Matt Ridley’s not-so-mythical “myth” of basic science

What Ridley gets right (and what he bulldozes by accident)

Let’s start with the part that deserves a polite nod. The classic “linear model”basic science → applied science → productcan be misleading.
Real innovation rarely behaves like a neat corporate org chart. Technologies often precede full scientific explanations. Industries frequently
develop tools that later enable new discoveries. Translation from lab insight to real-world impact can take decades, detour through dead ends,
and emerge from places nobody predicted.

So yes: if your favorite science story starts with “Then a professor had a brilliant thought,” and ends three pages later with “and that’s why
your smartphone exists,” you’re probably reading a bedtime story for grant committees. (No judgment. Bedtime stories are soothing.)

But where Ridley’s critique goes off-road is when “linear model is simplistic” quietly turns into “basic science is mostly decorative.”
Those are different claims. One is a warning about storytelling. The other is a policy stance with real consequences: less support for research
whose benefits are uncertain, long-term, or shared broadly rather than captured by a single company.

Basic science, in plain English: what it is and why it’s hard to “ROI” on purpose

Basic research is usually defined by its intent: it’s work aimed at understanding underlying principles, not building a product on a deadline.
Applied research aims at a specific practical objective. Experimental development turns knowledge into new or improved products and processes.
That sounds tidyuntil you meet real scientists, who will happily do all three before lunch and call it “a normal Tuesday.”

The false binary: “basic” vs “useful”

A better way to think about research is not “basic or applied,” but “what questions are we asking, and what constraints are shaping the work?”
Some research is curiosity-driven. Some is mission-driven (health, defense, energy, agriculture). Some is “use-inspired” and sits in the sweet
spot where deep understanding and real-world needs push each other forward.

Modern translational science frameworks say this out loud: basic research isn’t an optional preface you can skip; it’s the foundation that every
later stage builds oneven when the path is twisty and the payoff arrives wearing a fake mustache.

Why markets underfund the “weird but important” stuff

Ridley loves markets (and markets love many things back). But even hardcore economists agree on a stubborn reality: knowledge spills over.
When discoveries are easy for others to learn from, companies can’t capture all the benefitsso they tend to invest less than what’s best for
society overall. That’s not a moral failure; it’s a math problem. Public funding exists largely because some research behaves like a public good:
valuable, shareable, and not easily monopolized without slowing the whole system down.

If you want a simple metaphor: basic science is like building roads. A private company might build a road to its own factory. A nation builds
road networks so everyone can move, trade, respond to emergencies, and invent the next thing. Nobody demands that each mile of interstate
must personally deliver a quarterly returnbecause we understand the system-level value.

The “not-so-mythical” reality: how basic science actually fuels innovation

Basic science doesn’t usually hand you a finished gadget. It does something more powerful (and less Instagrammable): it expands what’s possible.
It creates tools, methods, measurements, materials, and mental models that make later breakthroughs cheaper, faster, and more reliable.

1) Basic research creates “option value”

Many breakthroughs are not planned outcomes; they’re the result of being prepared when the world changes. Pandemics happen. New materials become
manufacturable. Computing power explodes. A discovery that looked abstract suddenly becomes the key that fits a brand-new lock.

2) It builds instruments and techniques that unlock whole industries

One of the sneakiest ways basic science matters is through measurement. You can’t engineer what you can’t measure. Advances in microscopy,
spectroscopy, imaging, sequencing, and precision timing are like upgrading humanity’s senses. And once your senses get better, your ambitions
get bolder.

3) It trains people who become the innovation ecosystem

Research funding doesn’t just buy experiments; it builds talent. Graduate students, postdocs, lab staff, and early-career investigators learn
how to solve problems no one has solved before. Some stay in academia. Many move into startups, industry R&D, medicine, data science,
and government labs. The skills transfer even when the exact topic doesn’t.

Case studies that refuse to stay in a straight line

If Ridley’s story is “tinkerers first, theory later,” the real story is “it’s complicated, and both matter, often in alternating bursts.”
Here are a few examples where basic science is not a mythical unicornit’s the invisible scaffolding holding up very practical things.

GPS: your phone’s “blue dot” runs on deep physics and absurdly precise time

GPS is a master class in why “abstract science” is sometimes the most practical thing in the room. Satellite navigation depends on timing so
precise that even tiny relativistic effectshow gravity and motion affect clocksmust be corrected. Without accounting for these effects, errors
would accumulate and positioning would drift. GPS also depends on atomic clocks, the kind of technology that looks like pure physics until you
realize it’s what makes modern navigation possible.

In other words: if you’ve ever arrived at the correct coffee shop instead of the “congratulations, you’re now in a river” coffee shop, you’ve
benefited from the kind of science that once sounded like a philosopher’s fever dream.

The internet: mission-driven research, academic networks, and a lot of unglamorous plumbing

The early internet wasn’t born as a consumer product; it grew from research networks, standards development, and sustained investment in
networking infrastructure. Programs like ARPANET proved concepts and connected early nodes; later, broader academic networking efforts scaled
access and created the backbone that helped the network of networks become… well, the internet.

Markets eventually built massive businesses on top of that foundation. But the foundation itself depended on long-horizon funding, shared
protocols, and interoperability goals that no single private actor could easily justify on short timelines.

mRNA vaccines: “overnight success” after decades of molecular biology

mRNA vaccines looked sudden to the public because the crisis was sudden. But the scientific groundworkunderstanding nucleic acids, immune
responses, protein expression, delivery systems, and how to stabilize and package genetic instructionswas built over many years. When COVID-19
hit, that accumulated knowledge meant developers weren’t starting from zero; they were sprinting from a very high starting line.

This is exactly the kind of payoff basic biomedical science tends to deliver: not a predictable schedule, but a rapid response capacity when
society urgently needs it.

CRISPR: basic research into bacteria ends up rewriting biology’s toolkit

CRISPR is a poster child for “seemingly unrelated curiosity research becomes transformative.” It began as scientists trying to understand how
bacteria defend themselves against viruses. That basic insight evolved into an extraordinarily powerful genome-editing tool used across biology,
agriculture, and medicine research.

It’s hard to imagine a venture capitalist pitching “bacterial immune trivia” as a product line. Yet the downstream value is enormous. That gap
between “not obviously profitable today” and “world-changing tomorrow” is precisely where public investment has the strongest rationale.

Transistors and semiconductors: yes, engineers tinkeredand they also needed fundamental physics

The transistor is often told as an invention story (and it is). But it’s also a story about materials science, solid-state physics, and
painstaking work to understand and control impurities, interfaces, and fabrication techniques. The modern semiconductor era is built on both:
clever engineering and deep scientific understanding that kept shrinking, stabilizing, and scaling devices for decades.

The practical takeaway is not “basic science always comes first.” The takeaway is “when you combine understanding and engineering, you get a
compounding machine.”

The money question: “crowding out,” “crowding in,” and why this debate never dies

Ridley’s argument often leans on an economic concern: if government funds research, it might “crowd out” private R&Dmeaning companies invest less
because public money is doing the job (or because incentives shift toward grant-chasing).

That can happen in specific contexts. But it’s not the only possibility. Public research can also “crowd in” private investment by lowering the
cost of future innovation, producing foundational knowledge, and creating new opportunities for commercialization. Real-world evidence shows both
dynamics exist depending on the field, the funding mechanism, and the market structure.

The honest answer is not a slogan. It’s portfolio management: fund a mix. Evaluate outcomes thoughtfully. Build bridges between sectors without
forcing every project to pretend it’s three months away from an IPO.

A better framing than “basic vs applied”: build an innovation portfolio that can survive reality

If you want the best version of Ridley’s challenge without the overcorrection, it’s this: don’t worship a fairy tale about basic science
magically birthing products. Instead, build a system where:

• Curiosity-driven basic research keeps expanding the frontier of what’s possible.
• Use-inspired research connects deep questions to urgent human needs.
• Mission agencies and translational programs move knowledge toward real-world impact (without pretending translation is easy).
• Industry R&D turns opportunities into scalable products, manufacturing, and services.
• Open science norms (data sharing, reproducibility, standards) keep the ecosystem healthy and cumulative.

The punchline: innovation isn’t a straight line, but it is a system. And systems fail when you remove the parts that don’t look profitable
on a quarterly dashboard.

So, is the “myth of basic science” mythical? The myth is the simplistic story. The value is not mythical at all. It’s just inconveniently
distributed across time, institutions, and industrieswhich is exactly why it needs deliberate support.

Experiences from the real world: what this debate feels like on the ground (extra 500+ words)

To make this less abstract, here are experience-based vignettescomposites of patterns that researchers, engineers, and founders commonly describe
when they live inside the basic-to-applied ecosystem. These aren’t “and then I did X” stories. Think of them as the recurring scenes in the
long-running TV show called Innovation: Season 4,827.

1) The grant proposal that became a productby accident

A team writes a basic research proposal about a weird measurement problem: signals are too noisy, instruments drift, and the data is messy.
The goal isn’t a product; it’s a more reliable method. Two years later, the method is still not “market-ready,” but it works so well that other
labs start asking for the protocol. Then a company calls. Not because the original proposal promised a product, but because the company’s
manufacturing line has the same noise problemand the lab’s technique is the first thing anyone has tried that reliably untangles it.
The “innovation” wasn’t a planned invention; it was a tool that became useful when a different domain recognized its value.

2) The translational bottleneck nobody wants to pay for

Another common experience: everyone loves discovery, and everyone loves the finished therapy or device. The middle is where budgets go to die.
You’ll hear people describe the “valley of death” between promising science and practical deployment: validation studies, reproducibility work,
scale-up, safety testing, regulatory documentation, standards, and manufacturing constraints. None of this is glamorous. Much of it is essential.
This is where public-private partnerships and translational programs earn their keepby turning “interesting” into “trustworthy,” and then into
“usable at scale.” Without that middle layer, basic science can remain stranded as a set of papers admired mostly by people who already agree.

3) The industry R&D manager who quietly depends on academia

In industry, you’ll often find leaders who are publicly skeptical of academic research and privately addicted to it. Not because they want
“theories,” but because they want a pipeline of ideas, methods, and trained talent. When a company hires scientists with deep training, it isn’t
only hiring knowledgeit’s hiring problem-solving habits: experimental design, statistical caution, instrument literacy, and the ability to reason
through uncertainty. Those habits are forged in environments where not every question comes with a product roadmap.

4) The “tinkerer” who needs theory right when things get hard

Tinkering gets you faruntil it doesn’t. Many engineers describe a point where incremental trial-and-error hits a wall. Performance plateaus.
Failures become intermittent and weird. That’s when foundational understanding stops being a luxury and starts being a flashlight. Suddenly,
concepts that sounded academicmaterial defects, surface states, stochastic processes, immune pathways, signal-to-noise ratiosbecome the language
that lets teams diagnose what’s actually happening. The best innovation cultures don’t choose between tinkerers and theorists; they put them in
the same room and give them whiteboards.

5) The long-tail payoff that changes how people value “basic”

Finally, there’s a recurring emotional arc: early skepticism, later gratitude. People who spend years around research often describe a shift in
how they value uncertain work. At first, they want guarantees. After living through a crisisor watching a technology maturethey become less
obsessed with predicting the exact payoff and more focused on building capacity: instruments, networks, datasets, training, and shared knowledge.
They’ve seen enough “overnight successes” that took 20 years to stop asking basic science to act like a vending machine.

Put all these experiences together and you get a grounded conclusion: the real choice is not “basic science or innovation.” The real choice is
whether we want an innovation ecosystem that is resilientable to respond, adapt, and compoundor one that is optimized only for the next short
sprint and surprised every time reality shows up with a plot twist.