Editorial: the virtues of disruption

And you thought it was just superhero movies that kept telling the same story over and over again, offering a lot of flash and noise without moving the plot forward. It turns out no less venerable an institution than scientific research has run into the same problem.

In a systematic analysis of tens of millions of academic papers, published since the mid-20^th century, a team of American investigators concluded that fewer and fewer of these publications are in any way “disruptive”. This subjective conclusion followed an exhaustive count of how often any given paper refers to previously published work. If there are a lot of these citations, then the findings largely restate or consolidate existing knowledge. Although there may be a great deal of technical expertise and equipment brought to bear on a particular research question, the result merely replicates what is already known.

Far more of what happens in laboratories around the world falls under this heading of “consolidation” than most of us would like to imagine. The romantic view of science casts such work as probing the unknown, butting up against mystery, paving the way for fresh insights about nature — not checking the answer yet again. Nevertheless, decades and millions of publications later, this output has become less a parade of discovery than an ongoing confirmation that all is well.

As research has matured into a global enterprise, however, street-wise members of this elite community know a conservative approach to be the most practical. Whether you are a graduate student trying to match the whims of a supervisor, or a seasoned faculty member applying for a government grant, it is a huge advantage to be able to explain not only what you want to do, but assure your sponsor of what you will find. Covering known ground, in a clear, competent way, makes your proposal a far easier sell; venturing into unexplored territory, not so much.

Not surprisingly, therefore, more and more papers refer to what has gone before. What stands out, then, are the occasional efforts that generate entirely new evidence, referring almost exclusively to the work that is actually being described in the paper, rather than work conducted previously by others. Such publications touch upon those unknowns, mysteries, or fresh insights, and — according to the findings of this massive survey — they are in ever shorter supply.

Reflections on a genomic disappointment

When these findings appeared in Nature earlier this month, an accompanying article suggested that the causes of this trend remained elusive. But for University of Toronto Professor Aled Edwards, it already made perfect sense. He holds the Temerty Health Nexus Chair of Health Innovation and Technology in the university’s Temerty Faculty of Medicine. He also founded the Structural Genomics Consortium (SGC), a research body dedicated to research with no intellectual property attachments, with the aim of making scientific findings freely available for innovative applications in human health.

Edwards’ inspiration for the SGC was the complete sequencing of the human genome in 2003, which had been expected to usher in a new era of genetic-based approaches to medicine. Unfortunately, his expectations were dashed by the extremely limited way in which researchers tackled what should have been a new frontier. More specifically, two decades after some 20,000 human genes were revealed, a huge portion have been ignored. Instead, almost all published research has focused on narrow range of genetic proteins, which were already fairly well known.

Writing in the Canadian health care publication Healthy Debate, Edwards saw symptoms of a research ecosystem that supports productive science, but not innovative science. “Although scientists are ostensibly trained to be risk takers, empirically, they are not,” he said. “It may be that the ecosystem in which we work, from promotion, to peer recognition, securing grants and getting papers accepted in journals, makes taking risks difficult.”

In such a risk-averse setting, he noted, we should not expect science to serve as the driver of innovation, no matter how much science policy rhetoric insists. Scientists, it turns out, have become reluctant to innovate even within their research community, never mind seeking to influence the world beyond that community.

“These are troubling, but important issues because, despite massive amounts of research funding globally, we are making too little progress in the prevention, diagnosis and treatment of diseases,” Edwards concluded.

Where your flying car and jet pack wound up

Following the revelation about dwindling disruption, some observers flew into high dudgeon, insisting that this trend is short-changing society, which is why we have yet to see the kind of progress science fiction has eagerly promised for decades. We wanted flying cars and jet packs, but instead we got a dazzling array of globally accessible YouTube animations, showing what these gadgets might be like. Meanwhile, the commercial potential of these innovations remains undeveloped, and their impact on society is left only to the imagination.

Courtesy of YouTube, we could claim to be intellectually richer, having so many different ways to explore disruption. We could say much the same thing about a scientific establishment that has become very good at the process of science, but less clear about its purpose. Traditionally, that purpose has been to extract some order from the universe, but at its disruptive best, science provides a considerable amount of disorder, and discomfort.

For an outstanding example of what this looks like in practice, consider the James Webb Space Telescope(JWST). Located far beyond the orbit of the moon, this instrument is causing no end of disruption in the astronomical world. In addition to aesthetically pleasing images, the accompanying data is upending current models of how the universe evolved. Far from stand-pat stargazing, this is a view on things never before seen, just as we like to imagine all scientific endeavours.

At the same time, this entire undertaking was ridiculously expensive, time-consuming, and even its chosen name is still being hotly debated. There would surely have been a large contingent of more reasonable researchers, who would gladly have diverted all that time and money into other projects, covering much more familiar territory in science and not upsetting anyone.

The ultimate irony is that for all the public attention and acclaim dedicated to the JWST, it contributes little more to our world than those YouTube videos. It does not represent innovation of the kind that changes our lives in some fundamental way, however much we would like it to do so. For that to happen, risk, chaos, and controversy must not only be considered — they must be embraced.

The UK government has taken steps toward just such an embrace, with the introduction last year of its Advanced Research and Invention Agency (ARIA), which is targeted at “transformational” science and technology. ARIA was introduced as a new scientific engine the UK needs in the 21st century, “one that embraces the idea that truly great successes come from taking great leaps into the unknown.”

The United States, long home to this kind of risk-taking encouragement, last year introduced its Advanced Research Projects Agency for Health (ARPA-H), which similarly wields the notion of “transformational” in its description. Nor is there any hint that the expectations and goals include adding to the global supply of scientific papers. The language around ARPA-H bluntly specifies the creation of tools to tackle disease, solutions to manufacturing problems, and “reducing the likelihood that people will become patients”. Such language reflects a major investment that is supposed to affect people where they live, not scientists where they publish.

For the record, Canada has raised its own clarion call for such an initiative. Recommendation 6.6 of Canada’s Fundamental Science Review dealt with the need for “granting councils to encourage and better support high-risk research with the potential for high impact.”

That was in 2017. In 2023, we are looking forward to a federal budget, and some details about whether the Canada Growth Fund or the Canada Innovation and Investment Agency are prepared to deliver some of the disruption the country needs, and not just more work for scientific presses. Watch this space.

R$