What is your experience of using AI for research in condensed matter theory?

 I have been dabbling a little with using AI (at a very basic level) to help me with some research problems. For example, in a recent preprint, I did the maths in the Appendix with help from AI.

I am curious to hear from the experience of others of using it for research in condensed matter theory and theoretical chemistry. 

Peter Woit made a similar request to the high-energy theory community. The responses are helpful and worth reading. An earlier post about AI in math drew the following concrete suggestions from "Greg"

"In math, one thing I found useful to get my bearings was to go through some of my published papers and ask it to try and improve the main result.

Another is to do an adversarial audit of a paper you are very familiar with (yours if you have the stomach for it, someone else’s if not) and then see which of the complaints are pedantic and which are actual.

A third is to give you a history and references/citations for an argument that it provides.

A fourth is to suggest different ways to simplify an argument, even if they are only half-formed in your mind or if you are having trouble making them precise. (Try to be as precise as possible, but don’t use the same filter that you would use with a colleague. Sometimes vague is the best you can do.)

If you are thinking of proposing a question or conjecture in a grant application or preprint, ask it to stress test the conjecture. Don’t lean too much in one direction when prompting it and asking repeated questions (unless there is a particular thread you are trying to untangle, the use your nose, of course), but try to have it explore the positive direction in one query and the negative direction in a follow-up."

[A year ago, I posted about a benchmarking paper for condensed matter theory. I have not seen any updates on that work. For example, have the latest AI models improved significantly on these benchmarks?]

About a month ago, my office mate, Henry Nourse, gave a nice talk, "Is my skillset safe in an AI future?" at the UQ Condensed Matter group meeting. Henry gave a very concrete example of how he used it for a problem involving Raman scattering from quantum spin chains.

Henry stimulated me to see whether AI could help with a problem I was struggling with. It was basically a mathematics problem involving a Fourier transform. There were things that I thought were probably true, but after days of messing with algebra, my aging brain could not prove them. Google Gemini gave the answer in less than a minute. On the one hand, that is impressive. On the other hand, the absolute key was knowing precisely what question to ask it and knowing whether the anwer made sense. Furthermore, it took hours for me to check the result and to refine it, sometimes through further prompts. Besides getting the result, a nice benefit is that AI can quickly generate a LaTeX file that describes the mathematical argument and provides relevant references. Nevertheless, there are still hours more work of checking and polishing the text.

[Aside. The proof made use of various Bessel function identities that I had forgotten. I had used them extensively in a paper about thirty years ago.]

This success motivated me to use Google Gemini to develop, start to solve, and analyse a "simple" classical two-dimensional model of springs interacting with Ising spins. It is simple enough that I think most of the analysis can be done analytically. I have been working on related models on and off for the last couple of years. However, again due to my aging brain, I often get stuck on algebra or the subtleties of the theory of elasticity. AI does the algebra quickly. Furthermore, all the equations are formatted in LaTex. Great!

Again, it bears repeating that a thorough knowledge and intuition about the underlying physics is probably essential for knowing what prompts to make, and whether the answers make sense and are useful. The fact that I had struggled with related models offline beforehand really guided my interaction with AI.

Again, everything still needs to be checked. Sometimes AI makes suggestions that I think may be "hallucinations" [e.g. suggesting connections to the Svetitsky-Yaffe conjecture for lattice gauge theory] that may have arisen from earlier questions about unrelated topics. When I ask similar questions on a later day, it sometimes gives a different answer. For example, once it neglected to include the effect of one term in the Hamiltonian.

Initially, I struggled to keep records and manage the workflow. Eventually, I found it helpful to have a single LaTex file which is like a rough draft of a paper, into which I insert results. I found my engagements with AI both interesting, stimulating, and stressful. I now limit my sessions to short periods of time and then polish and edit the LaTex document in between sessions. This helps me consider what the next steps are. I am glad I am retired and can do all this at a leisurely pace.

What I am doing is incredibly basic compared to what may be possible. I am just using the free version of Google Gemini.

My limited experience raises all sorts of questions about the potential and pitfalls of graduate students using AI in research. Will it enhance or hinder their education? Potentially both?

There are some great quotes from Terence Tao in a recent New York Times article

“The effort needed to solve problems [without AI] is often very instructive. It teaches you something. It’s like going to the gym and having a goal to lift a weight a hundred times. Now, A.I. can solve questions without really getting any value out of them. It’s like having machines that can lift weights for you at the gym.”

“It feels to me like a really clever student who has memorized everything for the test but doesn’t have a deep understanding of the concept,..”

“This has some value,... But once someone shows a specific path to the waterfall, people just take that path. They don’t spend as much time looking for other paths.”

I also recommend Tao's recent lecture at the International Congress of Mathematicians.

For now, I would love to hear from others, particularly in condensed matter theory or theoretical chemistry.

What do you use AI for? What have you learned from the experience? How do you go about it? For example, how do you decide what prompts to use? How do you manage workflows? How do you document your work? What is it good at? What is it bad at? What potential benefits do you see? What pitfalls?

The disappointing story of superconductivity in Strontium Ruthenate

In 1994 superconductivity was discovered in strontium ruthenate (Sr2RuO4). This attracted considerable interest because it had a perovskite crystal structure, just like the cuprates. Furthermore, it was a stoichiometric compound and so not plagued by impurities like the cuprates.

In 1998, things got more interesting when NMR Knight shift measurements were interpreted as evidence for triplet superconductivity.

Analogues were made with triplet Cooper pairing in superfluid 3He mediated by ferromagnetic spin fluctuations.

Triplet pairing is associated with odd-parity (spatial) and time-reversal symmetry breaking. Evidence for the latter was claimed from muon spin relaxation (muSR) and the polar Kerr effect.

There are subtle questions about whether a bulk sample of a triplet superconductor exhibits spontaneous magnetisation. Leggett discussed this in an Appendix of his textbook. It turns out the magnetisation probably only exists on the edges.

Aside. The metallic phase is of interest because (unlike the cuprates) it is a Fermi liquid. More recently, it has been argued to be a Hund's metal.

Fueled by hype about topological quantum computing, the past two decades have seen even greater interest in the material due to proposals that it may be a topological superconductor. See for example, this paper.

Now we come to the disappointment. It turns out that the original Knight shift measurements were flawed, probably due to a problem with thermometry.

Recent, careful Knight shift measurements suggest spin-singlet pairing. They were described in a Physics Today article by Alex Lopatka in 2021, An unconventional superconductor isn’t so odd after all. The article describes all the intricacies and challenges of these measurements. Stuart Brown is to be commended for persisting with this problem.

What about the Kerr effect and muSR measurements suggesting time-reversal symmetry breaking?

The polar Kerr effect involves rotation of the plane of polarisation of the electromagnetic radiation by an angle of 65 nanoradians! There is only one group in the world (at Stanford) that can detect these ultra-minute rotations.

muSR may also be problematic. It is not really known where the implanted muon sits in the crystal or what effect it has on the surrounding crystal structure. In particular, these perturbations may produce a small local magnetic field which is nothing to do with the claimed global field due to the magnetism associated with the triplet superconductivity. A recent preprint by Warren Pickett considers some of the challenges associated with interpreting these experiments as evidence for time-reversal symmetry breaking.

What is disappointing about this?

Obviously, it would be nice to have a triplet superconductor and even more a topological one.

However, for me, the big disappointment is that it took almost thirty years for the original NMR measurements to be checked and shown to be wrong. This may reflect several sociological problems.

Kauzmann's maxim: people will tend to believe what they want to believe rather than what the evidence before them might suggest.

The condensed matter community tends to be infatuated with exotica.

There is not enough application of Occam's razor. Luxury journals don't want simple explanations or authors to raise doubts or ambiguities.

As far as I am aware, the 1998 Nature paper on the NMR Knight shift has still not been retracted.

This post was stimulated by a helpful colloquium at UQ given recently by James Annett. He has worked on strontium ruthenate for many years and is a co-author of a relevant review article.

Update. 23 April. James Annett pointed out to me that the authors for the 1998 NMR published a paper in 2020 which acknowledges that their original paper was incorrect.

Reduction of the 17O Knight Shift in the Superconducting State and the Heat-up Effect by NMR Pulses on Sr2RuO4

My review article on emergence

I just posted on the arXiv a long review article on emergence

Emergence: from physics to biology, sociology, and computer science

The abstract is below.

I welcome feedback. 

------

Many systems of interest to scientists involve a large number of interacting parts and the whole system can have properties that the individual parts do not. The system is qualitatively different to its parts. More is different. I take this novelty as the defining characteristic of an emergent property. Many other characteristics have been associated with emergence are reviewed, including universality, order, complexity, unpredictability, irreducibility, diversity, self-organisation, discontinuities, and singularities. However, it has not been established whether these characteristics are necessary or sufficient for novelty. A wide range of examples are given to show how emergent phenomena are ubiquitous across most sub-fields of physics and many areas of biology and social sciences. Emergence is central to many of the biggest scientific and societal challenges today. Emergence can be understood in terms of scales (energy, time, length, complexity) and the associated stratification of reality. At each stratum (level) there is a distinct ontology (properties, phenomena, processes, entities, and effective interactions) and epistemology (theories, concepts, models, and methods). This stratification of reality leads to semi-autonomous scientific disciplines and sub-disciplines. A common challenge is understanding the relationship between emergent properties observed at the macroscopic scale (the whole system) and what is known about the microscopic scale: the components and their interactions. A key and profound insight is to identify a relevant emergent mesoscopic scale (i.e., a scale intermediate between the macro- and micro- scales) at which new entities emerge and interact with one another weakly. In different words, modular structures may emerge at the mesoscale. Key theoretical methods are the development and study of effective theories and toy models. Effective theories describe phenomena at a particular scale and sometimes can be derived from more microscopic descriptions. Toy models involve minimal degrees of freedom, interactions, and parameters. Toy models are amenable to analytical and computational analysis and may reveal the minimal requirements for an emergent property to occur. The Ising model is an emblematic toy model that elucidates not just critical phenomena but also key characteristics of emergence. Many examples are given from condensed matter physics to illustrate the characteristics of emergence. A wide range of areas of physics are discussed, including chaotic dynamical systems, fluid dynamics, nuclear physics, and quantum gravity. The ubiquity of emergence in other fields is illustrated by neural networks, protein folding, and social segregation. An emergent perspective matters for scientific strategy, as it shapes questions, choice of research methodologies, priorities, and allocation of resources. Finally, the elusive goal of the design and control of emergent properties is considered.

Four scientific reasons to be skeptical of AI hype

The hype about AI continues, whether in business or science. Undoubtedly, there is a lot of potential in machine learning, big data, and large language models. But that does not mean that the hype is justified. It is more likely to limit real scientific progress and waste a lot of resources.

My innate scepticism receives concrete support from an article from 2018 that gives four scientific reasons for concern.

Big data: the end of the scientific method? 

Sauro Succi and Peter V. Coveney

The article might be viewed as a response to a bizarre article in 2008 by Chris Anderson, editor-in-chief at Wired, The End of Theory: The Data Deluge Makes the Scientific Method Obsolete

‘With enough data, the numbers speak for themselves, correlation replaces causation, and science can advance even without coherent models or unified theories’.

Here are the four scientific reasons for caution about such claims given by Succi and Coveney.

(i) Complex systems are strongly correlated, hence they do not (generally) obey Gaussian statistics.

The law of large numbers (central limit theorem) may not apply and rare events may dominate behaviour. For example, consider the power law decays observed in many complex systems. They are in sharp contrast to the rapid exponential decay in the Gaussian distribution. The authors state, "when rare events are not so rare, convergence rates can be frustratingly slow even in the face of petabytes of data."

(ii) No data are big enough for systems with strong sensitivity to data inaccuracies.

Big data and machine learning involve fitting data to a chosen function, such as a "cost function" with many parameters. That fitting involves a minimisation routine which acts on some sort of "landscape." If the landscape is smooth and minima are well-separated and not separated by too large of maxima then the routine may work. However, if the landscape is rough or the routine gets stuck in some metastable state there will be problems, such as over-fitting.

(iii) Correlation does not imply causation, the link between the two becoming exponentially fainter at increasing data size.  

(iv) In a finite-capacity world, too much data is just as bad as no data.

In other words, it is all about curve fitting. The more parameters used the less likely for insight to be gained. Here the authors quote the famous aphorism, attributed to von Neumann and Fermi, "with four parameters I can fit an elephant and with five I can make his tail wiggle."

Aside: an endearing part of the article is the inclusion of tow choice quotes from C.S. Lewis

‘Once you have surrendered your brain, you've surrendered your life’ (paraphrased)

‘When man proclaims conquest of power of nature, what it really means is conquest of power of some men over other men’.

I commend the article to you and look forward to hearing your perspective. Is the criticism of AI hype fair? Are these four scientific reasons good grounds for concern. 

Certain benefits of Bayes

Best wishes for the New Year! One thing I hope to achieve this year is an actual understanding of things "Bayesian".

I am particularly interested because it gives a way to be more quantitative and precise about some of the intuitions that I use in science. For example, I tend to be skeptical of new experimental results (often hyped) that claim to go against well-established theories, regardless of how good the "statistics" of the touted result.

In this vein, Phil Anderson argued that Bayesian methods should have been used to rule out the significance of "discoveries" such as the 10 keV neutrino and the fifth force. In 1992 he wrote a Physics Today column on the subject.

An interesting metric for mathematical formula is the ratio of profound and wide implications to the simplicity of the formula and its derivation. I suspect that Bayes' formula for conditional probabilities would win first place!

P(A|B) denotes the probability of A given B. 

The proof takes about two lines. If you multiply both sides of the equation about by P(B) the identity holds because both sides of the equation are just different ways of writing P(A and B).

My first attempt to understand the applications and implications of Bayes was reading the relevant sections in Phil Nelson's beautiful book, Physical Models of Living Systems. There is a helpful section entitled, "Bayes formula provides a consistent approach to upgrading our degree of belief in light of new data."

More recently, I found this wonderful and short video very helpful, as it clearly defines terms, uses graphical representations, and gives some concrete examples.

 

A Bayesian perspective highlights the importance of reporting negative results and is the basis of a seminal paper

Why Most Published Research Findings Are False by John P. A. Ioannidis

A measure of the profundity of Bayes is that the Stanford Encyclopedia of Philosophy has two articles on the topic

Bayes Theorem

Bayesian Epistemology

Could faculty benefit from a monastery experience?

A few months ago, the New York Times ran a fascinating Guest Opinion, Why Universities Should Be More Like Monasteries by Molly Worth, a historian at University of North Carolina. She describes a popular undergraduate course, the "monk" class, at the University of Pennsylvania. 

On the first day of class — officially called Living Deliberately — Justin McDaniel, a professor of Southeast Asian and religious studies, reviewed the rules. Each week, students would read about a different monastic tradition and adopt some of its practices. Later in the semester, they would observe a one-month vow of silence (except for discussions during Living Deliberately) and fast from technology, handing over their phones to him.

This got me wondering about whether universities and funding agencies might experiment with similar initiatives for faculty. It might be a bit like the Aspen Center for Physics and Gordon Research Conferences were before the internet. Faculty would surrender their phones and have the internet disabled on their computers. For one week they would be required to spend their time reading, writing, and thinking. Exercise, daydreaming, doodling, and just having fun are to be encouraged. No administrative work or grant writing. The emphasis would be coming up with new ideas, not finishing off old work. For one hour per day, participants could meet with others and talk about their new ideas. I think it might be refreshing, reinvigorating, and highly productive (in the true sense of the word).

After trialling the program for a week, why not then try it for month-long periods.

These are the conditions under which Newton wrote the Principia and Darwin The Origin of Species. In their case, they did it over a period of several years.

What do you think?

A few things condensed matter physics has taught me about science (and life)

We all have a worldview, some way that we look at life and what we observe. There are certain assumptions we tend to operate from, often implicitly. Arguably, our worldview is shaped by our experiences: family, friendships, education, jobs, community organisations, and our cultural context (political, economic, and social).

A significant part of my life experience has been working in universities as a condensed matter physicist and being part of a broader scientific community. Writing a Condensed Matter Physics: A Very Short Introduction crystallised some of my thoughts about what CMP might mean in broader contexts. I am more aware of how my experience in CMP has had a significant influence on the way I view not just the scientific enterprise, but also broader philosophical and social issues. Here are a few concrete examples.

Complex systems. The objects studied in condensed matter physics have many interacting components (atoms). Further, there is an incredible diversity of systems (materials and phenomena) that are studied. Many different properties and parameters are needed to characterise a system and its possible states. There are many different ways of investigating each system. Similarly, almost everything else of interest in science and life is a complex system.

Emergence. This is central to CMP. The whole is greater than the sum of the parts. The whole is qualitatively different from the parts. Related features include robustness, universality, surprises, and the difficulty of making predictions. An emergent perspective can provide insights into other complex systems: from biology to psychology to politics.

Differentiation and integration. A key aspect of describing and understanding a complex system is conceptually breaking it into smaller parts (differentiation), determining how those parts interact with one another, and determining how those interacting parts combine to produce properties of the whole system (integration).

Diversity: The value of multiple perspectives and methods. Due to the complexity of condensed matter systems, multiple methods are needed to characterise their different properties. Due to emergence, there are various scales and hierarchies present. Investigating and describing the system at these different scales provides different perspectives on the system. What does the scientist do with all these different perspectives? Interpretation and synthesis are needed. That is not an easy or clearcut enterprise.

 Navigating the middle ground. The most interesting CMP occurs in an intermediate interaction regime that is challenging theoretically. Insight can be gained by considering two extremes that are more amenable to analysis: weak interaction and strong interaction. I had fun using conservative-liberal political tensions as a metaphor for divisions in the strongly correlated electron community.

The Art of Interpretation. Everything requires interpretation: a phone text message, a newspaper article, a novel, a political event, data from a science experiment, and any scientific theory. With interpretation, we assign meaning and significance to something. How we do this is complex and draws on our worldview, both explicitly and implicitly. Regardless of our best intentions, interpretation always has subjective elements.

Synthesis. Given the diversity of data, perspectives, and interpretation, it is a challenge to synthesise them into some coherent and meaningful whole. All the pieces are rarely consistent with one another. Some will be ignored, some discarded, some considered peripheral, and others central. This synthesis is also an act of interpretation.

All models are wrong but some are useful. One way to understand complex systems is in terms of "simple" models that aim to capture the essential features of certain phenomena. In CMP significant progress (and many Nobel Prizes) has resulted from the proposal and study of such models. There is a zoo of them. Many are named after their main inventor or proponent: Ising, Anderson, Hubbard, Heisenberg, Landau, BCS,... All theories in CMP are also models since they involve some level of approximation, at least in their implementation. These models are all wrong, in the sense that they fail to describe all features and phenomena of the system. But, the best models are useful. Their simplicity makes them amenable to understanding, mathematical analysis, or computer simulation. Furthermore, the models can give insight into the essential physics underlying phenomena, predict trends, or be used to analyse experimental data. 

The autonomy of academic disciplines. Reality is stratified. At each level of the hierarchy, one has unique phenomena, methods, concepts, and theories. Most of these are independent of the details of what happens at lower levels of the hierarchy. Given the richness at each level, I do not preference one discipline as more fundamental or important than the others.

Pragmatic limits to knowledge. We know so much.  We know so little. On the one hand, it is amazing to me how successful CMP has been. We have achieved an excellent understanding, at least qualitatively of many emergent phenomena in systems that are chemically and structurally complex (e.g., liquid crystals and superconductivity in crystals involving many chemical elements). On the other hand, there are systems such as glasses and cuprate superconductors that have been incredibly resistant to understanding. Good research is very hard, even for the brilliant. Gains are often incremental and small. This experience leads me to have sober expectations about what is possible, particularly as one moves from CMP to more complex systems such as human societies, national economies, and brains.

Science is a human endeavour. Humans can be clever, creative, insightful, rational, objective, cooperative, fiercely independent and capable of great things. The achievements of science are a great testimony to the human spirit. Humans can also be stubborn, egotistical, greedy, petty, irrational, ruthlessly competitive, and prone to fads, mistakes and social pressures. Science always happens in a context: social, political, cultural, and economic. Context does not determine scientific outcomes but due to human nature, it can corrupt how science is done.

The humanity of scientists leads to a lack of objectivity captured in Walter Kauzmann's maxim: people will tend to believe what they want to believe rather than what the evidence before them suggests that they should believe. My decades of experience working as a scientist leads me to scepticism about extravagant claims that some scientists make, particularly hype about the potential significance (scientific, technological, or philosophical) of their latest discovery or their field of research. Too often such claims do not stand the test of time.

Humility. This brings together practically everything above. The world is complex, people are complex, and human-world interactions are complex. It is easy to be wrong. We often have a pretty limited perspective of what is going on. 

How can funders promote significant breakthroughs?

 Is real scientific progress slowing? Are funders of research, whether governments, corporations, or philanthropies, getting a good return on their investment? Along with many others (based largely on intuition and anecdote) I believe that the system is broken, and at many different levels. What are possible ways forward? How might current systems of funding be reformed?

The Economist recently published a fascinating column (in the Finance and Economics section!), How to escape scientific stagnation. It reviews a number of recent papers by economists that wrestle with questions such as those above.

Philanthropists... funding of basic research has nearly doubled in the past decade. All these efforts aim to help science get back its risk-loving mojo.

In a working paper published last year, Chiara Franzoni and Paula Stephan look at a number of measures of risk, based on analyses of text and the variability of citations. These suggest science’s reward structure discourages academics from taking chances.

Another approach in vogue is to fund “people not projects”. A study in 2011 compared researchers at the Howard Hughes Medical Institute, where they are granted considerable flexibility over their research agendas and lots of time to carry out investigations, with similarly accomplished ones funded by a standard NIH programme. The study found that researchers at the institute took more risks. As a result, they produced nearly twice as much highly cited work, as well as a third more “flops” (articles with fewer citations than their previously least-cited work). 

Despite the uncertainty about exactly how best to fund scientific research, economists are confident of two things. The first is that a one-size-fits-all approach is not the right answer,... DARPA models, the Howard Hughes Medical Institute’s curiosity-driven method, and even handing out grants by lottery, as the New Zealand Health Research Council has tried, all have their uses.

The second is that this burst of experimentation must continue. The boss of the NSF, Sethuraman Panchanathan, agrees. He is looking to reassess projects whose reviews are highly variable—a possible indication of unorthodoxy. He is also interested in a Willy Wonka-style funding mechanism called the “Golden Ticket”, which would allow a single reviewer to champion a project even if his or her peers do not agree.  ...many venture-capital partnerships employ similar policies, because they prioritise the upside of long-shot projects rather than seeking to minimise failure. 

The study that I would like to see done is along the following lines. Identify at what age and what type of institution and what type of funding environment, the biggest breakthroughs happen. I suggest that you will find in the U.S.A, that it was done by young faculty at the top 20 institutions in an era when they did not have to worry much about getting grants. If so, then I think most of the money should be given to them!

 

Systemic flaws that are undermining good science

Everyone likes to be right. But, sometimes I really wish I was wrong, particularly about problems I see in the world. I wish I was wrong about science being broken. Some of these issues I discuss in the final chapter of Condensed Matter Physics: A Very Short Introduction, due to the relevance of these problems to the future of the field.

Similar concerns were discussed with greater clarity, way back in 2014, by four scientists who are much more experienced and distinguished than I am. 

Rescuing US biomedical research from its systemic flaws 

Bruce Alberts, Marc W. Kirschner, Shirley Tilghman, and Harold Varmus

Positions the different authors have held include President of the US Academy of Sciences, President of Princeton University, and Director of the National Institutes of Health.

Although the article focuses on biomedical research I think the three words "medicine, biomedical, and biology" could be replaced respectively with "technology, materials science, and condensed matter physics" almost everywhere in the article. 

Here are a few quotes.

The long-held but erroneous assumption of never-ending rapid growth in biomedical science has created an unsustainable hypercompetitive system that is discouraging even the most outstanding prospective students from entering our profession—and making it difficult for seasoned investigators to produce their best work. This is a recipe for long-term decline, and the problems cannot be solved with simplistic approaches. Instead, it is time to confront the dangers at hand and rethink some fundamental features of the US biomedical research ecosystem.

... the remarkable outpouring of innovative research from American laboratories—high-throughput DNA sequencing, sophisticated imaging, structural biology, designer chemistry, and computational biology—has led to impressive advances in medicine and fueled a vibrant pharmaceutical and biotechnology sector. In the context of such progress, it is remarkable that even the most successful scientists and most promising trainees are increasingly pessimistic about the future of their chosen career.

... hypercompetition for the resources and positions that are required to conduct science suppresses the creativity, cooperation, risk-taking, and original thinking required to make fundamental discoveries.

The system now favors those who can guarantee results rather than those with potentially path-breaking ideas that, by definition, cannot promise success. Young investigators are discouraged from departing too far from their postdoctoral work, when they should instead be posing new questions and inventing new approaches. Seasoned investigators are inclined to stick to their tried-and-true formulas for success rather than explore new fields. 

One manifestation of this shift to short-term thinking is the inflated value that is now accorded to studies that claim a close link to medical practice. Human biology has always been a central part of the US biomedical effort... Many surprising discoveries, powerful research tools, and important medical benefits have arisen from efforts to decipher complex biological phenomena in model organisms. In a climate that discourages such work by emphasizing short-term goals, scientific progress will inevitably be slowed, and revolutionary findings will be deferred.

As competition for jobs and promotions increases, the inflated value given to publishing in a small number of so-called “high impact” journals has put pressure on authors to rush into print, cut corners, exaggerate their findings, and overstate the significance of their work. 

The development of original ideas that lead to important scientific discoveries takes time for thinking, reading, and talking with peers. Today, time for reflection is a disappearing luxury for the scientific community. 

...administrative tasks are taking up an ever-increasing fraction of the day and present serious obstacles to concentration on the scientific mission itself. 

The following is particularly true of luxury journals. 

Professional editors are increasingly serving in roles played in the past by working scientists and can undermine the enterprise when they base judgments about publication on newsworthiness rather than scientific quality. 

Even after they have landed a research position in academia or research institutes, new investigators wait an average of 4–5 y to receive federal funding for their work compared with 1 y in 1980 (2). Two stark statistics tell much of the tale—the average age at which PhD recipients assume their first tenure-track job is 37 y, and they are approaching 42 y when they are awarded their first NIH grant.

Although it varies across fields and individuals, I get the impression that most scientists do their best work in the rough age range of 35-45. Currently, people are spending most of these years looking for a permanent job and then applying for grants, rather than actually doing science.

The graph below shows just how much the system changed in just thirty years. NIH grants became "gentrified". In different words, all the grants now go to "old farts" doing the same old thing, rather than to "young turks" who want to try new things and have a real impact.

Percentage of NIH R01 Principal Investigators aged 36 and younger and aged 66 and older, 1980–2010

The authors did make some concrete proposals and in a follow-up article, they discuss a broader meeting held to discuss the issues.

Addressing systemic problems in the biomedical research enterprise

To what extent progress has been made in the biomedical community in the past eight years I do not know.

Emergence matters (in a nutshell)

Emergence is one of the most important concepts in the sciences: from physics to biology to sociology. Most of the big questions in science involve emergence. Yet there is no consensus about what emergence is, how to define it, or why it matters. This is my attempt to clarify some of the important issues and questions. For reasons of brevity, I give no references and only a few examples. They can come later. Here I am trying to take a path that is intermediate between the precision of philosophers and the looseness of condensed matter physicists' discussion of emergence. My goals are clarity and brevity.

Characteristics of emergent phenomena

Consider a system that is composed of many interacting parts. If the properties of the system are compared with the properties of the individual parts, a property of the whole system is an emergent property if it has the following characteristics.

1. Novelty 

An emergent property of the system is a property that is not present in the individual parts of the system.

2. Modification of parts 

An emergent property of the system is associated with a modification of the properties of and the relationships between the parts of the system. 

3. Universality

An emergent property is universal in the sense that it is independent of many of the details of the parts. As a consequence, there are many systems that can have the emergent property.

4. Irreducibility

An emergent property cannot be reduced to properties of the parts.

5. Limited predictability

An emergent property is difficult to predict solely from knowledge of the properties of the parts and how they interact with one another.

Here are a few issues to consider about the five characteristics above. 

First, “emergent property” could possibly be replaced with emergent phenomenon, object, or state.

Second, for each of the five characteristics is it necessary and/or sufficient for the system property to be emergent?

Third, one of the most contested characteristics concerns predictability. “Difficult to predict” is sometimes replaced with “impossible”, “almost impossible”, “extremely difficult”, or “possible in principle, but impossible in practice.” After an emergent property has been observed sometimes it can be understood in terms of the properties of the parts. An example is the BCS theory of superconductivity, which provided a posteriori, rather than a priori, understanding. A keyword in the statement above is “solely”.

Examples of properties of a system that are not emergent are volume, mass, charge, and number of atoms. These are additive properties. The property of the system is simply the sum of the properties of the parts.

Scales and hierarchies

Central to emergence is the idea of different scales. Emergent properties only occur when scales become larger. Scales that are simply defined, and might be called extrinsic, are the number of parts, length scale, and time scale. A more subtle scale, which might be called intrinsic, is a scale associated with the emergent property. This emergent scale is intermediate between that of the parts and that of the whole system.

Emergent scales lead naturally to hierarchies, such as those associated with different scientific disciplines, as shown below. Hierarchies also occur within individual disciplines.

At each level there are distinct phenomena, concepts, theories, and scientific methods.

Another important scale is that of complexity. Generally, as one goes up the hierarchy one says that the level of complexity increases. Giving a precise version of such statements is not simple.

Complexity

Simple rules can lead to complex behaviour. This is nicely illustrated by cellular automata. It is also seen in other systems with emergent properties. For example, the laws describing the properties of electrons and ions in a crystal or a large molecule are quite simple (Schrodinger’s equation plus Coulomb’s law). Yet from these simple rules, complex phenomena emerge: all of chemistry and condensed matter physics!

There is no agreed universal measure for the complexity of a system or with many components. One possibility is the Kolmogorov measure. Using such measures to elucidate emergence, such as how complexity changes with other scales, is an important challenge.

Other issues

There are a host of other issues and topics that enter discussions about emergence. Some of these are of a more philosophical nature. Here I just list them: robustness, quality vs. quantity, objective vs. subjective, universality vs. particularity, ontology vs. epistemology, discontinuities, incommensurability, theory reduction, asymptotic singularities, top-down causation, supervenience, differentiation and integration (not calculus) of system parts, reductionism, foundationalism, fundamentalism, strong versus weak emergence, and criteria for theory acceptance.

Discussion of some of these issues can be quite abstract but to make the discussion above more precise they may need to be considered. 

Emergence is relevant to practical matters such as scientific strategy, priorities, allocation of resources, and our dispositions as scientists. Too often views on these issues are implicit and not reflected upon. 

The practical matter of scientific strategy

When studying a system, the first choice that must be made is what scale or scales to focus on. For example, in materials science, the options range from the atomic scale to the macroscopic. This choice determines the tools and methods, both experimental and theoretical, that can be used to study the system. In different words, the scientist is making a choice of ontology: the object they choose to study. This then determines epistemology: the concepts, theories, and organising principles a scientist may use or hope to discover. Effective theories and toy models enter here. 

When systems have been studied by a range of methods and at a range of scales, a challenge is the synthesis of the results of these studies. Value-laden judgements are made about the priority, importance, and validity of such attempts at synthesis. Often synthesis is relegated to a few sentences in the introductions and conclusions of papers.

For known systems and emergent properties, there is the possibility of creating new methods and probes to investigate them at appropriate scales.

New systems can be created and investigated in the hope of discovering new emergent properties (e.g., new states of matter) or more modestly, that manifest an emergent property that is more amenable to scientific study or technological application.

As emergent properties involve multiple scales they are often of interest to and amenable to study by more than one scientific discipline. This creates opportunities and challenges for interdisciplinary collaboration.

Individual scientists must and do make decisions about the relative priority of the different strategies outlined above. Research groups, departments, institutions, professional societies, and funding agencies must and do also make decisions about such priorities. The decision outcomes are also emergent properties of a system with multiple scales from that of the individual scientist to global politics. I claim that too often these weighty decisions are made implicitly, rather than explicitly following debate and deliberation.

The disposition of the scientist

All scientists are human. In our professional life, we have hopes, aspirations, values, fears, attitudes, expectations, and prejudices. These are shaped by multiple influences from the personal to the cultural to the institutional. We should reflect on the past century of our study of emergent systems from physics to biology to sociology. If we honestly evaluate our successes and failures I think this may lead us to have certain dispositions that are interrelated.

Humility. There is so much we do not understand. Furthermore, we fail abjectly at predicting emergent properties. This is not surprising. Unpredictability is one of the characteristics of emergent properties. There is a hubris associated with grand initiatives such as “the theory of everything”, the Human Genome Project, “materials by design”, and macroeconomic modelling. 

Expect surprises. There are many exciting discoveries waiting. They will be found by curiosity and serendipity.

Wonder. Emergent phenomena are incredibly rich and beautiful to behold, from physics to biology to sociology. Furthermore, the past century has seen amazing levels of understanding. But this is a “big picture” and “coarse-grained” understanding, not the description that the reductionists lust for and claim possible. 

Realistic expectations. Given the considerations above I think we should have modest expectations of the levels of understanding possible, and what research programs, from that of individual scientists to billion-dollar initiatives, can achieve. We need to stop the hype. Modest expectations are particularly appropriate with respect to our ability to control emergent properties.

The holy grail

“The philosophers have only interpreted the world, in various ways. The point, however, is to change it.”

Karl Marx

Understanding complex systems with emergent properties is an ambitious scientific challenge. This enterprise has intrinsic intellectual merits. But a whole other dimension and challenge is to use this understanding to modify, manipulate, and control the properties of systems with emergent properties. This enticing prospect appeals to technologists, activists, and governments. Such promises feature prominently in grant applications, press releases, and reports from funding agencies. Diverse examples of this control goal include chemical modification of known superconductors to produce room-temperature superconductivity, drug design, social activism, the leadership of business corporations, and governments attempting to manage the economy. 

However, we should honestly reflect on decades of “scientifically informed” and “evidence-based” initiatives in materials science, medicine, poverty alleviation, government economic policy, business management, and political activism. Unfortunately, the fruit from these initiatives is disappointing, particularly compared to what has often been promised.

My goal is not to promote despair but rather to prevent it.  With more realistic expectations, based on reality rather than fantasy, we are more likely to make significant progress in finding ways to make some progress (albeit modest but worthwhile) in learning how to manipulate these complex systems.

This post contains many claims that require discussion, refinement or abandonment. I welcome suggestions on how to improve these ideas.

How to move towards doing Deep Work

"Shallow work is non-cognitive, logistical or minor duties, often performed while distracted. These efforts require little cognitive effort, tend to create little value, and are usually easy to replicate." Examples include replying to emails, browsing websites, looking at social media, filling in forms, and attending meetings.

"Deep work is the ability to focus without distraction on a cognitively demanding task. These efforts create new value, improve your skill, and are hard to replicate.”

A colleague told me that Cal Newport's book, Deep Work, revolutionised his professional life.  These two short videos give a nice summary, focusing on quite practical ways to implement the ideas. They are both based on this article by Dan Silvestre.

Einstein on big questions

The mere formulation of a problem is far more essential than its solution, which may be merely a matter of mathematical or experimental skills.

To raise new questions, new possibilities, to regard old problems from a new angle, requires creative imagination and marks real advance in science.

I am enough of an artist to draw freely upon my imagination. Imagination is more important than knowledge. Knowledge is limited. Imagination encircles the world.

Albert Einstein and Leopold Infeld (1938), The Evolution of Physics

I recently encountered this quotation in The Poetry and Music of Science: Comparing Creativity in Science and Art by Tom McLeish. I have heard many times the "Imagination is more important than knowledge" quote, sometimes as a dubious justification for dubious ideas. However, I did not know the context. 

My postdoctoral advisor, John Wilkins tried to drill into me, the idea in the first paragraph, that just coming up with a well-defined formulation of a problem could be a significant advance. This idea certainly had some impact on me, since I sometimes hear my non-scientist wife quote it!

On reflection, I am afraid that I too easily lose sight of this priority of defining problems, just like the method of multiple alternative hypotheses. Good science is hard.

Why am I reading this article? What question am I trying to answer?

Why am I writing this paper? What question am I trying to answer?

What is the problem I assigning a student to work on? Is it well-formulated?

Defining good research questions is hard work and requires discipline.

Sage wisdom on computational materials science

Roald Hoffmann and Jean-Paul Malrieu are two of my favourite living theoretical chemists. Both greatly value the role of concepts and intellectual clarity in theory. Hoffmann has featured in 22 posts on this blog.

They recently published a wonderful trilogy in  Angewandte Chemie.

Simulation vs. Understanding: A Tension, in Quantum Chemistry and Beyond. 

Part A. Stage Setting

Part B. The March of Simulation, for Better or Worse

Part C. Toward Consilience

I add this trilogy to my list of 5 papers every computational chemistry student should read, suggested by me a decade ago. [Malrieu is author of one of those and Hoffmann co-author of another.]

Although the trilogy addresses and uses specific examples from computational quantum chemistry it is just as relevant to anyone interested in computational materials science. Actually, I hope that anyone interested in materials science would read and digest it as it gives a sober and balanced perspective about the relationship between theory, simulation, and understanding.

Articles are timely as they address hype about how AI techniques will "revolutionise" materials theory. 

The articles are beautifully written and engage with broader themes such as philosophy of science, culture, art, and politics.

Finally, I just love this photo of the two authors, both in their eighties. the photo reflects some of the joy they find in science, so beautifully expressed in these articles.

I thank Ben Powell for bringing the papers to my attention.

The science and politics of the origins of covid-19

I want to begin by stating some hypotheses. Some may be obvious. Others may be contentious. I will number them so that people can easily make comments about specific ones. The underlying issues are illustrated in recent debates about the possible origins of covid-19.

1. Systematic critical thinking is essential to scientific progress and public policy. Healthy doses of skepticism can be valuable.

2. Science progresses well by making multiple hypotheses and examining carefully what evidence is consistent with each of the hypotheses. This is something that Murray Gell-Mann wished someone had told him when he was twenty years old.

3. Transparency is essential to science. People need to share data, including primary data. The more that such data is publicly available the better. This is what open science advocates. 

4. Science is built on ethical conduct, both implicit and explicit. It is important that declarations of conflicts of interest are not just a box-ticking exercise.

5. Scientists cannot have allegiance to some greater authority than truth and integrity. Problematic allegiances include to a company, a family, an institution, a political party, or to a nation. An example is the case of the recent change to the charter of Fudan University, indicative of the stranglehold that the Chinese Communist Party has over Chinese universities.

6. Given these issues about integrity and conflicts of interest luxury journals are problematic because there is a conflict of interest between the commercial success of the publisher in the short term (achieved by promoting hype, i.e. newsworthy sexy scientific breakthroughs, even if they are wrong) and the boring work of doing careful painstaking science.

7. One approach to solving some of these problems is self-regulation of scientific communities. However, when sub-communities (e.g. virologists, string theorists) self-regulate this may be impeded by conflicts of interest.

8. Given the issues above, science journalists need to be more critical and skeptical. Too often they seem in awe of scientists and want to promote hype as it sells. Journalists need to ask more hard questions about conflicts of interest, weak reasoning, claimed "breakthroughs", hype, proposed great technological applications, and the "science as saviour" narrative.

9. There is a fear among scientists about publically speaking about scientific uncertainty and ambiguity. This fear is understandably driven by the experience of "skeptics" latching onto uncertain statements to promote climate change denialism, young-earth creationism, and anti-vaccines. Thus, a great challenge in public engagement is to educate about the role of uncertainty in science.

10. Science always occurs in a political context whether it is in Australia, Romania, or China. The context will always have some influence, but it should not be determinative.

11. The greater the stakes (whether potential Nobel Prizes, company profits, government scandal, a disaster) in play, the greater the likelihood will be for mistakes, corruption, deception, and cover-up. Consequently, the level of scientific diligence and regulation needs to be proportionate to the possible benefits and risks. Extraordinary claims require extraordinary evidence.

12. Beware of the argument from authority. A hypothesis should be accepted or rejected based on the quality of the reasoning and evidence provided, not on the scientific prestige (or lack thereof) of the proponent.

All of the claims above I see played out recently in debates about the origins of covid-19. Two distinct hypotheses are dissected in a helpful and long article recently published in the Bulletin of Atomic Scientists.

The origin of COVID: Did people or nature open Pandora’s box at Wuhan?  Nicholas Wade 

Hypothesis 1. The virus spread from a wet market in Wuhan. The virus was zoonotic, i.e. as a result of evolution it crossed the species barrier from bats to humans.

Hypothesis 2. The virus spread from the Wuhan Institute of Virology where a research group was investigating bat viruses and doing "gain of function" research to see how the bat viruses might be modified genetically into a form that could infect humans. 

The article is worth reading because it carefully lays out the science while also raised many of the issues I mention above. A few things that I learned follow.

There is significant evidence that the MERS, SARS1, Ebola viruses are zoonotic. The evidence consists of finding intermediate genetic forms in intermediate species. Often this evidence was found within months of the disease outbreak. In contrast, after 18 months there is still no evidence of intermediate forms for SARS2.

The "gain of function" research in Wuhan was being funded by the USA National Institutes of Health, via a grant to the EcoHealth Alliance of New York, led by Peter Daszak. Wade writes

"We stand together to strongly condemn conspiracy theories suggesting that COVID-19 does not have a natural origin,” a group of virologists and others wrote in the Lancet on February 19, 2020, when it was really far too soon for anyone to be sure what had happened. Scientists “overwhelmingly conclude that this coronavirus originated in wildlife,” they said, with a stirring rallying call for readers to stand with Chinese colleagues on the frontline of fighting the disease.

Contrary to the letter writers’ assertion, the idea that the virus might have escaped from a lab invoked accident, not conspiracy. It surely needed to be explored, not rejected out of hand. A defining mark of good scientists is that they go to great pains to distinguish between what they know and what they don’t know. 

It later turned out that the Lancet letter had been organized and drafted by Peter Daszak, ... This acute conflict of interest was not declared to the Lancet’s readers. To the contrary, the letter concluded, “We declare no competing interests.”

Wade points out that there is no direct evidence for either of the two hypotheses (which he calls theories).

He also talks quite a bit about "who is to blame" and claims that we need to know the answer as to which hypothesis is correct in order to know how to prevent the next pandemic. However, I disagree. Based on the evidence we already have we can conclude the following.

A. New deadly viruses can be zoonotic. The best way to reduce their likelihood is to close wet markets and reduce environmental destruction.

B. Even if SARS2 did not spread from the "gain of function" research in Wuhan it is completely plausible that it could have. Thus, given such risks that research should be stopped until a case is made that the possible benefits outweigh the risks and that it is done with much greater transparency and regulation than currently.

For balance I include an extract from Wikipedia

In May 2021, Wade published an article which advanced the claim that COVID-19 likely originated from a leak at the Wuhan Institute of Virology.^[12][13] Wade's article generated significant controversy,^[14] and has become one of the most-cited pieces in support of the lab leak hypothesis.^[15] This claim is at odds with the prevailing view among scientists that the virus most likely has a zoonotic origin.^{[16][17][18][19]} Some experts have supported taking the lab leak possibility seriously, while the majority consider it very unlikely, calling it "speculative and unsupported".^[20][21] Others noted the explosive and implausible nature of Wade's allegations about virologists conspiring to avoid blame for causing the pandemic,^[22] with Science-Based Medicine among those calling Wade's argument a conspiracy theory.^[23]

Another article worth reading (recommended by a commenter on this blog) is

Beijing’s useful idiots: Science journals have encouraged and enforced a false Covid narrative by Ian Birrell.

Implicit versus explicit beliefs

 How can we design a room-temperature superconductor? How can a government stimulate economic growth? How can an NGO help reduce domestic violence? Why do communities become segregated on racial lines? How can I improve my mental health?

These important questions may seem unrelated. However, I propose that often there is a common issue about the strategies that people (whether individuals, professions, NGOs, funding agencies, governments, ...) propose to find answers or when definite answers are proposed.

Many strategies and answers involve a heavy dose of implicit beliefs. These are assumptions that are never stated. They may be elements of a worldview, which according to one definition, is

a commitment, a fundamental orientation of the heart, that can be expressed as a story or in a set of presuppositions (assumptions which may be true, partially true, or entirely false) which we hold (consciously or subconsciously, consistently or inconsistently) about the basic construction of reality, and that provides the foundation on which we live and move and have our being.

 James W. Sire, The Universe Next Door: A basic worldview catalog

These implicit beliefs may relate to values and morality. But I want to focus more on implicit beliefs that are related to academic disciplines such as philosophy of science, psychology, political science, theology, economics, anthropology,  sociology, ...  Most of us have never studied these disciplines and some of us may be skeptical about some of them. But, my point is that everyone has implicit ideas about what is true with regard to the objects these disciplines study. Everyone has a philosophy of science. Everyone has ideas about how minds work and how to change societies. It is just that these beliefs are rarely stated. 

Why does this matter? If implicit beliefs are never stated, they can never be tested, evaluated, critiqued, refined, or rejected. I believe that implicit beliefs are too often based on intuition, prejudice, common sense, or culture (social pressure to conform to accepted wisdom). This is not necessarily bad. Sometimes intuition, common sense, and culture are helpful and correct. We could not survive in life if we did not have them. We simply don't have the time, energy, and resources to constantly question and validate everything. On the other hand, if there is a vacuum, it will get filled with something. A major lesson from scientific history is that intuition, prejudice, common sense are sometimes wrong.

I now give three concrete examples of implicit beliefs. They cover computational materials science, public policy, and social activism.

Understanding materials using computers

Amongst others, there are two things, we would like more computational power to be able to do. One is to do reliable ab initio calculations of the properties of complex molecules and solids, from proteins to crystals with unit cells containing large numbers of atoms.  Another is to do exact diagonalisation (or some alternative reliable method) of many-body Hamiltonians, such as the Hubbard model, on large enough lattices that finite-size effects are minimal or can be reliably accounted for.

Over the past decade, there has been a lot of hype about how quantum computers and/or machine learning techniques will solve these problems and thus initiate a new era of materials understanding, discovery, and design with significant technological and economic benefits. My problem is that these claims usually seem to have the implicit belief that the only obstacle to progress is one of computational power. This fallacy has recently been deconstructed and critiqued in detail in three beautiful essays by Roald Hoffmann and Jean-Paul Malrieu, Simulation vs. Understanding: A Tension, in Quantum Chemistry and Beyond.

Public policy

National economies around the world have been battered by the covid-19 pandemic. In response, governments of prosperous countries are spending big on stimulus packages. This involves taking on massive amounts of debt and significant government intervention in "free" market economies. Will these initiatives achieved their goals, particularly in the long term? Could they actually make things worse? Responses from pundits, both for and against, are laden with implicit beliefs. Unfortunately, economists cannot agree on the answer to the basic question, "Does government stimulus spending actually produce economic growth?" This issue is nicely discussed in a pre-pandemic podcast at Econtalk. 

NGOs and social activism

Many NGOs are about change. They aim to build a better world, addressing problems such as domestic violence, poverty, climate change, corruption, racism,... They aim to promote education, human rights, good governance, democracy, health, transparency, .. I love NGOs. I support many: philosophically, financially, and practically. To survive most NGOs have to raise funds, whether from many small donors or large philanthropies. This requires a well-honed pitch that aims to inspire potential donors to give. Furthermore, the whole operation of most NGOs is laden with implicit beliefs, whether those of the founders, staff or donors.

Consider a hypothetical NGO whose goal is to reduce the number of murders in a country. I chose this example because it may at first appear less controversial and contentious than some. Almost everyone thinks murder is wrong (always) and societies should stop reduce it. But why do murders occur? Revenge, passion, drugs, alcohol, money, politics, racism, ... Will making the purchase of guns difficult reduce the murder rate? Gun lobbyists will claim "Guns don't kill people. Criminals do! Law-abiding citizens need guns for self-defense." (cringe). There are many other alternative strategies: increasing penalties (longer jail terms or even the death penalty), the number of police, weapons for police, community policing, drug rehabilitation, breaking up gangs, ... Wow! It's complicated. My main point is that the hypothetical NGO will probably have an implicit belief that one particular strategy is the best one. Furthermore, if you identify and question this belief reasonable debate may not follow, but it may even be claimed that you don't care about stopping murder.

Some NGOs and philanthropies have become mindful of these issues. In response to a grant application that I helped an NGO write we were asked what our "theory of change" was? I discovered that there is a whole associated "industry. According to theoryofchange.org (!)

One organisation which began to focus on these issues was the US-based Aspen Institute and its Roundtable on Community Change. ... [leading] to the publication in 1995 of New Approaches to Evaluating Comprehensive Community Initiatives. In that book, Carol Weiss, ... hypothesized that a key reason complex programs are so difficult to evaluate is that the assumptions that inspire them are poorly articulated. She argued that stakeholders of complex community initiatives typically are unclear about how the change process will unfold and therefore give little attention to the early and mid-term changes that need to happen in order for a longer term goal to be reached.  

This led to software designed to help organisations plan initiatives, with a particular emphasis on teasing out assumptions embedded in plans. A related method is construction of a logframe matrix [logical framework].

All models are wrong but some are useful. I first learnt this aphorism from Scott Page, in his wonderful course Model Thinking at Coursera.  Models help us think more clearly. Simple quantitative models, such as agent-based models, in the social sciences, have the value that their assumptions can be clearly stated, and then the consequences of these assumptions can be investigated in a rigorous manner.

What do you think? Are there examples that you think involve implicit beliefs that need to be stated explicitly?

Science begins and ends with humility

Science requires humility. Any scientific investigation starts with acknowledging ignorance. Scientific progress requires a willingness to admit mistakes and accept evidence, even when it goes against cherished and esteemed beliefs, theories, and colleagues.

As a scientist, I am fascinated by the science of covid-19, from the genetic code of the virus to the mathematical modeling of epidemics. I find it amazing how much we do know. It is also amazing how much we do not know.

The SARS-CoV2 virus is one of many coronaviruses; a name derived from the crown-like appearance of a virus particle in an electron microscope. The points on the crown are called spike proteins; they are attached to a spherical surface composed of other proteins. The diameter of the virus particle is about one-tenth of a micron. If you lined up ten thousand particles next to each other in a straight line they would be the size of a pinhead. The complete details of the atomic composition and geometrical arrangement of these spike proteins have been determined. This sphere (virus capsid) encapsulates the genetic information that is encoded in a single strand of an RNA molecule. The spike proteins allow a virus particle to attach itself to and enter a human respiratory cell. Inside the cell the virus particle bursts releasing the RNA molecule that then moves to the ribosome of the cell which then makes many copies of the RNA molecule. The information in each of these molecules is then used to manufacture the proteins that compose a virus particle. The copies of the RNA and proteins then reassemble into thousands more virus particles that then leave the host cell and move onto more cells.

It is amazing we know so much. Furthermore, we know the exact details of this genetic information. The RNA molecule in the SARS-CoV2 virus particle consists of a unique sequence of 33,000 letters (G, U, T, or C). In the laboratory, scientists can make a molecule with exactly this sequence and use these molecules to make artificial copies of the virus particles. We know so much. It is amazing. Aren’t we clever!

The flu epidemic of 1918-1920 killed more people than World War I. Back then we did not even know that RNA existed, the structure of any protein molecule, what the genetic code was, the mechanism of infection, how to mathematically model the spread of epidemics or the relative merits of different strategies for managing epidemics. We now know so much more. Today, this knowledge is saving thousands of lives.

Yet, we know so little. Although we know all the amazing details above, we cannot predict the structure of the virus particles. Furthermore, we don’t know the design of effective and safe drugs and vaccines to treat covid-19. A vaccine has never been developed for a coronavirus. This does not mean that it is not possible or even that it won’t happen in the next year. We also don’t really know how to balance the medical benefits and the economic and social costs of lockdowns.

Modeling, understanding, and describing social, political, and economic phenomena is even more difficult than physical, chemical, and biological phenomena. Scott E. Page is a Professor of Political Science, Complex Systems, and Economics at the University of Michigan. He teaches an online course, ``Model thinking’’ that has been taken by more than a million people. In his recent book, The Model Thinker, Page makes the case for using multiple models to describe human behavior.

We conclude with a plea for humility and empathy. In constructing models of people, a modeler must be humble. Given the challenges of diversity, social influence, cognitive errors, purpose, and adaptation, our models will inevitably be wrong, which is why we take a many-model approach.

Parenthetically, I note that this many-model approach is similar to the method of multiple hypotheses advocated by John Platt and that I have blogged about previously.

In 1932, Albert Einstein responded to a letter from Queen Elizabeth of Belgium, who complimented him on his lucid explanation to her of various topics in theoretical physics. Einstein wrote: 

It gave me great pleasure to tell you about the mysteries with which physics confronts us. As a human being, one has been endowed with just enough intelligence to be able to see clearly how utterly inadequate that intelligence is when confronted with what exists. If such humility could be conveyed to everybody, the world of human activities would be more appealing.

Quoted in Helen Dukas and Banesh Hoffman, Albert Einstein, The Human Side: Glimpses from His Archives. Princeton, NJ: Princeton University Press, 1979, 48.

The text above is an extract from a draft chapter that I have contributed to a forthcoming book produced with my friends in the "holy" scribblers group.

More to follow.