Analysis Science

What arguments against the ‘next LHC’ say about funding Big Physics

A few days ago, a physicist (and PhD holder) named Thomas Hartsfield published a strange article in Big Think about why building a $100-billion particle physics machine like the Large Hadron Collider (LHC) is a bad idea. The article was so replete with errors things that even I – a not-physicist and not-a-PhD-holder – cringed reading them. I also wanted to blog about the piece but theoretical physicist Matthew Strassler beat me to it, with a straightforward post about the many ways in which Hartsfield’s article was just plain wrong, especially coming from a physicist. But I also think there were some things that Strassler either overlooked or left unsaid and which to my mind bear fleshing out – particularly points that have to do with the political economy of building research machines like the LHC. I also visit in the end the thing that really made me want to write this post, in response to a seemingly throwaway line in Strassler’s post. First, the problems that Hartsfield’s piece throws up and which deserve more attention:

1. One of Hartsfield’s bigger points in his article is that instead of spending $100 billion on one big physics project, we could spend it on 100,000 smaller projects. I agree with this view, sensu lato, that we need to involve more stakeholders than only physicists when contemplating the need for the next big accelerator or collider. However, in making the argument that the money can be redistributed, Hartsfield presumes that a) if a big publicly funded physics project is cancelled, the allocated money that the government doesn’t spend as a result will subsequently be diverted to other physics prohects, and b) this is all the money that we have to work with. Strassler provided the most famous example of the fallacy pertinent to (a): the Superconducting Super Collider in the US, whose eventually cancellation ‘freed’ an allocation of $4.4 billion, but the US government didn’t redirect this money back into other physics research grants. (b), on the other hand, is a more pernicious problem: a government allocating $100 billion for one project does not implicitly mean that it can’t spare $10 million for a different project, or projects. Realpolitik is important here. Politicians may contend that after having approved $100 billion for one project, it may not be politically favourable for them to return to Congress or Parliament or wherever with another proposal for $10 million. But on the flip side, both mega-projects and many physics research items are couched in arguments and aspirations to improve bilateral or multilateral ties (without vomiting on other prime ministers), ease geopolitical tensions, score or maintain research leadership, increase research output, generate opportunities for long-term technological spin-offs, spur local industries, etc. Put another way, a Big Science project is not just a science project; depending on the country, it could well be a national undertaking along the lines of the Apollo 11 mission. These arguments matter for political consensus – and axiomatically the research projects that are able to present these incentives are significantly different from those that aren’t, which in turn can help fund both Big Science and ‘Small Science’ projects at the same time. The possibility exists. For example, the Indian government has funded Gaganyaan separately from ISRO’s other activities. $100 billion isn’t all the money that’s available, and we should stop settling for such big numbers when they are presented to us.

2. These days, big machines like the one Hartsfield has erected as a “straw man” – to use Strassler words – aren’t built by individual countries. They are the product of an international collaboration, typically with dozens of governments, hundreds of universities and thousands of researchers participating. The funds allocated are also spent over many years, even decades. In this scenario, when a $100-billion particle collider is cancelled, no one entity in the whole world suddenly has that much money to give away at any given moment. Furthermore, in big collaborations, countries don’t just give money; often they add value by manufacturing various components, leasing existing facilities, sharing both human and material resources, providing loans, etc. The value of each of these contracts is added to the total value of the project. For example, India has been helping the LHC by manufacturing and supplying components related to the machine’s magnetic and cryogenic facilities. Let’s say India’s Departments of Science and Technology and of Atomic Energy had inked contracts with CERN, which hosts and maintains the LHC, worth $10 million to make and transport these components, but then the LHC had been called off just before its construction was to begin. Does this mean India would have had $10 million to give away to other science projects? Not at all! In fact, manufacturers within the country would have been bummed about losing the contracts.

3. Hartsfield doesn’t seem to acknowledge incremental results, results that improve the precision of prior measurements and results that narrow the range in which we can find a particle. Instead, he counts only singularly positive, and sensational, results – of which the LHC has had only one: the discovery of the Higgs boson in 2012. Take all of them together and the LHC will suddenly seem more productive. Simply put, precision-improving results are important because even a minute difference between the theoretically predicted value and the observed value could be a significant discovery that opens the door to ‘new physics’. We recently saw this with the mass of a subatomic particle called the W boson. Based on the data collected by a detector mounted on the Tevatron particle accelerator in Illinois, physicists found that the mass of the W boson differed from the predicted value by around 0.12%. This was sufficient to set off a tsunami of excitement and speculation in the particle physics community. (Hartsfield also overlooked an important fact and which Strassler caught: that the LHC collects a lot more data than physicists can process in a single year, which means that when the LHC winds down, physicists will still have many years of work left before they are done with the LHC altogether. This is evidently still happening with the Tevatron, which was shut down in 2011, so Hartsfield missing it is quite weird. Another thing that happened to Tevatron and is still happening with the LHC is that these machines are upgraded over time to produce better results.) Similarly, results that exclude the energy ranges in which a particle can be found are important because they tell us what kind of instruments we should build in future to detect the same particle. We obviously won’t need instruments that sweep the same energy range (nor will we have a guarantee that the particle will be found outside the excluded energy range – that’s a separate problem). There is another point to be made but which may not apply to CERN as much as to Big Science projects in other countries: one country’s research community building and operating a very large research facility signals to other countries that the researchers know what they’re doing and that they might be more deserving of future investments than other candidates with similar proposals. This is one of the things that India lost with the scuttling of the India-based Neutrino Observatory (the loss itself was deserved, to be sure).

Finally, the statement in Strassler’s post that piqued me the most:

My impression, from his writing and from what I can find online, is that most of what he knows about particle physics comes from reading people like Ethan Siegel and Sabine Hossenfelder. I think Dr. Hartsfield would have done better to leave the argument to them.

Thomas Hartsfield has clearly done a shoddy job in his article in the course of arguing against a Big Physics machine like LHC in the future, but his screwing up doesn’t mean discussions on the need for the next big collider should be left to physicists. I admit that Strassler’s point here was probably limited to the people whose articles and videos were apparently Hartsfield’s primary sources of information – but it also seemed to imply that instead of helping those who get things wrong do better next time, it’s okay to ask them to not try again and instead leave the communication efforts to their primary sources. That’s Ethan Siegel and Sabine Hossenfelder in this case – both prolific communicators – but in many instances, bad articles are written by writers who bothered to try while their sources weren’t doing more or better to communicate to the people at large. This is also why it bears repeating that when it comes to determining the need for a Big Physics project of the likes of the LHC, physics is decidedly one non-majority part of it and that – importantly – science communicators also have an equally vital role to play. Let me quote here from an article by physicist Nirmalya Kajuri, published in The Wire Science in February 2019:

… the few who communicate science can have a lopsided influence on the public perception of an entire field – even if they’re not from that field. The distinction between a particle physicist and, say, a condensed-matter physicist is not as meaningful to most people reading the New York Times or any other mainstream publication as it is to physicists. There’s no reason among readers to exclude [one physicist] as an expert.

However, very few physicists engage in science communication. The extreme ‘publish or perish’ culture that prevails in sciences means that spending time in any activity other than research carries a large risk. In some places, in fact, junior scientists spending time popularising science are frowned upon because they’re seen to be spending time on something unproductive.

All physicists agree that we can’t keep building colliders ad infinitum. They differ on when to quit. Now would be a good time, according to Hossenfelder. Most particle physicists don’t think so. But how will we know when we’ve reached that point? What are the objective parameters here? These are complex questions, and the final call will be made by our ultimate sponsors: the people.

So it’s a good thing that this debate is playing out before the public eye. In the days to come, physicists and non-physicists must continue this dialogue and find mutually agreeable answers. Extensive, honest science communication will be key.

So more physicists should join in the fray, as should science journalists, writers, bloggers and communicators in general. Just that they should also do better than Thomas Hartsfield to get the details right.

Analysis Science

On resource constraints and merit

In the face of complaints about how so few women have been awarded this year’s Swarnajayanti Fellowships in India, some scientists pushed back asking which of the male laureates who had been selected should have been left out instead.

This is a version of the merit argument commonly applied to demands for reservation and quota in higher education – and it’s also a form of an argument that often raises its head in seemingly resource-constrained environments.

India is often referred to as a country with ‘finite’ resources, often when people are discussing how best to put these resources to use. There are even romantic ideals associated with working in such environments, such as doing more with less – as ISRO has been for many decades – and the popular concept of jugaad.

But while fixing one variable while altering the other would make any problem more solvable, it’s almost always the resource variable that is presumed to be fixed in India. For example, a common refrain is that ISRO’s allocation is nowhere near that of NASA, so ISRO must figure how best to use its limited funds – and can’t afford luxuries like a full-fledged outreach team.

There are two problems in the context of resource availability here: 1. an outreach team proper is implied to be the product of a much higher allocation than has been made, i.e. comparable to that of NASA, and 2. incremental increases in allocation are precluded. Neither of these is right, of course: ISRO doesn’t have to wait for NASA’s volume of resources in order to set up an outreach team.

The deeper issue here is not that ISRO doesn’t have the requisite funds but that it doesn’t feel a better outreach unit is necessary. Here, it pays to acknowledge that ISRO has received not inconsiderable allocations over the years, as well as has enjoyed bipartisan support and (relative) freedom from bureaucratic interference, so it cops much of the blame as well. But in the rest of India, the situation is flipped: many institutions, and their members, have fewer resources than they have ideas and that affects research in a way of its own.

For example, in the context of grants and fellowships, there’s the obvious illusory ‘prestige constraint’ at the international level – whereby award-winners and self-proclaimed hotshots wield power by presuming prestige to be tied to a few accomplishments, such as winning a Nobel Prize, publishing papers in The Lancet and Nature or maintaining an h-index of 150. These journals and award-giving committees in turn boast of their selectiveness and elitism. (Note: don’t underestimate the influence of these journals.)

Then there’s the financial constraint for Big Science projects. Some of them may be necessary to keep, say, enthusiastic particle physicists from being carried away. But more broadly, a gross mismatch between the availability of resources and the scale of expectations may ultimately be detrimental to science itself.

These markers of prestige and power are all essentially instruments of control – and there is no reason this equation should be different in India. Funding for science in India is only resource-constrained to the extent to which the government, which is the principal funder, deems it to be.

The Indian government’s revised expenditure on ‘scientific departments’ in 2019-2020 was Rs 27,694 crore. The corresponding figure for defence was Rs 3,16,296 crore. If Rs 1,000 crore were moved from the latter to the former, the defence spend would have dropped only by 0.3% but the science spend would have increased by 3.6%. Why, if the money spent on the Statue of Unity had instead been diverted to R&D, the hike would have nearly tripled.

Effectively, the argument that ‘India’s resources are limited’ is tenable only when resources are constrained on all fronts, or specific fronts as determined by circumstances – and not when it seems to be gaslighting an entire sector. The determination of these circumstances in turn should be completely transparent; keeping them opaque will simply create more ground for arbitrary decisions.

Of course, in a pragmatic sense, it’s best to use one’s resources wisely – but this position can’t be generalised to the point where optimising for what’s available becomes morally superior to demanding more (even as we must maintain the moral justification of being allowed to ask how much money is being given to whom). That is, constantly making the system work more efficiently is a sensible aspiration, but it shouldn’t come – as it often does at the moment, perhaps most prominently in the case of CSIR – at the cost of more resources. If people are discontented because they don’t have enough, their ire should be directed at the total allocation itself more than how a part of it is being apportioned.

In a different context, a physicist had pointed out a few years ago that when the US government finally scrapped the proposed Superconducting Supercollider in the early 1990s, the freed-up funds weren’t directed back into other areas of science, as scientists thought they would be. (I couldn’t find the link to this comment nor recall the originator – but I think it was either Sabine Hossenfelder or Sean Carroll; I’ll update this post when I do.) I suspect that if the group of people that had argued thus had known this would happen, it might have argued differently.

I don’t know if a similar story has played out in India; I certainly don’t know if any Big Science projects have been commissioned and then scrapped. In fact, the opposite has happened more often: whereby projects have done more with less by repurposing an existing resource (examples herehere and here). (Having to fight so hard to realise such mega-projects in India could be motivating those who undertake one to not give up!)

In the non-Big-Science and more general sense, an efficiency problem raises its head. One variant of this is about research v. teaching: what does India need more of, or what’s a more efficient expense, to achieve scientific progress – institutions where researchers are free to conduct experiments without being saddled with teaching responsibilities or institutions where teaching is just as important as research? This question has often been in the news in India in the last few years, given the erstwhile HRD Ministry’s flip-flops on whether teachers should conduct research. I personally agree that we need to ‘let teachers teach’.

The other variant is concerned with blue-sky research: when are scientists more productive – when the government allows a “free play of free intellects” or if it railroads them on which problems to tackle? Given the fabled shortage of teachers at many teaching institutions, it’s easy to conclude that a combination of economic and policy decisions have funnelled India’s scholars into neglecting their teaching responsibilities. In turn, rejigging the fraction of teaching or teaching-cum-research versus research-only institutions in India in favour of the former, which are less resource-intensive, could free up some funds.

But this is also more about pragmatism than anything else – somewhat like untangling a bundle of wires before straightening them out instead of vice versa, or trying to do both at once. As things stand, India’s teaching institutions also need more money. Some reasons there is a shortage of teachers include the fact that they are often not paid well or on time, especially if they are employed at state-funded colleges; the institutions’ teaching facilities are subpar (or non-existent); if jobs are located in remote places and the institutions haven’t had the leeway to consider upgrading recreational facilities; etc.

Teaching at the higher-education level in India is also harder because of the poor state of government schools, especially outside tier I cities. This brings with it a separate raft of problems, including money.

Finally, a more ‘local’ example of prestige as well as financial constraints that also illustrates the importance of this PoV is the question of why the Swarnajayanti Fellowships have been awarded to so few women, and how this problem can be ‘fixed’.

If the query about which men should be excluded to accommodate women sounds like a reasonable question – you’re probably assuming that the number of fellows has to be limited to a certain number, dictated in turn by the amount of money the government has said can be awarded through these fellowships. But if the government allocated more money, we could appreciate all the current laureates as well as many others, and arguably without diluting the ‘quality’ of the competition (given just how many scholars there are).

Resource constraints obviously can’t explain or resolve everything that stands in the way of more women, trans-people, gender-non-binary and gender-non-conforming scholars receiving scholarships, fellowships, awards and prominent positions within academia. But axiomatically, it’s important to see that ‘fixing’ this problem requires action on two fronts, instead of just one – make academia less sexist and misogynistic and secure more funds. The constraints are certainly part of the problem, particularly when they are wielded as an excuse to concentrate more resources, and more power, in the hands of the already privileged, even as the constraints may not be real themselves.

In the final analysis, science doesn’t have to be a powerplay, and we don’t have to honour anyone at the expense of another. But deferring to such wisdom could let the fundamental causes of this issue off the hook.