., 2023), and the movement away from traditional agricultural and pastoralist livelihoods in areas of central and southwestern Africa (Akinbami, 2021; Thorn et al., 2023). This can result in cultural heritage loss (Pearson, 2023). These compounding and reinforcing effects can exacerbate pre-existing social inequities and determine the pattern of displacement (e.g. short-term or long-term to permanent) among different populations (Lama, 2021; Boas et al., 2022). Additionally, with slow-onset events, decisions to migrate can be driven by social networks and connections; when members of a community migrate, others may make the decision to follow (Manchin and Orazbayev, 2018; Thorn et al., 2023; Tubi and Israeli, 2023). This can, in and of itself, be subject to tipping dynamics; when a certain percentage of a community has left, this has been observed to negatively impact those left behind, potentially triggering subsequent outmigration (Rai, 2022).

In the absence of appropriate governance mechanisms and protocols for how and where to relocate displaced communities, negative feedback consequences for the Earth systems are possible (Islam et al., 2021; Thorn et al., 2023). Hosting communities may face strains on their natural resources and/or sinks to meet the additional needs of the displaced. For example, Tafere (2018) identified environmental degradation resulting from the influx of displaced populations in eastern Africa, often in environmentally sensitive (e.g. protected forests) or already strained regions (e.g. arid or semi-arid areas). Such straining of ecological systems to accommodate increased eco-service demand due to forced migration could contribute to accelerating regional ecological destabilization at the very least.

2.4 Conflict

Despite growing concerns about conflict, the causal link between climate change and conflicts and their underlying dynamics remain debated (Burke et al., 2009; Buhaug, 2010; Buhaug et al., 2014; Solow, 2013; Kelley et al., 2015; Selby et al., 2017). While statistical models inferred either significant coincidences of particular civil conflict events with concurrent climate extreme events or significant associations of warming and drought trends with civil conflict trends, many qualitative in-depth assessments of the particular civil conflict events and their underlying mechanisms dismiss such coincidences and associations (e.g. Buhaug, 2010; Selby et al., 2017). Though not the only cause (Sakaguchi et al., 2017; Mach et al., 2019; Scartozzi, 2020; Ge et al., 2022), climate change undermines human livelihoods and security because it increases the vulnerability of populations (e.g. to extreme events, food and water scarcity), grievances, and political tensions through an array of indirect – at times non-linear and latent (i.e. not measurable) – pathways, thereby increasing human insecurity and the risk of violent conflict (Scheffran et al., 2012; van Baalen and Mobjörk, 2017; Koubi, 2019; von Uexkull and Buhaug, 2021; Ide et al., 2023). It is difficult to separate mutually enforcing vulnerabilities to both climate and conflict that trigger an escalating spiral of violence and amplify cascading crisis events beyond critical thresholds (Buhaug and von Uexkull, 2021) and connected through telecoupling (Franzke et al., 2022).

Many conflicts can be described in terms of social tipping mechanisms, which can be triggered by Earth system destabilization, where causal mechanisms are inferred using data (Sun et al., 2022) and can be modelled through socially connected tipping dynamics, for instance using the logistic map approach (see Table 1) (Guo et al., 2018; Aquino et al., 2019; Ge et al., 2022; Guo et al., 2023). Using a complex systems lens and connecting the human–environmental–climate security (HECS) nexus framework (Daoudy, 2021; Daoudy et al., 2022; Scheffran et al., 2012) and the social feedback loop (SFL) framework (Kolmes, 2008) can help clarify conflict tipping mechanisms in coupled social–ecological systems. The HECS framework infers that climatic drivers of civil conflicts are best understood as a result of policy decisions and governance that reflect the ideology and preferences of ruling elites or ethnic bias instead of investigating the direct functions of climate extremes. SFL suggests that initial social disruptions directly caused by gradual climate change and climate extreme events can themselves generate a distinct positive feedback loop leading to self-accelerating rates of societal disintegration and to civil conflicts (Kolmes, 2008). In turn, using a combined HECS–SFL lens, civil conflicts can be perceived as amplified social disintegration and disruption resulting from societal and political responses to the initial disintegration and disruptions caused directly by climate extremes and climate change (Scheffran et al., 2023). Self-reinforcing feedbacks emerge in social–ecological systems as a result of complex interactions among socio-economic, environmental, and political events and variables, such as institutional capacity for solving social–ecological problems initially caused by climate change (Daoudy et al., 2022). These complex interactions result in the amplification of social–ecological shocks that climate change and extremes initially caused and potentially disrupt and negatively tip the system in concern to a conflict state. The affected system becomes entrapped in the conflict state until sufficient incentives can move it out. However, there are gaps in our understanding of latent mechanisms that introduce variable delay (e.g. slow social transformations), confounding factors, non-linear bifurcations (e.g. some transformations are irreversible), and regional variability.

When conflicts escalate, exhibiting a tipping dynamic, they can in turn impact the Earth system, either directly as warfare itself is producing excessive greenhouse gas (GHG) emissions or by destroying vital ecosystems such as forests, as is, for instance, currently the case of Russia's war in Ukraine (de Klerk et al., 2022). For example, the Kakhovka Dam was destroyed in 2023 during the Russia–Ukraine conflict. Early assessments (UKCEH and HRW, 2023; UNEP, 2023) indicated a maximum downstream flood extent of around 83 000 ha (6–9 March 2023), including inundation of downstream urban areas and disruption of irrigation for agriculture, water supply, and sanitation systems. Over half a million hectares of habitat of conservation importance was estimated to have been affected by the dam breach, from the upstream Kakhovka Reservoir and its wetland habitats to the downstream Black Sea Biosphere Reserve. This impact area covered the distribution of 567 species that have a listing on the IUCN European Red List, 28 of these species have a threat status of vulnerable or worse. There were also concerns about the supply of cooling water to the upstream Zaporizhzhia Nuclear Power Plant, i.e. one war-induced ecological disaster could have resulted in another ecological disaster. Illegal logging, deforestation, and charcoal production also support militia in many protracted conflicts throughout Africa (Branch et al., 2023). However, even beyond involvement in war activities, everyday military operations directly generate vast emissions of GHGs (Kester and Sovacool, 2017; Crawford, 2019). The feedback impact of conflicts on the Earth system can also occur indirectly through impeding humanity's ability to collaborate to find solutions to global challenges such as climate change. Within societies entangled in a conflict, resources are diverted to winning the conflict rather than to mitigate climate change, also affecting a country's environmental governance mechanisms. Finally, the continued presence of a large number of tactical and nuclear weapons represents a significant threat to global climate and other Earth system processes (Turco et al., 1983; Xia et al., 2022).

2.5 Financial destabilization

The impacts of Earth system destabilization on the financial sector are now receiving increasing attention (Ameli et al., 2023; Chenet, 2024), with studies suggesting that climate-related damages will impact the stability of the global banking system significantly (Lamperti et al., 2019), as will biodiversity loss (Kedward et al., 2023). For instance, stocks of capital at risk due to climate-induced extremes and more frequent weather events such as floods would adversely affect insurance companies (Lamperti et al., 2019). Reinsurance companies are withdrawing increasingly from areas exposed to high climate change risks, e.g. areas vulnerable to wildfires and floods (Frank, 2023). Earth system destabilization is likely to result in stranded assets (Caldecott et al., 2021). Escalating climate change can also destroy the capital of firms, reduce their profitability, deteriorate their liquidity, reduce the productivity of their workforce, leading to a higher rate of default, harming the financial sector and the economy in general (Dafermos et al., 2018; Dietz et al., 2021). One issue with the existing empirical evidence and models that try to estimate climate damage for the financial sector is, however, that they do not account for ESTPs (Keen et al., 2022; Kedward et al., 2023; Trust et al., 2023; Marsden et al., 2024).

Still, the first advances are being made. Martin et al. (2024) propose an integrated dynamic environment–economic model for the coupling of an Earth model of intermediate complexity and a non-linear macroeconomic model in continuous time. Using this model, they found that above a warming of about +2.3 °C, damages drastically foster the need for additional investments in productive capital for adaptation, which could potentially lead to the emergence of private-debt tipping points and a worldwide cascade of defaults. The inability to repay obligations generates non-performing loans (or bad debt) in the balance sheets of banks and other financial institutions, with possible systemic implications such as those experienced during the 2008 global financial crisis. It is estimated that climate change will increase the frequency of banking crises by 26 % to 248 % depending on the extent of climate change (Lamperti et al., 2019). If the banks' equity deterioration due to economic imbalances reaches a certain threshold, secondary systemic effects can be triggered. Financial institutions exposed to troubled banks would suffer losses in the market value of their assets, potentially triggering contagion phenomena (Kiyotaki and Moore, 2002; Yan et al., 2010; Roukny et al., 2013; Chinazzi and Fagiolo, 2015). These contagion phenomena can result in a financial tipping point being reached, where contagion becomes self-perpetuating due to feedback loops in the system that amplify the initial shocks (May et al., 2008; Gai and Kapadia, 2010; Haldane and May, 2011). If ESTPs are triggered, destroying assets and the economic productivity of whole regions, we can expect rapid non-linear tipping effects in the coupled financial sector (Battiston et al., 2017). The financial and economic system would eventually settle into a new state, although this state may be characterized by recession, high unemployment, austerity, and other deteriorating economic conditions. The consequences of such a financial upheaval are often a rapid increase in social instability (i.e. interaction with anomie) and radicalization (i.e. interaction with radicalization) as more people are forced to compete for basic needs (i.e. interaction with conflict) (Dietz et al., 2021).

This could also impact societies' abilities to mitigate climate change, thus risking the derailment of sustainability transition (Laybourn et al., 2023). Governments will likely try to stabilize financial markets through bailing-out policies, such as providing fresh capital and saving insolvent banks, and it is predicted that climate change will likely increase the frequency of bailouts (Lamperti et al., 2019). Recent government bailouts in response to COVID-19 have shown a distinct lack of sustainability focus (Rockström et al., 2023). Bailouts negatively affect the public budget and lead to increasing government debts, leaving decreasing resources for addressing Earth system destabilization, for instance through effective climate change mitigation measures. Financial destabilization would also deplete businesses and individuals of resources to invest in post-carbon transition (Laybourn et al., 2023).

3 Cascading negative social tipping dynamics

The basis for many tipping point behaviours in social–ecological systems is a non-linear relationship between critical pairs of variables. Non-linearities create disproportionate relationships between cause and effect, potentially leading to change that is faster, more intense, or more extensive than expected (and hence, harder to reverse or control). Cascades, as defined by Klose et al. (2021), are sequential occurrences of events in which an initial event triggers a series of subsequent events and are one important attribute of systemic risk (Sillmann et al., 2022). Cascades are more likely when multiple variables within a given system exhibit and transform non-linear relationships to each other; i.e. when coupled, these relationships transform in ways that often cannot be understood. Crossing multiple negative tipping points in diverse systems increases the likelihood of (partial or localized) societal collapse.

In the context of migration, this can manifest as a domino effect, where an environmental or socio-political event causes involuntary displacement or voluntary migration as people search for improved living conditions and better economic opportunities. This is well documented in the Lake Chad basin case where climate change and unsustainable resource management affect the sustainability of natural resources, increasing vulnerability and leading to coping strategies such as migration (McLeman et al., 2021). In Ukraine, the war-induced ecological devastation in the aftermath of the Kakhovka Dam destruction has displaced thousands of people, and a major humanitarian programme was initiated in response (WHO, 2023).

A possible tipping cascade can be identified between climate change, food insecurity, and migration. The last 5 years have seen an increase in food insecurity, representing a problematic reversal of the progress done since the 1990s to reduce world hunger (FAO et al., 2022). Climate tipping points could dramatically impact food security through direct impacts on production (availability) and indirect impacts on access to food when displacement occurs. One of the most direct ways in which tipping points can affect food insecurity is through changes in rainfall distribution, which would render agricultural livelihoods in rainfed regions unfeasible without irrigation (or other) technologies (Giannini et al., 2017; Benton et al., 2017). Indeed, even in the most optimistic climate mitigation scenarios, which would lead to a temporary overshoot over 1.5 °C and then return to temperatures below that threshold, a tipping point might occur in precipitation patterns, which can result in adverse food security impacts (cf. Ritchie et al., 2020). Additionally, recent studies suggest that escalating climate change could result in concurrent weather extremes driven by a strongly meandering jet stream, which could trigger simultaneous harvest failures across major crop-producing regions, posing a serious threat to global food security (Kornhuber et al., 2023). Food security can change seasonally. As such, food security does not exhibit traditional bifurcation in the sense of irreversibility. However, a permanent change towards a state of food insecurity would be catastrophic, representing a permanent food crisis. Krishnamurthy et al. (2022) offer a framework to identify “transitions” as prolonged periods of food insecurity (Fig. 3), using the Integrated Food Security Phase Classification (IPC), the leading global metric for standardized food security assessment, which combines data on agricultural production, food prices, nutrition rates, weather patterns, and other variables to determine the general food security situation in a given location. With these metrics, a tipping point in a food system can be thought of as a shift between periods with minimal food insecurity (IPC 1 or 2) to periods of sustained food crisis (IPC 3 or higher). An example of a potential tipping point using the IPC categories was found in eastern Africa in 2015/2016 due to anomalously low rainfall in both the summer and autumn. This trend, combined with insufficient drought preparedness, resulted in crop failures and livestock mortality and consequently a depletion of livelihood assets, food stocks, and overall food security in northern and eastern regions of Ethiopia (Fig. 3).

Figure 3 Example of a “tipping point” in the context of food security, showing the transition from stable food security conditions to a food crisis resulting from drought in Ethiopia (Source: Krishnamurthy et al., 2020; FEWS NET 2018).

The links between food insecurity and migration are complex; severe food insecurity has been found to trap people that wish to migrate but are unable to (Sadiddin et al., 2019), but there is also evidence that migration can be driven by food insecurity (Smith and Wesselbaum, 2022). Migration flows are also impacted by climate change directly (i.e. the local environment becomes unsuitable for favourable habitation) and indirectly (i.e. by impacting relative wages through effects on farmers' crop yields). A climate disaster, for instance triggered by a climate tipping point being breached, may also lead to sudden displacement, whether temporary or permanent. To summarize, a cascading dynamic plays out when various tipping points become coupled, for instance when the tipping in an Earth system triggers the tipping in food insecurity and potentially simultaneously a tipping in displacement, which may in turn reinforce food insecurity.

Other potential cascading links exist as well. For instance, societies may tip into a state of conflict because of competition over dwindling resources as tipping in food insecurity occurs, and conflicts in turn may reinforce food insecurity, a cascade made likely when institutions are weak and governance fails (Martin-Shields and Stojetz, 2019; Anderson et al., 2021; Shemyakina, 2022). Radicalization and polarization can fuel conflicts (McNeil-Willson et al., 2019; Rousseau et al., 2021). Radicalization and polarization has been also observed in countries hosting displaced communities (Ravndal, 2018), a link often moderated by socio-economic inequality and perceived insecurity. Radicalization, polarization, and anomie can reinforce each other as well. Research suggests, for instance, that people trust each other less in countries with greater polarization (Rapp, 2016). On the other hand, people with mental health issues are more susceptible to conspiracy theories, which can fuel radicalization (Green et al., 2023). Finally, financial destabilization can be a driver for radicalization, polarization, and anomie (Funke et al., 2016; Bygnes, 2017; Doerr et al., 2022). However, these and other potential cascading links and processes are still little researched and understood.

Table 2 Negative (social) tipping points and options for prevention and impact management.

4 Emerging research questions and intervention options

4.1 Methods and models and emerging data questions

Various methods and approaches have been suggested for the study of tipping processes in social and socio-ecological systems, which can be used to study negative social tipping points and the cascading interactions between them. In Table 1 we discuss the most prevalent methods and some new emerging approaches. We would like to emphasize here that we are not suggesting that negative social tipping points are knowable in advance, in terms of determining or predicting either the exact threshold or the time when a tipping will occur. In fact, the knowability of tipping points is a challenge not only for social tipping points but also for ESTPs (Boulton et al., 2023). It is usually only possible to determine a tipping point subsequently. However, even then there is often not a single negative social tipping point, the exact threshold may vary, for instance, from one country to another or from one community to another (e.g. Spaiser et al., 2018, who derive data-specific segregation tipping points for various schools located on a curve), as the setup of reinforcing and dampening feedbacks will be different in every context. This is also true for some ecological tipping points; e.g. different lakes will have different tipping points (Hessen et al., 2023). The methods we are suggesting here are useful (1) to study tipping processes once they have occurred or to generate various model-based scenarios to build our general understanding of tipping processes so we are better equipped to respond to them and (2) to build early warning systems that could potentially capture a system becoming more unstable, chaotic, or exhibiting more unusual behaviour before a tipping point has been reached (Dakos et al., 2023). The purpose is to increase our agency (see Sect. 4.2).

We are also conscious that all models are oversimplifications of many stories, perspectives, and detailed mechanisms. Tipping models can be higher-dimensional, but even a low-dimensional tipping model, such as a neural network (see Table 1), can be used to estimate tipping parameters. In effect, a simple model provides a projection of more complex mechanisms in a function space. The main questions are how much information we lose in projecting to a tipping model compared to a projection to a different model and how useful the projection is in enhancing our understanding of underlying mechanisms and in determining agency pathways. We believe tipping model development is important to advance our understanding and enhance our agency, but we also advocate the comparison of different models to identify the most useful model.

Further emerging data questions include the following research avenues.

• What are the most relevant and appropriate datasets for early warning of negative social tipping points? Social tipping points are more complex than physical tipping points due to the interacting relationships between climate parameters and social responses. Given this complexity, there is a need to identify relevant data sources and methods that can be used to detect and anticipate tipping points. Recent advances in machine learning and increasing digital social data all offer an unprecedented opportunity to understand early warning signals for social tipping points. Once datasets are identified, ensuring that these are accessible and usable for analysis is highly important. Moving forward, it will be important to consider sharing platforms to ensure access.

• What are the characteristics of datasets that can render them more (or less) useful for detecting social tipping points? A key practical question for tipping point analysis is whether there are specific characteristics that make datasets more appropriate for detection of critical transitions. Early warning of tipping points ultimately depends on reliable, high-frequency data (Scheffer et al., 2009; Dakos et al., 2015). For example, in an analysis of data requirements for early warning of food security tipping points, Krishnamurthy et al. (2020) highlighted the importance of temporal resolution over spatial resolution to detect autocorrelation or flickering in coupled climate–food systems. However, research has shown that even limited datasets such as Soil Moisture Active Passive (SMAP) can provide game-changing opportunities for detecting food security transitions (Krishnamurthy et al., 2022).

• Which early warning signals are more meaningful for different applications? Identifying the most useful metrics and statistics for early warnings of tipping points translates to actionable information, but it requires a clear understanding of underlying system functioning and mechanisms. For instance, in food security applications, autocorrelation is the key metric used to detect a transition in food security states, with the rolling average statistic indicating the direction of the transition (Krishnamurthy et al., 2022). Such insights can help leverage resources in a timely fashion to avert negative effects associated with social systems that exhibit tipping points. Moreover, probabilistic insights from research on collective social dynamics may complement insights from new early warning signals for social tipping. These approaches identify measurable qualities of social systems or networks, such as heterogeneity, connectivity, and individual-based thresholds that make social tipping points more likely (Bentley et al., 2014). For maximum efficacy, these modelling efforts should derive from both qualitative and quantitative methods so that they benefit from both data and lived experience.

4.2 Intervention options and emerging policy questions

Given that negative social tipping points are under-researched, there is little knowledge on how they can be prevented or managed. As noted for instance by Milkoreit et al. (2024), social tipping point governance has not really been developed yet. In Table 2, we nevertheless provide a preliminary overview of potential intervention options linked to the discussed negative social tipping points and their main potential interactions. Future research needs to focus on identifying other potential intervention options and tying these together into a coherent tipping points governance framework. Ultimately, effective governance of negative (social) tipping points will hinge upon the understanding of collective social dynamics and proactive resource-based interventions. The main line of agency we would like to emphasize is the strengthening of societal institutions and polycentric governance mechanisms (Carlisle and Gruby, 2019; Morrison et al., 2023). We also would like to emphasize agency in driving positive social tipping processes that improve the long-term sustainability and wellbeing of people the and planet (Gaupp et al., 2023) and prevent societies sliding into negative social tipping dynamics.

Further emerging policy questions include the following avenues.

• How do multiple climate extremes and other shocks and stressors combine, especially regarding slow-onset climate change processes that drive systemic changes and tipping points? Evidence provided here suggests that severe climate events, such as droughts and hurricanes, can result in highly complex social change, including negative social tipping points. Additional research is required to understand if and how climate and social tipping points interact and whether one tipping point can result in a plethora of other transitions.

• As critical transitions unfold, how does the risk landscape shift in response? Societies respond to environmental stress and resource scarcities. However, these responses may lead to new risks. Understanding how critical transitions affect the current (and future) risk landscape can provide essential information for decision-makers to prioritize investments in adaptation and mitigation.

• What are the processes required to integrate research into policy-making? There is growing research on early warning signals for tipping points. However, once suitable datasets and early warning diagnostics are identified, what are the enabling processes and steps required to integrate actionable early warning systems into decision-making? New data analytics, dashboards, and communications material may go a long way towards facilitating the transition to early warning systems of tipping points that can translate into action.

5 Conclusions

We mapped selected key potential negative social tipping points and their potential cascading interactions. We have also briefly discussed potential intervention options and provided examples of methods and models that need to be advanced in the future. We do not claim to have captured all possible social negative tipping points in the context of Earth system destabilization, and we acknowledge that other social subsystems could experience negative tipping points as well, e.g. breakdown of (certain) global supply chains (Marcucci et al., 2022) or breakdown of the public health system (at least in certain areas) triggered by a massive freak heat event or the breakout of a disease due to climate change (Sharma, 2023; Skinner et al., 2023). Our goal is to highlight that if societies fail to stabilize the Earth system through decarbonization, land use reallocation, and other measures, societies will not merely stay in the business-as-usual state. Through mechanisms of negative social tipping accompanying further Earth system destabilization, they instead risk transitioning into a new social system state, which may be characterized by greater impoverishment, authoritarianism, hostility, discord, violence, conflict, and alienation. Societies more vulnerable to climate change are likely to experience such negative social tipping sooner, but this will inevitably have knock-on effects globally. It is increasingly likely that in some regions large-scale climate adaptation will need to be undertaken to reduce vulnerabilities to the current and future magnitude of climate change.

The acceleration of climate tipping points perpetuates a vicious cycle that weakens societies and their abilities to respond, feeding further Earth system destabilization. This vicious cycle is also fed by widening socioeconomic inequalities (Millward-Hopkins, 2022). As the consequences of climate change intensify, societal trust, cooperation, and altruism may erode due to increased competition for scarce resources, displacement of populations, and other climate-related challenges. Our knowledge of negative social tipping points is still very patchy and fragmented, with many estimations and models likely to be underestimating the effects of breaching Earth system tipping points. This is particularly true for economic and financial sector models (Marsden et al., 2024). Researchers (Keen et al., 2022) are advocating for developing future loss calculations in close collaboration with climate scientists to ensure adequate representation of climate catastrophes.