Contesting truth
Why can’t we all agree about science?
Re-Assembling Reality #11, by Mike Brownnutt and David A. Palmer
The nice thing about science is that it rests on evidence. Information which is publicly available, shared, and agreed upon. The basic facts of science, while they might fall short of objectivity in its purest form, at least get very close to that ideal. This is why science, unlike religion, commands universal acceptance.
OK, not quite universal. There are weirdos and wackos who don’t accept the truth that is staring them in the face. Creationists and anti-vaxxers and climate-change deniers. But anyone who kicks against the simple objective truth of science can be ignored. Science commands universal acceptance among all reasonable people.
Except it doesn’t.
However much we would like to ignore this fact, we must face it: science does not command universal acceptance. In this Essay we will take the ideas outlined in true for you but not for me (Re-Assembling Reality #10) and see how they work out in a specific example. It turns out that the more nuanced idea of scientific objectivity which we developed in Essay #10 has significant implications for understanding why science is not universally accepted, and for what we may wish to do about it.
A quick recap of objectivity
In the Enlightenment vision of science (Re-Assembling Reality #5) it is possible to obtain purely objective knowledge. This is achieved by ensuring that people stay out of the process. People come with biases and feelings and hopes and fears. As long as all social factors can be eliminated, as long as we can have sufficiently few people involved in the process, the knowledge will be objective.
As we have watched the Enlightenment vision gradually unravel, we found that it is not possible to have perfectly objective knowledge, untainted by human fingerprints (Re-Assembling Reality #7). We do not need to wallow in unbridled subjectivity, though. Instead, we achieve something approximating objectivity by a social process (Re-Assembling Reality #10). As long as we can have sufficiently many people involved in the process, the knowledge will be reasonably objective.
We will call knowledge objective if it has gone through a collective process of critical reflection which brings in many perspectives. The process should be transparent, public, and verifiable; based on publicly available, shared, agreed-upon information. To the extent all relevant parties in a knowledge community accept that a conclusion was attained by this method, the conclusions can be considered objective. If the process is more transparent, more public, more shared, and more agreed upon, by people from more different perspectives, then the conclusions are more objective.
Climate change and objectivity
Consider the statement, “Anthropogenic climate change is happening.”
There exists a lot of solid data and analysis showing that the world is getting warmer, and that human beings made (and continue to make) it happen. How can this be anything less than utterly objective?
Let us look at whether the claim’s truth can be established on the basis of publicly available, shared, agreed-upon information.
Consider a recent report by the International Panel on Climate Change (IPCC) [1]. The report takes a large amount of scientific data, and presents it clearly and concisely. It provides an introduction which expresses the key findings with bold graphics and careful explanations. The IPCC takes the public communication of science very seriously. It goes to much greater lengths than most scientific organisations to ensure that its work is accessible to all. It is an exemplar of best practice. But is that enough to save us from subjectivity? Can the truth of claims about climate change be established based on publicly available, shared, and agreed upon information?
Let us take an example:
This figure is taken from the Introduction of the IPCC’s report. The Introduction is the bit that gives you the gentle run-up, before hitting you with the really technical stuff later. For those wondering what the dots and squiggles mean, the figure caption explains:
“Estimated changes in the observed globally and annually averaged surface temperature anomaly relative to 1961–1990 (in °C) since 1950 compared with the range of projections from the previous IPCC assessments. Values are harmonized to start from the same value in 1990. Observed global annual mean surface air temperature anomaly, relative to 1961–1990, is shown as squares and smoothed time series as solid lines (NASA (dark blue), NOAA (warm mustard), and the UK Hadley Centre (bright green) reanalyses).…”
Is this analysis publicly available, shared, and agreed upon? You can download it for free from the IPCC website, so anyone with an internet connection can get hold of a copy. But holding the information in your hand is not the same as making meaning of the information in your head. If you don’t understand what the “annual mean surface air temperature anomaly relative to 1961–1990” is, the meaning that the report intends to convey is not available to you. Maybe you could find out what it means if you spent a few days reading the report or, better still, a few months studying atmospheric physics. But “available to everyone who has been able to take time out to study atmospheric physics” is very different from “available to all.”
The scientists who carried out a particular piece of research know the details of their own work. Over time they have grown to trust their apparatus and the results it gives. Why would someone who does not share their years of experience, who has not worked with the equipment personally, trust those results? They trust them because the scientists who have worked on it say “trust us.” A group of researchers writes a paper together. None of them has checked all of the work of all of the others, but they trust each other to be honest and diligent and careful. The IPCC has not checked all of the work they were given. They didn’t go in to each and every lab and ask to see the original data and check the calibrations on the original apparatus on which the data was taken. They took it on trust. Then they compiled it, presented it to you, and asked you to trust them.
This doesn’t sound very objective. Not even by the weakened standard of objectivity that we have adopted since the collapse of our Enlightenment optimism. At the heart of what should be a communal process of collective critical reflection is a group of people who do not have the time to collectively check up on each other, and who wave results through with a “yeah, I trust them.” This may “meet the epistemic standards of that knowledge community,” but people outside of that community may feel that the epistemic standards have been set rather low. Maybe we, as outsiders, should take up that critical reflective role. So let us ask some questions.
What does it mean to “harmonize” the values? Does it mean they fudged it? “No,” they say. So what does it mean? “Trust us,” they say.
What was wrong with the data that they needed to “reanalyze” it? Does that mean they fudged it? “No,” they say. “Trust us.”
And this is only three sentences, from a caption on p.131 of the Introduction. The full caption is three times that long, but the rest of the caption does not so much answer our questions as raise more. (“TAR results are based on the simple climate model analyses presented and not on the individual full three-dimensional climate model simulations.” Why? What did the full simulations show? Does that mean they fudged it? “No,” they say. “Trust us.”)
You could go and read 1136 pages of the rest of the report. And you could spend the thirty years in academia that it would take to understand the rest of the report. Except you couldn’t. Firstly because you have other things to do with the next thirty years of your life. And secondly because, by the time you had understood it all, we would have blown past 2°C of warming and have locked in even more catastrophic levels of warming to come. You do not have time to understand it all, and the world does not have time for you to understand it all. Please trust the scientists and accept their conclusions.
Do you trust them? How gullible are you? How many times has the government taken you for a ride before? How many people in power have lied to you? Who is lining their pockets from this? Sure, the scientists say they are only in it for the citations, but only a crazy person would be motivated by citations. So either the scientists are crazy or they are liars. Either way, would you trust them? They must laugh so hard at the poor dumb fools who lap up their nonsense.
Scientific conclusions are only as believable as the least believable link in the chain of dissemination. And climate science — excellent science that it is — is no different. Not everyone is at the same place along the chain of dissemination. Not everyone shares the same opinion of how trustworthy any given link in that chain might be. The conclusions of climate science are — understandably — not shared and agreed upon by all.
People who disagree with the scientific consensus on climate change are not for that reason crazy. Consider someone whose life experience shows that the people most likely to take advantage of them are politicians, elites, people that went to university, people who think they know better than you, people who want to tell you what to do, people who will take away your livelihood because the data tells them to, people who hide behind lecterns or books but never come to your part of town. If a person with that life experience were crazy, they would say, “Maybe this time will be different.” If they were rational, they would say, “When 97% of government-funded scientists agree on something, I am with the 3%.”
Objectivity requires more subjectivity
When a skeptic objects to the scientific consensus because it is the scientific consensus, throwing more science at the situation will not help. More data will not make a dent on the issue, and showing that 99% of scientists agree will only make matters worse. The science is bound up with politics, social policy, trust in humanity, even religion.
The Enlightenment solution to this problem was to purge science of all of these contentious aspects. When pure, objective science stood apart — unsullied by politics, or religion, or questions of human nature — then, and only then, would (and should) people accept science. We now know, however, that the Enlightenment solution cannot work [2]. Science cannot divest itself of politics or religion, and certainly not of questions regarding human trustworthiness. They are inextricably bound together.
Maybe, then, we can try going in the opposite direction to the Enlightenment. Rather than attempting to strip subjectivity from science, we acknowledge science’s subjectivity. We can recognise that showing someone the data is not the same as having that person understand what the data means. We can recognise that something which constitutes a convincing explanation for one person may not be a convincing explanation for another. We can recognise that reasonable people can disbelieve scientific results because they have seen corruption in politics. We can then infer that cultivating virtue among politicians can lead to greater acceptance of science. We can recognise that scientists cannot show every step of their working; that every scientific claim, at some point, is under-girded by the appeal “trust me”.
As objective as possible
Let us accept that people who are unconvinced by science’s claims on climate change may not be crazy. Let us accept that we do not need to deride them. Can we just ignore them, instead? Some people will never be convinced. And, surely, if their objection is that politicians are out to screw them over, healing such political wounds cannot reasonably be part of science.
As nice as that sounds, to ignore such people would be to miss the implications of what we mean by increasing objectivity. We can see this by following through each stage of how scientists forge objective claims.
Convincing a colleague
Imagine you are in the lab and you get a result. A dot appears on your monitor, exactly where theory had predicted. You call over a colleague to show them. They squint at the screen. They are not impressed.
“Nah. I used this monitor yesterday. I think it has a broken pixel just there.”
You check and, sure enough, there is a problem with the screen. Your personal excitement at getting the predicted result had blinded you to more mundane explanations for the dot. By getting the input from your colleague, who is not so invested in demonstrating your theory, you make your analysis more objective.
You get a different monitor. You check that it works. You run the experiment again. You get a dot, just where your theory said you should. Just where you got it last time. You show your colleague.
“Nah. Same problem. Broken pixel.”
“It’s not!” you insist. “It’s a different monitor.”
“Maybe the same pixel keeps blowing. Try another one just to make sure.”
You are fuming. How can your colleague be so dumb! The result is staring them in the face. Now you have to spend an hour hooking up another monitor and re-running the experiment. Any idiot can say “broken pixel,” then head off to lunch. You are the one that has your time wasted.
Still, they were right last time. Their intervention did make the claim more objective. It excluded your confirmation bias. Part of the scientific process is getting colleagues, even obtuse colleagues, to accept what you say.
Convincing your group
You print out your results and present them at your research group meeting. Everyone sits round, staring at the dot on the print-out.
“Did you have the power set correctly? I changed it last week.”
— “Yes.” You had noticed, and you had changed it back.
“Was it in the right range? It goes out of calibration at the end of the range.”
— “Yes. It was calibrated.” How stupid do they think you are? Of course it was calibrated!
“Did you take the magnetic field into account?”
— “Um… What magnetic field?”
Another week wasted. You go back to the lab. You measure the magnetic field and take it into account in your analysis. You adjust your theory. You notice, in checking the magnetic field, that the temperature drifts. You stabilise the temperature. You take the results again.
This is how group meetings work. You have to bat away twenty dumb questions from your group but, once in a while, just very occasionally, no matter how smart you are, it is you that was dumb, not them. And so listening to your colleagues, waiting for that rare nugget of their insight, makes your science more objective.
Convincing a conference
You put your findings in a presentation and give a talk on it. The Q&A afterwards is helpful: “Does this apply to other charge states as well?” “Is this analogous to a similar effect in neutral systems [that you had never heard of, but which you make a note to go and read up on]?”
The conversations during the coffee breaks are mixed. One person got hung up on a minus sign in one of your slides that you think is probably a typo. Someone else just laid into you:
“You can’t apply that formalism here.”
“Yes I can, because [reasons].”
“No! The paper that published those reasons was retracted because they fabricated all their data.”
“Sure, the data was fabricated. But the theory they presented was sound.”
“So now you expect everyone to comb through retracted papers to pick out the good bits? That’s a waste of my time!”
“Just because you can’t be bothered to do it, doesn’t mean no one can! The theory is sound! And I am using it!”
You could have enjoyed a nice quiet coffee, but instead you spent the break arguing with someone who apparently cannot see beyond their own interests. Sometimes we take their ideas on board. Sometimes we walk away and forget about them. Always we have another set of conversations at the next coffee break. Why do we do it? Because that is how we build objectivity in science.
Convincing the referees
You write everything up as a paper and submit it to a journal for publication. The first journal rejects it immediately because they say it is outside the scope of their publication. The second journal bounces it because the editor does not think the result is novel. The third journal sends it to three referees: One thinks it is great and should be published as is; one thinks it is rubbish and should be rejected outright; and one thinks it is great, but adds six pages of questions. Most of the questions simply demonstrate that the referee, however positive, has misunderstood what you are saying. Maybe one or two questions point to something you’d never thought about before. So although you fume at the reviewer, you have to admit that the questions lead to revisions that substantially improve the paper.
Why do we do it? Why do we go through the pain and trauma of trying to explain things to referees who are clearly idiots? It would be the work of five minutes to put the paper up on our own website. No paywall. No editorial constraints. No referees making dumb recommendations. Every hour spent arguing with journals about how to present the last set of results is an hour you are not spending in the lab getting the next set of results. Why do we do this?
We do it because of objectivity. We involve a wider community of voices, discourse with people from increasingly different perspectives, we more explicitly lay out each idea for critique. These are the very conditions we specified in Essay #10 for the process by which claims become increasingly objective. Step by step, the scientific process establishes objective positions by bringing an ever broader group of people into the debate.
Convincing the public
So let us not stop with journal referees. The wider the group of people we convince, the more disparate their views, and the more diverse their background assumptions then the more objective we can consider any conclusion on which we forge agreement.
Attempting to convince the general public that climate change is real is not science overstepping its bounds. It is not a dangerous shift from “making objective statements of fact” to “proselytizing”. On the contrary: it is through these discussions we are making — that is to say forging— objective statements of fact. This is part of the process by which science generates facts which it can hold to be objective.
In this process, the general public is not to be seen as the passive recipients of settled scientific knowledge. Rather, they are co-creators of scientific knowledge. It is in arguing with climate-change deniers — no less than in arguing with referees or arguing with colleagues — that we hone the objectivity of our position. My colleague who dismissively grunts “broken pixel” and slouches off to lunch is annoying, as is the internet troll who says “Big Solar pays the scientists to lie” and leaves the chat. They are both probably wrong. Both complaints are easy to make and hard to address. But engagement with them makes the final position more objective.
Objectivity meets pragmatism
To summarise: In order to make our scientific claims as objective as possible, scientists should engage tirelessly in public discussion of their research, not least with those most strongly opposed to them.
At this point, a little realism may creep into our analysis: This is not what scientists do. The IPCC is exceptional in their public dissemination efforts, and even they draw the line somewhere. Climate scientists believe their time is better spent taking measurements than arguing with people on the internet. Why do scientists shy back from a challenge like this? Are they not truly committed to the cause of making science as objective as possible?
Of course scientists are not committed to making science as objective as possible!
Once again, pragmatism steps in. There is a trade-off to be made. Every hour I spend arguing with peers at a conference is an hour I am not spending in the lab taking data. An argument with my peers may lead to greater insight, greater objectivity, a stellar breakthrough in which I discover my theory is wrong or (hopefully) even more right than I had previously conceived. Or it might not. And I will never know until I have had the argument. In any event, I will never convince all of my peers, so at some point I say I have convinced enough for my own satisfaction and I return to the lab.
Where each scientist sets the bar for “convinced enough people for their own satisfaction” will vary.
Some scientists consider arguing with colleagues and journals to be a waste of time, so they simply plough through, producing results, unheeding of any criticism from their colleagues. Some such scientists, unfettered by convention and consensus, generate prodigious results. Many such scientists, however, produce few major results, if any. Some are sidelined as crackpots.
Some scientists engage with as many different perspectives beyond the laboratory as possible. They talk to politicians, teachers, farmers, theologians, children, and conspiracy theorists. Some such scientists gain deep insights which transform science. Many such scientists, however, gain little, if anything. Some are commended as great popularisers. (Noting, of course, that in research circles, being “commended as a populariser” is harsh criticism indeed: what were they doing popularising things, when they should have been doing research?) Some are sidelined as crackpots.
So where should scientists make the cutoff? Or where should they strike the balance? It’s hard to say.
Maybe talking to one more person will be helpful, and maybe it won’t. But you won’t know until you have talked to them. And by then it is too late.
Maybe it is good to take the time to convince the editor of an established journal to publish your paper. And maybe it is good to save the effort and start your own journal instead.
We divide science up into disciplines that don’t talk to each other because forging agreement between disparate perspectives is hard, and there are enough hard things in science to keep us occupied, without throwing interdisciplinarity into the mix. Maybe a lifetime of narrow focus will yield great results. And maybe a lifetime of reaching across boundaries will yield great results. It is a shame that each scientist is only given one lifetime in which to work.
A choice must be made, and we will never know how things would have turned out on the path not taken.
Where do we go from here?
A major reason for there not being more agreement in science is that forging agreement takes time and is hard. It is often thought that such time and effort could be better spent on other things, and residual disagreements are a price worth paying for being able to do those other things. Of course, how much time and how much effort we are willing to dedicate to seeking agreement may change.
We don’t spend much time seeking agreement on interpretations of quantum mechanics. It’s just not that important. If you want to adopt the many-worlds interpretation, that’s fine. I think it is a silly interpretation, but it’s your life. We don’t seek consensus because we don’t care enough to make the effort. We don’t organise public outreaches on the topic, because no one needs to care that much.
We do spend a lot of time seeking agreement on climate change. Because it is that important. If you want to just keep cranking your air-conditioner higher, that’s not fine. We seek consensus because we care enough to make the effort. We organise public outreaches on the topic because everyone needs to care that much.
At the moment we draw our pragmatic line and say that, while taking time and effort to heal political divisions in the general population would help us do better science, it is not an effort that is worth making. We can do good enough science at the moment, and our time is better spent in the lab than in the community.
Maybe in future the situation will change. Maybe the fall-out from the pandemic will convince people that it is worth making the effort to understand each other and forge agreement. Maybe it won’t. We shall see.
In this essay, we have explored the implications of what we mean by objectivity, and how they play out in science. In the next Essay, we will see what it means for religion.
[1] Ulrich Cubasch et al. (2013). “Introduction.” In Climate Change 2013: The Physical Science Basis. WGI, AR5, of the IPCC.
[2] If you are new to this series and not sure why we can simply declare that Enlightenment science is not an option, this assertion rests on arguments we made in Essays #5, #6, #7, and #8.
This essay and the Re-Assembling Reality Medium series are brought to you by the University of Hong Kong’s Common Core Curriculum Course CCHU9061 Science and Religion: Questioning Truth, Knowledge and Life, with the support of the Faith and Science Collaborative Research Forum and the Asian Religious Connections research cluster of the Hong Kong Institute for the Humanities and Social Sciences.
