Does Computer Science Need Computers? | Quanta Magazine

The famous computer scientist Edsger Dijkstra once said that using computers in research is like forcing medical scientists to suffer the diseases they study. He also called the programming language Fortran “the infantile disorder” and said that teaching COBOL should be a criminal offence. Dijkstra’s harsh words were meant to show that computer science is about deep ideas, not just machines.

The author of the article liked Dijkstra’s comparison. It made the field feel timeless and important. But the author also wondered if the comparison was right. Is computer science really about computers, or is it about something else?

The question has been debated for a long time. In the 1930s, mathematicians began to think about the limits of calculation. In the 1940s, engineers built the first general‑purpose electronic computers. By the 1950s and 1960s, people started to call the new field “computer science.” Soon, scholars argued about what the field should study.

In 1967, three computer scientists—Allen Newell, Alan Perlis, and Herbert Simon—wrote a letter to the journal Science. They said, “Wherever there are phenomena, there can be a science to describe and explain those phenomena. There are computers. Ergo, computer science is the study of computers.” They were replying to critics who said that a science must study natural things. They pointed to chemistry, which studies artificial substances, as an example of a science that looks at man‑made things.

Simon’s 1969 book The Sciences of the Artificial made the idea stronger. He argued that computer science is a science of intentionally designed systems. The field is special because it studies things that humans build.

In 1974, Donald Knuth offered a different view. He said computer science is the study of algorithms—step‑by‑step procedures that solve problems. Algorithms can be written in many languages, just like ideas can be expressed in English, Mandarin, or Arabic. Humans use algorithms too, for tasks like sorting a list of names. From this view, the math behind computation is most important. Computers are useful because they let us solve problems that would take us too long to do by hand.

William Rapaport, a professor of computer science and philosophy, says the field has two parents: math and engineering. He thinks computer science is a mix of the two. He also says the field is about two big questions. The first is, “What can be computed?” The second is, “How do you compute it?”

These questions help explain why some people think computers are essential and why others do not. The author is most interested in the theoretical side, where people rarely touch a real computer. The question is: do computers play a key role in that theory?

To answer the first question, researchers need a model of computation. In the 1930s, several models were proposed. In 1937, mathematician Alan Turing described a machine that could read and write symbols on an infinite tape according to simple rules. Turing called it a “Turing machine.” Later, other researchers proved that the Turing machine was equivalent to the other models. This gave the field a single, universal theory of computation.

Turing’s work was not about computers. He did not have a general‑purpose computer, and he was not trying to design one. He was trying to solve a deep problem in mathematics. He imagined the machine as a way to model the mental work of a human doing calculations.

The theory of computation is useful for many things that are not computers. Researchers use it to model natural processes. For example, they study how unpredictable behavior can arise in physical systems, how evolution works, and how to solve puzzles in quantum gravity. The field that Simon called a “science of the artificial” now helps us understand the natural world.

Tom Gur, a theoretical computer scientist at Cambridge, says, “You can view the other sciences through computation. It’s this underlying logical pattern that shows up everywhere.”

The second question is about algorithms. In the late 1960s and early 1970s, researchers built a framework to measure how long algorithms take to solve problems. They did this at an abstract level, ignoring the details of hardware.

They found that some problems have clever algorithms that solve them quickly. Other problems only have slow algorithms. The study of why some problems are hard and others are easy is called computational complexity theory. It is the part of theoretical computer science that looks at the inherent difficulty of problems. It also underlies modern encryption.

Christopher Moore, a theoretical computer scientist, says, “Mathematical problems have a fundamental structure that makes them easier or harder to solve. It’s not about how fast your computer is, and it’s not about how clever you are.”

Valentine Kabanets, a complexity theorist, said that discovering this hidden structure feels like “discovering the laws of the universe.” He told the author about the most famous open problem in complexity theory.

Later, complexity theory led to new ideas about proof. Gur says that by rethinking proof as an interactive process, researchers found ways to prove a statement is true without revealing why it is true. They also found ways to check a proof by looking at only a few tiny parts.

Scott Aaronson, a theoretical computer scientist, says that many fundamental questions could have been asked hundreds of years ago. “It’s just that no one thought to ask them,” he says.

The 19th‑century inventor Charles Babbage imagined a general‑purpose calculating machine called the Analytical Engine. In his autobiography, he wrote that the machine would need a new theory of algorithms. He asked, “By what course of calculation can these results be arrived at by the machine in the shortest time?” Babbage never finished the machine, and it is unclear how he would have solved the problem. He may have thought that the machine’s design would make some methods faster than others. He did not foresee the rich mathematical structure that complexity theorists discovered.

Matti Tedre, a computer scientist, says that Dijkstra’s comparison to astronomy is almost right. “Dijkstra is absolutely right; it’s just that he’s wrong about the importance of telescopes to astronomy,” Tedre says. “We wouldn’t know a thing about the universe if we didn’t have telescopes.”

The history of computer science shows a close link between deep theory and practical machines. Aaronson points to the second law of thermodynamics, which says that disorder, or entropy, tends to increase over time. He says, “It’s maybe the most fundamental thing that you can say about the evolution of the entire universe. And yet it’s not something that anyone thought of until they were building steam engines.”

Ryan Williams, a complexity theorist at MIT, says, “Sufficiently interesting problems in practice generate great theoretical questions.”

These ideas show that computer science is not just about computers. It is about the ideas that computers help us explore, and the questions that arise when we build better machines.

The field continues to grow. New areas like artificial intelligence, quantum computing, and data science build on the same ideas. These areas show that the core questions of what can be computed and how to compute it keep evolving.

For students, learning about algorithms and models of computation gives a strong foundation. It helps them think about problems in a clear, logical way, even if they never build a computer.

In the end, computer science is a blend of math, engineering, and imagination. It asks big questions about logic, information, and the limits of what we can do. Computers are tools that help us explore those questions.

So, while computers are essential for many tasks, the heart of computer science lies in ideas that can exist without a machine.