Core Idea
Somewhere between the research laboratory and the evening news, the comfortable uncertainty of science is solidified into smooth talking points.
What begins as tentative findings in academic papers transforms into unqualified certainties in press releases, then into simplified slogans in advocacy campaigns, and finally into received wisdom that no one questions anymore.
The name for this is system drift.
Ideas travel through institutions, media, and public discourse the way rumours travel through a crowded room. Each retelling strips a little nuance, and by the time the finding reaches public conversation, the original caveats are gone.
Occasionally, though, where jargon or excessive hedging had buried the practical implication, the simplified version lands closer to what the evidence actually showed. Just “wash your hands” is a good example of complex and contested research contracted into both a distortion and an improvement.
More typically, the original conditions, caveats, and context that gave a claim to any validity disappear. What’s left is often a partial or corrupted truth.
Counterpoint
The comfortable assumption we all like very much is that mainstream institutions act as quality filters. Universities vet research, journalists fact-check claims, government agencies base policy on evidence.
When something becomes widely accepted, the assumption runs that the system worked. The better argument won, and the better argument is what people now believe.
The reality is messier.
Institutions have incentives that bend toward consensus rather than accuracy. Academic careers depend on publishing, not on being right. Media outlets need stories that grab attention, not stories that capture complexity. Advocacy groups require simple narratives that motivate action, not complicated truths that inspire paralysis.
Each step in the information chain adds its own distortions.
The peer reviewer who waves through familiar conclusions. The science journalist who cuts the hedging to meet word count. The policy advisor who cherry-picks studies that support predetermined positions. The public intellectual who packages uncertainty into confident predictions. The large language model trained on the drifted version rather than the source.
But all these sources have structural incentives to simplify the evidence. System drift happens because humans prefer simple clarity to complexity they are less likely to understand.
Most of us want conclusions delivered, not questions handed back. So we sand away the rough edges of doubt until we’re left with polished certainties that feel reassuring but may bear little resemblance to what the evidence actually supports.
The most dangerous drifted ideas are those that feel obviously true. They carry the weight of institutional authority without the burden of intellectual rigour.
Thought Challenge
Trace the decay... Pick a claim you’ve heard repeated recently that sounds authoritative. Follow it backwards through the citation chain. Find the original research. What qualifications did the authors include? What limitations did they acknowledge? How many degrees of separation exist between their careful conclusions and the confident assertions you encounter in popular discourse?
Test the foundations... Choose a policy position you support or oppose. Map out the evidence base that supposedly justifies it. How much of that evidence consists of other people’s interpretations rather than primary sources? How many of the key claims trace back to the same small set of studies? What happens to your confidence when you strip away the accumulated authority and look at the raw materials?
Follow the incentive chain... For any simplified claim you have traced, map who benefited from the simplification at each stage. What did the press office gain from removing the caveats? What did the journalist gain from cutting the hedging? What did the advocacy group gain from the confident assertion? Incentives do not establish that a claim is wrong. They identify where pressure entered the system and which translation steps were most likely to distort.
Closing Reflection
Humans need simple stories to navigate complex realities, and institutions need clear messages to justify their existence.
A mindful sceptic accounts for this machinery. When everyone agrees about something, that agreement itself becomes a fact worth investigating. Not because consensus is always wrong, but because the process that creates consensus often has little to do with truth.
The correction is not perpetual suspension of judgement. Follow the chain. Find the primary source. Note what qualifications the authors placed on their own findings. The confidence of the version that reached you is the thing to distrust.
The most dangerous sentence in any field is “Everyone knows that.” It usually means someone stopped checking.
Evidence Support
Ioannidis, J. P. A. (2005). Why most published research findings are false. PLoS Medicine, 2(8), e124.
TL;DR… most published research claims are likely to be false due to biases, flexibility in study design, publication pressures, and selective reporting. It demonstrates that consensus and “received wisdom” often rely on findings that may not replicate or stand up to re-analysis.
Relevance to insight… foundational for the concept of system drift: popular narratives in science become detached from original logic and evidence as unreliable findings are repeated and cemented into “truth” by institutional and media endorsement, regardless of their validity.
Greenhalgh, T., Snow, G. L., Ryan, S., Rees, S., & Salisbury, H. (2015). Six biases against patients and carers in evidence-based medicine. BMC Medicine, 13, 200.
TL;DR… systemic biases that emerge as evidence-based medicine moves from theory to practice, showing how simplified concepts become institutional dogma that overlooks diverse voices and contexts.
Relevance to insight… how institutionalisation and drift occur as foundational logic is lost amid popularisation and advocacy, reinforcing the importance of continual scepticism and reassessment.
Lazer, D., et al. (2018). The science of fake news. Science, 359(6380), 1094-1096.
TL;DR… misinformation amplifies and mutates as it moves through social networks, often starting from minor misinterpretations that ultimately become dominant, widely believed narratives.
Relevance to insight… how logic decays and illusion solidifies as narratives gain popularity and repeat exposure, untethered from original empirical reality.
Lewandowsky, S., Ecker, U. K. H., & Cook, J. (2017). Beyond misinformation: Understanding and coping with the “post-truth” era. Journal of Applied Research in Memory and Cognition, 6(4), 353–369.
TL;DR… psychological and social mechanisms by which originally careful, nuanced statements are reworked and simplified, especially via social media, culminating in widespread post-truth narratives.
Relevance to insight… system drift is systemic and continual—not a single error, but the outcome of persistent oversimplification as complex evidence is molded into comforting stories for mass consumption.
Sumner, P., Vivian-Griffiths, S., Boivin, J., Williams, A., Venetis, C. A., Davies, A., ... & Chambers, C. D. (2014). The association between exaggeration in health related science news and academic press releases: retrospective observational study. Bmj, 349.
TL;DR… A systematic study of UK academic press releases and their corresponding news coverage found that exaggeration of claims was most commonly introduced at the press release stage — before journalists processed the material. Approximately 40% of press releases contained causal claims that went beyond what the original research supported.
Relevance to insight… This is the on-mechanism citation the entry requires. It locates the first major stage of degradation within the institution that produced the original finding, provides quantified evidence that exaggeration is systematic rather than incidental, and directly supports the sequential translation model the insight describes.