The Degree of Shared Context, the Explicit Society, and the Theory of Diplomatic Negotiation

The Theory of Context Externalization and Its Research Program

— Transaction Costs, Game Theory, and Design for a Society of Strangers

Abstract

What is needed for people of different backgrounds to live together in a single society? This paper reformulates that classical question as a problem not of “trust” but of “information.” The core concept is context externalization — the process of moving information stored in people's heads and relationship networks into external systems: language, documents, contracts, rules, and institutions. The paper develops this concept at four levels. First, theory: Hall's high-context/low-context distinction is reinterpreted as a question of where information is stored rather than of cultural essence, and connected to the institutional economics of Coase, North, and Greif. Second, micro-foundations: using the Myerson–Satterthwaite theorem, we show game-theoretically the mechanism by which externalization shrinks the region of trade failure, and show that in negotiation, transparency has two diametrically opposed effects — an “information-rent channel” (a liability) and a “credibility channel” (an asset). Third, evidence: we array the extreme collapse of context in postwar Busan (about 470,000 at the end of 1949 → 844,134 in 1951 → 1,049,363 in 1955), the United States as a society sustained by immigration for 170 years, and the sub-national natural experiment across the 50 U.S. states (Vandello & Cohen, Nisbett & Cohen, the Gelfand line of work, Rentfrow). Fourth, the frontier: we show that the relationship between a society's internal transparency and its external diplomatic bargaining power is an open problem, and propose four competing hypotheses and a treaty-type-split verification design. Every proposition is marked with its status as Fact or Derived, all citations were checked and verified against original sources (Appendix), and the falsification conditions for each hypothesis are stated. This paper aims not at the test of a single hypothesis but at the declaration of a research program.

Keywords: context externalization, transaction costs, high-/low-context, information asymmetry, credibility, audience costs, two-level games, diversity, institutional trust, Busan, U.S. regional cultures

Abstract (author's original English)

How do strangers cooperate? This paper reframes the classical question as a problem of information rather than trust, and proposes context externalization: the migration of information from heads and relationships into language, documents, contracts, rules, and institutions. We develop the concept at four levels. (1) Theory: Hall's high-/low-context distinction is reinterpreted as the storage location of information, and connected to Coase–North–Greif institutional economics. (2) Micro-foundations: via Myerson–Satterthwaite, externalization shrinks the region of trade failure; in bargaining, transparency operates through two opposing channels — the information-rent channel (a liability) and the credibility channel (an asset, via Schelling commitment, two-level games, and audience costs). (3) Evidence: postwar Busan's abrupt context collapse (population ~470,000 in late 1949 → 844,134 in 1951 → 1,049,363 in 1955), the 170-year immigrant society of the United States, and a sub-national natural experiment across 50 U.S. states. (4) Frontier: whether domestic transparency translates into external diplomatic bargaining power remains an open problem; we formalize four competing hypotheses and a treaty-type-split verification design. Facts and inferences are explicitly marked, all sources verified against originals (Appendix), and falsification conditions stated. This paper declares a research program, not a single hypothesis.

1. Introduction: The Problem of Strangers

A society is a collection of strangers. In a small community this is not a problem. Members know one another's families, pasts, reputations, and ways of behaving. The premise “it's them, so it'll be fine” holds.

As cities grow and populations move, this premise breaks down. You cannot know where the person you met today came from, which rules they take for granted, or whether they put the same meaning into the same word. At this point a society must choose one of two strategies.

• The inference-strengthening strategy: demand from every member a higher capacity for situational interpretation — “nunchi,” the ability to read the room.

• The information-externalization strategy: reduce the amount of information that must be interpreted. Speak precisely, write it down, make rules, and apply the same procedure to everyone.

This paper's overall thesis is as follows. As the heterogeneity of a society's members grows, Strategy 1 runs into the limits of scalability, and a society that wants sustained cooperation faces a growing incentive to move toward Strategy 2 — the externalization of context. This move brings a gain (reduced transaction costs) and a cost (the loss of information rent in negotiation) at the same time, and the structure of that cost appears asymmetrically between domestic transactions and negotiations between states.

This paper's methodological principles are three.

• Separation of fact and inference: every key sentence is marked [Fact] or [Derived].

• Verified citations only: cited literature and statistics were checked against original sources or accredited bibliographic databases (Appendix).

• Explicit falsifiability: for each hypothesis, we state “what observation would cause this hypothesis to be discarded.”

2. Theoretical Background

2.1 Hall: Where Is Information Stored?

[Fact] Edward T. Hall introduced high-context and low-context communication in Beyond Culture (1976). A high-context message is one in which “most of the information is either in the physical context or internalized in the person, while very little is in the coded, explicit, transmitted part of the message” (Hall, 1976, p. 91). Japanese, Korean, and Arab cultures are placed on the high-context side; German, Swiss, Scandinavian, and American cultures on the low-context side.

[Derived] This paper's reinterpretation: the high/low-context distinction is a statement not about the character of a culture but about the storage location of information. High context = information stored in relationships, situations, and the inner self. Low context = information stored in explicit codes. This reinterpretation avoids cultural essentialism while formalizing Hall's insight in information-theoretic terms.

2.2 Accepting the Criticism: Kittler et al. (2011)

[Fact] Kittler, Rygl, and Mackinnon (2011) systematically reviewed the academic use of the high/low-context concept and pointed out its limitations: (1) the fixing of a continuum into a dichotomy, (2) the neglect of within-country diversity, and (3) the fact that Hall's original evidence rests on field observation rather than controlled experiment.

[Derived] This paper's response: (1) treat context dependence not as a fixed attribute but as condition-responsive behavior; (2) take not the nation but “a change in conditions” (an abrupt drop in shared context) as the unit of analysis; and (3) use within-country variation as an analytic asset, as in the U.S. state-level comparison (Chapter 8). Each of the three points of criticism is thereby either avoided or turned to advantage.

2.3 The High-Context Character of Korean Communication

[Fact] Kang (2018), through qualitative analysis of interviews, showed that the most frequent trait in Korean communication is “providing less information than necessary, and expecting the listener to understand unstated information from shared context” (Discourse and Cognition, vol. 25, no. 2, pp. 1–24). Also reported were the intentionality of ambiguous utterances, the regulating force of relationships, and the suppression of emotions and opinions.

2.4 Diversity and Social Trust: A Robust Correlation, an Empty Mechanism

[Fact] The representative evidence is as follows.

• Putnam (2007): 41 U.S. communities, more than 29,000 people (SCCBS). Residents of ethnically diverse neighborhoods trusted not only other groups but their own group less, and showed lower trust in local government and media, less neighborly cooperation, less giving, and fewer friends. The “constrict” thesis — “hunkering down” like a turtle (Scandinavian Political Studies, vol. 30, no. 2, pp. 137–174).

• Alesina & La Ferrara (2002): the strong predictors of lower trust were (1) traumatic experience, (2) membership in a discriminated-against group, (3) economic failure, and (4) residence in a racially mixed community; religion and origin themselves were not significant (Journal of Public Economics, vol. 85, no. 2, pp. 207–234).

• Dinesen, Schaeffer, and Sønderskov (2020): a meta-analysis of 87 studies and 1,001 estimates. The negative diversity–trust relationship is statistically significant and stronger at the neighborhood level and with more localized measurement (Annual Review of Political Science, vol. 23, pp. 441–465).

[Derived] The gap in this literature: the correlation is robust, but the mechanism is unsettled. This paper's candidate mechanism is “information before emotion (hostility)” — in a mixed environment, the shared information needed to read the other party is lacking, and the decline in trust may be the result of rising inference costs. Alesina & La Ferrara's finding — that it is the mixed “environment,” not origin, that operates — is consistent with this interpretation.

2.5 Institutional Economics: The Classics of Externalization

[Fact]

• Coase (1937): market transactions carry costs, and the reason firms exist is to reduce those costs (“The Nature of the Firm,” Economica).

• North (1990): the shift from personal exchange to impersonal exchange requires formal rules and third-party enforcement (Institutions, Institutional Change and Economic Performance, Cambridge UP).

• Greif (1994): a comparison of two merchant groups in the medieval Mediterranean. The Maghribis (collectivist) enforced through intra-group reputation, communication networks, and informal sanctions; the Genoese (individualist) traded outside the group and developed formal, legal enforcement devices (Journal of Political Economy, vol. 102, no. 5, pp. 912–950).

[Derived] Greif's contrast is the historical prototype of the externalization theory. The Maghribi method = the high-efficiency, low-scalability solution under conditions rich in shared context. The Genoese method = the high-scalability solution under conditions lacking shared context. Historically, the side that expanded was Genoa, and the reason was not morality but scalability.

3. The Externalization of Context: Core Theory

3.1 Definition

Context externalization: the process of moving information that existed in people's heads or relationship networks into external systems such as language, documents, contracts, rules, procedures, and standards.

Examples: (1) price tag — the externalization of the price in the seller's head; (2) contract — the externalization of a promise in memory; (3) traffic signal — making mutual guessing of intentions unnecessary; (4) law — a common standard of judgment in place of relational negotiation; (5) administrative procedure — identical transactions between officials and citizens who do not know each other.

The path along which the storage location of information moves: relationships → language → documents → rules → institutions.

3.2 Shared Context Is a Social Resource

The composition of shared context: (1) common history and experience, (2) linguistic nuance, (3) norms and tacit rules, (4) relationship networks and reputation, (5) frames for interpreting situations. When this resource is abundant, few words suffice; when it is depleted, the same cooperation requires more explicit information. Conceptual formula: shared context ↓ ⇒ cost of inferring intentions ↑.

“Nunchi” is not free. Decoding unstated meaning is a composite inference from relationships, the past, facial expression, tone, and situation, and it demands information. A society that demands high-level inference from every member does not scale to a society of strangers.

3.3 The System of Hypotheses

• H1. Social and cultural heterogeneity ↑ ⇒ shared context ↓.

• H2. Shared context ↓ ⇒ cost of inferring intentions ↑.

• H3. Inference cost ↑ ⇒ relative value of explicit communication ↑.

• H4. An environment of high heterogeneity in which continued interaction is compelled ⇒ relative importance of externalized coordination devices (contracts, standards, rules, procedures) ↑.

• H5. An environment in which relational trust is hard to form ⇒ institutional trust based on standardized, predictable procedures functions as a substitute coordination device.

The falsification conditions for each hypothesis are stated in Chapter 13.

4. Why Externalization Works

4.1 The Trade-Failure Theorem

[Fact] The Myerson–Satterthwaite theorem (1983, Journal of Economic Theory): when a buyer and a seller bargain with their reservation prices as private information, there exists a region in which even trades that benefit both sides structurally fail to occur (efficiency, budget balance, individual rationality, and incentive compatibility cannot all be achieved at once).

[Derived] The micro-level meaning of externalization: the externalization of context is an operation that turns private information into public information. It moves a game of incomplete information toward a game of complete information, and as a result the region of trade failure shrinks. The proposition that “transparency lowers transaction costs” is not an intuition but a direct corollary of this theorem.

4.2 Externalization = Mechanism Design

[Derived] In the language of mechanism design theory, externalization devices (standards, rules, procedures) are “rules of a game that work without knowing the participants' types.” Good externalization either makes the reporting of types incentive-compatible (a direct mechanism) or makes type information itself unnecessary (the price tag: the same price regardless of type). The quality of an institution can thus be measured as “the reduction in dependence on type information.”

5. The Two-Channel Theory: Transparency in Negotiation

5.1 Channel 1 — The Information-Rent Channel (Liability)

[Fact/Derived] The standard result of bargaining theory: private information creates information rent. The less observable my budget, my urgency, and my bottom line are to the other side, the larger my share of the bargaining surplus. [Derived] A society with an ingrained habit of documenting everything structurally forgoes this rent. In distributive bargaining that divides a pie, transparency can be a liability.

5.2 Channel 2 — The Credibility Channel (Asset)

[Fact]

• Schelling (1960, The Strategy of Conflict): the party that demonstrates its own inability to retreat gains the advantage in negotiation (the paradox of commitment; delegation; the strategy of the ultimatum).

• Putnam (1988, International Organization): international negotiation is a two-level game between the international table (Level I) and the domestic ratification table (Level II). The smaller the domestic win-set and the more observable it is to the other side, the more the claim “our legislature will not pass this” becomes a verified fact — and thus bargaining power.

• Fearon (1994, American Political Science Review): in a society with an open press, an opposition, and open procedures, a leader's public threats and promises carry domestic costs (audience costs). That is why such a society's signals have credibility.

[Derived] Synthesis: in negotiations where credibility is the asset — promises, alliances, deterrence, norm compliance — a society that is transparent internally (that is, one whose domestic constraints and audience costs are observable) is structurally strong.

5.3 The Theory of Negotiation Types: Where the Sign Flips

[Derived] This paper's core theoretical propositions:

• Commitment games — alliances, deterrence, treaty compliance, norm-setting: transparency is an asset (the credibility channel dominates).

• Bargaining-over-surplus games — trade terms, cost-sharing, prices: transparency is a liability (the information-rent channel dominates).

• Therefore, if “the relationship between internal transparency and diplomatic bargaining power” is measured without splitting by negotiation type, a correlation ≈ 0 will be observed, and that zero is not “irrelevance” but the cancellation of + and −. Empirical work without a type split gives the wrong answer to this problem.

6. Case I: Postwar Busan — The Abrupt Collapse of Shared Context

6.1 Observed Facts

[Fact]

• Population surge: the Busan Encyclopedia of History and Culture (the city's digital reference work) — after the outbreak of the Korean War, the refugee influx and the designation of Busan as the temporary capital brought the population to 844,134 in 1951 and 1,049,363 in 1955 (the first time it passed one million). The city's official chronology also records “population passes one million” for 1955. By the census figures cited by a Busan cultural tourism guide, about 470,000 in December 1949 → 844,000 in February 1951 — nearly double in about one year and two months (KTV Citizen Reporters, 2024).

• Academic confirmation of the scale of increase: Kong (2022) — despite the government's refugee dispersal policy (the “Matter Concerning the Dispersal of Refugees”), concentration in Busan continued, and as of 1951 the population was about 850,000, an increase of roughly 370,000 over the prewar level (Journal of Asian Studies, vol. 25, no. 1, pp. 425–444).

• Total mobilization of coordination devices (Kong, 2022): (1) the Temporary Measures Act on the Accommodation of Refugees of August 4, 1950 — extending accommodation from vested property to ordinary houses, inns, entertainment houses, and other buildings; (2) occupancy orders in the name of the Minister of Social Affairs; (3) the “give up one room” movement of Busan's native residents; (4) food rationing (rice balls → 3 hop of grain per person plus 50 won for side dishes), with UNKRA and UNCACK relief; (5) shantytowns clustered around Gukje Market and Yongdusan, and an average of three to four fires a day.

• Cultural heterogeneity: Choi (2015) — 1950s Busan was a “melting pot of heterogeneous cultures” where “the residue of Japanese culture, the newly rising American culture, and the cultures of refugees and returning compatriots” intersected (Media & Society, vol. 23, no. 1, pp. 47–95).

• Reorganization of urban space: Konduri & Lee (2023) — GIS analysis confirmed that the inflow of refugees and migrants continuously reshaped Busan's population composition, physical development, and housing policy (Sustainability, vol. 15, no. 24, 16857).

6.2 The Boundary Between Fact and Inference

[Fact] What is confirmed: (1) a large-scale inflow in a short period; (2) a change in population composition and cultural heterogeneity; (3) coordination problems in housing, food, administration, and labor, and the total mobilization of improvised devices; (4) the continued absorption of migrants thereafter.

[Derived] Not proven: “These conditions made the communication of Busan's people more direct.” This paper submits this not as a fact but as the Busan hypothesis: in Busan, where shared context was abruptly weakened, the value of explicit and observable information would have risen relatively, to enable rapid cooperation among strangers. The first four links of the causal chain (migration → heterogeneity → decreased context → inference cost → value of explicit information → externalization devices → cooperation) are supported by the historical record; the last three are unverified (verification design: Chapter 12).

7. Case II: The United States — Immigration as a Constant Condition

[Fact]

• Demography: the U.S. Census Bureau's historical summary (U.S. Census Bureau, 2026) — the foreign-born share was 9.7% in 1850 (about 2.2 million), 14.4% in 1870, 14.8% in 1890 (the peak); it fell after immigration restrictions to 4.7% in 1970 (the low), then rose to 50.2 million, 14.8%, in 2024. MPI tally: 47.8 million (14.3%) in 2023; the 2022→2023 increase (about 1.6 million, 3.6%) was the largest since 2010.

• Hall's classification: the United States is the archetype of low-context communication (Hall, 1976).

• U.S. evidence on diversity and trust: Putnam (2007) and Alesina & La Ferrara (2002) are both based on U.S. data (Section 2.4).

[Derived] For nearly 170 years the United States has maintained “a society in which one in ten to one in seven people came from somewhere else.” It is no coincidence that Hall's low-context classification reads as an adaptation to this demographic condition. Since Hall's classification itself is observation-based (Section 2.2), however, this is an assessment of consistency, not a proof of causation.

8. Case III: The 50 U.S. States — A Natural Experiment

8.1 Why Regional Comparison?

[Derived] Cross-national comparison makes causal inference difficult because language, legal systems, and history are all mixed together. The 50 U.S. states share federal law, language, and currency while differing in settlement history, population composition, and norm strength. If “the amount of shared context” determines the mode of coordination, then systematic differences should be observed among states whose conditions for forming it differ.

8.2 Four Independent Measurements, One Terrain

[Fact]

• The geography of individualism–collectivism (Vandello & Cohen, 1999, JPSP vol. 77, no. 2, pp. 279–292): a 50-state collectivism index built from eight archival indicators (share living alone, elderly living alone, grandchildren living in, divorce/marriage ratio, no religious affiliation, Libertarian vote share, carpooling rate, self-employment rate). Collectivism strongest in the Deep South; individualism strongest in the Great Plains and Mountain West. State scores range from Montana 31 (most individualist) to Hawaii 91 (most collectivist). The state index predicted variance in individual attitudes (convergent validity).

• The Southern culture of honor (Nisbett & Cohen, 1996, Culture of Honor, Westview; Cohen, Nisbett, Bowdle & Schwarz, 1996, JPSP vol. 70, no. 5, pp. 945–960): insulted white men from the South, unlike those from the North, perceived a stronger threat to masculine reputation, showed elevated cortisol and testosterone, and showed more aggressive and dominant behavior. Cohen (1996, JPSP vol. 70, no. 5, pp. 961–978) showed that honor norms are inscribed in Southern law, social policy, and the votes of federal legislators.

• Tightness–looseness (Gelfand et al., 2011, Science vol. 332, no. 6033, pp. 1100–1104; Harrington & Gelfand, 2014, PNAS vol. 111, no. 22, pp. 7990–7995): in an analysis of 6,823 people in 33 countries, tight societies have strong norms plus low tolerance for deviance and are associated with ecological and historical threats. The 50-state index is built from nine indicators including the legality of corporal punishment, executions, marijuana penalties, access to alcohol, religiosity, and the foreign-born share of the population. In international comparison, the United States is classified as a relatively loose culture.

• The geography of personality (Rentfrow et al., 2013, JPSP vol. 105, no. 6, pp. 996–1012): a cluster analysis of Big Five data from more than 1.5 million people in five samples over 12 years. (1) Friendly & Conventional (north-central Great Plains and the South); (2) Relaxed & Creative (the West and parts of the East Coast; the most diverse, wealthy, and educated); (3) Temperamental & Uninhibited (New England and the Mid-Atlantic).

8.3 Synthesis

[Derived] Different research teams, eras, and measurements (archival indicators, physiological experiments, norm strength, personality clusters) converge.

• The South: collectivist, internalized honor norms (norms stored in the nervous system rather than in documents), relatively tight, friendly and conventional. The mode of coordination for conditions rich in shared context.

• The West, frontier regions, and immigrant coasts: individualist, loose, relaxed and creative, highly diverse. The mode of coordination for conditions lacking shared context.

[Derived] The implication of this convergence: high and low context are not ethnicity but adaptation to conditions (indirect evidence for the proposition of Section 2.2). But there is one honest tension — in the Harrington & Gelfand index, states with more immigration fall on the “loose” side. At first glance this conflicts with H4 (heterogeneity → explicit rules), but tightness–looseness measures the strength and enforcement of norms, whereas this paper's variable is the storage location of information. The American South is not loose, yet its norms are stored in honor, reputation, and socialization rather than in documents (8.2-2). “The strength of norms” and “the externalization of norms” are orthogonal dimensions, and this distinction is a task for future measurement research.

9. Immigrants' Institutional Trust: Empirical Support for H5

[Fact]

• Fadillah (2026, Sociological Research Online, published online August 17, DOI 10.1177/13607804261473694): World Values Survey wave 7, 21 countries, N = 33,929. Compared with natives, immigrants show +0.14 in institutional trust and +0.03 in social trust — an asymmetry of roughly fivefold (“differential trust incorporation”). The author's interpretation: institutional trust is formed through transactions with legible, standardized procedures accessible even to newcomers, while social trust requires relational experience that takes time to accumulate. The gap is 51% smaller for naturalized citizens, and second-generation immigrants are indistinguishable from natives. The better a country's institutional environment, the larger the gap — immigrants respond to actual institutional characteristics, not vague optimism.

• Maxwell (2010, International Migration Review vol. 44, no. 1, pp. 25–52): in a cross-generational comparison of immigrants in Europe, first-generation immigrants' trust in political institutions was higher than natives'.

• Röder & Mühlau (2012, Journal of Ethnic and Migration Studies vol. 38, no. 5, pp. 777–792): immigrants' high trust in host-country institutions is explained by low expectations or by a comparison frame relative to the country of origin.

• Scoping review (Frontiers in Political Science, 2025): confirmed the repeatedly reported pattern of first generation > second generation in institutional trust, and convergence toward native levels with length of residence.

[Derived] The picture this literature paints as a whole: knowing people takes time, but rules can be provided first. In a highly diverse society, institutions are not the “remaining problem” but the first infrastructure of cooperation (corresponding exactly to H5).

10. Formalizing the Open Problem

[Derived] What has been established so far is the story of a society's “inside”: externalization → reduced transaction costs (Chapter 4), and the substitute function of institutional trust (Chapter 9). But the following is unknown:

Research question: What is the relationship between a society's internal transparency (its level of externalization) and its external diplomatic bargaining power?

The reason this is unknown is that, as we saw in Chapter 5, game theory provides two diametrically opposed channels. Transparency creates credibility (an asset) and at the same time reveals one's hand (a liability).

11. Four Competing Hypotheses and a Verification Design

[Derived]

• The transfer hypothesis: the internal capacity for externalization (standardization, documentation, proceduralization) transfers into the ability to make treaties, international norms, and standards → the credibility channel dominates. International standard-setting is “the externalization of context between states.”

• The training-ground hypothesis: a society with high internal diversity rehearses “settling terms with an unknown counterpart” every day and thus has a broad human base for external negotiation. (Caution: theoretically the weakest link. Micro-level evidence that repeated-game experience scales up to the state level is lacking.)

• The irrelevance hypothesis: diplomatic bargaining power is a function of BATNA (the best alternative if negotiation fails) — military strength, economic size, alliance structure — and is orthogonal to domestic information structure.

• The paradox hypothesis: an excessively explicit society is at a disadvantage in diplomatic games that require detecting and concealing intentions → the information-rent channel dominates.

[Derived] Verification design (reflecting the warning of Section 5.3):

• Dependent variables: records of treaty conclusion and compliance, leadership in international standard-setting, outcomes of trade and cost-sharing negotiations.

• Independent variables: each country's internal transparency indicators (contract enforcement, rule of law, disclosure).

• Required split: analyze treaties separately as commitment-type (alliances, security, norms) and bargaining-type (trade, cost-sharing, prices). A + in the commitment type and a − in the bargaining type means both channels coexist. If only one is significant, that channel dominates. If the aggregate correlation without the split is zero, this must be read not as irrelevance but as cancellation.

12. Return to Busan: An Empirical Design for Communication

[Derived] “Busan directness” is a hypothesis, not a folk notion. Paths to verification:

• A comparative corpus of historical texts: Busan-area newspapers, radio scripts, commercial advertisements, union documents, corporate documents, administrative documents, and market and port transaction records from the 1950s to the 1970s, versus Seoul, Daegu, and Gwangju in the same period. Indicators: frequency of direct speech acts, density of euphemism, and the proportion of explicit statements of quantity, deadline, and conditions.

• Within-city contrast: refugee-dense settlement areas (around Gukje Market, Yongdusan, and Ami-dong) versus areas of native continuity — controlling fixed effects through within-region variation in the intensity of exposure to the refugee influx.

• Secondary analysis of oral history: re-coding transaction, negotiation, and conflict scenes in existing refugee oral records (the frequency and context of statements of the type “we couldn't understand each other” or “you had to say it plainly”).

• Controls: urbanization rate, income, education, industrial structure. Following Alesina & La Ferrara's (2002) warning (that it is the mixed environment, not origin, that operates), separate compositional effects from contextual effects.

13. The Hard-Core Propositions

[Derived] The hard core of this research program (propositions whose falsification would require revision of the entire program):

• P1 (the storage-location proposition): every coordination device stores information either in “people” (relationships, the inner self) or in “systems” (documents, rules, institutions). This is a continuum, not a binary choice.

• P2 (the condition proposition): the amount of shared context determines the relative value of system storage. Shared context ↓ ⇒ marginal benefit of externalization ↑.

• P3 (the cost proposition): externalization reduces transaction costs (the Myerson–Satterthwaite line). This is the benefit item of externalization.

• P4 (the price proposition): externalization forces the surrender of information rent in negotiation. This is the cost item of externalization.

• P5 (the asymmetry proposition): the benefit of P3 is realized mainly in internal games, and the cost of P4 mainly in external bargaining-type games. The credibility channel adds a benefit in external commitment-type games.

14. Falsification Conditions

[Derived] This program is to be discarded or revised upon the following observations.

• The discovery of a case in which shared context dropped sharply, yet externalization devices did not develop and interaction did not collapse either → revise P2.

• A case in which externalization devices developed without any reduction in information asymmetry → the mechanism of P2 is in error.

• Systematically unfavorable outcomes for countries with high internal transparency in commitment-type negotiations → the credibility channel is refuted.

• Systematically favorable outcomes for countries with high internal transparency in bargaining-type negotiations → the information-rent channel is refuted.

• Repeated findings of high institutional trust among immigrants even in countries with poor institutional environments → revise H5 (Fadillah 2026's “moderating effect of the institutional environment” is already a meaningful passing case for this falsification condition).

15. Summary of Counterexamples and Limitations

• The diversity → conflict path: diversity may end not in externalization but in separation and hunkering down (Putnam, 2007). [Derived] Externalization is not inevitable but one adaptation under “conditions in which continued interaction is compelled.” The branching condition (what separates externalization from separation) is unresolved.

• The problem of directionality: it cannot be determined whether explicit institutions are the result of diversity, or whether societies equipped with institutions are the ones that can bear diversity (selection and reverse causation). Fadillah's (2026) moderating effect suggests that institutional quality is a precondition.

• The methodological limitations of Hall's original work (Kittler et al., 2011): this paper partly avoids them by redefining the concept, but there is no instrument for directly measuring “context dependence.”

• The unverified causation in Busan: as stated in Section 6.2, the change in Busan's communication is a hypothesis.

• The orthogonality problem with tightness–looseness (Section 8.3): separate measurement of norm strength and externalization level is needed.

16. Why This Program Is New

[Derived] This program's claims to originality are three.

• The rewiring of concepts: Hall (communication), Coase–North (institutions), Myerson–Satterthwaite (information theory), and Schelling–Putnam–Fearon (negotiation) existed independently within their own schools. This paper connects them through a single variable: “the storage location of information.”

• The identification of where the sign flips: the question “is transparency good?” is discarded and replaced with “in which type of game, and in which direction?” The 2×2 structure of internal/external × commitment/bargaining is the theory's unit of prediction.

• A testable research agenda: the Busan corpus study (Chapter 12) and the treaty-type-split analysis (Chapter 11) are two concrete empirical projects — the first work orders of this program.

17. Conclusion: A Society That Does Not Require Mutual Understanding, and Its External Diplomatic Bargaining

The question this paper posed is not “which culture is better.” The problem is scale and heterogeneity. In a society where one cannot know everyone, a society must either compel inference or externalize information. The devices modern society has developed — contracts, law, price tags, standards, procedures, timetables, ID cards, protocols — all perform the same function. They move the context that people carried into systems.

The gain is certain at the level of a theorem. Transaction costs fall, dead trades come back to life, and strangers cooperate. Busan in 1951 was the extreme test bed, 170 years of the United States the constant experiment, and the 50 states the internal control group.

The cost is unknown. An internally transparent society may be strong where promises are sold and weak where prices are haggled. Which of these determines the society's external fate more, no one has yet measured.

So this paper ends not with an answer but with a measurement plan. A good theory is not one that reaches a conclusion, but one that points precisely at what must be measured.

(Every item was checked and verified against the original source or an accredited bibliographic database. See the appendix.)

• Alesina, A., & La Ferrara, E. (2002). Who trusts others? Journal of Public Economics, 85(2), 207–234.

• Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405.

• Cohen, D. (1996). Law, social policy, and violence: The impact of regional cultures. Journal of Personality and Social Psychology, 70(5), 961–978.

• Cohen, D., Nisbett, R. E., Bowdle, B. F., & Schwarz, N. (1996). Insult, aggression, and the southern culture of honor: An "experimental ethnography." Journal of Personality and Social Psychology, 70(5), 945–960.

• Dinesen, P. T., Schaeffer, M., & Sønderskov, K. M. (2020). Ethnic diversity and social trust: A narrative and meta-analytical review. Annual Review of Political Science, 23, 441–465.

• Fadillah, M. I. (2026). Differential trust incorporation: How immigrants develop institutional trust faster than social trust. Sociological Research Online. Published online ahead of print (August 17, 2026). https://doi.org/10.1177/13607804261473694

• Fearon, J. D. (1994). Domestic political audiences and the escalation of international disputes. American Political Science Review, 88(3), 577–592.

• Gelfand, M. J., Raver, J. L., Nishii, L. H., et al. (2011). Differences between tight and loose cultures: A 33-nation study. Science, 332(6033), 1100–1104.

• Greif, A. (1994). Cultural beliefs and the organization of society. Journal of Political Economy, 102(5), 912–950.

• Hall, E. T. (1976). Beyond Culture. New York: Anchor Books/Doubleday.

• Harrington, J. R., & Gelfand, M. J. (2014). Tightness–looseness across the 50 United States. PNAS, 111(22), 7990–7995.

• Kittler, M. G., Rygl, D., & Mackinnon, A. (2011). Beyond culture or beyond control? Reviewing the use of Hall's high-/low-context concept. International Journal of Cross Cultural Management, 11(1), 63–82.

• Konduri, S., & Lee, I.-H. (2023). Refugee and migrant integration in urban spatial structures and city development: Case study of Busan, South Korea. Sustainability, 15(24), 16857.

• Maxwell, R. (2010). Evaluating migrant integration: Political attitudes across generations in Europe. International Migration Review, 44(1), 25–52.

• Myerson, R. B., & Satterthwaite, M. A. (1983). Efficient mechanisms for bilateral trading. Journal of Economic Theory, 29(2), 265–281.

• Nisbett, R. E., & Cohen, D. (1996). Culture of Honor: The Psychology of Violence in the South. Boulder, CO: Westview Press.

• North, D. C. (1990). Institutions, Institutional Change and Economic Performance. Cambridge: Cambridge University Press.

• Putnam, R. D. (1988). Diplomacy and domestic politics: The logic of two-level games. International Organization, 42(3), 427–460.

• Putnam, R. D. (2007). E pluribus unum: Diversity and community in the twenty-first century. Scandinavian Political Studies, 30(2), 137–174.

• Rentfrow, P. J., et al. (2013). Divided we stand: Three psychological regions of the United States and their political, economic, social, and health correlates. Journal of Personality and Social Psychology, 105(6), 996–1012.

• Röder, A., & Mühlau, P. (2012). Low expectations or different evaluations: What explains immigrants' high levels of trust in host-country institutions? Journal of Ethnic and Migration Studies, 38(5), 777–792.

• Schelling, T. C. (1960). The Strategy of Conflict. Cambridge, MA: Harvard University Press.

• Kang, Hyeon-suk [강현숙] (2018). Traits of high-context culture in Koreans' communication. Discourse and Cognition, 25(2), 1–24. https://doi.org/10.15718/discog.2018.25.2.1

• Kong, Mi-hee [공미희] (2022). The settlement process of refugees in Busan during the Korean War and government policy. Journal of Asian Studies, 25(1), 425–444.

• Choi, Yi-sook [최이숙] (2015). Postwar 1950s Busan as a postcolonial city, and radio. Media & Society, 23(1), 47–95.

• Busan Metropolitan City. “History of Busan: Chronology”; Busan Encyclopedia of History and Culture, entry “Modern Era.”

• U.S. Census Bureau (2026). Foreign-Born Population in the United States Changed Significantly in Both Numbers and Countries of Origin.

• Migration Policy Institute. Frequently Requested Statistics on Immigrants and Immigration in the United States.

• (Scoping review of immigrants' institutional trust) Frontiers in Political Science (2025).

Every citation in this research program was checked against original sources or accredited bibliographic databases (KCI, official journal pages, SAGE, APA PsycNet, PNAS, U.S. Census Bureau, MPI).

• Verified (bibliographic data and content match): Kang (2018, Discourse and Cognition vol. 25, no. 2, pp. 1–24, DOI 10.15718/discog.2018.25.2.1 — “providing less information than necessary” confirmed as the most frequently mentioned trait) / Kong (2022, Journal of Asian Studies vol. 25, no. 1, pp. 425–444 — +370,000 / 850,000, the Temporary Measures Act of August 4, 1950, and the “give up one room” movement confirmed) / Choi (2015, Media & Society vol. 23, no. 1, pp. 47–95 — “melting pot of heterogeneous cultures” confirmed) / Konduri & Lee (2023, Sustainability 15:16857, 2023-12-14) / Dinesen et al. (2020, ARPS 23:441–465 — 87 studies, 1,001 estimates, stronger for local measures) / Putnam (2007 — 41 communities, 29,000+, the constrict thesis) / Alesina & La Ferrara (2002, JPubE 85:207–234 — the mixed-environment effect) / Greif (1994, JPE 102:912–950) / Hall (1976 — the p. 91 definition) / Kittler et al. (2011, IJCCM 11:63–82) / Vandello & Cohen (1999, JPSP 77:279–292 — the Deep South most collectivist, Montana 31, Hawaii 91) / Nisbett & Cohen (1996, Westview) / Cohen et al. (1996, JPSP 70:945–960 — the physiological-measure experiments) / Rentfrow et al. (2013, JPSP 105:996–1012 — 1.5 million+, three regions) / Gelfand et al. (2011, Science 332:1100–1104 — 33 countries, 6,823 people) / Harrington & Gelfand (2014, PNAS 111:7990–7995 — nine indicators including the foreign-born share) / Maxwell (2010, IMR 44:25–52) / Röder & Mühlau (2012, JEMS 38:777–792).

• One correction: the venue of Fadillah (2026) — the draft's Sociology was an error. The actual venue is Sociological Research Online (ISSN 1360-7804), SAGE / British Sociological Association, published online 2026-08-17, DOI 10.1177/13607804261473694. From the abstract, WVS wave 7, 21 countries, N = 33,929, +0.14 / +0.03 (fivefold asymmetry), a 51% smaller gap for naturalized citizens, the disappearance of the gap in the second generation, and the moderating effect of the institutional environment were directly confirmed.

• Statistical cross-verification: Busan's population was triple-checked across the Busan Encyclopedia of History and Culture, the city's chronology, and the KTV report (844,134 in 1951; 1,049,363 in 1955, the first time past one million; about 470,000 at the end of 1949). U.S. immigration statistics were cross-checked between the U.S. Census Bureau (2026) and MPI.

• Classic literature: Coase (1937), Schelling (1960), North (1990), Myerson–Satterthwaite (1983), Putnam (1988), and Fearon (1994) are cited for edition information only, and the propositions drawn from them are limited to content that can be regarded as standard results of each school (no numerical citations).

• Caution: the figure from Dinesen et al. (2020) that “the neighborhood-level effect ≈ three times the national-level effect” was confirmed through a secondary-citation route, and that route is noted in the text.