Index Biases

The systemic, often invisible skews built into the methodologies of influential global indices (e.g., Democracy Index, Corruption Perceptions Index, Ease of Doing Business). These biases can include: conceptual bias (defining "democracy" only as multi-party liberal democracy), source bias (relying on surveys of Western-educated elites), methodological bias (weighting factors that favor neoliberal policies), and political bias (producing results that align with the geopolitical interests of the organizations' home countries). Index biases turn quantitative measurement into a powerful tool for ideological normalization.

Example: The Corruption Perceptions Index is often criticized for Index Biases. It tends to rate poorer countries as more corrupt, often because it measures the perception of Western business elites, not the reality of, say, legalized corruption (lobbying, regulatory capture) in wealthy nations. This bias shapes investment flows and political discourse, punishing the Global South for forms of corruption the index is blind to in the West.

by Dumu The Void February 5, 2026

Biases of Biases

The systematic, structural distortions in which biases get recognized and critiqued within a society or institution. The biases of the powerful (e.g., pro-corporate, status-quo bias) are often rendered invisible or "neutral," while the biases of the marginalized (e.g., advocacy, protest bias) are hyper-visible and pathologized. It's a hierarchy of perceived distortion.

Example: In mainstream political commentary, a politician's bias towards protecting Wall Street is framed as "pragmatic realism," while a activist's bias towards wealth redistribution is framed as "ideological extremism." This is the operation of biases of biases—the rules that determine which perspectives are allowed to be "objective" and which must wear the label of bias.

by Dumu The Void February 9, 2026

Biases about Biases

The preconceived, often simplistic, notions we hold about what bias is and who has it. These are stereotypes about bias itself—e.g., "liberal media bias," "conservative denialism," "academic elitism"—that people use as pre-packaged filters to dismiss information without engaging. It's bias at a meta-level: being wrong about how wrongness operates.

Biases about Biases Example: A viewer immediately dismisses a documentary on climate change because "It's from Netflix, and Netflix has a woke bias." This is not an analysis of the film's evidence, but the application of a bias about bias—a canned ideological shortcut that prevents any actual evaluation of content.

by Dumu The Void February 9, 2026

Law of Absolute and Relative Biases

The principle that bias operates in two modes: absolute bias (distortions that are always and everywhere problematic) and relative bias (perspectives that are problematic in some contexts but valuable in others). The law acknowledges that some biases are universally harmful—racism, sexism, any distortion that systematically harms based on irrelevant characteristics. Other biases are context-dependent—a researcher's commitment to a theory can bias their interpretation (bad) or drive productive inquiry (good). The law of absolute and relative biases reconciles the need to reduce harmful bias with the recognition that complete bias-freedom is impossible and that some "biases" are just perspectives.

Law of Absolute and Relative Biases Example: "He accused her of bias because she approached the topic from her cultural background. She invoked the law of absolute and relative biases: some biases are universally harmful (she wasn't expressing those), others are just perspective (her cultural lens was inevitable, not malicious). The question wasn't whether she had bias—everyone does—but whether her bias was distorting or merely situating."

by Abzugal February 16, 2026

Law of Spectral Biases

The principle that biases exist on a spectrum between absolute and relative, with infinite gradations and multiple dimensions. Under this law, no perspective is simply biased or unbiased—each occupies a position in spectral space defined by its sources of distortion, its areas of clarity, its cultural situatedness, its epistemic vices and virtues. The law of spectral biases recognizes that bias is not binary but continuous, that we can be more or less biased in different dimensions, and that the goal is not elimination (impossible) but awareness and mitigation. This law is the foundation of epistemic humility, the recognition that your perspective is always partial, always situated, always capable of improvement.

Law of Spectral Biases Example: "She analyzed her own thinking using spectral biases, mapping it across dimensions: cultural assumptions (present but identified), emotional influences (acknowledged), cognitive shortcuts (working on them), institutional pressures (naming them). The spectral coordinates showed where her bias was most distorting and where it was manageable. She didn't become unbiased—no one does—but she became more aware, which is the point."

by Abzugal February 16, 2026