Theory of the Evidence Spectrum

The theory that evidence exists on a spectrum, not as a binary category of "evidence" vs. "not evidence." The Evidence Spectrum recognizes that claims can be supported by evidence to varying degrees, in different dimensions, from different sources. A single anecdote is evidence—weak evidence, low on the spectrum, but still evidence. A randomized controlled trial is stronger evidence, higher on the spectrum. A meta-analysis of many trials is stronger still. The spectrum includes many dimensions: strength, relevance, reliability, independence, replicability. The Theory of the Evidence Spectrum calls for evaluating where evidence falls on multiple axes, not simply asking "is there evidence?" The question is never whether evidence exists but how good it is, how relevant, how reliable—where it sits on the spectrum.

Example: "He dismissed her anecdote as 'not evidence.' The Theory of the Evidence Spectrum showed why that was wrong: it was evidence, just low on the spectrum—weak, but still evidence. Dismissing it entirely was itself unscientific. She wasn't claiming it proved anything; she was claiming it pointed somewhere. The spectrum let them discuss where it fell, not whether it counted."

by Dumu The Void March 7, 2026

Fooled by Evidence Theory

A framework revealing how evidence itself can mislead—not because it's false, but because of how it's produced, selected, and interpreted. Fooled by Evidence Theory shows how publication bias (only positive results published), selection bias (only convenient populations studied), and interpretation bias (only confirming evidence noticed) create an evidence base that systematically misrepresents reality. We are fooled when we trust "the evidence" without asking how it was made, who made it, and what was left out.

Fooled by Evidence Theory "The evidence supports our policy, they announced. But the evidence was funded by corporations, published in pay-to-play journals, and selected from dozens of studies that showed the opposite. Fooled by Evidence: trusting what's presented without asking what's missing. Evidence can lie—not by falsifying, but by selecting. We are fooled by what we're shown, never seeing what's hidden."

by Dumu The Void March 8, 2026

Fallacy of Impossible Evidence

The strategic demand for evidence that cannot exist in principle, often used to dismiss claims that are nevertheless well-supported by the evidence that does exist. Unlike demanding more evidence (which can be reasonable), this fallacy demands evidence of a fundamentally different kind—usually the kind that would require time travel, omniscience, or violation of physical law to obtain. "Where were you at 3:17 AM on June 12th, 2008?" when discussing a general pattern of behavior. "Show me a fossil of the exact moment one species became another" when discussing evolution. It weaponizes the impossibility of perfect records against the possibility of any knowledge at all.

Example: "He demanded security footage from a store that burned down in 1985 to prove I shopped there—pure Fallacy of Impossible Evidence, since the evidence he required was literally ashes."

by Dumu The Void March 11, 2026

decision-based evidence making

An alternative to evidence-based decision making, "decision-based evidence making" is the practice of collecting evidence to support a pre-existing decision, rather than using evidence in any way to inform the decision itself.

This technique typically involves cherry-picking sources or relying on bad or unreliable authorities to justify or legitimise (after the fact) a questionable decision which has already been made.

Dr. Daniel Jernigan resigned from CDC in 2025, telling reporters that Robert F Kennedy Jr seems to be "going from evidence-based decision making to decision-based evidence making" as the Centres for Disease Control began citing debunked studies to support a dodgy claim that vaccines cause autism.

by bitchuck November 20, 2025

Law of Absolute and Relative Evidence

The principle that evidence operates in two modes: absolute evidence (facts that are evidence regardless of perspective, context, or interpretation) and relative evidence (facts that serve as evidence only within particular frameworks, for particular purposes, to particular audiences). The law acknowledges that some evidence is universally compelling—a video of a crime, a DNA match, a document with a signature. Other evidence is context-dependent—statistics that prove one point to one audience and the opposite to another, testimony that's credible in one culture and suspect in another. The law of absolute and relative evidence reconciles the intuition that evidence should be objective with the reality that its force depends on who's judging. Good arguments use both kinds, building on undeniable facts while understanding that interpretation is always relative.

Example: "They argued about whether the data was evidence of climate change. Absolute evidence: the temperature readings were real, measurable, undeniable. Relative evidence: whether those readings proved catastrophic warming depended on models, interpretations, and assumptions. The law of absolute and relative evidence said: the data was absolute; its meaning was relative. They stopped arguing about the data and started arguing about interpretation."

by Abzugal Nammugal Enkigal February 16, 2026

The 8 Axes of the Evidence Spectrum

A framework for evaluating evidence along eight key dimensions, providing a comprehensive map of where any piece of evidence falls. The 8 axes are: 1) Strength (how powerfully the evidence supports the claim), 2) Reliability (how trustworthy the source/method is), 3) Relevance (how directly the evidence addresses the claim), 4) Independence (how free the evidence is from conflict of interest), 5) Replicability (how consistently the finding can be reproduced), 6) Sample/Population Fit (how well the sample represents the population of interest), 7) Methodological Rigor (how well the study was designed and executed), and 8) Consilience (how well the evidence coheres with other established knowledge). These axes allow for nuanced evaluation rather than binary judgments.

The 8 Axes of the Evidence Spectrum Example: "They stopped arguing about whether the study was 'evidence' and started mapping it on the 8 axes. Strength: moderate. Reliability: high. Relevance: low (different population). Independence: questionable (industry funded). The axes showed where the evidence was strong and where it was weak—and why they disagreed about what it meant."

by Dumu The Void March 7, 2026

The 16 Axes of the Evidence Spectrum

An expanded framework for even more nuanced evaluation, adding eight dimensions to the original eight. The additional axes include: 9) Temporal Relevance (how current the evidence is), 10) Ecological Validity (how well the evidence reflects real-world conditions), 11) Mechanistic Understanding (whether we know why the evidence works), 12) Alternative Explanations (how thoroughly competing explanations have been ruled out), 13) Effect Size (how large the observed effect is, not just whether it's statistically significant), 14) Precision (how narrow the confidence intervals are), 15) Generalizability (how well the findings apply across contexts), and 16) Transparency (how fully the methods and data are available for scrutiny). The 16 axes provide a nearly complete picture of evidential quality, useful for high-stakes decisions where nuance matters.

The 16 Axes of the Evidence Spectrum *Example: "The policy debate was high-stakes, so they used all 16 axes. The evidence was strong on reliability and rigor, weak on ecological validity and generalizability. The 16 axes showed exactly where the uncertainty lay—not in whether the evidence existed, but in how well it applied. The policy was informed, not determined, by evidence—which is how it should be."*

by Dumu The Void March 7, 2026