The Apple Tree

01 May 2026

The Apple Tree

“No category of sciences exists to which one could give the name of applied sciences. There are science and the applications of science, linked together as fruit is to the tree that has borne it.” - Louis Pasteur (1871)

Introduction

Whenever I think about human knowledge, the same image comes back to me: an apple tree. It stands in good soil. Rain falls on it, the sun warms it, and over the years it grows, it blooms, and eventually it gives fruit. Civilisations, I think, grow in much the same way - except that a civilisation also needs something that a tree does not. It needs a state, and institutions, to set up the conditions in the first place, to keep the whole thing working, and to hand out the fruit once it appears.

That is what I want to argue here: the state should fund science, and when it does, it should put fundamental research first - because this is what keeps the tree itself alive.

The Soil

The first and most important condition for the development of human knowledge is to create fertile soil. In ancient times, this role was assigned to the ruler. Perhaps the most indicative example is the figure of Al-Ma’mun, a 9th-century caliph and a contradictory figure, who ordered the translation of Aristotle and at the same time persecuted scientists for their beliefs. However, he is remembered for having built the House of Wisdom, a place where Greek, Persian, and Indian texts were translated into Arabic. The next generation did not simply preserve what it had inherited, but also reflected on it, expanded it and spread it throughout much of the known world.

Today we have parliaments, presidents, ministers, and committees. Behind the complex terms of priorities, efficiency, and accountability to taxpayers lies a simple fact - funding for science is woefully inadequate. Unlike in Al-Ma’mun’s time, today’s rulers must set aside money for socially disadvantaged citizens, for healthcare, for a huge administration, for public education, for the maintenance of infrastructure. Budgets are always stretched - and every expense weighed against the public benefit. Put this way, the question seems settled - it is more important to build a new hospital or a school than to spend money on quantum mechanics research, a scientific field unknown to 99.99% of the population, which may yield tangible results in 5 years, in 50, or never. Although most citizens would accept such logic, it hides a fundamental error: it assumes that the benefit of science can be measured on a time horizon no longer than a single parliamentary term.

The opposite logic was formulated as early as 1945 by Vannevar Bush in his work Science: The Endless Frontier - investment in science leads, step by step, to public welfare. The Human Genome Project is a good example. It was funded with public money, and the databases and sequencing methods it produced later opened the door to genetic testing, personalised medicine, and the mRNA vaccines used against COVID-19. One estimate put the economic effect of the project, and of the genomics industry it triggered, at about $1 trillion - many times more than what the state originally spent (Wadman, 2013).

The Tree

The role of the state is not limited to allocating funds. Once it has provided the soil, it must also ensure the proper growth of the tree of knowledge. This means creating an ecosystem in which even unexpected knowledge has a chance to be supported, and ensuring a fair distribution of scarce funds among scientists. In 1997, Donald Stokes challenged Bush’s view, arguing that we can only restore the relationship between government and the scientific community if we understand what is wrong with it. Stokes reinterprets the widely accepted division between theory and application, citing as a case in point the work of Louis Pasteur - fundamental but inspired by practice. Thus arose the ‘Pasteur Quadrant’ - research inspired by practical application but seeking fundamental understanding. Many examples from the last century confirm this thesis - the creation of the transistor, for example, in which inventors simultaneously developed the theory of quantum mechanics and applied it in engineering.

More recently, AlphaFold has directly embodied this idea: it addresses the fundamental question of how the spatial structure of a protein arises from the amino acid sequence, and within a few years it achieved results with direct relevance for biomedicine and drug design. But the leap was only possible because decades of publicly funded crystallography had created the Protein Data Bank - an open archive of about 200,000 experimentally determined structures on which the network was trained. DeepMind harvested the fruit from the orchard planted by the state.

Stokes’ theory challenges the very formulation of the question of whether governments should fund moonshot scientific projects, or incremental improvements with immediate effect. The two are not identical, but they are not opposites either - they are connected like the fruit and the tree that bears it.

The Flow of Sap

The question remains: what is the most appropriate method for this funding? How can the sap of the tree reach every branch and every leaf to bear fruit?

I was particularly impressed by the work of a young Cambridge PhD student, Shahar Avin. He argues that the way the state distributes money for science today - experts read the proposals and decide which ones are good (peer review) - does not actually work well. The reason is not the reviewers. It is deeper: at the time of submission, there is simply not enough information to judge which project will lead to a major discovery. Science is too complex and changes too quickly. Using computer simulations, Avin shows that in the real world, a lottery would select better projects than expert evaluation. Therefore, he proposes a hybrid model: experts quickly sort the proposals into three piles - the obviously strong ones are funded directly, the obviously weak ones are rejected, and the ‘average’ and ‘obscure’ ones enter a lottery.

To make it clear what Avin is talking about, we can go back to AlphaFold. If the problem of the spatial structure of proteins had been put before a standard committee twenty years ago, it would most likely have been rejected - the task seemed insoluble, and the idea that a neural network could replace decades of crystallography would have sounded frivolous. AlphaFold emerged outside academic peer review, in a private laboratory that allowed itself the risk of speculative research.

Going back to the apple tree: if we only plant the seeds that look most promising, we risk losing the ones that would have given the richest and most unexpected fruit. According to his view, this simply cannot be foreseen at the moment of planting.

The Fruit

And so we come to the fruit of our tree - the apples of knowledge that ripen into public welfare. This ripening has been visible for the past two centuries, as a direct result of the Industrial Revolution. Joel Mokyr, who received the Nobel Prize in Economics in 2025, explains why modern economic growth - the steady, cumulative growth since the late 18th century - is the exception, not the rule, in human history. For sustained growth, practical skill is not enough. Propositional knowledge - the abstract understanding of the laws and principles behind phenomena - is mainly needed. Mokyr calls this fusion ‘useful knowledge.’ Growth is self-sustaining only when there is a constant exchange between them within the fusion: theory explains why practice works, practice improves, improvement raises new theoretical questions. The Industrial Revolution was possible because the Enlightenment created a culture in which knowledge was shared openly, publicly verified, and considered legitimate only if it was justified. Without this institutional and cultural foundation, technologies would have exhausted their momentum.

Mokyr’s work convincingly demonstrates that abstract science is not a luxury added to the economy, but a substantial part of the engine that drives it. Therefore, if we apply a conventional distinction between fundamental and applied science, the emphasis of government funding should be precisely on the abstract - because without a constant replenishment of propositional knowledge, practical improvements gradually exhaust their potential, just as they exhausted themselves in all pre-modern civilisations.

There is another strong argument in this direction. According to the BERD Survey (NCSES, 2025), only 6% of the total private sector spending on science was allocated to basic science, while 94% was directed to applied science and R&D.

If the state abandons its role in supporting abstract theoretical science as a priority, the tree will continue to bear apples for some time - but only because it feeds on a reserve accumulated in previous generations. The moment this reserve is exhausted, progress will lose momentum.

The Forbidden Fruit

Finally, I want to draw attention to a disturbing trend. A growing share of both the resources and the fruits goes to military research, which is mainly applied - from 18.5% on average across the OECD to nearly 50% of government R&D funding in the US (OECD, 2026). When a significant portion of scientific funds passes through the military sector, the very nature of knowledge changes. It ceases to be public domain and becomes a resource with limited access - classified, licensed, locked behind security perimeters. After 2022, this trend has intensified: almost all major economies - the EU, Japan, South Korea - have redirected significant resources from civilian research to defence.

But such restriction is foreign to the nature of knowledge, because it becomes powerful only when it spreads. Mokyr’s work on the Republic of Letters shows that open science is not just a convenient way to exchange information - it is the mechanism by which knowledge becomes cumulative.

Open science is a safeguard against the concentration of knowledge in the hands of those who pay for it. It is also the first casualty when scientific resources are diverted to closed military programmes - and the hardest to recover once lost.

Conclusion

Science is a unified body of knowledge that cannot be divided into sectors or neatly categorised. It belongs to our intellectual life, and it is indivisible. The state, as a public institution and the organisational core of civilisation, must first of all provide a fertile environment - soil, water, and light - in which the tree of science can grow. It must control its growth through its institutions and direct the sap in the right, but sometimes unpredictable, direction. When weighing these directions, priority should be given to the fundamental abstract part of science - due to its particular importance for the accumulation of propositional knowledge and due to the fact that it is underfunded by the private and military sectors. Finally, the state must support open science in order to maximise the effect of new discoveries on public welfare.

Coda

I am seventeen years old as I write these lines. Before I turn forty, the world will probably face a shortage of resources that my generation, and those before us, have never known.

History offers two responses to such pressure - and only two.

We can fund science, or we can fund war.

The path depends on the decisions being made now.

That is why I cannot think of scientific funding as something abstract. It will shape the world I shall live in - or the one that ends before I do.

Works Cited

Abramson, Josh, et al. “Accurate Structure Prediction of Biomolecular Interactions with AlphaFold 3.” Nature, vol. 630, 2024, pp. 493–500, doi:10.1038/s41586-024-07487-w.

Avin, Shahar. Breaking the Grant Cycle: On the Rational Allocation of Public Resources to Scientific Research Projects. 2014. University of Cambridge, PhD dissertation. Apollo – University of Cambridge Repository, www.repository.cam.ac.uk/handle/1810/247448.

Bush, Vannevar. Science, the Endless Frontier: A Report to the President on a Program for Postwar Scientific Research. United States Government Printing Office, 1945.

Crawford, Jason. “Pasteur’s Quadrant.” The Roots of Progress, 5 June 2020, blog.rootsofprogress.org/pasteurs-quadrant.

Jeung, Yumi. “Government Cuts R&D Budget, Scientific Community Stunned.” University World News, 13 Sept. 2023, www.universityworldnews.com/post.php?story=20230913174051170.

Jumper, John, et al. “Highly Accurate Protein Structure Prediction with AlphaFold.” Nature, vol. 596, 2021, pp. 583–589, doi:10.1038/s41586-021-03819-2.

Mokyr, Joel. The Enlightened Economy: An Economic History of Britain 1700–1850. Yale UP, 2009.

---. The Gifts of Athena: Historical Origins of the Knowledge Economy. Princeton UP, 2002.

National Center for Science and Engineering Statistics. “Business R&D Performance in the United States Increases to $722 Billion in 2023.” U.S. National Science Foundation, 29 Sept. 2025, ncses.nsf.gov/pubs/nsf25353.

Organisation for Economic Co-operation and Development. “OECD Overall R&D Growth Stable; Government R&D Budgets Decline and Reorient towards Defence.” OECD, Mar. 2026, www.oecd.org/en/data/insights/statistical-releases/2026/03/.

Pasteur, Louis. “Pourquoi la France n’a pas trouvé d’hommes supérieurs au moment du péril.” Revue Scientifique, 1871.

Stokes, Donald E. Pasteur’s Quadrant: Basic Science and Technological Innovation. Brookings Institution Press, 1997.

Wadman, Meredith. “Economic Return from Human Genome Project Grows.” Nature, 12 June 2013, doi:10.1038/nature.2013.13187.