For those not paying attention, David Naylor is leading “Canada’s Fundamental Science Review Panel” which is looking at the state of fundamental science in Canada. Last week, I had a chance to participate at a roundtable of experts in Calgary, where the consultation was about Big and Expensive Science infrastructure. Being director of a big but relatively cheap science research infrastructure, I wondered first whether I really belonged to that club. Indeed, other participants included directors of Canada’s landmark mega-projects: Triumf, SNOLAB, Canadian light source, Compute Canada, the Thirty-Meter Telescope, etc. Revealing on Twitter where I was and with whom, solicited this tweet from a colleague: “That’s like inviting the big 5 banks to a meeting on interest rates & ATM fees!” In fairness it wasn’t that bad, though the consultation could have benefited from a few more independent observers of big Canadian Science. I considered myself one of those – one panelist confided that I was invited because I’m known to speak my mind. But, in spite of valiant efforts by Art MacDonald who was chairing, the format was not conducive to speaking minds and extensive analysis, hence, this post.
Before the consultation, we were given a well-prepared document which outlined the main challenges facing big science in the Canadian context. But to expect a coherent discussion by 20 participants in a 90 minutes span was not reasonable. The questions were a-propos:
- What is the definition of Big Science infrastructure in a Canadian context?
- How to prioritize, decide, advocate, and manage Big Science infrastructure?
- Who speaks for Canada on the international stage? How do we determine when it makes sense for Canada to build its own facility versus participating in an international facility?
The incoherence of the funding landscape for big and expensive science infrastructure in Canada is mind-boggling. Here we have to think beyond CFI funded projects to include lots of one-off projects as well as the CERCs and the latest CFREFs. I was shocked that one participant wanted our adulation for their CFREF. No way, not when it turned out so defective! Here was an initiative that was sold to the late Jim Flaherty in one guise, morphed in the hands of Ottawa’s bureaucrats, mishandled by university administrators, unfolded in one way under the Conservatives in Round one, then in another under the Liberals in Round two. If anything this was a textbook on how large scale initiatives should not be funded.
But back to the main question. How to define “big science infrastructure”? Implicit was an assumption that these were projects with “a rough minimum capital costs starting around $100 million.” Already, I was on a different track.
The Banff International Research Station (BIRS) that I direct, annually hosts –without counting CMO, its Mexican affiliate– more than 2200 mathematicians, statisticians, computer scientists, engineers, economists and other scientists. They come from more than 70 countries. BIRS success has already led to an offshoot in Oaxaca, Mexico. But the total BIRS budget is a mere $2.6 million per year, with about 25% from the federal government (NSERC) while the US National Science Foundation (NSF), the Alberta Government, and CONACyT cover the rest. Should BIRS be considered “big science infrastructure”? How do you compare this with a particle accelerator, a telescope, or a super computer? The reality is that “big science” by its very nature is difficult to compare side-by-side.
Besides the lack of an appropriate definition, the incoherence also stems from the many ways Canadian big science infrastructure is funded, owned, and operated. The modes of operations are as varied as their “owners” and as diverse as their funders: Direct government funds, NRC, CFI, Universities, provinces, etc.
I believe we should start by distinguishing between 2 types of big science infrastructure:
The first are the relatively small or medium-sized projects that are university based, and located on campuses, mainly to benefit a local scientific community. These should be peer reviewed and funded by –a reformed and improved- CFI. They are relatively straightforward and have a fixed funding formula (currently 40% Federal and 60% from universities, provinces, industry, etc). This formula also applies to their operating grants. The only problem is the irregularity of funding the CFI envelope which creates on-the-ground planning issues. Here, the Naylor panel should recommend a more orderly annual recurring funding envelope from the federal government.
The second, and more challenging, are the large, collaborative, multi-university projects. These have national and international scope with stakeholders crossing provincial or national boundaries. Examples include the Amundsen icebreaker, Canadian Light Source (CLS), SNOLAB, Compute Canada (CC), Ocean Networks Canada (ONC), Ocean Tracking Network (OTN), and the Vaccine and Infectious Disease Organization-International Vaccine Centre (VIDO-InterVac). Apparently they have all faced difficulties as they seek to secure adequate funding from a variety of sources.
There are a number of issues that need to be addressed:
- Prioritization and Advocacy: Projects of this significance should not be funded based on who has better access to Ottawa. Lobbying is not a functional route to funding good science.
- Cradle to Grave funding mechanism: The 40%-60% formula doesn’t make sense for projects whose beneficiaries are dispersed. Instead the Federal government must consider full-cycle costing.
- Competition or not: These projects are too costly and risky to rely on whether our best scientists manage to write a “fundable” proposal. A peer review system that is both interactive and iterative is required.
- The interlocutors: Who speaks for Canada’s role in international collaborations. Ditto for provincial government involvement.
The discussion for a clean mechanism for handling all this, led to some consensus that can be traced back to a 2006 recommendation by Arthur Carty, when he was the National Science Advisor to the Prime Minister. It called for the creation of a Major Science Initiatives Panel (MSIP) that would be charged with making recommendations regarding government support for Big Science infrastructure from cradle to grave. This panel of experts would be in charge of vetting initial proposals, making recommendations on capital funding and/or decommissioning after peer review.
I see this as a good start but there is still one outstanding problem: How can we “protect” such a panel from political interference so as to ensure a long-term perspective on Canadian Science that survives changes in government. After all, Arthur Carty himself was sidelined, and the position of “National Science Advisor to the Prime Minister” was cancelled by the Harper government. The current Liberal government has promised to restore the position, but the jury is still out on whether the country needs an advisor on Science to the PM, or a more independent “Chief Science Officer,” that reports to Parliament.
My suggestion is to create something more robust, specifically a new Council similar to the tri-council. This new “Council for Major Science Initiatives,” would have a base budget to cover operating costs for major science infrastructure. Administratively it would be quasi-independent so as to protect it from political winds and give it more permanence. It would rely on the most outstanding Canadian and international scientists for scientific decision making, and be chaired by the “Chief Science Officer.”
The new council would be empowered in two directions. First, it would evaluate existing infrastructure and award operating funds. Where necessary, it would recommend the wind-down of existing facilities and ensure this happens in an orderly fashion.
Second, it would be in charge of the prioritization, the selection, the advocacy, the peer review, and the negotiations with the various national and international partners. It would recommend new projects of highest priority to government either from new funds or recouped funds from decommissioned facilities.
This leaves one dangling string – the NRC. The National Research Council also operates several large science infrastructures that are focused on fundamental research (including telescopes). In this new construct, these pieces of the NRC would be reassigned to the new Council, creating a robust administrative structure and cohesion across all big science. This would align nicely with recommendations by the Jenkins Panel on how to handle the other pieces of the NRC, namely the business facing institutes, IRAP, and the regulatory mandate.
Done right, the proposed council would streamline the process of maintaining, decommissioning, and creating big science research infrastructure for Canada, in a fair, transparent, honest, efficient, coherent, and most of all, relevant for research and for Canada.