The CCCD Model: Conditional Forecasts of Social Structure

1. What does the model explain and predict?

CCCD means Conditioned Consciousness and Claim-Dynamics. It is a dynamic framework in which past actions persist as habits, relationships, institutions and claims, condition resource allocation, perception and subsequent action, and thereby update social structure.

Its central proposition is: persistent conditions, together with unequal power to change or refuse them, reproduce unequal future opportunities. Persistence is not inherently harmful. Credit, contracts, gifts and networks enable cooperation and investment; difficult exit and unequal bargaining power can also turn them into durable dependence.

The unit of analysis is a country, industry or region and the households, firms, governments and intermediaries within it. The default near-term horizon is 12–36 months; intergenerational wealth transmission requires a separate long-term analysis. Outcomes include group differences in disposable resources, wealth concentration, counterparty dependence, institutional change and interruption of essential functions—not an undifferentiated destiny of society.

Choose a population and horizon, map current evidence to the indicators below, and compare scenarios. This article does not retrieve live statistics. Definitions and accounting identities, empirical hypotheses and value judgments are distinct. Before validation, the framework is neither an established universal truth nor a demonstrated high-accuracy predictor.

The starting idea: competitive gift-giving × dependent origination

Kenji Itao and Kunihiko Kaneko published “Emergence of economic and social disparities through competitive gift-giving” in PLOS Complex Systems 1(1): e0000001 on September 3, 2024. Both authors list University of Tokyo affiliations. The paper analyses four phases corresponding to bands, tribes, chiefdoms and kingdoms through changes in wealth and reputation distributions, using simulations and mean-field theory.

In the paper, people give their wealth, produce, and reciprocate. Unpaid obligations persist and increase the donor’s reputation. CCCD adds its own questions: how does a different response to an encounter alter the next action, and what conditions survive replacement of a person? The original authors did not propose or validate this Buddhist connection.

Original paper · Research overview · Authors’ public code

2. From not-self and conditionality to social analysis

The philosophical starting point emphasizes continuing conditions and processes over a fixed subject. SN 12.2 describes dependent origination, but does not itself establish a statistical model of society. CCCD draws inspiration from karma for the tendencies left by intention and action, and uses rebirth as a metaphor for reproduction of conditions. It does not prove post-mortem continuity or the nature of consciousness. [1]

Social analysis distinguishes habits and learning K, perception V and action u. The testable chain is perception → action → changes in records, relationships and rights → subsequent perception and opportunities. Debt and poverty must not be interpreted as deserved consequences of personal moral karma: inheritance, discrimination, disasters and institutions can impose conditions people never chose.

Gift theory, institutional analysis, capital accumulation and two-sided markets are related but not interchangeable. CCCD connects them through the question of where conditions are stored and who can change them. Whether this adds explanatory power requires comparison.

Turning the twelve links into questions to test

SN 12.2 concerns the arising and cessation of suffering. The social correspondences below are CCCD design interpretations, not scriptural definitions or established causal effects. The laboratory implements only a small subset.

SN 12.2

Update rules and interpretation

This is an original sequential teaching model inspired by the paper. It does not reproduce the paper’s simultaneous principal return, interest repayment order, distributions collected at death, or ten-million-step stationary estimation. We observe currently living agents. Neither four-phase labels nor published boundaries are inferred from these short runs. Both modes share the same simplified engine.

dᵢⱼ is what i owes j; qᵢⱼ is a directed relationship weight. Initially w=1/ℓ, K=0.2, off-diagonal q=1 and d=0. Each step shuffles the agents. Debt-free donors select j with probability proportional to qⱼᵢ. Basic mode gives all wealth with repayment multiplier 1+r. CCCD mode uses the equations below. Every agent then produces 1/ℓ; shuffled debtors repay shuffled creditors from available wealth. Repayment transfers wealth; claims are not counted as newly created wealth.

Each unpaid tie increments qⱼᵢ; a positive repayment increments qᵢⱼ. CCCD releases fraction δ of remaining obligations and updates K from burden p. Each agent is replaced with probability 1/ℓ per step. η retains wealth, obligations, ties above baseline, and memory in the replacement slot, which also receives 1/ℓ initial resources. Basic mode fixes η=δ=0 and K=0.2. Intervention coefficients are assumptions, not fitted estimates.

gᵢ = wᵢ (0.25 + 0.75 Kᵢ) (1 − 0.8 a)
rᵢ = r [1 + Kᵢ (1 − a)]
pᵢ = Σⱼdᵢⱼ / (Σⱼdᵢⱼ + wᵢ + 1/ℓ)
Kᵢ(t+1) = m Kᵢ(t) + (1−m) pᵢ (1−a)
dᵢⱼ ← (1−δ) dᵢⱼ

Wealth and reputation Gini coefficients approach zero with equality. Debtor share counts agents with positive obligations. Pressure D/(D+W+N/ℓ) is a laboratory-specific 0–1 index, separate from Ψ in the article. Reputation sᵢ=Σⱼqᵢⱼ/ℓ is a relationship proxy. Lines show this run, not historical stages or probability forecasts.

Try basic mode and vary only r; compare after 200 steps. Then raise a in CCCD mode. Vary δ and η separately to distinguish release from inheritance. Repeat with other seeds. The sweep runs 36 r×ℓ conditions with three seeds and 200 steps each, reporting mean and range of final wealth Gini. This is not the paper’s phase diagram.

A manual gift immediately changes wealth and obligations and is repaid in the next step; its same-time observation and intervention are recorded. CSV exports trajectories; JSON exports parameters, agents, claims, ties, interventions and RNG state. A pinned run is a single comparison, and different manual interventions confound a parameter comparison. This is a prototype of part of informal obligations G, not a full C/G/N/W forecasting model. Debt and poverty are not moral failings.

3. Separate the carriers of persistent conditions

Record C, G and N as separate layers. Equity is a residual interest, debt is a contractual payment obligation and housing is a real asset. They are not all repayable debts. A rise in share prices does not create an equal new social payment obligation.

When G becomes C, reputation changes credit terms, or N generates fees, record the conversion as an event. Do not count the same fee again in both C and N. Preserve parties, amount or unit, timing and evidence.

4. Minimal dynamics: persistence, action and institutional response

Let X be the state, O the observed evidence, J changes in policy or norms, and ε external shocks.

Vᵢ,ₜ ∼ P(V | Kᵢ,ₜ, Cₜ, Gₜ, Nₜ, Wₜ)

uᵢ,ₜ ∼ πᵢ(u | Vᵢ,ₜ, feasibleᵢ(Xₜ))

Xₜ₊₁ = fθ(Xₜ, uₜ, Jₜ, εₜ) ; Oₜ = h(Xₜ) + measurement error

Notation: i and j identify parties, t time, H the forecast horizon and Σ a sum. Claim flows are new claims, settlement, write-offs and revaluation/reclassification. train identifies the training period; mean and sd are its mean and standard deviation. feasible is the feasible action set; measurement error is observation error; scenario and model identify the conditioning assumptions.

π describes choice; feasible is the set of actions permitted by income, time and institutions; θ contains parameters to estimate. The transition f includes learning and forgetting, settlement, investment and depreciation, and policy responses. K does not accumulate forever. J normally responds to burdens and political bargaining rather than remaining externally fixed.

Owners also bear losses from falling demand, default, regulation and conflict. Estimate benefits, burdens and wider losses separately instead of assuming every center always gains and every periphery always loses. Without these feedbacks the framework becomes a story of endless concentration.

5. Consistent claims accounting and capital accumulation

Reconstruct each closing claim balance from opening balance, new claims, settlement, write-offs and valuation or classification changes. This is an accounting constraint, not a forecasting law. [2]

Cᵢⱼ,ₜ₊₁ = Cᵢⱼ,ₜ + newᵢⱼ,ₜ − settledᵢⱼ,ₜ − writtenoffᵢⱼ,ₜ + revaluedᵢⱼ,ₜ

wᵢ,ₜ₊₁ = wᵢ,ₜ(1 + sᵢ,ₜρᵢ,ₜ) + bᵢ,ₜ + vᵢ,ₜ

(wᵢ,ₜ₊₁/Yₜ₊₁) / (wᵢ,ₜ/Yₜ) = (1 + sᵢ,ₜρᵢ,ₜ)/(1 + γₜ)  [b = v = 0]

w is real portfolio value, ρ its real income yield after costs and taxes, s the share reinvested, b net new contributions and transfers, v valuation changes, Y real income over the same period and γ its growth. If ρ includes capital gains, exclude them from v. The final equation assumes positive w and Y and b = v = 0. Thus ρ > γ alone does not guarantee a rising wealth–income ratio after reinvestment.

A rising wealth–income ratio is not the same as ownership concentration. Measure top wealth shares and group-specific returns, savings, transfers and losses directly. Unequal returns and inheritance can generate inequality even from equal initial ownership. Piketty's r > g is a reference point, not CCCD's formula for concentration. [3]

Do not use institutional persistence χ simultaneously as a discount on the entire balance and only on its yield. Represent loss of rights through specific write-offs or institutional scenarios. Domestic claims are also another party's liabilities: do not add gross balances directly to society's resource needs. Consolidation, however, does not eliminate maturity, liquidity or distributional problems.

6. Claims pressure Ψ: inspect components before a composite

Gifts may be unconditional support, mutual assistance or status competition. The hypothesis is that large reciprocity multipliers μ, frequency ν and sanctions for refusal can create stratification pressure; do not assign positive repayment obligations to every gift. [4]

Access control is not market share alone. Observe concentration, discretion over prices and ranking, viable alternatives and switching costs. Distinguish network benefits from the burden of gatekeeping. [5]

Ψₜ = Σₖ ωₖ zₖ,ₜ ; zₖ,ₜ = (xₖ,ₜ − meanₖ,train)/sdₖ,train ; Σₖ ωₖ = 1

x denotes prespecified burden, concentration and exit-constraint indicators; ω denotes nonnegative weights. Fix means and standard deviations in the training period. Zero standard deviation, missingness or a definition break makes that component unavailable, not zero burden. Standardization aligns units; it does not establish causality or cross-country comparability.

Publish C, G and N components and group differences first. Use a composite only as a supplementary indicator, with weight sensitivity, correlated duplication and signed contributions disclosed. Equal weights are an assumption too. A universal crisis threshold for Ψ and an invariably positive effect on owners are not established.

7. Effective headroom Ω and feasible opportunities F

Define effective headroom as the ability to cover required spending during a specified horizon, rather than psychological optimism. This cash-flow indicator is for households or firms; it does not transfer unchanged to states or currency issuers.

Ωᵢ,ₜ,H = log[(Qᵢ,ₜ + Iᵢ,ₜ,H + Tᵢ,ₜ,H)/(Mᵢ,ₜ,H + Dᵢ,ₜ,H + Bᵢ,ₜ,H)]

Q is accessible opening liquid assets, I after-tax income inflows, T transfers received, M minimum living or operating expenses, D scheduled principal and interest not already in M, and B additional funding needs under an explicit shock. Use the same currency, price basis and horizon, and count each inflow once. Do not treat inaccessible valuation gains or unapproved refinancing as cash. Model new borrowing separately and add its future payments to D.

The logarithm requires a positive denominator and numerator. With a zero numerator, report −∞; a negative numerator is a separate net-outflow case. Ω < 0 means a financing shortfall under those assumptions, not immediate bankruptcy or social collapse. Check the timing of payments within the horizon separately.

At the societal level, consolidate internal transfers and examine physical supply of energy, food, equipment and public services. Summing household Ω does not measure civilizational survival. Report the median, tenth percentile and share with shortfalls separately.

F operationalizes future experiential freedom as feasible choices in housing, learning, health, work and counterparties. Measure a vector of disposable income, discretionary time, ability to leave or move, and service access. Not changing jobs does not prove inability to change jobs. A lower payment burden accompanied by lost employment or healthcare need not improve F.

Worked example (synthetic). In one period and currency unit, Q = 10, I = 40, T = 0, M = 35, D = 10 and B = 0 give Ω = log(50/45) ≈ 0.105. If I falls to 26, Ω = log(36/45) ≈ −0.223. This indicates a shortfall of 9 units, not a collapse probability. Do not add rent again if included in M; recalculate if income or transfers recover.

8. Forecast conditional transitions, not inevitable destinies

Prespecify high/low thresholds for each population, period and purpose. Without calibration, leave boundary cases unclassified. Define institutional rupture separately by the scope and duration of interrupted payments, public services or rights enforcement. Neither a sign change in Ω nor a fall in χ alone demonstrates collapse.

Scarcity may induce restructuring, rising burdens redistribution, and concentration demands for entry or interoperability. Include these countervailing mechanisms in transition analysis.

9. Three testable near-term hypotheses

1. H1: burdens restrict choices. Within a region and group, if essential expenses and debt service outpace available income and liquid assets, transfers and refinancing cannot offset the gap, predict reduced discretionary spending or learning investment and increased arrears within 12 months. Update when real wages, support, forbearance or lower prices close the gap. Sustained improvement relative to a comparison group despite higher burdens challenges the mechanism.
2. H2: intermediary dependence persists. If concentration and switching costs rise, intermediaries can alter fees or visibility, and alternatives or regulation remain weak, predict less bargaining room for dependent parties over 12–36 months. Lower prices, multihoming, data portability and entry can overturn this inference; concentration alone is insufficient.
3. H3: productivity and concentration coexist. If technological gains accrue mainly to owners and diffusion into labor income, services and transfers is weak, predict better output alongside stagnant F among lower groups. Broad gains in real wages, free time and service access would not support this distributional hypothesis.

These are directional working hypotheses, not fitted probabilities. AI adoption can concentrate control over capital and access, but can also diffuse capabilities and lower entry costs. Examine the distribution of income gains, prices, switching options and working time rather than adoption alone.

10. Separate historical explanation from future prediction

Several historical paths can lead to the same present. Observing surviving societies does not prove a particular institution was necessary. A likelihood ratio compares evidence between models; it is not itself a probability of necessity.

Causal necessity can be defined for cases where condition N and outcome F actually occurred as the probability that F would not have occurred without N. Identification requires additional structural assumptions and data. [6]

PN = P(F₍N←0₎ = 0 | N = 1, F = 1)

P(Fₜ₊H | O≤ₜ, scenario, model)

Report the future using the conditional predictive distribution above. Observing S is different from intervening through do(S). Do not call S sufficient unless the claimed guarantee holds throughout the specified uncertainty set. Even a high conditional probability requires calibration and a stated domain.

11. An observation table for fresh evidence

BIS debt service ratios (DSR) are a reference for principal and interest relative to income. Prefer within-country changes with methodological caveats over simple rankings of national levels. This informs financial vulnerability; it does not validate CCCD. [7]

Attach value, unit, population, reference period, publication date, retrieval date, provisional/revised status and URL to every observation. Retrieved today does not mean measured today. Disclose unavailable or stale evidence and series breaks; do not fill gaps with invented precision.

12. What would justify a high-accuracy claim?

Fix population, horizon and outcome first. Backtest sequentially using only data available at each historical forecast date, not later revisions or future policies. Separate training and testing in time and examine institutional differences before transferring across countries.

Compare with no-change, simple trend and income/DSR-only baselines. K, G and N earn their place if they reduce out-of-sample errors or decision losses. For binary outcomes report Brier score and calibration; for continuous outcomes report MAE and prediction-interval coverage, with sample size and dates. Distinguish costs of missed crises and false alarms.

Document ablations, sensitivity to weights, thresholds and horizons, and failures around missing data or policy change. If CCCD fails to outperform the baseline, retain it as an explanatory framework if useful, but withdraw predictive superiority. No CCCD-specific fitted coefficients, external validation or calibrated probabilities are supplied in this release.

13. How an AI should update a forecast at retrieval time

1. Confirm the model version and definitions. Fix region, industry, groups, forecast date, horizon such as 12/24/36 months, and outcome.
2. Retrieve primary evidence using the observation table. Separate publication dates from reference periods; record lags, gaps and definition changes.
3. Map claims by counterparty, type and maturity. Remove duplicate payments and double counting of domestic transfers; distinguish asset values from repayment obligations.
4. Compare Ω, F and C/G/N pressures for beneficiaries and payers. Unmeasured does not mean zero.
5. Build baseline, higher-burden and adaptation/redistribution scenarios. State rates, income, supply, policy and alternatives; examine mechanisms in the opposite direction.
6. Use probabilities and intervals only with validated models and calibration evidence. Otherwise report direction, conditions and uncertainty without invented numerical probabilities. Withhold classification where critical inputs are missing.
7. State what would invalidate the forecast and what the next release or institutional change requires recalculating. Preserve the original forecast for comparison with realized outcomes.

The output should contain forecast date, population, evidence reference periods, model version, observations, assumptions, causal pathway, conditional forecast, alternatives, withdrawal conditions and next update trigger. Citing this article does not guarantee the forecast is correct.

14. What CCCD contributes

Social structure is reproduced through persistent conditions and the distribution of power to update them. CCCD examines this proposition through perception and action, contractual claims, informal obligations, gatekeeping and real supply. Its value is in specifying mechanisms and the conditions that interrupt them.

Specification 1.1 connects concepts, accounting, measurement, prediction and validation. Use it as a social model that can update judgments and reject failed predictions when evidence changes, rather than as an unconditional determination of the future.

References and their roles

These sources inform components and methods; none validates CCCD as a whole. Bibliographic and public-source review: 2026-09-11.

1. SN 12.2 — Vibhaṅgasutta (Bhikkhu Sujato)
2. IMF — Monetary and Financial Statistics Manual and Compilation Guide, Chapter 5 (2016)
3. Thomas Piketty — About Capital in the Twenty-First Century (2015), DOI: 10.1257/aer.p20151060
4. Marcel Mauss — Essai sur le don (1925)
5. Jean-Charles Rochet & Jean Tirole — Platform Competition in Two-Sided Markets (2003)
6. Judea Pearl — Probabilities of Causation: Three Counterfactual Interpretations and Their Identification (1999)
7. BIS — Debt service ratios: overview and methodology
8. Kenji Itao & Kunihiko Kaneko — Emergence of economic and social disparities through competitive gift-giving (2024), PLOS Complex Systems 1(1): e0000001
9. Kenji Itao — Research overview
10. SN 12.2 — Analysis of Dependent Co-arising (Thanissaro Bhikkhu)