A message from our sponsors

The main national funding body for science and engineering in the UK is The Engineering and Physical Sciences Research Council, or EPSRC. Recently there have been some important changes announced via their website. (Whether there was any other warning about these changes I do not know — I confess to having rather taken EPSRC for granted.) Rather than giving my opinion about the changes, which will be obvious, I thought I’d let EPSRC speak for itself: I think that what they are doing should be widely publicized, and in this post I shall restrict myself as much as I can to quoting from their website, with a few words of explanation here and there to make the context clear. If you have time, I recommend visiting and exploring the website itself, and seeing what you can deduce about EPSRC’s attitude to mathematics.

A quick disclaimer: though I imagine that my implied opinions are widely shared, I am writing this in a personal capacity and without the endorsement of anybody else.

To give an idea of the kind of language to be found throughout the website, here is the most prominent paragraph from the main page:

Shaping Capability: EPSRC’s Research Portfolio 2011

EPSRC released a map of its full research portfolio alongside further information on how it will implement the Shaping Capability strand of its delivery plan for 2011 onwards.

What is the EPSRC? The answer can be found on its “About us” page:

EPSRC is the main UK government agency for funding research and training in engineering and the physical sciences, investing more than £850 million a year in a broad range of subjects – from mathematics to materials science, and from information technology to structural engineering.

For more information about what we do, our priorities and how we implement our strategy, please read through our Implementing the Delivery Plan section.

Now, if you look at that section you find that they have published a timeline. This seems a good way of finding out what “Implementing the Delivery Plan” means. The first item (or rather, the top left-hand item — the items are displayed two-dimensionally) is this, for June 2011.

Publish Shaping Capability Landscape by July, describing current and future shape of investments.

It’s starting to look as though we need to understand what “Shaping Capability” is all about. There isn’t a handy link to follow, but after a small search I find something like a definition on their Executive Summary page. This page is, by the way, a subpage of their Delivery plan page.

Shaping Capability: Ensuring we have the right people, with the right resources, in the right places to deliver the highest quality long-term research in areas where the UK leads internationally and where there is current or future national need.

Maybe it’s also worth looking the introduction page. It’s nice and short, so here’s the lot.

Our vision and commitment

Driven by visionary leaders, the research we sponsor will both be internationally excellent and deliver long-term impact for the health, prosperity and sustainability of the nation and the world.

Our strategic priorities

Our 2010 Strategic Plan, endorsed by leading members of our research and business communities, is driving and accelerating our change agenda as we take a more proactive role in shaping research and training to meet national need.

The EPSRC approach: from funder to sponsor

We will work closely with our university and business partners, and with other stakeholders and sectors, as we further align the engineering and science base for the good of the UK as a whole.

OK, a message is beginning to emerge pretty clearly. Stripped of the jargon, it’s this: the EPSRC plans to think harder about what research the country actually needs, and therefore what it should be supporting.

Since the EPSRC’s job is to decide how to spend the money that the government gives it, and since that will clearly involve evaluating research proposals before deciding which ones to support, it isn’t immediately obvious what has changed. What is the “change agenda” they mention? Let’s try to find out by clicking on Our Strategic Priorities.

That takes us to a page that helpfully gives us an overview by telling us what the three “high-level priorities” are. And they are

Delivering Impact

Shaping Capability

Developing Leaders

This is starting to feel like reading a maths textbook, because I’ve forgotten the definition of Shaping Capability. But at least I know I can look it up. What about Delivering Impact? There doesn’t seem to be a concise definition, so here are a couple of paragraphs about it, selected from several more (which can be found on the Strategic Priorities page linked to above).

Help researchers target impact – by actively emphasising priorities against national needs, stating expectations, requirements and responsibilities.

Encourage our key university, business and government partners to align their strategies to a national agenda and priorities, and to create spaces for researchers and users to work together as normal business within that strategic framework.

Develop enduring relationships between ourselves, business and the research community, by bringing companies together on a sector basis, agreeing important national research challenges and aligning the research we fund and the people we sponsor to meet those challenges.

So, broadly speaking, the idea is to identify certain national priorities and to channel research funding into research that directly advances those priorities. Or at least, that’s what I think the idea is.

I’d also like to quote a passage from the Developing Leaders section, since it is directly relevant to what actually provoked me to write this post.

In this Delivery Plan period we will develop the next generation of scientists and engineers, ensuring that early career researchers with the greatest potential are well-supported.

To understand this point better there are two things I want to know. First, how does EPSRC support those future research leaders that it identifies, and second, how does it identify them? And as a mathematician, I have a particular interest in how EPSRC identifies the future leaders in mathematical research. More generally, in the light of much of the rest of the EPSRC web page, I want to know where mathematics fits in to the EPSRC’s assessment of research priorities and national need.

Before I get on to that, here is a section from their page on Maintaining the flow of skilled researchers.

Our vision is to support the most talented and forward-thinking researchers, investing in the next generation of scientists and engineers. This Delivery Plan builds on our successful approach to training, instigated in 2008, which focused on the £300 million investment in Centres for Doctoral Training.

These centres support students in four-year cohorts in highly innovative, research-excellent environments where both depth and breadth are championed. Students trained in this way are much sought-after by business and academia for their quality, productivity, and employability. More broadly, we will develop the next generation of scientists and engineers, ensuring the highest potential early career researchers have access to the support they need.

Now 300 million pounds is a lot of money. Is that all going on supporting PhD students in science and engineering? And what is a Centre for Doctoral Training? How does it differ from, say, a university department? I ought to be able to answer that since we’ve got a mathematical one in Cambridge. But I think my answer would be less good than you will get by looking it up on Wikipedia. Just to add my pennyworth, the Doctoral Training Centre here in the Cambridge mathematics department now takes a large proportion of all the mathematics PhD students here (I’m afraid I don’t know the exact numbers), and focuses on analysis, and in particular applied analysis. What’s in it for our department? The short answer is money: a large amount of money from EPSRC. It’s not quite as simple as that though, since EPSRC doesn’t fund the Centre for Doctoral Training in its entirety. So the department had to divert some of its own resources into the CDT in order to get the external money. Here “resources” mainly means people and posts. Again I don’t know the details.

I’ve gone off at a slight tangent here but I want to draw attention to the main point: that the policy has been to identify a certain area, in this case applied analysis, and try to encourage lots of people to come together to work in that area.

Incidentally, something I don’t quite understand is how they know that “Students trained in this way are much sought-after by business and academia for their quality, productivity, and employability” when they are in “four-year cohorts” and their “successful approach to training” was “instigated in 2008”. Doubtless there is a rational explanation.

OK, back to the main question: how does the new “change agenda” affect mathematicians in the UK?

From a page on Global, economic and societal challenge themes:

We will set a balance between national capability and challenge themes of approximately 60:40.

We cannot afford to invest in everything, so we are focusing our resources on supporting four main themes: Manufacturing the Future, Energy (an RCUK cross-Council initiative), the Digital Economy (an RCUK cross-Council initiative), and Healthcare Technologies.

I don’t know, but I hope that mathematics counts as “national capability”, since it doesn’t seem to fit very well into any of the above themes. (Of course we have our contributions to make to the digital economy, but if you look at the page on that you see that they seem to have other things in mind.)

OK, what does the EPSRC think of maths? Let’s go to a page called Our portfolio. On that page, we find a diagram entitled Research Infrastructure, on which there is a circle with Mathematical Sciences written inside it. That looks promising, but before I click on it (which I can, it turns out), let me give another quote about the general plans. It comes from a page called Developing the shape of the portfolio.

We are now assessing research areas within the portfolio and taking decisions about the relative scale of our investment in these areas. These decisions will state whether we intend to ‘grow’, ‘maintain’ or ‘reduce’ a research area as a proportion of the total EPSRC portfolio, indicating the overall trend we would like to see.

Grow – this research area will be grown relative to other areas in the portfolio

Maintain – this research area will be maintained relative to other areas in the portfolio

Reduce – this research area will be reduced relative to other areas in the portfolio

Aha. We seem to be getting down to nuts and bolts here. It will be interesting to see which areas of mathematics are to be grown, which maintained and which reduced. OK, time to click on mathematical sciences, which brings us to a diagram of interrelationships between areas of mathematics.

The various areas are represented by blue circles of differing sizes, with the exception of Statistics and Applied Probability, which has a dark green circle. And a little key on the left tells us that dark green means “grow” and blue means “under review”. If you click on a circle, you can find out what proportion of mathematical funding goes to that area. For instance, I find that Logic and Combinatorics gets 3.4%, Mathematical Analysis 10.4%, Algebra, Geometry, Topology and Number Theory (quite an area that) 17.1%, and so on. Statistics and Applied Probability is currently on 14.1%.

OK, how does EPSRC support mathematicians, other than through its Doctoral Training Centres? One very important aspect of its support has always been fellowships. Some of these have been postdocs, helping to keep people in academia when they have finished their PhDs but do not yet have an academic job. Some have been to enable researchers to take time out from their teaching and concentrate just on research for a few years. All have been very valuable to mathematics in Britain. So let’s see what the website tells us about fellowships now.

From the page entitled Fellowships we get the following clues.

EPSRC Fellowships – a new approach

EPSRC’s Strategic Plan committed us to providing greater support to the world-leading individuals who are delivering the highest quality research to meet UK and global priorities.

Areas in which fellowships are available

EPSRC’s Capability and Challenge Themes have gone through an initial exercise of prioritising the areas in which fellowships will be offered. This will be an ongoing process and will be updated on a regular basis throughout the Delivery Plan period.

Clicking on Areas in which fellowships are available takes us to a very informative table: it does exactly what it promises and tells us in which areas fellowships are available. And we find that in mathematical sciences the only fellowships on offer are in Statistics and Applied Probability. This is quite a shock, since they used to be available throughout the mathematical sciences. It also appears to contradict the following statement, which can be found at the bottom of their page called Our portfolio.

Any changes in the portfolio will take place gradually over time. We will regularly review the shape of the portfolio and the scale of investment in different areas as it evolves.

Putting it together with the paragraph quoted earlier,

In this Delivery Plan period we will develop the next generation of scientists and engineers, ensuring that early career researchers with the greatest potential are well-supported.

it would appear that they have judged that the next generation of leaders in mathematics will all be statisticians and applied probabilists.

That’s basically the punchline of this post. But while writing it, I came to think that maybe my shock was premature. After all, they do say that they have gone through “an initial exercise” and that it is “an ongoing process” and that “priority areas will be updated on a regular basis”. So perhaps once they have finished reviewing the other areas of mathematics we will find that fellowships become available in those as well. If so, it would be nice if they could say so in clear unambiguous terms. Here’s another statement that may possibly back up that interpretation.

In the first instance, we expect the below areas to be valid for a year. Where priorities are removed we will give advance notice.

If the optimistic interpretation is correct, it is still a bit strange. Suppose you are a bright mathematician just finishing your PhD and you work in an area currently under review. It may be that that area will end up as one of the lucky dark green growth areas, but until EPSRC actually decides this, you cannot apply for a postdoctoral fellowship. And even if your area is earmarked for shrinking, should it be shrunk to zero? That too seems a bit odd. And suppose that the objective of EPSRC is to encourage people into certain areas. Wouldn’t it be better to let people know this before they decided on their PhD subject so as to attract the best people you can?

The rationale for what EPSRC is doing seems to be that there are economies of scale: if you herd lots of people into one area, then they can benefit from each other and operate more efficiently. I don’t know, but I’m prepared to believe that this may be true for several areas of science and engineering. But for mathematics it is much less clear. (I do of course believe strongly that mathematical cooperation can lead to substantial gains in efficiency. But there’s something called the internet that means that we don’t have to be in the same place to cooperate.)

Anyhow, watch this space. The timeline tells us that “new investment decisions in line with Landscape begin” in October. Perhaps then we’ll find out whether Logic and Combinatorics is going to get an increase on its current 3.6%, and even a fellowship or two.

Enough for now. I’d better get back to implementing my enhanced evidence-based strategic thematic impact capability landscape delivery outcomes plan agenda work.

[Addendum: I’ve recently come across this interesting and relevant article in the Times Higher Education Supplement.]