Not Enough Disruption?

For the past few weeks, a paper in Nature has provoked vapors among those who want more scientific disruption. Bill Broad summarized it in the New York Times.

The paper defined “disruption” by a pattern of citations in the scientific literature. Horrifyingly to some, those disruptions have decreased since 1950.

There are layers of assumptions. First, that a pattern of citations in scientific papers is a measure of disruption. Next, that disruption is good and necessary for science. The overall goal of this disruption is not made explicit. To better our lives? In what way? For an abstract ideal of progress? For personal glory?

Four examples are given in Johnson’s thread: the DNA helix, relativity, quantum mechanics, and space flight. None were developed in a single paper. All emerged after long histories of related work. The first two are associated with single historic papers. The last two aren’t.

DNA helix: Darwin, late 19th century genetics, cancer and infectious disease studies, the development of x-ray diffraction for molecular structure. Crick and Watson’s paper of 1954 laid out the correct structure, but Linus Pauling guessed incorrectly.

Relativity: Planck, Stefan, Boltzmann, black body radiation, astronomical observations.

Quantum Mechanics: Some of the same things that went into relativity. Discovery of x-rays and radioactivity. No one single paper, rather a group of scientists in the 1920s who had major disagreements over how to express and interpret.

Space Flight: Imagined for millennia. Chinese gunpowder. Tsiolkovsky, Goddard, Oberg. V-2 rockets (now scuds), ICBMs. Incremental building on empirical findings.

None of the four bore fruit immediately. Years and decades were necessary for their practical use, as well as intellectual transformation. All have bad as well as good sides: nuclear weapons and writing rich men’s names across the night sky, for example.

What disruptions, then, are we missing?

Those four examples all 1) have long histories of development leading up to them, 2) relied on the contemporaneous work of multiple people, and 3) did not bear fruit for years and decades. What current areas of research and development show similar patterns?

Broad suggests three possible breakthroughs that I would argue follow those patterns: CRISPR gene-editing technology, mRNA vaccines, and gravitational waves, the last two of which he pooh-poohs as incremental.

I’ll suggest three more: What is called artificial intelligence but is in fact machine learning; computing and the internet; and dark matter and energy.

The Nature study looks backward to define disruptive papers, whereas we are in the middle of the development of the next disruptions. We don’t know which ones will survive or change our thinking, although some changes are beginning to be apparent. Additionally, much of the discussion lumps scientific breakthroughs (relativity) with technological breakthroughs (space travel).

A more basic assumption in the Nature paper is that a single paper, usually by a single (white male cishet) genius can, by itself and out of nowhere, change the world. Nothing in science comes out of nowhere. Few scientific papers are the product of a single person’s mind.

For a very long time, everyone but white cishet male scientists and engineers was excluded or treated very badly if they somehow managed to get past the gatekeepers (Alan Turing, Rosalind Franklin). That is changing, and perhaps the way we think about how science develops needs to change too.

The word “disruption” starts a rapid climb in use around 1960, but “disruptive technology” comes later, in the 1990s. That’s when white male cishet computer guys used that term about the good they felt they were bringing to the world.

If we are looking for a single paper that can make an enormous difference in how we look at the world, perhaps we should consider Carol Cohn’s 1985 paper deconstructing the male gaze in defense technology. Defense technology has always spurred science.

Cohn wrote her observations of the use of words as part of a broader gendered orientation to defense technologies (“patting the bomb”). Those observations have been partially, ad hoc, turned into a program of analysis that uses gender and other characteristics as ways to gain insight into human activity, including technical progress.

Surely this is a disruptive paper that opens the way to innovative multidimensional analysis of human activity that could result in new ideas and creative uses of old. But Reviewer 2 often shunts aside any suggestion of such a program: Cohn did that already. Once and done.

That kind of analysis is what I’m doing in this piece. We need more of it.

Photo from UC Berkeley Physics Department. All those young men looking forward to their own breakthroughs.

Cross-posted to Lawyers, Guns & Money

4 comments

  1. mlerdau · 2 Days Ago

    You make very important points. Thanks. I do, however, wish you had taken more space to unpack and discuss your ideas. I hope a longer version of this post is in the works.

    Like

    • Cheryl Rofer · 2 Days Ago

      I will be writing about my ideas all through the coming year.

      Like

  2. Jim Lund · 2 Days Ago

    Cheryl, the paper’s results are even less interesting. The ‘slowing’ is per paper, the number of breakthroughs is constant!

    “Despite substantial increases in the numbers of papers and patents published each year, there is little change in the number of highly disruptive papers and patents, as evidenced by the relatively flat red, green and orange lines. This pattern helps to account for simultaneous observations of both aggregate evidence of slowing innovative activity and seemingly major breakthroughs in many fields of science and technology.”

    Like

    • Cheryl Rofer · 2 Days Ago

      Yes, I think that is what the paper says, although it is a little confusing. Most commentators look at the declines from earlier periods, though, to say that highly disruptive papers have declined. I am dubious enough about many aspects of the paper that I wouldn’t use it to make any strong claims, except that it contains many assumptions that the authors do not acknowledge.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s