Evolving the Scientific Method

By Kevin Kelly Evolving the Scientific Method Technology is changing the way we conduct science. Images: Wikipedia (from top): Library of Alexandria; Francis Bacon; Robert Boyle; Karl Popper (courtesy of LSE library); Placebo (courtesy of Elaine and Arthur Shapiro); Zuse Z3 computer (courtesy of Deutschen Museum in München) Science is our most potent invention because it has given us a method to keep reinventing it. All our collective knowledge and experti

By | December 1, 2010

Evolving the Scientific Method

Technology is changing the way we conduct science.

Images: Wikipedia (from top): Library of Alexandria; Francis Bacon; Robert Boyle; Karl Popper (courtesy of LSE library); Placebo (courtesy of Elaine and Arthur Shapiro); Zuse Z3 computer (courtesy of Deutschen Museum in München)

Science is our most potent invention because it has given us a method to keep reinventing it. All our collective knowledge and expertise (that’s science) began with relatively simple arrangements of information. The simplest organization was the invention of the fact. Facts became codified not by science, but by the European legal system in the 1500s. In court lawyers had to establish agreed-upon observations as evidence that could not shift later. Science adopted this useful innovation. Over time, the novel ways in which knowledge could be ordered increased. This complex apparatus for determining the factual correctness of information, and relating it to old knowledge, is what we call science.

The scientific method is not one uniform “method.” It is a collection of scores of techniques and processes that has evolved over centuries (and continues to evolve). Each method is one small step that incrementally increases the unity of knowledge in society. A few of the seminal inventions that furthered the development of the scientific method are shown in the accompanying illustration.

Together these landmark innovations created the modern practice of science. (I am ignoring various alternative claims of priority because the exact dates don’t matter for this illustration.) A typical scientific discovery today will rely on facts (data) and a falsifiable hypothesis; be tested in repeatable, controlled experiments, perhaps with placebos and double-blind controls; and be reported in a peer-reviewed journal and indexed in a library of related reports.

The scientific method, like science itself, is accumulated structure. New scientific instruments and tools add new ways to gather and organize information. Recent methods build upon earlier techniques. Technological advances keep adding connections among facts and more complex relations among ideas. As this short timeline makes clear, many of the key innovations of what we now think of as “the” scientific method are relatively recent. The classic double-blind experiment, for instance, in which neither the subject nor the tester is aware of what treatment is being given, was not named or widely used until the 1950s. The placebo was not used until the 1930s. It is hard to imagine science today without these methods.

The scientific method,
like science itself,
is accumulated structure.

This relative newness makes one wonder what “essential” method in science will be invented next. The nature of science is still in flux; the technium is rapidly discovering new ways to know.

What is the technium? As described on my Web site, TheTechnium.org, it designates the greater sphere of technology—one that goes beyond hardware to include culture, law, social institutions, and intellectual creations of all types. In short, it’s anything that springs from the human mind. It includes hard technology, but much else of human creation as well. I see this extended face of technology as a whole system with its own dynamics.

Given the acceleration of knowledge, the explosion of information, and the rate of progress, the nature of the scientific process is on a course to change more in the next 50 years than it has in the last 400 years. What might be some new processes in the future? One could imagine that the inclusion of negative results will become routine. And that computer proofs will become more reliable, common, and trusted. And that wiki journals will contain reports that are not fixed, but are continually modified and edited in real time—all could become part and parcel of the scientific method.

At the core of science’s self-modification is technology. New tools enable new ways of discovery, different ways of structuring information. We call that organization knowledge. With technological innovations the structure of our knowledge evolves. The achievement of science is to discover new things; the evolution of science is to organize the discoveries in new ways. Even the organization of our tools themselves is a type of knowledge. Right now, with the advance of communication technology and computers, we have entered a new way of knowing. The thrust of the technium’s trajectory is to further organize the avalanche of information and tools we are generating and to increase the structure of the made world.

Kevin Kelly is Senior Maverick at Wired magazine, which he cofounded in 1993. He is also editor and publisher of the Cool Tools website. Formerly, he was publisher and editor of the Whole Earth Review, a journal of unorthodox technical news; he cofounded the ongoing Hackers’ Conference; and was involved with the launch of the WELL, a pioneering online service started in 1985. He is author of the best-selling New Rules for the New Economy and the classic book on decentralized emergent systems, Out of Control.

This essay is adapted from Kelly’s latest book, What Technology Wants (Viking/Penguin), published in October 2010. You may read the first chapter at http://www.kk.org/books/what-technology-wants.php


Avatar of: Paul Bernstein

Paul Bernstein

Posts: 1

December 8, 2010

Readers of this article might find interesting the recent innovation of adding three more layers of blinding on top of the now-traditional two layers (using a total of five layers of blinding to increase rigor in the data-gathering) in certain experiments, as described at www.windbridge.org/papers/JP71_2007-2008_Beischel_Methods.pdf on pages 11 ["47"] through 13 ["49"] of that document.
Avatar of: Jerry Jones

Jerry Jones

Posts: 12

December 8, 2010

Wikipedia was invented in 1403?
Avatar of: PAUL STEIN


Posts: 61

December 9, 2010

This is the way it has been and always will be. As students, how often have our senior investigators told us, "When I was in graduate school, we never had ..."? As principal investigators ourselves, how many times have we been stifled from doing things because the proper tools or technology haven't been invented yet? And, for those in the autumn of their scientific careers, how often have we said, "If we had then what we have now, ...?" Every practicing scientist should exceedingly thank those developers of our everyday tools.
Avatar of: Mark Riggle

Mark Riggle

Posts: 3

December 10, 2010

Karl Popper's view was that a THEORY had to be falsifiable. For him (and philosophy of Science in general), here is a huge difference between a theory and a hypothesis which often is blurred by scientists themselves. A theory is a body of connected ideas and that body generates various testable hypotheses. The notion is to use the testable hypotheses for supporting one of several competing theories. That is complicated when the theory's body actually determines the way in which experimental evidence is interpreted and even conceived. When a widely accepted theory body accumulates too many errors and is replaced by a new theory that does not suffer the errors, that replacement was called a paradigm shift by Kuhn. All sciences have had many paradigm shifts, but they just have been edited out of our history.\n\nThis reminds me of a famous Crick quote: "Any theory that explains all the data is likely to be wrong because some of the data will be wrong."\n\n\n
Avatar of: David Hardman

David Hardman

Posts: 4

December 13, 2010

Kevin Kelly is pointing out the obvious. They are not his ideas. Wired is not a peer-reviewed journal but a magazine. Modern science is in some ways like old time religion. Often there is no "falsifiable hypothesis", for example DNA methylation. It is believed that it is important, yet has not so far been specifically modified to test the hypothesis.\n\n"At the core of science?s self-modification is technology" is a statement. What about testable hypotheses? The experiment? When was the last time Kevin Kelly did an experiment?\n\nI think he should have left his thoughts on the back of the matchbox where he found them. Sorry, that's called an opinion.\n\n

Popular Now

  1. Major German Universities Cancel Elsevier Contracts
  2. Running on Empty
    Features Running on Empty

    Regularly taking breaks from eating—for hours or days—can trigger changes both expected, such as in metabolic dynamics and inflammation, and surprising, as in immune system function and cancer progression.

  3. Most of Human Genome Nonfunctional: Study
  4. Identifying Predatory Publishers