The Vacuum

In 325 CE, the Roman Emperor Constantine convened the Council of Nicaea to resolve a dispute about the nature of Christ. The question was whether Christ was homoousios — of the same substance as the Father — or homoiousios — of similar substance. The difference between these two Greek words is a single letter: iota.

The dispute was not small. Athanasius of Alexandria, champion of the homoousios position, was exiled five times by successive emperors who favored the Arian alternative. The empire's administrative apparatus, military commands, and civic appointments were entangled with the theological question for over a century. Edward Gibbon observed that the Roman world was convulsed by a diphthong.

But the relevant feature of the dispute is not its intensity. It is the relationship between its intensity and its evidence base. The question of whether Christ shared the Father's substance or merely resembled it could not be investigated. No experiment could distinguish the two positions. No observation could falsify either. Each was internally consistent, supported by selective readings of the same scriptural texts, and defended by theologians of equal sophistication. The dispute could not be narrowed by evidence because there was no evidence to narrow it with. And because no evidence constrained the space of permissible argument, every argument was permissible. The debate expanded to fill the epistemological vacuum.

In 1958, Ancel Keys launched the Seven Countries Study, a longitudinal investigation of the relationship between dietary fat and coronary heart disease. Keys proposed that saturated fat raised serum cholesterol, which caused atherosclerosis, which caused heart attacks. The chain was plausible at each link. It was also observational, confounded by a dozen variables — sugar intake, fiber, smoking prevalence, exercise, genetics, national healthcare systems — that the study design could not isolate.

The hypothesis became policy before it became settled science. The 1977 McGovern Committee report recommended reduced fat intake for all Americans. The 1980 USDA Dietary Guidelines made low-fat eating the national standard. Food manufacturers responded by replacing fat with sugar in reformulated products. An entire industry of reduced-fat foods emerged.

The counter-evidence accumulated slowly. The Framingham offspring cohort found no clear relationship between dietary fat and heart disease in the next generation. The Women's Health Initiative — the largest randomized dietary trial ever conducted, following 48,835 women over eight years — found no significant reduction in cardiovascular disease from a low-fat diet. Siri-Tarino and colleagues published a 2010 meta-analysis of twenty-one prospective studies finding no significant association between saturated fat intake and cardiovascular disease.

The debate was among the most heated in medical science. Careers were built and destroyed on fat-heart positions. Journals published accusation and counter-accusation. Congressional hearings revisited the guidelines. Popular books sold millions of copies arguing each side. The heat did not reflect the strength of the evidence. The heat reflected its weakness. Because the observational data could support multiple interpretations, each interpretation attracted defenders whose commitment intensified as the ambiguity persisted. If the data had been decisive in either direction, the debate would have collapsed into textbook knowledge. Instead, the equivocal evidence sustained a sixty-year argument that the evidence itself could not resolve.

In 1747, James Lind conducted a trial on HMS Salisbury. He divided twelve sailors suffering from scurvy into six pairs and administered a different supplement to each: cider, elixir of vitriol, vinegar, seawater, oranges and lemons, or a paste of garlic, mustard seed, and myrrh. The citrus pair recovered within six days. The others did not.

The result was clear, repeatable, and decisive. Lind published it in 1753. The British Navy did not adopt citrus rations until 1795 — a forty-eight-year gap that is usually attributed to institutional inertia, bureaucratic resistance, and the difficulties of preserving citrus on long voyages. What the gap was not attributed to was scientific controversy. No serious counter-argument held that citrus did not cure scurvy. The delay was logistical, not epistemic.

Once adopted, the effect was immediate. Scurvy disappeared from the British fleet within years. The debate ended — not because opponents were persuaded, but because the evidence left no room for alternative positions. There was no vacuum to fill. The data occupied the entire space.

Gregory Benford observed in 1980 that passion is inversely proportional to the amount of real information available. The observation is sometimes treated as a witticism, but it describes a mechanism.

Data is a dampener. In a system where information constrains the range of defensible positions, disagreement narrows to interpretation of shared evidence. The evidence does not eliminate disagreement, but it limits the field. Only certain positions are consistent with the data. Advocates for inconsistent positions lose credibility, lose institutional support, and eventually stop publishing.

In a system where information is absent, no position is inconsistent with the data, because there is no data to be inconsistent with. Every position is equally defensible. Every argument is equally irrefutable. And because no argument can be refuted, no argument is abandoned. The participants do not converge, because convergence requires a shared constraint, and the constraint is exactly what is missing.

The result is not ignorance. Ignorance would be an absence of opinion. The result is the opposite: a proliferation of opinion, each opinion held with increasing conviction, because the absence of disconfirming evidence is experienced subjectively as the presence of confirming evidence. The vacuum does not feel empty. It feels like certainty.

This is why the most heated disputes in public life tend to cluster around topics with the weakest empirical foundations. It is not that people care more about unanswerable questions. It is that answerable questions have a built-in dampener — the next experiment, the next dataset, the next observation — that constrains the oscillation. Unanswerable questions lack the dampener. The system's natural tendency toward amplification runs unchecked. And what fills the vacuum is not knowledge. It is the energy that knowledge, if it existed, would have absorbed.