“Who monitors whether foundational research fields are strengthening or weakening? Who connects research trajectory to strategic intent?… Someone needs to own disciplinary health.”
The new allocations provides UKRI with the funding to create a proper capability to analyse disciplinary health (and to develop the same level of understanding around technology areas and sectors) - something I know both of us have been calling for for a long time.
Thanks Ben, this is a fantastic article which captures many of the concerns I have about the new funding model, especially the muddled way in which the 'curiosity-driven' category is applied.
This is perhaps a peripheral comment to the main thrust of the article, with most of which I basically agree. You say: "The STFC disciplines are precisely the areas where UK research share is already in rapid retreat. Particle physics saw its UK share fall from approximately 4.7% in the 5-year period before 2021 to 2.3% the period to 2024, with momentum reversing from slightly positive historically to clearly negative in recent years. In the ‘galaxy formation and evolution’ research topic, UK share fell from around 15.5% to 6.5%. These trends predate the current cuts. The PPAN reductions will likely accelerate declines that were already under way." and illustrate this with accompanying graphs. These figures cannot be correct, as anyone working in these areas would immediately see. Particle physics and astronomy experimental activity is intrinsically international. Let me concentrate on particle physics, which I know best. Essentially everything in experimental particle physics is carried out in large international collaborations; publications in the periods you consider are dominated by the LHC, in which ~ every scientific country in the world is involved in every publication. Theory is different and publications have small numbers of authors but basically is proportional to numbers of academics and there has been no major change in this in the period under consideration. It is impossible to believe this extraordinary drop in your cumulative averages between 2017 and 2018. Confused by this I recalculated the numbers from Openalex by hand. My numbers for the selection Particle physics theoretical and experimental studies: UK/world in rolling averages over five-year periods starting in 2012 are : 8.0%, 6.7%, 7.8%, 7.8%, 7.9%, 8.2%, 8.1%, 8.5%, 8.5%, 8.1% - the final number ending in 2021-2025, one year further than your graphs. Other than a drop in 2013, which is not statistically significant, these numbers are essentially flat, precisely what, for the reasons I cite above, I would a priori expect. Although I haven't calculated the numbers for astrophysics, I am sure they would also be flat. There must be some error in your algorithm.
Thank you for this very helpful comment. Since publishing the post, I have discovered that the OpenAlex snapshot I used for this analysis contained a data quality error: a number of arXiv preprints spanning multiple years had been given an incorrect publication date of 2022. This explains the drop in PP as 2022 gets added to the rolling five-year windows. This appears has been fixed in the latest snapshot, which may be why you couldn’t replicate the result. I am currently in the process of writing a short addendum to this post with updated graphs. (I probably won’t include 2021-2025 as the data still looks incomplete for 2025 - it tends to take them up to a year to add records.) Thanks again!
can I encourage you to correct the article, as you indicate above, soon? The graphs and your comments on them from the OpenAlex data are totally misleading and I worry that people reading the article but not these comments will get completely the wrong idea about the health of particle physics and astronomy in the UK. In the current disastrous situation engineered by UKRI, this is a serious matter.
Yes indeed. Many thanks. I have to admit to being a bit puzzled about the absolute values you quote for particle physics share of publications - my calculations "by hand" were about half yours. Certainly it seems to me very unlikely that UK astronomers are involved in 30% of all published papers. Again, about 1/2 that would have seemed more reasonable to me. However, this doesn't really matter - the important thing is the trends, which your revised graphs now show as ~ flat, which agrees with my results and what I would expect.
Yes, I've been back and forth on this with myself for a while. I can broadly replicate your numbers if I compare the UK share with everything. But it created a huge drop in astronomy in 2025. When I dug into it, I saw that nearly 2/3 of papers in 2025 had no country affiliation at all (a data coverage issue). So I decided to compare the UK share against the subset of the corpus with robust geographic data in OpenAlex. This likely inflates the UK share somewhat, but not doing it almost certainly deflates the UK share greatly (hence the 2025 drop). Ultimately it's a methodological choice that one could argue either way - but, as you rightly say, the trend is the important thing.
On this specific point
“Who monitors whether foundational research fields are strengthening or weakening? Who connects research trajectory to strategic intent?… Someone needs to own disciplinary health.”
The new allocations provides UKRI with the funding to create a proper capability to analyse disciplinary health (and to develop the same level of understanding around technology areas and sectors) - something I know both of us have been calling for for a long time.
Thanks Ben, this is a fantastic article which captures many of the concerns I have about the new funding model, especially the muddled way in which the 'curiosity-driven' category is applied.
This is perhaps a peripheral comment to the main thrust of the article, with most of which I basically agree. You say: "The STFC disciplines are precisely the areas where UK research share is already in rapid retreat. Particle physics saw its UK share fall from approximately 4.7% in the 5-year period before 2021 to 2.3% the period to 2024, with momentum reversing from slightly positive historically to clearly negative in recent years. In the ‘galaxy formation and evolution’ research topic, UK share fell from around 15.5% to 6.5%. These trends predate the current cuts. The PPAN reductions will likely accelerate declines that were already under way." and illustrate this with accompanying graphs. These figures cannot be correct, as anyone working in these areas would immediately see. Particle physics and astronomy experimental activity is intrinsically international. Let me concentrate on particle physics, which I know best. Essentially everything in experimental particle physics is carried out in large international collaborations; publications in the periods you consider are dominated by the LHC, in which ~ every scientific country in the world is involved in every publication. Theory is different and publications have small numbers of authors but basically is proportional to numbers of academics and there has been no major change in this in the period under consideration. It is impossible to believe this extraordinary drop in your cumulative averages between 2017 and 2018. Confused by this I recalculated the numbers from Openalex by hand. My numbers for the selection Particle physics theoretical and experimental studies: UK/world in rolling averages over five-year periods starting in 2012 are : 8.0%, 6.7%, 7.8%, 7.8%, 7.9%, 8.2%, 8.1%, 8.5%, 8.5%, 8.1% - the final number ending in 2021-2025, one year further than your graphs. Other than a drop in 2013, which is not statistically significant, these numbers are essentially flat, precisely what, for the reasons I cite above, I would a priori expect. Although I haven't calculated the numbers for astrophysics, I am sure they would also be flat. There must be some error in your algorithm.
Thank you for this very helpful comment. Since publishing the post, I have discovered that the OpenAlex snapshot I used for this analysis contained a data quality error: a number of arXiv preprints spanning multiple years had been given an incorrect publication date of 2022. This explains the drop in PP as 2022 gets added to the rolling five-year windows. This appears has been fixed in the latest snapshot, which may be why you couldn’t replicate the result. I am currently in the process of writing a short addendum to this post with updated graphs. (I probably won’t include 2021-2025 as the data still looks incomplete for 2025 - it tends to take them up to a year to add records.) Thanks again!
Dear Ben,
can I encourage you to correct the article, as you indicate above, soon? The graphs and your comments on them from the OpenAlex data are totally misleading and I worry that people reading the article but not these comments will get completely the wrong idea about the health of particle physics and astronomy in the UK. In the current disastrous situation engineered by UKRI, this is a serious matter.
Thank you for the encouragement, Brian. I hope today's updates to the article address this.
Yes indeed. Many thanks. I have to admit to being a bit puzzled about the absolute values you quote for particle physics share of publications - my calculations "by hand" were about half yours. Certainly it seems to me very unlikely that UK astronomers are involved in 30% of all published papers. Again, about 1/2 that would have seemed more reasonable to me. However, this doesn't really matter - the important thing is the trends, which your revised graphs now show as ~ flat, which agrees with my results and what I would expect.
Yes, I've been back and forth on this with myself for a while. I can broadly replicate your numbers if I compare the UK share with everything. But it created a huge drop in astronomy in 2025. When I dug into it, I saw that nearly 2/3 of papers in 2025 had no country affiliation at all (a data coverage issue). So I decided to compare the UK share against the subset of the corpus with robust geographic data in OpenAlex. This likely inflates the UK share somewhat, but not doing it almost certainly deflates the UK share greatly (hence the 2025 drop). Ultimately it's a methodological choice that one could argue either way - but, as you rightly say, the trend is the important thing.