Why More Data Does Not Automatically Mean More Knowledge
Modern science possesses tools that previous generations could scarcely imagine. Sensors measure changes at microscopic scales, algorithms process billions of observations, and standardized procedures generate results that can be replicated across laboratories, institutions, and countries. Precision has become one of the defining strengths of contemporary research.
Yet this success conceals a rarely discussed paradox: precision is not the same as completeness.
We often assume that increasingly accurate measurements bring us ever closer to reality itself. Likewise, we tend to believe that larger datasets naturally produce deeper understanding. Both assumptions are only partly true. Every scientific measurement captures only a fragment of the reality it seeks to describe.
The Spotlight of Science
Scientific methods can be imagined as a powerful spotlight.
In a dark room, a spotlight illuminates a particular area with remarkable clarity. Details become visible that would otherwise remain hidden. This is precisely what scientific methods do. They create clarity, comparability, and precision.
 |
| Picture by Boris Izmaylov on Unsplash |
But every spotlight has an unavoidable limitation: everything outside its beam remains in the dark.
Every measurement therefore determines not only what becomes visible, but also what remains unseen. A thermometer measures temperature, but not the full complexity of a weather system. A life-satisfaction survey captures responses to predefined questions, but not the entirety of a person's lived experience. An algorithm may detect patterns in data without understanding their meaning for the individuals behind them.
This necessary process of selection can be described as structural selectivity. Scientific methods are not neutral windows onto reality. They are instruments that make certain aspects of the world visible while leaving others beyond view.
This is both their strength and their limitation.
From Object to Data Point
For a phenomenon to become scientifically analyzable, it must first become measurable.
Real-world phenomena are translated into forms that can be quantified, compared, and statistically analyzed. Experiences become scale values. Behaviors become variables. Biological processes become numerical measurements.
This transformation is indispensable. Without it, modern science would be impossible.
Yet every translation alters what is being translated.
When individuals rate their life satisfaction on a scale from one to ten, a data point is created. That data point can be analyzed, compared, and aggregated with thousands of others. But it is not identical to the experience itself.
A gap emerges between the phenomenon and its representation. Complexity is reduced in order to make analysis possible. Science gains precision, but it simultaneously leaves part of the original richness behind.
Measurement does not replace the phenomenon. It creates a methodologically controlled representation of it.
Ontological Partiality: Why Every Measurement Captures Only a Fragment
This leads to a more fundamental insight: scientific knowledge is ontologically partial.
By ontological partiality, I mean that every measurement can capture only part of the phenomenon under investigation. Even the most precise method remains confined to the particular dimensions it has been designed to observe.
This does not mean scientific findings are false.
Quite the opposite.
Scientific results can be highly reliable, reproducible, and extraordinarily useful. Yet their validity remains tied to the boundaries of the measurement process itself.
The problem arises when clarity within the spotlight is mistaken for reality as a whole.
A sharply illuminated fragment can easily create the illusion that the entire landscape has been understood.
The Additivity Fallacy
At this point, a particularly important misunderstanding emerges—one that has become increasingly relevant in the age of Big Data.
If a single data point captures only a fragment of reality, it seems reasonable to assume that enough data points will eventually reconstruct the whole.
But this conclusion does not necessarily follow.
Consider a puzzle. Millions of perfectly measured puzzle pieces are of little help if entire sections of the image were never included in the first place. Increasing the number of pieces does not automatically eliminate the missing parts.
The same logic applies to data.
Millions of survey responses may improve statistical confidence, but they do not overcome the limitations of the original questions. Billions of digital traces may reveal behavioral patterns without capturing motivations, meanings, or subjective experiences.
More data often increase the resolution within the spotlight. They do not automatically expand the reach of the spotlight itself.
This mistaken assumption can be described as the Additivity Fallacy: the belief that accumulating ever larger quantities of data necessarily leads to a more complete understanding of reality.
On the Additivity Fallacy we have an own contribution ➡️see oct.26
Epistemic Honesty
What follows from this observation?
Certainly not a rejection of scientific methods.
The challenge is not to abandon precision but to complement it with reflection.
Researchers should be aware not only of what their methods reveal, but also of what those methods leave unobserved. They should communicate the scope and limits of their operationalizations and remain conscious of the distinction between a phenomenon and its measurable representation.
This attitude can be described as epistemic honesty.
Epistemic honesty means recognizing the power of scientific methods without exaggerating their reach. It means treating data as valuable tools rather than complete mirrors of reality.
Far from weakening science, such transparency strengthens it.
Conclusion: Precision Requires Reflection
The success of modern science rests on its ability to transform complex phenomena into forms that can be systematically analyzed. Without this reduction, many scientific achievements would be impossible.
Yet every measurement remains a fragment.
The spotlight of science can illuminate the world with extraordinary clarity. It reveals what was previously hidden and expands the boundaries of human knowledge. But it remains a spotlight nonetheless.
The challenge, therefore, is not simply to collect more and more data. The challenge is to remain aware of where the light ends.
Knowledge does not emerge from precision alone. It emerges when precision is joined by reflection.
References
Bachelard, G. (2002). The formation of the scientific mind: A contribution to a psychoanalysis of objective knowledge. Clinamen Press. (Original work published 1938)
Cartwright, N. (1983). How the laws of physics lie. Oxford University Press.
Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed again. Cambridge University Press.
Kuhn, T. S. (2012). The structure of scientific revolutions (4th ed.). University of Chicago Press. (Original work published 1962)
Latour, B. (1987). Science in action: How to follow scientists and engineers through society. Harvard University Press.
Morin, E. (2008). On complexity. Hampton Press.
Polanyi, M. (2009). The tacit dimension. University of Chicago Press. (Original work published 1966)
Taleb, N. N. (2007). The black swan: The impact of the highly improbable. Random House.
Weinberg, S. (1994). Dreams of a final theory. Vintage Books.
Watzlawick, P. (1976). How real is real? Communication, disinformation, confusion. Random House.