[Paper Review] A physicist's view of the notion of "racism"

Charles Jego, Bertrand M. Roehner|ArXiv.org|Apr 22, 2007

Opinion Dynamics and Social Influence17 references3 citations

TL;DR

This paper uses intermarriage statistics from U.S. censuses to argue that segregation is not inherently tied to group identity but is instead driven by the proportion of a minority in a population. It shows that high-minority proportions (e.g., 29% Black in Mississippi) lead to lower intermarriage rates, while low proportions (e.g., 0.56% American Indian in Louisiana) result in higher integration, challenging group-based labels of 'racism' as scientifically unfounded.

ABSTRACT

It is not uncommon, e.g. in the media, that specific groups are categorized as being racist. Based on an extensive dataset of intermarriage statistics our study questions the legitimacy of such characterizations. It suggests that, far from being group-dependent, segregation mechanisms are instead situation-dependent. More precisely, the degree of integration of a minority in terms of the frequency of intermarriage is seen to crucially depend upon the the proportion p of the minority. Thus, a population may have a segregative behavior with respect to a high-p (p>20%) minority A and at the same time a tolerant attitude toward a low-p (p<2%) minority B. This remains true even when A and B represent the same minority; for instance Black-White intermarriage is much more frequent in Montana than it is in South Carolina. In short, the nature of minority groups is largely irrelevant, the key factor being their proportion in a given area.

Motivation & Objective

• To investigate whether the term 'racism' can be objectively applied to entire populations or nations based on social indicators.
• To test the hypothesis that segregation patterns are not group-specific but depend on the proportion of a minority in a given area.
• To use intermarriage rates as a quantitative proxy for integration and assess its consistency with residential and school segregation data.
• To challenge the common assumption that certain regions or groups are inherently more 'racist' based on historical or cultural stereotypes.
• To demonstrate that the same population may exhibit tolerant behavior toward one minority while segregating another, depending solely on demographic proportions.

Proposed method

• The study analyzes 5% random samples from the 1980 U.S. Census to compute intermarriage rates between White individuals and members of various minority groups (Black, American Indian).
• It introduces a normalized integration index Γ, defined as the ratio of intermarriage rates in low- vs. high-proportion minority regions, to compare tolerance across different minority groups.
• The analysis compares intermarriage patterns across U.S. states with varying minority proportions, using a 'null-experiment' to test whether observed patterns deviate from random expectations.
• The researchers cross-validate findings using data on residential segregation (δ index), school integration, and hate crimes from official U.S. Census and FBI sources.
• They apply a comparative framework inspired by statistical physics, modeling inter-ethnic relationships as bond formation between two types of units, emphasizing proportionality over identity.
• The study uses a ceteris paribus approach to isolate the effect of minority proportion from other confounding factors like socioeconomic status or historical legacy.

Experimental results

Research questions

• RQ1 Is the characterization of a population as 'racist' consistent with quantitative data on intermarriage, residential segregation, and school integration?
• RQ2 To what extent does the proportion of a minority in a population determine intermarriage rates, independent of the minority's ethnic identity?
• RQ3 Can the same population exhibit both high and low levels of integration toward different minorities, depending solely on their numerical representation?
• RQ4 How do intermarriage rates compare across regions with similar minority identities but different population proportions?
• RQ5 Does the frequency of hate crimes correlate with minority population proportion, supporting or contradicting the integration hypothesis?

Key findings

• The integration of American Indians into White communities is significantly higher in areas with low minority proportions (e.g., Montana, South Dakota) compared to high-proportion areas, with a 6.4-fold increase in intermarriage rates when proportion drops from 7.2% to 0.56%.
• For Black populations, the difference in intermarriage rates between high- and low-proportion regions is even more dramatic, with a 118-fold increase in integration when the minority proportion drops from 29% to 2.0%.
• In Louisiana, despite a high level of racial tension and low Black-White intermarriage (32% Black population), American Indian-White intermarriage is high (0.56% American Indian population), indicating that the same population can be both segregative and tolerant depending on minority proportion.
• The study finds a strong correlation (r=0.82) between low school integration and high Black population proportion in Southern states, confirming that proportionality drives segregation across multiple indicators.
• Hate crime data from 2000 show 104 per million against Blacks and only 27 per million against American Indians, consistent with higher intermarriage and integration rates for the latter group.
• Residential segregation indices (δ) confirm the pattern: Milwaukee (δ=0.89, p=25% Black) is far more segregated than Orange County (δ=0.52, p=2.0% Black), reinforcing the role of proportion in segregation.