The Degree of Shared Context, and the Explicit Society

Transparency Lowers Transaction Costs — But Does That Society Win at Negotiation?

Transactions are expensive

Every transaction carries costs that never appear on the price tag.

The cost of finding out whether the other party can be trusted. The cost of settling the terms. The cost of making them keep their promise. Economics calls these transaction costs. The question Ronald Coase posed in 1937 — if the market is so good, why do firms exist? — was, in the end, a story about this cost as well.

In a transaction with someone you know, this cost has already been paid. The relationship is the collateral, and reputation is the guarantee. A single phrase — “it's them, so it'll be fine” — replaces hundreds of lines of contract.

A transaction with a stranger has no such collateral. So it is expensive. Not because the other party is malicious. Because there is no information.

Modern society is a collection of strangers. So how has society handled this cost?

Transparency — a device for lowering the total cost

If you cannot use collateral, you have to use information.

A price tag takes information that lived in the seller's head and posts it on the wall. A contract moves a promise held in memory onto paper. Standards, procedures, regulations, timetables, ID cards — they all do the same thing. They move information that was stored in people's heads and relationship networks out of those people and into a system.

I call this process the externalization of context.

The anthropologist Edward T. Hall's distinction between high-context and low-context is, in the end, this same problem. In a high-context society, information is stored not in words but in context. In a low-context society, information is carried in explicit language and codes. It is no coincidence that in a study analyzing Korean communication (Kang, 2018), the most frequently observed trait was “saying less than necessary and expecting the rest to be understood from shared context.” Only the storage location of the information differs.

Externalized information is open to anyone. The need to find out about the other party shrinks, the need to verify shrinks, the need to guess shrinks.

Game theory shows exactly why this works. According to the Myerson–Satterthwaite theorem (1983), when a buyer and a seller bargain without knowing each other's true reservation price, there is a region in which even trades that benefit both sides structurally fail to happen. The very fact that information is hidden kills trades. The externalization of context is an operation that reveals this hidden information, and so it revives trades that were dead.

In other words, the essence of transparency is not morality. It is a device for reducing transaction costs. The more transparent a society, the cheaper its transactions with strangers.

Busan, 1951 — a city where transparency was forced

There is a city where this logic was tested to the extreme.

War broke out, and Busan became the temporary capital. According to the city's official records (the Busan Encyclopedia of History and Culture), the population swelled from about 470,000 at the end of 1949 to 844,134 in 1951 — nearly doubling in about a year — and reached 1,049,363 in 1955, the first Korean city outside the capital region to pass one million. Under the Temporary Measures Act on the Accommodation of Refugees, the government designated even inns and entertainment houses as accommodation, and when that was still not enough, Busan's native residents launched a “give up one room” movement (Kong, 2022). It was a movement to hand an empty room to a stranger.

Jeolla, Hamgyeong, Pyeongan — people who did not know one another's dialects or norms began living in the same alleys. Researchers have called 1950s Busan a “melting pot of heterogeneous cultures” (Choi, 2015).

Shared context had physically collapsed. A city where “it's them, so it'll be fine” no longer worked. For transactions to continue in such a city, information has to be stored somewhere else. In prices, in terms, in rules.

Let me be honest. There is as yet no evidence that this changed Busan's way of speaking or of doing business. I leave this as a hypothesis, not a fact. But I want to ask: in a city where shared context evaporated overnight, wouldn't it be stranger if the value of explicit information had not risen?

The United States — the same experiment, run for 170 years

The United States is a society that has lived with this condition as a constant. The foreign-born share of its population was 14.8% in 1890, fell at one point to 4.7% (1970), and as of 2024 is back at 14.8% — 50.2 million people (U.S. Census Bureau). A country where, for nearly 170 years, 10 to 15% of the population has been “people who came from somewhere else.”

Perhaps that is why, in Hall's classification, the United States is the archetypal low-context society. It writes things into contracts, states terms explicitly, and builds standards.

And one recent study shows the price of this structure. Fadillah (2026), analyzing 33,929 people across 21 countries, showed that immigrants trust institutions (police, courts, government) significantly more than natives do (+0.14), while their trust in people is almost the same (+0.03). A fivefold asymmetry. Immigrants trust the system before they know the people. Because the system is open even to newcomers.

Transparency is infrastructure that stands in for trust.

So far, this has been the story of a society's “inside.”

But what about between nations?

Here I stop. And I pose a question to which no one has an answer.

That a society has built a transparent system internally, and whether that society is strong in negotiations with other societies, are separate matters.

The relationship between internal transparency and external diplomatic bargaining power is unknown. Is a transparent country also strong in diplomacy? Or does internal transparency instead trip it up in external negotiations — a society that lays all its cards on the table can hardly be at an advantage in a negotiation? Or are the two unrelated?

Fortunately, there is a tool for this question. Game theory. And game theory answers this question not with “we don't know” but with “it splits.” Because it shows exactly where the sign flips.

The two channels game theory reveals

Treat negotiation between states as a game, and transparency operates through two diametrically opposed channels.

Channel 1 — information rent: what transparency takes away.

The basic result of bargaining theory is this. Private information is money. The less the other side knows about how far I can concede, how urgent I am, and where my real bottom line lies, the larger the share I take from the negotiation. Economics calls this information rent.

A society in which the habit of writing everything down is ingrained has, in effect, given up this rent. In negotiations that divide a pie — trade terms, defense cost-sharing, price haggling — transparency can amount to “showing my hand.”

Channel 2 — credibility: what transparency creates.

There is an opposite channel as well. The paradox Thomas Schelling showed in The Strategy of Conflict (1960) — the one who ties their own hands wins the negotiation. Prove that you cannot retreat, and the other side concedes.

Robert Putnam's two-level game (1988) connected this to domestic politics. International negotiation unfolds at two tables at once: the international table and the domestic ratification table. The more observable a country's domestic constraints are from outside, the more “our legislature will never pass this” becomes a verified fact rather than a bluff. The range of possible agreement narrows — and precisely because of that, the position improves.

James Fearon's audience costs (1994) bring this to its peak. In a society with an open press, an opposition, and transparent procedures, a threat or promise a leader makes into the air carries a domestic price. Say it and fail to keep it, and you pay at the next election. So that society's words carry a different weight. In negotiations where the credibility of promises is the asset — alliances, deterrence, treaty compliance — the transparent society is strong.

To sum up:

Transparency is a liability in bargaining games and an asset in commitment games.

So the answer is “it splits”

Now the four hypotheses can be translated into the language of game theory.

• The transfer hypothesis — the internal capacity for externalization transfers into the ability to make treaties, standards, and norms → the case where the credibility channel dominates.

• The paradox hypothesis — an excessively explicit society is at a disadvantage in games where one must hide one's hand → the case where the information-rent channel dominates.

• The training-ground hypothesis — a society with high internal diversity rehearses “settling terms with an unknown counterpart” every day → theoretically the weakest link. Micro-level evidence that repeated-game experience scales up to state-level negotiation is still lacking.

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

And from this structure comes the single most important warning. If both Channel 1 and Channel 2 are operating, then lumping all negotiations together and measuring “the correlation between transparency and diplomatic power” will yield zero. That zero is not “no relationship.” It is a plus and a minus canceling out. Empirical work that does not split by type produces the wrong answer to this problem.

The verification design looks like this. Take each country's internal transparency indicators (contract enforcement, rule of law, disclosure) and its negotiation outcomes, but analyze treaties separately as commitment-type (alliances, security, norms) and bargaining-type (trade, cost-sharing, prices). A plus in the commitment type and a minus in the bargaining type means both channels are real; if only one appears, that channel dominates.

Instead of a conclusion, a research note

The internal structure of a society that does not require people to understand one another is now coming into outline. Transparency that moves information from people into systems lowers transaction costs, and so cooperation among strangers becomes possible. Busan in 1951 and 170 years of the United States are its test beds.

But what happens when that society meets another society — even game theory does not answer with a single sign. The answer is “it splits,” and the point where it splits is the type of negotiation.

Transparency gives words their weight, and at the same time reveals one's hand.

Where promises are being sold, it is an asset; where prices are being haggled, it is a liability.

Which is larger, no one has yet measured.

So this piece ends not with an answer but with a measurement plan. And perhaps that is what good research is in the first place — not reaching a conclusion, but pointing precisely at what must be measured.

The theoretical background, full citations, and verification design for this piece are set out in the research paper “The Collapse of Shared Context and the Explicit Society.”

Key references: Hall (1976) Beyond Culture / Schelling (1960) The Strategy of Conflict / Coase (1937, Economica) / Myerson & Satterthwaite (1983, Journal of Economic Theory) / Putnam (1988, International Organization) / Fearon (1994, American Political Science Review) / Greif (1994, Journal of Political Economy) / Kang (2018, Discourse and Cognition) / Kong (2022, Journal of Asian Studies) / Choi (2015, Media & Society) / Fadillah (2026, Sociological Research Online) / Busan Encyclopedia of History and Culture, U.S. Census Bureau

-------- Related paper ------ (by Donghun Lee)----

The Collapse of Shared Context and the Explicit Society

Diversity, Communication, and Institutions — A Comparative Study of Postwar Busan and the United States (and Its Regions)

Abstract

What is needed for people of different backgrounds to live together in a single society? Traditionally, this question has been answered with the concepts of social trust, a sense of community, social capital, and shared values. This paper starts from a different point: in a society where people do not know one another well enough, how do members understand one another's intentions and coordinate their behavior? Linking Edward T. Hall's theory of high-context and low-context communication with meta-analytic evidence from the literature on ethnic diversity and social trust, this paper proposes a theoretical model of context externalization. As social heterogeneity increases, shared context among members decreases, the cost of inferring the other party's intentions rises, and as a result the incentive grows to move information out of relationship networks and into external systems — language, documents, contracts, rules, and institutions. The model is applied to two historical cases. The first is postwar Busan, whose population surged from about 470,000 at the end of 1949 to 844,134 in 1951 as refugees poured in during the Korean War. The second is the United States, an immigrant nation whose foreign-born share has ranged between 9.7% and 14.8% since 1850; going further, the paper uses the state-level empirical work of Vandello and Cohen (1999), Nisbett and Cohen (1996), and Harrington and Gelfand (2014) to analyze differences among U.S. regions as a single “natural experiment.” The paper strictly separates observed facts from theoretical inference, and reformulates folk notions such as “Busan directness” or “American directness” as testable hypotheses rather than premises. Every cited source and statistic was verified directly against the original, and the verification record is disclosed in the appendix.

Keywords: high-/low-context communication, context externalization, social trust, institutional trust, diversity, refugees, Busan, U.S. regional cultures

Abstract (author's original English)

What allows strangers to cooperate? Traditional answers invoke social trust, community, and shared values. This paper instead asks how people coordinate when they cannot know one another. Linking Hall's high-/low-context communication theory with meta-analytic evidence on ethnic diversity and social trust, we propose a model of context externalization: as social heterogeneity erodes shared context, the cost of inferring others' intentions rises, creating incentives to move information from heads and relationships into language, documents, contracts, rules, and institutions. We apply the model to two cases: postwar Busan, whose population roughly doubled within months of the Korean War's outbreak, and the United States, an immigrant society whose sub-national regions supply a natural experiment in coordination styles. We strictly separate observed facts from theoretical inference and reframe folk claims about "directness" as testable hypotheses rather than premises. All sources were verified against original records (see Appendix).

Keywords: high-/low-context communication, context externalization, social trust, institutional trust, diversity, refugees, Busan, U.S. regional cultures

1. Introduction: How Do People Who Do Not Know One Another Live Together?

A society is a collection of strangers. In a small community, this problem is not large. Members know one another's families, pasts, reputations, and ways of behaving. The premise “it's them, so it'll be fine” holds. But as cities grow and populations begin to move, this premise weakens. You cannot know where the person you met today came from, which cultural rules they take for granted, or whether they attach the same meaning to the same word.

At this point, a society has broadly two strategies to choose from.

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

2. The information-externalization strategy: reduce the amount of information that must be interpreted in the first place. Speak precisely, write terms into documents, make rules, and apply the same procedure to everyone.

The central claim of this paper concerns the latter. As the cultural and social heterogeneity of a society's members grows, Strategy 1 runs into the limits of scalability, and if the society is to sustain continuous cooperation, the incentive increases to move toward Strategy 2 — drawing information out of people's heads and relationship networks and making it explicit in external systems. This paper names that process context externalization.

To examine this claim, the paper arranges two cases.

1. Postwar Busan: a city whose population nearly doubled in little over a year after it became the temporary capital following the outbreak of the Korean War. A case of the extreme condition in which shared context “collapsed abruptly.”

2. The United States: a society constituted by immigration since its founding, and the archetype of low-context communication in Hall's classification. Moreover, because its 50 states show different settlement histories, population compositions, and norm strengths, the United States offers a “natural experiment” in which inter-regional comparison can be carried out while holding fixed many of the variables that cross-national comparison struggles to control.

From the outset, however, this paper declares one ethical and methodological principle. It does not claim that the demographic change of postwar Busan “created” the communication culture of Busan's people. No data currently exist that directly test such a causal link. What this paper does is (1) present verifiable historical and empirical facts, (2) propose the theoretical mechanism linking those facts as an explicit hypothesis, and (3) present a verification design capable of determining whether that hypothesis is true or false. The distinction between Fact and Derived inference determines the credibility of this paper, and so that readers can check the status of any sentence, the text marks [Fact] and [Derived] throughout.

2. Theoretical Background

2.1 High Context and Low Context: Hall's Original Definition

[Fact] Edward T. Hall introduced the concepts of high-context and low-context communication in Beyond Culture (1976) (Hall, 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,” and a low-context message is the reverse, one in which “the mass of the information is vested in the explicit code” (Hall, 1976, p. 91). In Hall's classification, Japanese, Chinese, Korean, and Arab cultures fall toward the high-context side, and German, Swiss, Scandinavian, and American cultures toward the low-context side.

Simplified, it looks like this.

1. High context: message = words + shared context

2. Low context: message ≈ explicitly expressed information

What matters here is not that one side is superior, but that the information is stored in a different place. High-context communication is extremely efficient in an environment rich in shared context. It is why two people who have worked together for a long time can understand each other without saying everything. But when the same method is applied between strangers who share no context, it fails.

2.2 Criticism of Hall's Theory and This Paper's Response

[Fact] Hall's theory, as widely used as it is, has accumulated criticism. Kittler, Rygl, and Mackinnon (2011) systematically reviewed the use of the high-/low-context concept in the international business literature and pointed out its limitations: (1) a tendency to freeze a continuum into a dichotomy, (2) neglect of diversity within a single country, and (3) the fact that Hall's own evidence rests on field observation and experience rather than controlled experiment.

[Derived] This paper accepts these criticisms but does not take them to mean the theory must be discarded. The paper's usage differs from Hall's original in two respects.

1. This paper does not treat “culture” as an essence. Context dependence is treated not as ethnicity but as situational adaptation to an information structure. The same individual behaves in a high-context way or a low-context way depending on the level of shared context with the transaction partner.

2. This paper focuses not on a fixed national-level classification but on a change in conditions — specifically, an abrupt decrease in shared context. This not only sidesteps the criticism of ignoring within-country diversity, but turns within-country variation into an analytic asset, as in the state-level comparison of the United States (Section 6).

2.3 The High-Context Character of Korean Communication: Empirical Evidence

[Fact] Kang (2018), through qualitative analysis of interview data, analyzed the high-context characteristics of Korean communication within Hall's framework. The most frequently mentioned trait was providing less information than necessary, while expecting unstated information to be understood from shared context through the listener's effort. Also reported were (2) the prevalence and intentionality of ambiguous utterances, (3) the way the importance of relationships regulates many aspects of communication, (4) the suppression of emotions and opinions rooted in the Confucian tradition, and (5) a less rationalist mode of thinking compared with low-context cultures (by respondents' own statements) (Kang, 2018, Discourse and Cognition, vol. 25, no. 2, pp. 1–24).

[Derived] What this study suggests is not the folk notion that “Koreans speak indirectly,” but that the storage location of information differs. Some information is in the sentence, some in the relationship, some in the situation, and some is assumed to be already known by both parties. For two people who share a great deal of context, this structure is efficient. The problem arises when the same structure is applied unchanged to strangers.

2.4 Diversity and Social Trust: Meta-Analytic Evidence

[Fact] The relationship between ethnic diversity and social trust has been studied extensively. The representative evidence is as follows.

1. Putnam (2007): Analyzing the Social Capital Community Benchmark Survey of more than 29,000 people in 41 U.S. communities, he reported that residents of ethnically diverse neighborhoods trust not only other groups but even their own group less, and also show lower trust in local government and media, less neighborly cooperation, less charitable giving, and fewer friends. He described this as “hunkering down,” like a turtle, and named it the “constrict” thesis (Scandinavian Political Studies, vol. 30, no. 2, pp. 137–174).

2. Alesina and La Ferrara (2002): Using individual-level data from U.S. communities, they showed that the factors most strongly associated with low trust were (1) a recent traumatic experience, (2) membership in a historically discriminated-against group, (3) economic failure, and (4) residence in a racially mixed community. By contrast, religious belief and ethnic origin themselves had no significant effect on trust (Journal of Public Economics, vol. 85, no. 2, pp. 207–234). This last point — that the problem is the mixed “environment,” not origin — matters for this paper's interpretation of the mechanism.

3. Dinesen, Schaeffer, and Sønderskov (2020): In a meta-analysis of 1,001 estimates from 87 studies, they confirmed a statistically significant negative relationship between ethnic diversity and social trust, and found the relationship stronger for trust in neighbors and for more localized measures of diversity (Annual Review of Political Science, vol. 23, pp. 441–465). According to the authors' report, the magnitude at the neighborhood level was roughly three times that at the national level.

[Derived] But this literature has a common gap. These studies show that diversity correlates with trust, yet there is no consensus on the mechanism — why it does. This paper proposes one candidate mechanism for that gap. The problem may be information before it is emotion (hostility). That is, as people of different backgrounds increase, the shared information needed to infer the other party's intentions, norms, and expectations becomes scarce, and the decline in trust may be the result of that rise in inference cost.

2.5 Institutions and Impersonal Exchange: Evidence from Economic History

[Fact] The contrast between the internalization and externalization of context has already been demonstrated empirically in economic history. Greif (1994) compared two merchant groups in medieval Mediterranean trade (Journal of Political Economy, vol. 102, no. 5, pp. 912–950).

1. The Maghribi traders: Grounded in collectivist cultural beliefs, they deterred agent misconduct through reputation within the group and collective sanctions. Information spread through the group's internal communication network, trade was confined within the group, and enforcement devices were informal.

2. The Genoese traders: Grounded in individualist cultural beliefs, they traded with outsiders as well, and in exchange developed formal, legal enforcement devices.

[Fact] North (1990) generalized this contrast, formalizing the point that when personal exchange (exchange based on repeated dealings and mutual knowledge) shifts to impersonal exchange (one-off exchange with strangers), informal conventions alone can no longer restrain opportunism, and formal rules and third-party enforcement become necessary (Institutions, Institutional Change and Economic Performance, Cambridge University Press).

[Derived] Greif's contrast illustrates this paper's model exactly. The Maghribi method is the high-efficiency, low-scalability solution for conditions rich in shared context; the Genoese method is the low-density, high-scalability solution for conditions lacking shared context. What deserves attention is that the side that “won” historically was the latter — and that it did so not through moral superiority but through scalability.

3. Theoretical Model: The Externalization of Context

3.1 Shared Context Is a Social Resource

This paper defines shared context as a social resource. It includes the following.

1. Common history and experience

2. Common linguistic nuance

3. Common norms and tacit rules

4. Relationship networks and social reputation

5. A common frame for interpreting situations

When this resource is abundant, a great deal of information is conveyed in few words. When it is scarce, the same cooperation requires more explicit information. Conceptually:

- Shared context ↓ ⇒ cost of inferring the other party's intentions ↑

3.2 “Nunchi” Is Not Free

The ability to sense another person's intentions is commonly regarded as a virtue. But from the perspective of society as a whole, nunchi has a cost. To understand unstated meaning, one must interpret the other person's relationships, past, expression, tone, and situation — an inference that demands information. The scarcer the shared information, the greater the amount of inference required. A method that demands a high level of contextual interpretation from every member works at a scale where members know one another, but does not scale to a society of strangers. Here a society can make a different choice: to make the information that would have to be inferred explicit in the first place.

3.3 Definition: The Externalization of Context

The externalization of context refers to 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 are as follows.

1. Price tag: the externalization of price information that was in the seller's head

2. Contract: the externalization of a promise held in the parties' memories

3. Traffic signal: the externalization of a behavioral rule that spares drivers from guessing one another's intentions

4. Law: the externalization of a common standard of judgment that replaces personal retaliation or relational negotiation

5. Standardized administrative procedure: the externalization of a procedure that lets an official and a citizen transact identically without knowing each other

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

3.4 Hypotheses

From the discussion above, the following testable hypotheses are derived.

1. H1. As social and cultural heterogeneity increases, shared context among members decreases.

2. H2. As shared context decreases, the inference cost of interpreting the other party's intentions and behavior increases.

3. H3. As inference cost increases, the relative value of explicit communication increases.

4. H4. In environments of high social heterogeneity, the relative importance of externalized coordination devices — contracts, standards, rules, procedures — increases.

5. H5. In environments where relational trust is hard to form, institutional trust, which operates through standardized and predictable procedures, can function as a substitute coordination device.

[Derived] Where this model differs from the existing “diversity → distrust” literature is the endpoint. The outcome of diversity is not necessarily conflict, separation, or hunkering down; under conditions where interaction must continue, a productive path — the development of explicit coordination systems — is possible. H5 corresponds exactly to the recent large-scale empirical study (Fadillah, 2026) examined in Section 5.3.

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

4.1 Observed Facts

4.1.1 The Population Surge

[Fact] According to the Busan Encyclopedia of History and Culture, the city's digital reference work, after the outbreak of the Korean War refugees poured in from all over the country and Busan became the temporary capital, so that the population surged to 844,134 in 1951 and reached 1,049,363 in 1955, passing one million for the first time (Busan Encyclopedia of History and Culture, entry “Modern Era”). The city's official chronology likewise records 1950: “became the refugee (temporary) capital as government agencies relocated after the outbreak of the Korean War,” and 1955: “Busan's population passes one million” (Busan Metropolitan City, “History of Busan: Chronology”).

[Fact] The following figures convey the speed of the increase. By the census figures cited by a Busan city cultural tourism guide, the population of Busan in December 1949 was a little over 470,000, but the survey of February 1951, immediately after the January 4 Retreat, showed 844,000 — nearly double in about one year and two months (KTV Citizen Reporters, “Exploring the ‘1,023 Days’ of Busan, the Refugee Capital,” 2024).

[Fact] Kong (2022), in a study using government documents, reported that even though the government promulgated dispersal policies such as the “Matter Concerning the Dispersal of Refugees,” concentration in Busan continued, and as of 1951 Busan's population reached about 850,000, an increase of roughly 370,000 over the prewar figure (Journal of Asian Studies, vol. 25, no. 1, pp. 425–444).

4.1.2 Problems of Settlement and Coordination

[Fact] According to Kong's (2022) account, the responses of the government and the city of Busan were as follows.

1. On August 4, 1950, the Temporary Measures Act on the Accommodation of Refugees was promulgated, extending the scope of accommodation beyond vested property to ordinary houses, inns, entertainment houses, and other buildings.

2. The city of Busan issued occupancy orders in the name of the Minister of Social Affairs to houses with spare capacity, and asked village residents to offer even an empty room or a mill.

3. Busan's native residents spontaneously launched a private “give up one room” movement.

4. Food began with the provision of rice balls and later shifted to rations of 3 hop (about half a liter) of grain per person per day plus 50 won for side dishes, with the UN Korean Reconstruction Agency (UNKRA) and the UN Civil Assistance Command Korea (UNCACK) supplementing relief.

5. Even so, housing and food were short; many refugees built shacks and huts themselves, shantytowns clustered around Gukje Market and Yongdusan, and during this period Busan saw an average of three to four fires a day.

4.1.3 Cultural Heterogeneity

[Fact] Choi (2015) analyzed postwar 1950s Busan as “a melting pot of heterogeneous cultures — the residue of Japanese culture from the colonial period, the newly rising American culture, and the cultures of refugees and returning compatriots” (Media & Society, vol. 23, no. 1, pp. 47–95). In a city so strongly influenced by Japanese culture that it was called “Little Japan,” a heterogeneous broadcasting environment overlapped, in which the central government's propaganda broadcasts coexisted with listening to Japanese radio.

[Fact] Konduri and Lee (2023), in an urban analysis using GIS, showed that the inflow of refugees and migrants after liberation continuously changed Busan's population composition, physical development, and housing policy, and that their settlement reshaped the city's socioeconomic and spatial characteristics (Sustainability, vol. 15, no. 24, 16857).

4.2 The Boundary Between Fact and Inference

[Fact] What has been confirmed so far is as follows.

1. A large number of refugees flowed into Busan in a short period (844,134 in 1951, roughly +370,000 over the prewar figure).

2. The composition of the population itself changed, and the city was culturally heterogeneous.

3. Serious coordination problems arose in housing, food, administration, and labor, and both the state and civil society mobilized improvised coordination devices.

4. Busan continued to absorb migrants during industrialization thereafter (1,049,363 in 1955).

[Derived] But the following has not been proven: “Because of these conditions, the communication of Busan's people became more direct.” This causal claim cannot be made with the data currently available. This paper therefore submits it not as a fact but as a hypothesis.

The Busan hypothesis: In Busan, where shared social context was abruptly weakened by war and mass migration, the value of more explicit and observable information would have risen relatively, so that people of different backgrounds could cooperate quickly.

The causal chain of this hypothesis is as follows.

1. Mass migration → 2. Increase in social heterogeneity → 3. Decrease in shared context → 4. Increase in the cost of inferring intentions → 5. Increase in the value of explicit information → 6. Increase in the importance of language, contracts, rules, and procedures → 7. Increase in the possibility of cooperation among strangers

Links 1–4 are strongly supported by the historical record, but links 5–7 are hypotheses to be tested, and a verification design is proposed in Section 8.

5. Case II: The United States — An Immigrant Society and Explicit Coordination

5.1 The Demography of Immigration

[Fact] According to the historical summary of the U.S. Census Bureau, the foreign-born share of the population started at 9.7% in 1850 (about 2.2 million), reached 14.4% in 1870 and peaked at 14.8% in 1890, fell continuously after the immigration restrictions of the 1920s to a low of 4.7% in 1970, and then rose again, reaching a record 50.2 million in 2024 — 14.8% of the total population of 340.1 million (U.S. Census Bureau, 2026). By the Migration Policy Institute's count, the figure was 47.8 million (14.3%) as of 2023, and the increase between 2022 and 2023 (about 1.6 million, 3.6%) was the largest since at least 2010 (MPI, Frequently Requested Statistics on Immigrants). As of 2023, more than three-quarters of immigrants held lawful status as naturalized citizens, permanent residents, or lawful temporary residents.

[Derived] In other words, for nearly 170 years the United States has maintained a society in which roughly one in ten to one in seven people is “someone who came from somewhere else.” The cities of the great migration of the late nineteenth century — dense with immigrants who did not even share a language — were environments structurally deprived of shared context. This paper's interpretation is that the placement of the United States on the low-context side of Hall's classification is not coincidence, but can be read as an adaptation to these demographic conditions. Since Hall's classification itself is observation-based (Section 2.2), however, this interpretation is a matter of consistency, not of causation.

5.2 Diversity and Trust: Evidence from Within the United States

[Fact] In the U.S. case, two of the three studies in Section 2.4 (Putnam, 2007; Alesina & La Ferrara, 2002) are based directly on U.S. data. Restated, the gist is as follows.

1. The more diverse the neighborhood, the lower the trust, and this lowers trust not only in other groups but in one's own group (Putnam, 2007).

2. More than an individual's origin or religion, the environment of a mixed community predicts the decline in trust (Alesina & La Ferrara, 2002).

3. The effect is stronger at the neighborhood level than at the national level (Dinesen et al., 2020).

[Derived] This pattern contradicts the simple contact hypothesis — that contact reduces prejudice — and is instead consistent with this paper's interpretation: in a mixed environment, the shared information needed to read the other party is lacking. To be clear, this is an assessment of consistency, not a direct proof of the mechanism.

5.3 Immigrants and Institutional Trust: Empirical Support for H5

[Fact] Recent large-scale evidence exists that corresponds directly to this paper's H5 (in environments where relational trust is difficult, institutional trust functions as a substitute device).

1. Fadillah (2026): Analyzing 33,929 people across 21 countries in the seventh wave of the World Values Survey, the study showed that immigrants score 0.14 points higher than natives in institutional trust but only 0.03 points higher in social trust — an asymmetry of roughly fivefold (“differential trust incorporation”). The author interprets this as the two kinds of trust having different “production functions.” Institutional trust is formed through transactions with legible, standardized procedures accessible even to newcomers, whereas social trust requires relational experience that takes time to accumulate. The gap narrows with integration: naturalized citizens show a gap 51% smaller than non-citizens, and second-generation immigrants are indistinguishable from natives. Moreover, the better a country's institutional environment, the larger the gap, suggesting that immigrants respond to actual institutional characteristics rather than vague optimism (Sociological Research Online, published online August 17, 2026).

2. Maxwell (2010): In a comparison of political attitudes across immigrant generations in Europe, first-generation immigrants' trust in political institutions was reported to be higher than natives' (International Migration Review, vol. 44, no. 1, pp. 25–52).

3. Röder and Mühlau (2012): They explained immigrants' high trust in host-country institutions in terms of “low expectations” or “different evaluation standards” (a comparison frame relative to the country of origin) (Journal of Ethnic and Migration Studies, vol. 38, no. 5, pp. 777–792).

4. A subsequent systematic review (a 2025 scoping review) also confirmed the repeatedly reported pattern that first-generation immigrants' institutional trust is higher than the second generation's, and that it converges (declines) toward native levels as length of residence increases (Frontiers in Political Science, 2025).

[Derived] The picture this literature as a whole paints corresponds exactly to this paper's model. Knowing people takes time, but rules can be provided first. Immigrants trust the system before they trust people, and the more legible the system (the more standardized and transparent), the faster that trust forms. This shows that in a highly diverse society, institutions can be not the “remaining problem” but the first infrastructure of cooperation.

6. Regional Comparison Within the United States: The Natural Experiment of 50 States

6.1 Why Regional Comparison?

[Derived] Cross-national comparison makes causal inference difficult because language, legal system, and history all differ. The 50 U.S. states, by contrast, share federal law, language, and currency while differing in settlement history (the timing and origins of immigration), population mobility, and economic structure. If “the amount of shared context” determines the mode of coordination, then systematic differences should be observed among states whose conditions for forming shared context differ. Remarkably, independently developed measurement systems report a consistent geographic pattern.

6.2 The Geography of Individualism–Collectivism: Vandello and Cohen (1999)

[Fact] Vandello and Cohen (1999) standardized and summed eight state-level archival indicators — (1) the share of people living alone (reversed), (2) the share of people aged 65 and over living alone (reversed), (3) the share of households with grandchildren living in, (4) the ratio of divorces to marriages (reversed), (5) the share with no religious affiliation (reversed), (6) the Libertarian vote share in presidential elections (reversed), (7) the ratio of carpooling to driving alone, and (8) the self-employment rate (reversed) — to construct a collectivism index for the 50 states (Journal of Personality and Social Psychology, vol. 77, no. 2, pp. 279–292). The results were as follows.

1. Collectivist tendencies were strongest in the Deep South (the heartland of the former Confederacy — Louisiana, South Carolina, Mississippi, and so on), and individualist tendencies were strongest in the Great Plains and the Mountain West.

2. State-level scores ranged from Montana at 31 (most individualist) to Hawaii at 91 (most collectivist).

3. The index predicted variance in individual attitudes in national surveys (convergent validity) and was systematically associated with demographic, economic, cultural, and health variables.

[Derived] This pattern is consistent with settlement history. The Great Plains and Mountain West are regions with a frontier history — environments in which strangers were scattered thinly and had to solve problems on their own — while the Deep South was an environment of strong settledness and high community density. The former tends to choose individualist, explicit coordination that lets one “transact with anyone”; the latter, relational coordination that means “transacting for a long time with people one knows.”

6.3 The Internalization of Norms: The Southern Culture of Honor

[Fact] Nisbett and Cohen (1996) explained the high rate of honor-related violence in the American South as a “culture of honor” originating in a herding economy (Culture of Honor: The Psychology of Violence in the South, Westview Press). In three experiments by Cohen, Nisbett, Bowdle, and Schwarz (1996) (the “experimental ethnography” studies), white men from the South who had been insulted, unlike those from the North, (1) perceived a stronger threat to their masculine reputation, (2) showed elevated cortisol (emotional arousal), (3) showed elevated testosterone (physiological readiness for aggression), and (4) showed more aggressive and dominant behavior (Journal of Personality and Social Psychology, vol. 70, no. 5, pp. 945–960). Cohen (1996) went further, showing that honor norms are inscribed in Southern law and institutions — gun control, self-defense statutes, and the foreign-policy votes of federal legislators.

[Derived] In this paper's terms, the culture of honor is a case in which “the internalization of context” has developed to the extreme. What one must do in response to an insult is written nowhere, yet everyone in the community knows it. The norm is stored not in documents but in people's nervous systems (cortisol and testosterone). This is evidence, down to the physiological level, of the cost structure of high-context coordination — efficient, but opaque and at times dangerous to outsiders who do not share the norm.

6.4 The Strength of Norms: Tightness–Looseness

[Fact] Gelfand et al. (2011), using data on 6,823 people in 33 countries, placed cultures on a continuum from “tight” (many strong norms plus low tolerance for deviance) to “loose” (weak norms plus high tolerance) (Science, vol. 332, no. 6033, pp. 1100–1104). Tightness is associated with ecological and historical threats such as population density, resource scarcity, territorial conflict, and disease, and tight societies had more laws and controls and stronger surveillance and punishment. In this study, the United States was classified as a relatively loose culture, with Japan, Singapore, and Austria as examples of tight cultures.

[Fact] Harrington and Gelfand (2014) applied this scale to the 50 U.S. states, constructing a state-level tightness–looseness index from nine indicators — the legality of corporal punishment, the execution rate, the severity of marijuana penalties, access to alcohol, religiosity, the foreign-born share of the population, and so on — and showed that tightness is associated with ecological and human-made threats as well as with state-level outcome variables (PNAS, vol. 111, no. 22, pp. 7990–7995).

[Derived] What deserves attention is that the foreign-born share of the population enters this index as an indicator on the “loose” side. That is, the more immigration a state has, the looser its norms and the greater its tolerance. This stands in tension with this paper's model and requires honest treatment (it is addressed as a counterexample and limitation in Section 7.2).

6.5 The Geography of Personality: Rentfrow et al. (2013)

[Fact] Rentfrow et al. (2013) cluster-analyzed Big Five personality tests from more than 1.5 million people across five samples over 12 years and divided the United States into three psychological regions (Journal of Personality and Social Psychology, vol. 105, no. 6, pp. 996–1012).

1. Friendly & Conventional: the north-central Great Plains and the South. High sociability, agreeableness, and conscientiousness; low openness. Politically conservative.

2. Relaxed & Creative: the West and parts of the East Coast. Very high openness; the most racially and ethnically diverse; high wealth and education.

3. Temperamental & Uninhibited: New England and the Mid-Atlantic. High neuroticism; low agreeableness and conscientiousness.

6.6 Synthesis: Four Independent Maps Pointing to the Same Terrain

[Derived] Maps drawn by different research teams, in different eras, with different measurements (archival indicators, physiological experiments, norm strength, personality clusters) converge.

1. The South (especially the Deep South): collectivist (Vandello & Cohen), internalized honor norms (Nisbett & Cohen), relatively tight (Harrington & Gelfand), friendly and conventional in personality (Rentfrow). The mode of coordination for conditions rich in shared context — information is stored in relationships and norms.

2. The West, frontier regions, and immigrant-dense coasts: individualist, loose, relaxed and creative in personality, highly diverse. The mode of coordination for conditions lacking shared context — information is stored in individuals and explicit systems.

[Derived] This convergence is consistent with the predictions of this paper's hypotheses. Regions with low diversity and mobility and strong settledness maintain context-internalizing (high-context) coordination, while regions with deep histories of migration and frontier settlement developed context-externalizing (low-context) coordination. Both ends of Hall's spectrum coexist within the single country of the United States, which supports this paper's interpretation (Section 2.2) that high and low context are not “ethnicity” but adaptation to conditions.

7. Comparative Discussion: Busan, the United States, and Counterexamples

7.1 The Common Structure of the Two Cases

[Derived] Postwar Busan and the United States differ in scale and context, but they share the following structure.

1. An abrupt deficit of shared context: Busan through war, American cities through the great migration, both experienced “a dense concentration of people who did not know one another.”

2. The need for immediate coordination: Busan's housing and food rationing, the labor and commerce of American immigrant cities — neither had the leisure to “wait until people knew one another.”

3. The mobilization of externalization devices: in Busan, administrative documents — occupancy orders, ration cards, accommodation regulations — and in the United States, contracts, standards, law, and price systems filled the void of relational knowledge.

7.2 Counterexamples and Limitations: What This Theory Does Not Explain

There are at least four serious challenges to this paper's model. They are stated openly rather than hidden.

1. The diversity → conflict path. [Fact] The negative relationship between diversity and trust is robust (Dinesen et al., 2020), and diversity can also end in distrust, separation, and group closure (Putnam, 2007). [Derived] The externalization of context is not an inevitable consequence of diversity but one adaptation under the constraint that interaction must continue. Under conditions where transactions can be severed (where separation is possible), hunkering down may be chosen instead of externalization. The branching condition — what sends a society toward “externalization” versus “separation” — is an unresolved research problem.

2. The tension with the tightness–looseness index. [Fact] In Harrington and Gelfand's (2014) index, states with more immigration fall on the “loose” side. [Derived] At first glance this conflicts with H4, “diversity → explicit rules.” But what tightness–looseness measures is the strength and enforcement of norms, not the storage location of information. The American South is not loose (its norms are strong), yet those norms are stored not in documents but in honor, reputation, and socialization (Section 6.3). That is, “the strength of norms” and “the externalization of norms” may be orthogonal dimensions, and this paper's variable is the latter. This distinction must be separated out in future measurement research.

3. The problem of directionality. [Derived] With current evidence, it cannot be determined whether explicit institutions are the “result” of diversity, or whether societies equipped with explicit institutions are the ones “able to bear” diversity (selection and reverse causation). Fadillah's (2026) finding — that the better the institutional environment, the larger the immigrant–native trust gap — suggests that institutional quality is a precondition, leaving open the possibility that externalization is not a cause but a conditional outcome.

4. The limitations of Hall's theory itself. [Fact] The high-/low-context concept has been systematically criticized (Kittler et al., 2011). [Derived] This paper avoids some of that criticism by redefining the concept as a condition-responsive behavioral strategy rather than the essence of a culture, but there is as yet no instrument that directly measures “context dependence” at the individual, organizational, or regional level.

7.3 What the Theory Excludes

[Derived] We state explicitly what this paper's claim is not.

1. It is not the claim that “diverse societies are direct.” The testable claim is more limited: “When a diverse society requires continuous interaction and cooperation, coordination through reliance on shared context alone becomes difficult, so the incentive to rely on explicit communication and externalized institutions increases.”

2. It is not a claim of the moral inferiority of high context. In a group rich in shared context, tacit communication is extremely efficient, and the Maghribi traders' method was rational under its conditions (Greif, 1994).

3. It is not the claim that any particular linguistic habit of Busan is a product of the war. As stated in Section 4.2, that is an untested hypothesis.

8. Verification Design: How to Study “Busan Directness”

[Derived] The sentence “Busan people are naturally direct” may be a stereotype rather than research. An empirical design is proposed for turning it into a hypothesis.

1. A comparative corpus of historical texts. Collect Busan-area newspapers, radio scripts, commercial advertisements, labor union documents, corporate documents, administrative documents, and market and port transaction records from the 1950s to the 1970s, and compare them with corresponding materials from Seoul, Daegu, and Gwangju in the same period. Measures: the frequency of direct speech acts (demands, refusals, price offers), the density of euphemistic expression, and the proportion of explicitly stated terms (quantities, deadlines, conditions).

2. Within-city contrast. Within Busan itself, use refugee-dense settlement areas (around Gukje Market, Yongdusan, and Ami-dong) and areas with strong continuity of native residents as a control group. Because the “intensity” of exposure to the refugee influx varies within the city, comparison within the same city can control for regional fixed effects.

3. Secondary analysis of oral history. Re-code the language of transaction, negotiation, and conflict scenes in existing oral records of refugees. Analyze the frequency and context of statements of the type “we couldn't understand each other” or “you had to say it plainly.”

4. State-level regression analysis for the United States. Link the Vandello–Cohen collectivism index, the Harrington–Gelfand tightness–looseness index, and each state's foreign-born share, population mobility, and historical settlement record, and test whether the proxies for shared context predict the proxies for externalization (contract litigation density by state, volume of regulatory documents, standardization indicators).

5. Control variables. Urbanization rate, income, education, and industrial structure are confounders; in particular, reflecting Alesina and La Ferrara's (2002) warning — that it is the mixed “environment,” not origin itself, that operates — compositional effects and contextual effects must be separated.

9. Conclusion: A Society That Does Not Require Mutual Understanding

The question this paper posed is not “which culture is better.” High context is not bad, and low context is not good. The problem is scale and heterogeneity. In a society where one cannot know everyone, and where not everyone shares the same history and norms, a society must choose: demand higher inference from its members, or make the information that would have to be inferred explicit on the outside.

The devices modern society has developed — contracts, law, price tags, standards, procedures, timetables, ID cards, protocols, interfaces — all perform the same function. They move the context that people carried into systems. Seen this way, the core achievement of modern society may not be that people came to understand one another better, but rather that it built systems allowing people to cooperate without fully understanding one another.

Postwar Busan was a city where this transition took place under extreme pressure; the United States is a society where the transition proceeded for 170 years on the constant condition of immigration; and the 50 U.S. states are a natural experiment showing that the degree of the transition varied with regional history. In all three cases, what this paper claims is not the confirmation of causation, but testable hypotheses built on facts, and the path to testing them.

There is no need to turn diverse people into one culture. Instead, one need only build a society in which different people can work together without fully understanding one another. That is the starting point for seeing diversity not as a cost but as a design problem.

References

(Every item was verified against the original source or an accredited bibliographic database (KCI, official journal pages, PubMed, PNAS, SAGE, APA PsycNet, U.S. Census Bureau, MPI). See the appendix for verification details.)

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

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

3. 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.

4. 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.

5. 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

6. 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.

7. Greif, A. (1994). Cultural beliefs and the organization of society: A historical and theoretical reflection on collectivist and individualist societies. Journal of Political Economy, 102(5), 912–950.

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

9. Harrington, J. R., & Gelfand, M. J. (2014). Tightness–looseness across the 50 United States. Proceedings of the National Academy of Sciences, 111(22), 7990–7995.

10. 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.

11. 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.

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

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

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

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

16. Rentfrow, P. J., Gosling, S. D., Jokela, M., Stillwell, D. J., Kosinski, M., & Potter, J. (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.

17. 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.

18. 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

19. 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.

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

21. Busan Metropolitan City. “History of Busan: Chronology”; Busan Encyclopedia of History and Culture, entry “Modern Era” (844,134 in 1951; 1,049,363 in 1955).

22. U.S. Census Bureau (2026). Foreign-Born Population in the United States Changed Significantly in Both Numbers and Countries of Origin (9.7% in 1850, 14.8% in 1890, 4.7% in 1970, 50.2 million / 14.8% in 2024).

23. Migration Policy Institute. Frequently Requested Statistics on Immigrants and Immigration in the United States (47.8 million / 14.3% as of 2023).

24. Scoping review of immigrants' trust in host-country social institutions (first generation > second generation; convergence with length of residence). Frontiers in Political Science (2025).

Appendix: Source Verification Notes

Before writing this paper, every source scheduled for citation was checked against the original or an accredited bibliographic database. The results are as follows.

1. Verified (content and bibliographic data match): Kang (2018) — Discourse and Cognition, vol. 25, no. 2, pp. 1–24, DOI 10.15718/discog.2018.25.2.1; the five traits in the abstract confirmed (especially “providing less information than necessary” as the most frequently mentioned). Kong (2022) — Journal of Asian Studies, vol. 25, no. 1, pp. 425–444; +370,000 / about 850,000 in 1951, the Temporary Measures Act on the Accommodation of Refugees of August 4, 1950, and the “give up one room” movement confirmed. Choi (2015) — Media & Society, vol. 23, no. 1, pp. 47–95; the description “melting pot of heterogeneous cultures” confirmed. Konduri & Lee (2023) — Sustainability 15(24):16857, publication on December 14, 2023 confirmed. Dinesen et al. (2020) — ARPS vol. 23, pp. 441–465; 87 studies / 1,001 estimates, stronger effect for local measures confirmed. Putnam (2007) — 41 communities / 29,000+, the “hunker down” constrict thesis confirmed. Alesina & La Ferrara (2002) — JPubE 85(2):207–234; the mixed-community effect and the non-significance of religion/origin confirmed. Greif (1994) — JPE 102(5):912–950 confirmed. Hall (1976) — Beyond Culture (Anchor/Doubleday), the high-context definition (p. 91) confirmed. Kittler et al. (2011) — the points of criticism confirmed. Vandello & Cohen (1999) — JPSP 77(2):279–292; the Deep South as most collectivist, Montana 31 / Hawaii 91 confirmed. Nisbett & Cohen (1996) — Westview confirmed. Cohen et al. (1996) — JPSP 70(5):945–960; experimental design and physiological measures confirmed. Rentfrow et al. (2013) — JPSP 105(6):996–1012; 1.5 million+ / three regions confirmed. Gelfand et al. (2011) — Science 332:1100–1104; 33 countries / 6,823 people confirmed. Harrington & Gelfand (2014) — PNAS 111(22):7990–7995; nine indicators (including the foreign-born share) confirmed. Maxwell (2010) — IMR 44(1):25–52 confirmed. Röder & Mühlau (2012) — JEMS 38(5):777–792 confirmed.

2. One correction: the venue of Fadillah (2026). The draft listed it as Sociology, but the actual venue is Sociological Research Online (ISSN 1360-7804), published by SAGE and the British Sociological Association, online ahead of print on August 17, 2026, DOI 10.1177/13607804261473694. From the abstract, WVS wave 7, 21 countries, N = 33,929, institutional trust +0.14 versus social trust +0.03 (fivefold asymmetry), a 51% smaller gap for naturalized citizens, and the disappearance of the gap in the second generation were directly confirmed. Reflected in the text.

3. Statistical cross-verification: Busan's population (844,134 in 1951; 1,049,363 in 1955) was confirmed as consistent across the Busan Encyclopedia of History and Culture, the city's official chronology (passing one million in 1955), and the KTV report (about 470,000 at the end of 1949 → 844,000 in February 1951). U.S. immigration statistics were cross-checked between the U.S. Census Bureau (2026 historical summary) and the MPI tally.

4. Caveats: The figure from Dinesen et al. (2020) that “the neighborhood-level effect is about three times the national-level effect” was confirmed through a secondary-citation route (an explanatory source citing the study), and that route is noted in the text. North (1990) is a classic monograph cited for edition information only, with no numerical citation.