Science Advances One Funeral at a Time

I am currently reading Simon Singh's Big Bang, a fascinating account of how humanity arrived at its modern understanding of the universe. From the ancient Greeks and their geocentric models to Copernicus, Galileo, Newton, Einstein and the eventual triumph of the Big Bang theory, the book traces centuries of observation, disagreement, experimentation and discovery.

Amidst all this, one observation caught my attention: sometimes, science progresses because scientists die.

It sounds rather morbid, particularly for an enterprise dedicated to understanding the universe. But the more I thought about it, the more fascinating it became. We tend to imagine science as an orderly accumulation of knowledge. Someone makes an observation, proposes an explanation, tests it against evidence, and updates the explanation when better evidence emerges. Eventually, we arrive at a more accurate understanding of reality.

That is how science is supposed to work. Except there is one rather inconvenient variable in this process.

The scientist.

The Inconvenient Human

We generally associate scientists with curiosity, rationality, objectivity and an openness to evidence. These are, after all, the foundations of scientific inquiry. But scientists are also human beings, with all the psychological characteristics that come with being human.

They develop beliefs. They become attached to explanations. They build careers around particular theories. Their professional reputations, intellectual relationships and sometimes their very sense of self become intertwined with their understanding of the world.

Consider the geocentric model of the universe. For centuries, the Earth was understood to occupy the centre of the cosmos. The model was not merely an arbitrary belief that people refused to question. It was a sophisticated intellectual framework, refined over generations, that provided reasonably useful predictions of planetary movements. It was supported by the philosophical and observational understanding available at the time.

Then came the heliocentric alternative. Copernicus placed the Sun near the centre of the planetary system, and subsequent observations and mathematical developments progressively strengthened the case. Yet the transition was neither immediate nor straightforward. Even among astronomers, there were legitimate observational and physical difficulties to resolve before heliocentrism could fully displace the older model.

From our vantage point, the historical progression seems obvious. Earth-centred was wrong; Sun-centred was closer to reality; science moved forward.

But imagine being a scholar who had spent forty years studying, teaching and refining the earlier model. Accepting the new explanation was not necessarily a matter of replacing one astronomical diagram with another. It could mean reconsidering decades of intellectual work and some deeply held assumptions about how the world operated.

The evidence might challenge the theory. But the theory had become part of the person.

One Funeral at a Time

The physicist Max Planck made a famously uncomfortable observation about scientific progress. In his Scientific Autobiography, he suggested that new scientific truths do not always prevail by convincing their opponents. Sometimes they gain acceptance because an older generation passes away and a younger one grows up familiar with the new ideas.

The sentiment is popularly paraphrased as:

Science advances one funeral at a time.

It would be easy to dismiss this as the cynicism of a scientist frustrated with his contemporaries. Except there is some empirical evidence supporting the underlying phenomenon.

In 2019, researchers Pierre Azoulay, Christian Fons-Rosen and Joshua Graff Zivin published a study in the American Economic Review titled Does Science Advance One Funeral at a Time?. They examined how scientific fields responded to the premature deaths of 452 eminent life scientists.

Their findings were revealing. Following the death of an influential scientist, contributions from researchers outside the scientist's established circle increased substantially. These outsiders drew on different bodies of scientific knowledge, explored different research directions and produced work that was disproportionately likely to receive significant academic attention.

The established scientist had not necessarily been wrong. Nor does the study establish that younger researchers are inherently more rational. What it suggests is that intellectual authority can shape which questions are pursued, which approaches attract attention and whose ideas get an opportunity to flourish.

When that authority disappears, the field sometimes opens up.

Scientific progress, it seems, is influenced not only by the quality of evidence but also by the social and institutional structures through which evidence is evaluated.

Even the Universe Had to Overcome a Worldview

Singh's account of the Big Bang offers another fascinating example, separated from the geocentric debate by several centuries.

For much of modern scientific history, the universe was assumed to be eternal and, on the largest scale, unchanging. Even Einstein initially favoured a static universe, introducing the cosmological constant into his equations to permit such a model.

As observations of distant galaxies began to establish that the universe was expanding, a rather unsettling possibility emerged. If galaxies were moving apart, perhaps they had once been much closer together. Perhaps the universe itself had a history.

The emerging Big Bang model suggested that the observable universe had evolved from an earlier hot, dense state. Its rival, the steady-state model, accommodated cosmic expansion while maintaining that the universe had no beginning and remained statistically unchanged over time.

Fred Hoyle, one of the leading defenders of the steady-state theory, was an exceptionally capable physicist. His contributions to understanding how stars manufacture heavier elements were fundamental to modern astrophysics. Yet he remained committed to alternatives to the Big Bang even as accumulating observations, particularly the discovery of cosmic microwave background radiation in 1965, strongly favoured the evolving-universe model.

There is an important distinction here. Hoyle's original opposition was not simply irrational resistance. The steady-state model was a serious scientific proposal, and its supporters made valuable contributions by developing testable predictions and challenging their competitors. Scientific progress requires such challenges.

But the episode also illustrates how difficult it can be to abandon an explanatory framework into which one has invested years of intellectual effort.

Two scientific revolutions, separated by centuries, involving some of the most capable minds of their respective eras. Different theories, different instruments and vastly different levels of scientific sophistication.

Yet a familiar human difficulty seems to appear in both: the challenge of separating what we know from who we believe ourselves to be.

The Difficulty of Changing a Worldview

There is something fascinating about how deeply our understanding of the world becomes embedded in us.

A worldview is not something we consciously construct from scratch every morning. It develops gradually, through education, experience, observation and repeated reinforcement. Over time, the explanations we accept become the assumptions through which we interpret new information. We rarely notice this happening, because those assumptions eventually stop feeling like assumptions. They simply feel like reality.

Consider someone who has spent decades understanding the universe through a particular scientific framework. Every new observation is interpreted in relation to that framework. Every successful prediction reinforces confidence in it. Years of learning, teaching and research gradually produce an internally consistent understanding of how the world works.

Then something comes along that doesn't fit.

One might imagine that a scientist, trained to follow evidence, would simply revise the framework. But a worldview is not an isolated hypothesis that can be replaced without disturbing anything else. It is an interconnected system of explanations, accumulated knowledge and long-held assumptions. Changing one fundamental element can require reconsidering many others.

And the longer a worldview has served someone well, the harder it may be to recognise its limitations. What appears to an outsider as an obvious contradiction can appear to someone deeply invested in the framework as an anomaly that must have another explanation. New evidence is not necessarily rejected outright. It may simply be interpreted in ways that preserve the existing understanding.

This is not unique to science, nor is it necessarily a failure of intelligence or curiosity. The human mind relies on stable mental models to make sense of an enormously complicated world. Constantly reconstructing our understanding of reality would make ordinary thinking nearly impossible. Our ability to preserve and build upon existing knowledge is enormously useful.

But the same stability that makes a worldview useful can make it resistant to change.

And somewhere along the way, the distinction between this is how I understand the world and this is who I am can become surprisingly difficult to maintain.

When Knowledge Becomes Identity

This is where the history of science becomes, for me, a much broader story about human behaviour.

We often treat beliefs as things we possess, rather like books on a shelf. New evidence arrives, we examine it, remove an outdated belief and replace it with a better one.

But beliefs do not always remain independent of the person holding them.

Over time, repeated experiences, social reinforcement, professional success and personal investment can turn a belief into part of our identity. What began as an explanation becomes a way of understanding ourselves and our place in the world.

At that point, contradictory evidence creates a problem that extends beyond the original question.

If the explanation is wrong, what does that say about the years I spent defending it? About the decisions I made? About the expertise for which others respect me? About my judgment?

Psychology offers a useful concept here: cognitive dissonance, the discomfort that arises when conflicting beliefs, evidence or behaviours become difficult to reconcile. One way to reduce that discomfort is to revise our understanding. Another is to reinterpret, discount or avoid the information creating the conflict.

Neither response is unique to scientists. The same mechanisms can appear in professional relationships, families, organisations and personal decisions.

I explored this more deeply in a piece on Neurocient about self-identity and cognitive dissonance. It examines a seemingly unrelated setting: the conflict between a mother-in-law and daughter-in-law in an Indian household. Yet beneath the domestic disagreement lies a surprisingly similar psychological pattern. When someone has spent decades building an identity around a particular way of living, the existence of a credible alternative can feel less like new information and more like a challenge to the meaning of their own choices.

The circumstances could hardly be more different from a scientific dispute. The underlying difficulty, however, may be familiar.

Sometimes we are not defending an idea because the evidence overwhelmingly supports it. We are defending it because abandoning it would require us to reconsider something about ourselves.

This does not mean every strongly held belief is an identity defence, or that disagreement is evidence of closed-mindedness. But it gives us another dimension to consider when apparently intelligent, well-informed people arrive at very different conclusions from the same information.

Curiosity Has an Interesting Blind Spot

There is another distinction worth making.

We usually describe curiosity as a general characteristic. Someone is either curious or not. Scientists, researchers and intellectuals are typically assumed to possess it in abundance.

But perhaps curiosity is more selective than that.

A person may be extraordinarily curious about the universe while being considerably less curious about the assumptions through which they understand it. They may spend decades asking new questions within an established framework without seriously examining the framework itself.

This was one of Thomas Kuhn's central insights in The Structure of Scientific Revolutions. Scientific communities operate within paradigms: shared ways of defining problems, interpreting evidence and deciding what constitutes a legitimate explanation.

A paradigm is indispensable because it makes systematic inquiry possible. Scientists do not need to question every foundational assumption before conducting an experiment. They can build on accumulated knowledge and investigate increasingly specialised questions.

But the same framework that makes progress possible can also constrain what researchers consider worth investigating.

There is, therefore, a difference between curiosity that explores the world within an existing explanation and curiosity that questions the explanation itself.

The former can produce extraordinary scientific advances. The latter occasionally makes scientific revolutions possible.

Neither is inherently superior. Science needs both. But the distinction matters because someone can be exceptionally capable of the first while finding the second deeply uncomfortable.

Science Is Self-Correcting. Scientists Are Human.

None of this should be interpreted as an argument against science. Quite the opposite.

What makes science extraordinary is not that scientists are immune to the cognitive biases, social pressures or identity concerns that influence everyone else.

They aren't.

What makes science extraordinary is that it has developed methods for progressively correcting our understanding despite these limitations. Experimental evidence, reproducibility, mathematical consistency, peer criticism and competing hypotheses provide mechanisms through which explanations can be challenged and improved.

Those mechanisms are imperfect. They operate through institutions, incentives and people. Progress is not always smooth, and some ideas take longer than they should to receive serious attention.

But over sufficiently long periods, the process has demonstrated an extraordinary capacity to replace inadequate explanations with more powerful ones.

The qualification matters: not every scientific advance requires a generational change, and not every generational change produces better science. Established scientists can change their minds. Younger scientists can be just as committed to their own assumptions. Even resistance to emerging ideas can be valuable when it forces their proponents to produce stronger evidence.

The point is not that older scientists must disappear for knowledge to advance. It is that the scientific process sometimes advances despite the reluctance of individuals to revise their convictions.

That is an important distinction.

What Are We Certain About Today?

Reading the history of science creates a peculiar feeling.

It is easy to admire the intellectual courage of Copernicus, Galileo, Lemaître and others who challenged prevailing explanations. It is equally easy to wonder why their contemporaries did not recognise what now seems obvious.

But we are making the same mistake whenever we assume our own understanding is somehow exempt from history.

Our present theories are the best explanations supported by the available evidence. That does not make them final explanations. Two centuries from now, some assumptions that seem entirely reasonable today may be regarded as peculiar limitations of our era.

And this applies far beyond cosmology.

Every profession has its established models, accepted practices and accumulated wisdom. Organisations develop preferred ways of working. Leaders become associated with strategies. Experts build reputations around particular approaches. Over time, the distinction between an idea being useful and an idea being ours can become surprisingly difficult to maintain.

Which brings me back to Planck's uncomfortable observation.

Perhaps science really does advance one funeral at a time, at least occasionally. But the more interesting lesson is not about death, or even about scientific progress.

It is about the relationship between knowledge and identity.

We are remarkably capable of questioning the world around us. We are considerably less consistent at questioning the framework through which we interpret that world, particularly when that framework has become part of who we are.

And that leaves me with a rather uncomfortable thought.