# WP35 — Literature review and novelty matrix (Phase 1)

Verification status: every reference cited here was checked against Crossref by DOI or bibliographic query on 2026-11-01 (src/wp35/validation/verify_bib.py → research/prior_art.csv; manual corrections and books in research/bib_overrides.csv). The abstracts and key results of the 18 closest works were read from primary or repository sources by a dedicated verification pass (summary archived in docs/novelty_search_report.md). Items marked [secondary] were known only through reviews or citing works and are cited in the paper only for their general thesis.

# 1. What is already known, by strand

# 1.1 Scalar stress and information-processing hierarchies (archaeology/anthropology)

The oldest form of the idea is Johnson's scalar stress [secondary, chapter not read]: as the number of units that must be coordinated grows, the communication and decision load grows faster than linearly, and groups respond by fission or by adding hierarchical levels, with a modal span of about six units per level (Johnson 1982; Alberti 2014 tests a related fission threshold on Hutterite colonies). Wright and Johnson (1975) defined the state as a specialized administrative hierarchy of at least three decision-making levels and rejected population growth and exchange as single sufficient causes of its emergence in Susiana. Carneiro (1967) documented the log-log regularity between population and the number of organizational traits; Feinman (2011) reviews the size–complexity relation; Flannery (1972) framed institutional evolution as changes in information-processing architecture. None of this strand builds a demand/capacity ratio or tests it against instability.

# 1.2 Declining returns and collapse

Tainter (1988) [secondary] argues that complexity is a problem-solving investment with declining marginal returns, so that at some point collapse (simplification) becomes rational and the society loses its capacity to absorb shocks. Middleton (2017), Butzer (2012) and Cline (2014) review the collapse literature and warn against mono-causal accounts. Turchin's structural-demographic theory (Turchin 2003; Turchin and Nefedov 2009) models secular cycles driven by population pressure, elite overproduction and state fiscal distress; Goldstone (1991) is the historical origin of that theory. Sng (2014) is the closest economic precedent: in late-imperial China, the cost of transmitting information over distance creates a principal–agent problem whose severity rises with territorial size, so that fiscal capacity decays in the prosperous eighteenth century; this is a demand (territory) versus capacity (monitoring) argument tested within a single polity. Hoyer et al. (2024, preprint) use the Seshat CrisisDB to show that crisis consequences are heterogeneous, largely uncorrelated with each other, and not well predicted by territory, religion or administrative complexity taken one at a time ("no typical societal crisis").

# 1.3 Seshat-based quantitative history

Turchin et al. (2018) found that nine complexity characteristics (population, territory, capital population, hierarchical levels, government, infrastructure, writing, texts, money) load on a single principal component explaining about three quarters of variance across 414 polity-centuries. Shin et al. (2020) decomposed the remaining variance and identified a second axis opposing scale characteristics to information-processing characteristics, defined a "scale threshold" and an "information threshold" along PC1, and wrote — explicitly as a conjecture left for future research — that "mismatch among aspects of social development should be considered as a generic cause of instability". Miranda and Freeman (2020) clustered Seshat polities into two super-types with an unstable transitional cluster in which polities dwell only a few centuries. Turchin et al. (2022) compared more than 100,000 predictor combinations for the growth of complexity and found that agricultural productivity interacting with military technology (iron, cavalry) was best supported; they did not model instability. Whitehouse et al. (2019) was retracted after Beheim et al. (2021) showed that recoding missing moralizing-god data as "absent" drove the result; the Seshat team's "retake" (Whitehouse et al. 2023) reports qualitatively similar conclusions with corrected coding. The methodological lesson (never treat unknown as absent, never forward-fill, report results on extant data) is adopted in our pre-registration.

# 1.4 State capacity and its informational foundations (political economy)

Mann (1984) distinguished despotic from infrastructural power; Soifer (2008) and Hanson and Sigman (2021) discuss its measurement, the latter with a Bayesian latent model over 21 indicators (1960–2015). Scott (1998) describes legibility — standardized names, cadastres, censuses, weights and measures — as the state's way of making society readable; Lee and Zhang (2017) operationalize legibility through age-heaping in censuses and show it predicts tax contributions; Brambor et al. (2020) build an "information capacity" index (census, statistical agency, yearbooks, civil and population registers) for 85 polities since 1789; D'Arcy and Nistotskaya (2017) use cadastral records. Besley and Persson (2009, 2011), Dincecco (2017), Johnson and Koyama (2017), Mayshar et al. (2022), Borcan et al. (2018) study the origins and consequences of fiscal and legal capacity; Stasavage (2010) shows that geographic scale limited representative assemblies in Europe — a direct instance of distance-driven coordination cost. Gennaioli and Voth (2015) link capacity to military competition. None of these operationalize a ratio of coordination demand to capacity or test it against instability across pre-modern polities.

# 1.5 Transaction costs, codes and communication in organizations (economics)

Coase (1937), Williamson (1985) and North (1990, 1991) treat institutions as devices that economize on transaction costs; Hayek (1945) on knowledge dispersion and Simon (1962) on near-decomposable hierarchies are the classic information-processing views. Arrow (1974) introduced organizational codes; Crémer, Garicano and Prat (2007) model the firm's language as a code of bounded size in which precise words go to frequent events — literally lossy compression that trades scope against precision. Radner (1993), Bolton and Dewatripont (1994), Van Zandt (1999) and Garicano (2000) model hierarchies as information-processing networks with explicit communication costs; Dessein (2002) and Dewatripont and Tirole (2005) study authority and modes of communication. Mechanism-design results bound the message-space size needed to implement allocations (Mount and Reiter 1974; Segal 2007; Nisan and Segal 2006), and communication complexity (Yao 1979; Kushilevitz and Nisan 1997) gives the computer-science counterpart. These results are the formal basis for the propositions in Phase 3, but they concern firms, markets or abstract protocols, not polities.

# 1.6 Coordination, common knowledge and conventions

Lewis (1969) and Schelling (1960) define conventions and focal points; Young (1993), Kandori, Mailath and Rob (1993) and Ellison (1993) show how conventions are selected and how local interaction speeds convergence; Morris (2000) on contagion. Chwe (2000, 2001) argues that coordination requires common knowledge and that rituals and public communication structures exist to manufacture it cheaply; Rubinstein (1989) shows that almost-common knowledge is not enough. Morris and Shin (2002) and Angeletos and Pavan (2007) quantify the social value of public signals; Dewan and Myatt (2008) model leaders as coordination devices. Ostrom (1990, 2010) and Olson (1965) are the collective-action anchors; Boyd and Richerson (1985, 1988), Henrich (2004) and Richerson et al. (2016) supply the cultural-evolution mechanisms for norm maintenance in large groups; Greif (1993, 2006) and Milgrom, North and Weingast (1990) are the historical institutional cases of reputation and record-based enforcement.

# 1.7 Scaling and information theory

Bettencourt (2007, 2013) and Ortman et al. (2014), Lobo et al. (2020) show superlinear scaling of socioeconomic outputs with settlement population, ancient and modern. Scaling theory implies that interaction demand grows faster than population, but it assumes infrastructure and institutions co-scale; it predicts benefits, not instability. DeDeo (2017) is the closest explicit statement that transitions in political order are driven by information-processing innovations and that coarse-graining (lossy compression) is the central mechanism; Marzen and DeDeo (2017) give a rate-distortion model of perception with a high-fidelity regime (cost linear in environmental complexity) and a low-fidelity regime; Flack (2012, 2017), Daniels, Krakauer and Flack (2017) and Krakauer et al. (2020) develop coarse-graining and collective computation in animal societies. Shannon (1948) and Cover and Thomas (2006) are cited only for the definitions we use in simulations where messages and distributions are actually defined.

# 1.8 Instability forecasting and methods

Goldstone et al. (2010) forecast instability onsets 1955–2003 with regime type dominating; Fearon and Laitin (2003) and Hegre et al. (2001) are the civil-war baselines; Beck, Katz and Tucker (1998) and Carter and Signorino (2010) on binary time-series–cross-section data; Cox (1972) and Box-Steffensmeier and Jones (2004) for event history; Cameron and Miller (2015) and Conley (1999) for clustered and spatial inference; Oster (2019) and Cinelli and Hazlett (2020) for sensitivity to unobservables; Mace and Pagel (1994) and Dow (2007) for Galton's problem; Nosek et al. (2018) for pre-registration.

# 2. Novelty matrix

# 3. Precise contributions claimed (and not claimed)

1. Empirical: the first pre-registered test of the scale/information mismatch conjecture of Shin et al. (2020) against an independently coded instability outcome (CrisisDB), with a transparent and a latent measurement model, uncertainty propagation, and out-of-sample validation; replicated in form on a modern country-year panel.
2. Theoretical: formal propositions stating when a shared code/record system lowers expected communication cost per coordinated task as a function of scale, and when fixed institutional overhead makes it worse; stated as theorems with proofs or as conjectures, never as empirical facts.
3. Simulated: a factorial agent-based comparison of five institutional regimes with directly measured messages, bytes, success rates and overhead, searching for regimes where institutions reduce efficiency or add fragility.
4. Not claimed: causal identification in the historical panel; literal Shannon compression of institutions; a universal theory if H2 fails (see pre-registration §7). The name "Social Compression Hypothesis" has no prior academic use that we could find (search 2026-11-01), but the idea has clear ancestors (§2), which the paper credits.

# 4. Points reviewers will raise (anticipated, recorded before analysis)

• Collinearity: scale and capacity load on the same PC1 in Seshat; $P$ is by construction a residual. We report the correlation of $z(\log D)$ and $z(K)$ and the share of variance $P$ retains, and we test $P$ against $D$-only and $K$-only baselines (pre-registration §5.2).
• Crisis-selected outcome (CrisisDB covers crisis cases only) → severity-conditional-on-crisis interpretation, plus polity-hazard and modern-onset designs.
• Coding circularity: Seshat coders may infer administrative features from narrative sources that also describe crises; we use only the Polaris codes and flag inferred codes (IP/IA) in a sensitivity run.
• Missing data: Beheim et al. (2021) template; missing is never absent.
• Temporal precedence: most Polaris facts are undated → "pre-crisis" is weak; stated as a limitation, with a dated-only sensitivity.