Le texte de l’ouvrage est en anglais, tel qu’il a été écrit ; seule l’interface du site est en français.

11. Preventive peace and resilience in the face of crises

Chapitre 11 sur 15 · ≈ 16 min de lecture · 38 paragraphes · 17ᵉ des 40 sections de lecture · rechercher dans l’ouvrage

Instrumented accountability is, upstream, an infrastructure of peace.

This chapter shows how to transform cross-border grievances into evidence, dialogue, proportionate responses and peaceful avenues of redress. It extends the chain to two neighbouring functions — conflict prevention and crisis resilience — by arguing that instrumented accountability is, upstream, an infrastructure of peace.¶ 1

11.1 Executive summary

This chapter specifies the conditions under which the Global Accountability Chain may contribute to peace without being wrongly presented as a treaty putting an end to a war or as a substitute for the existing mechanisms of diplomacy, mediation, collective security or peacebuilding. Its potential contribution lies upstream: making displaced effects visible, distinguishing harm from attribution, opening a structured adversarial process, financing a response and offering avenues of redress before the grievance is absorbed into the balance of power.¶ 2

The thesis is conditional. An economic, environmental, technological or informational externality does not lead mechanically to violence. It may, however, feed a sequence of destabilisation where its costs are concentrated, its causes opaque, the responses asymmetric and the institutions of contestation weak. The architecture can interrupt this sequence at several points: independent measurement, public qualification, mediation, voluntary or financial correction, financing of prevention, certification of digital systems and judicial review. It produces peace only if these functions are exercised with consent, impartiality, proportionality, participation and evidentiary discipline.¶ 3

This proposal is in line with the prevention orientation carried by the United Nations, the World Bank and the Pact for the Future: development processes, diplomacy, mediation, human rights and institutional capacities must be articulated before escalation, not only after the crisis (United Nations and World Bank, 2018; United Nations Secretary-General, 2023; United Nations General Assembly, 2024). The Global Accountability Chain adds to this orientation a procedural grammar: who measures, who qualifies, who decides, who finances, who certifies, and before which forum the decision may be contested.¶ 4

11.2 Purpose, scope and status of the proposal

Purpose. The present chapter makes explicit the possible contribution of the 7+3 architecture to preventive and structural peace. It does not add an eighth institution. It identifies a cross-cutting purpose and a protocol for the use of the existing functions where cross-border or intergenerational effects risk feeding a durable contestation, a retaliation, a polarisation or an escalation.¶ 5

Perimeter. The disputes concerned are those in which a decision that is lawful or rational within its space of origin displaces costs, risks or vulnerabilities onto actors who have neither participated in the decision nor access to a credible adversarial procedure. Armed conflicts, ceasefires, borders, disarmament and peace operations remain governed by the competent institutions. The chain can only complement these arrangements by measuring the economic, financial, environmental, technological and distributive causes that precede them, prolong them or complicate their settlement.¶ 6

Epistemic status. This is a design proposal, not an established causal result. The link between accountability and the reduction of violence must be tested by pilots, comparisons and independent evaluations. Success cannot be inferred from the coherence of the architecture alone.¶ 7

Principle of non-substitution. The chain replaces neither national authorities, nor regional organisations, nor the preventive diplomacy of the United Nations, nor mediation, nor existing jurisdictions. It provides a common, versioned and contestable case file capable of improving their action.¶ 8

11.3 From peace as the absence of violence to peace as institutional capacity

Peace may be understood at three levels which must not be confused. The first is the absence or the cessation of organised violence. The second is prevention: the capacity to address tensions before they cross a threshold of escalation. The third is structural: the existence of legitimate institutions making it possible to render grievances visible, to apportion burdens, to correct errors and to preserve the dignity of the parties. The Global Accountability Chain intervenes principally at the second and third levels.¶ 9

This reading does not reduce peace to development, nor conflict to poverty. The work on prevention places more emphasis on configurations of exclusion, perceived injustice, low legitimacy, shock, opportunity for mobilisation and political competition than on a single cause. The contribution of the architecture is therefore procedural: it seeks to prevent a material dispute from becoming an identity-based or sovereigntist confrontation for want of a common language and a credible forum.¶ 10

Preventive peace does not mean suppressing all disagreement. It means making disagreement manageable. A pacifying accountability regime must allow the parties to contest the data, the models, the attribution, the values and the consequences without being obliged to contest the existence or the legitimacy of the other party. The quality of the procedure then becomes an asset of peace.¶ 11

11.4 Theory of change: from the ungoverned grievance to escalation, then to de-escalation

The theory of change rests on a cautious causal chain. A displaced effect does not automatically produce a crisis. It becomes politically dangerous when it is large or repeated, when it affects groups that are already vulnerable, when its causes are opaque, when responsibilities are narrated in incompatible ways and when no credible avenue of correction exists. The grievance may then be aggregated by political actors, linked to a history of injustice and converted into polarisation, retaliation or violence.¶ 12

Télécharger ce tableau (CSV)

The chain does not require all the stages to be activated. A case file may stop after measurement if the attribution is too weak; after qualification if no consequence is admissible; after mediation if a voluntary correction is accepted; or before the court if the contestation bears on legality or procedure. The capacity to stop is a condition of peace: an institutional machine that must always conclude and sanction creates its own incentives to escalate.¶ 13

11.5 The seven functions as capacities for peace

Télécharger ce tableau (CSV)

11.6 The three instruments and pacifying legitimacy

The three cross-cutting instruments do not produce the same form of legitimacy. The Carthage Charter sets the rights and the limits, in particular dignity, non-discrimination, participation, proportionality and the non-derogable floors. The chain's operational framework agreement organises the interfaces, the obligations of cooperation, the data, the audit, accession and withdrawal. The Call of Carthage opens public accession, qualified petition, deliberation and agenda-setting. Its full text is reproduced in Annex P.¶ 14

For peace, their complementarity is decisive. A purely technical procedure may lack legitimacy; a charter of principles without capacity for execution may disappoint; participation without rules of evidence may be captured by the most powerful mobilisation. Preventive peace requires simultaneously a normative limit, an operational capacity and a civic opening.¶ 15

• The Charter protects. It prohibits an emergency, a majority or a promise of financing from neutralising essential rights.
• The framework agreement makes cooperation predictable. It sets who answers, within what time limit, with what data and under what control.
• The Call makes the grievance audible. It makes it possible to reach the agenda before the only visible route is the street, closure or diplomatic confrontation.

11.7 A preventive-peace protocol based on the chain

Where a case file presents a plausible risk of escalation, the chain may activate a specific pathway. This pathway does not create a security competence. It organises the accelerated and adversarial treatment of a cross-border grievance, in liaison with the competent peace and development authorities.¶ 16

Télécharger ce tableau (CSV)

11.8 Articulation with existing peace infrastructures

The value of the architecture depends less on the creation of a new organisation chart than on its capacity to supply missing functions to the institutions that exist. The interfaces must be designed before the institutions. Within the United Nations system, a working hypothesis may apportion the roles as follows, subject to the mandates and to the consent of the States concerned.¶ 17

Télécharger ce tableau (CSV)

11.9 Three families of peace-oriented demonstrators

The data-centre demonstrator verifies the continuity of the interfaces, but it does not suffice to demonstrate a contribution to peace. Three families of cases could test the added value of the chain without waiting for an open crisis.¶ 18

11.9.1 Monetary shock, fiscal space and social tensions

A monetary tightening in a large economy may be transmitted through the exchange rate, debt service, capital outflows, credit, imported prices and the responses of recipient central banks. The pilot would presume neither fault nor claim. It would test whether decomposition into mechanisms, partial attribution, the taking into account of domestic vulnerabilities and targeted preventive financing make it possible to avoid the shock being narrated as a total external aggression or, conversely, as an exclusively domestic failure.¶ 19

11.9.2 Cross-border resource: water, energy or pollution

The case file would concern a measurable change in availability, quality, cost or risk affecting several jurisdictions. The chain would distinguish hydrological or environmental measurement, distributive qualification, the applicable obligations, the compensation options and the joint investments. Mediation should precede any pricing, and vital uses would remain protected by the principle that some things are not for sale.¶ 20

11.9.3 Digital infrastructure, information and trust

A platform, an artificial intelligence model or a data infrastructure may alter access to information, to credit, to work or to public services beyond its place of decision. The pilot would test the provenance of the data, distributive biases, redress mechanisms, the certification of systems and publication risks. The objective would not be to govern political debate by an algorithm, but to make contestable the pipeline that influences civic space.¶ 21

11.10 Safeguards against the conflict-generating effects of the architecture

Télécharger ce tableau (CSV)

11.11 Evaluation framework: how to know whether the chain really contributes to peace

An architecture of peace must not be evaluated by the number of meetings, notes or financings. It must be judged on its capacity to alter the trajectory of a dispute, with an explicit counterfactual. The following indicators are proposed for a pilot; they must be adapted to the context and are never added together into a single score of peace.¶ 22

Télécharger ce tableau (CSV)

11.12 Proposed pilot: Peace Accountability Track

A pilot may be conducted without creating a new institution and without producing any binding effect. It would use teams hosted by existing bodies and would compare the complete chain with a simplified architecture. Its purpose would be to test the value of procedural continuity, not to legitimise the 7+3 in advance.¶ 23

• Duration. Eighteen months: three months of preparation, nine months of processing, three months of follow-up and three months of independent replication.
• Cases. Two or three consented disputes, belonging to different families and not involving a situation where publication could aggravate an active conflict.
• Consequences. Rating, pricing and adjustment in notional mode only during the first phase; only voluntary corrections and consented financing may be implemented.
• Comparator. One team processes the same case file by means of a lighter protocol in order to estimate the cost and the marginal value of each function.
• Pre-registration. Assumptions, admissibility criteria, indicators, stopping thresholds, confidentiality rules and evaluation method published before processing.
• Governance. Independent committee, representation of the affected parties, declaration of interests, non-circular budget and the right to publish a dissenting opinion.

The pilot must be able to conclude that the chain is too heavy, too slow or too conflictual. It is successful if it produces useful information on the validity of the architecture, even where it leads to simplifying or rejecting a component.¶ 24

11.13 Antithesis, limits and criteria for rejection

The strongest antithesis holds that conflicts do not come from a lack of measurement but from incompatible interests, power relations, ideologies, security dilemmas and political calculations. An accountability chain could then produce more data without altering the fundamental incentives. It could even offer new instruments of pressure, delay an urgent decision or turn a negotiable disagreement into a dispute of principle.¶ 25

This objection is partly founded. The architecture does not claim to dissolve power. It seeks to alter the range of available options and the political cost of opacity, of arbitrariness and of the refusal to answer. It must be rejected as an infrastructure of peace if the pilots show any one of the following results:¶ 26

• the publication of ratings systematically increases polarisation or retaliation;
• the parties do not place greater trust in the procedure than in the existing mechanisms;
• the costs of coordination and delay exceed the gains from prevention;
• powerful actors durably escape obligations while the weak bear the consequences;
• mediation is marginalised in favour of pricing and litigation;
• digital systems cannot be sufficiently explained, audited or contested;
• affected populations do not have effective participation and independent counter-assessment;
• a simpler architecture obtains equivalent or better results at lower human and institutional cost.

11.14 Decision conclusion

The Global Accountability Chain is not a peace treaty in the conventional sense. It may nevertheless constitute an infrastructure of preventive peace if it makes it possible to address, before escalation, the grievances linked to cross-border and intergenerational effects. Its contribution is neither to suppress political conflict nor to substitute technique for diplomacy. It is to produce a common space in which the facts, the uncertainty, equity, financing and redress can be discussed without power being the sole arbiter.¶ 27

The proposal must therefore be presented as a testable hypothesis, not as a validated architecture. Its qualification as an infrastructure of peace depends on the results of the pilot.¶ 28

11.15 Resilience and crisis response: rapid modes within fixed mandates

The body of this document deals with chronic harm — the externalities that accumulate slowly. Acute shocks constitute the other half of the problem, and they are no longer rare. On cautious readings of the usual risk registers, the probability that at least one severe global shock materialises within the decade is judged high; that of three or more shocks combining within a generation is higher still; and there is a non-negligible probability that one shock triggers others (World Economic Forum, 2025). An architecture designed solely for slow harm would fall silent precisely at the moment when it would be most needed. This section sets out how the seven institutions and the three instruments behave in the face of an acute shock.¶ 29

11.16 The cost of the crisis to come

If the probability of severe shocks is now high, their price is in no way hypothetical. The episodes of the previous generation point the way to the next: the financial crisis of 2008 and the COVID-19 pandemic each imposed output losses running into trillions of dollars, and natural catastrophes alone, predominantly climate-driven, inflicted $318 billion of economic losses in 2024 alone, of which 57% uninsured — on a rising trend observed over four decades (Swiss Re Institute, sigma 1/2025). These are not independent draws. The compounding described by the risk registers — one shock raising the probability of the next (World Economic Forum, 2025) — means that the expected annual bill from acute crises is large, recurrent and growing, and not a rare event to be absorbed and then forgotten.¶ 30

As with the chronic harms of section 13, this burden is not shared equally. The economies that lack the cushions needed to self-insure — no reserve currency, no common fiscal safety net, restricted access to capital precisely when a downturn arrives — absorb the largest relative losses, exactly where a shock most easily turns into a debt crisis or a lost decade. This is the acute counterpart of the cost of inaction, and it must be read as a distinct magnitude: the chronic flow of unpriced externalities and the expected losses from extreme shocks are different measures, and they are never added together here. What they have in common is a single conclusion — that the least costly moment to build a response is before the shock, not during it.¶ 31

This caution has a precise theoretical basis. Under fat-tailed uncertainty — where catastrophic scenarios, even if unlikely, cannot be excluded — expected damage may become so large that ordinary cost-benefit analysis ceases to guide the decision, and the only rational response is to insure against the worst (Weitzman, 2009). What the architecture draws from this is not catastrophism — it predicts nothing — but the justification for pre-commitment: one guards against a tail one cannot date, precisely because one cannot date it.¶ 32

11.16.1 Cost discipline.

These orders of magnitude are drawn from third-party risk registers, and not from a forecast: they are reported as ranges attached to indicated sources. They justify a pre-commitment — the construction of the rules-based rapid modes set out below, while the political and budgetary openings still exist — and not a discretionary emergency power once the crisis is under way. A high expected cost is an argument in favour of rule-bound preparation, never a mandate to suspend it.¶ 33

11.17 Rapid modes within fixed mandates

The guiding principle is that of resilience by pre-commitment, not by discretion. Each institution has a rapid mode that is triggered by objective and published indicators rather than by a judgement of expediency; pre-authorised in the constitutive instruments, so that no institution seizes a new power in the moment; strictly bounded to that institution's single mandate; time-limited, with automatic expiry unless renewal is entered in the case file; and subject to review by the Humanity Court. This is the logic of a circuit breaker, not that of an emergency executive. It is a deliberate response to the oldest failure mode of crisis governance — the fact that emergency powers ratchet forward and never ratchet back — and it is consistent with the document's negative-mandate backbone: speed within a mandate, never an emergency power beyond it.¶ 34

The rapid-response model is in no way hypothetical. Pre-arranged, indicator-triggered financing already outperforms ex post appeals (Clarke and Dercon, 2016), and the Fund's Resilience and Sustainability Trust shows that standing shock facilities can be built within existing mandates (International Monetary Fund, 2022). Table 46 maps each institution's peacetime function to its rapid mode, the indicator that activates it, the discipline that closes it, and the closest existing precedent.¶ 35

The three instruments carry the safeguards. The Global Accountability Framework contains the crisis protocol itself — the catalogue of rapid modes, their triggers and their sunset clauses — so that emergency behaviour is authorised in advance and in writing, never improvised. The Carthage Charter sets the non-derogable floors that no rapid mode may cross, on the model of non-derogable rights in a state of emergency compliant with law (United Nations, 1966, article 4). The Call keeps crisis measures accountable to the populations concerned, so that speed never becomes a blank cheque for technocratic government.¶ 36

So framed, resilience is not a new mandate grafted onto the architecture; it is a property that emerges from completing the chain and from giving each institution a rapid mode bounded by rules. It also redistributes something that is today held unequally. The capacity to absorb shocks — rapid decision, common fiscal cushions, integrated response — is concentrated in the largest blocs and largely absent for smaller economies and regions. By providing rapid financing, shock measurement and crisis legality as global functions rather than as privileges attached to size, the architecture offers a minimum floor of resilience to those that are not, and may never be, a bloc.¶ 37

Two disciplines bound this claim. First, what is at stake is enabling resilience and crisis accountability, not a global emergency government: the architecture replaces neither central bank swap lines, nor the Fund's liquidity role, nor humanitarian response, and it must not claim to do so. Second, a rapid mode is only as safe as its sunset clause; the design therefore entrusts the enforcement of expiry to the Court and treats any rapid mode that cannot be switched off as a defect, not a feature.¶ 38

Télécharger ce tableau (CSV)

@incollection{yaiche_hc_chapitre_11,
  author    = {Yaïche, Nizar},
  title     = {11. Preventive peace and resilience in the face of crises},
  booktitle = {The Humanity Compact},
  year      = {2026},
  version   = {2.3},
  doi       = {10.5281/zenodo.22746533},
  publisher = {Zenodo},
  url       = {https://humanitycompact.org/ouvrage/chapitre-11/}
}

Ouvrage de Nizar Yaïche, ancien ministre des Finances de la République tunisienne. Qui est l’auteur ?