As COVID-19 rushes the United States yet again and the Delta variant has taken over, culture wars in America continue to throw misinformation around. Recently I’ve seen definitions of science being memed at science deniers. While I would have a general agreement with some of those definitions, there are a few details that bother me. Particularly recent memes that science is “just a process”. No – I believe it to include the products of the process, even if they may change over time.
I am also bothered by the dictionary definition of science at Merriam-Webster.com, which I find to be shockingly vague. By that definition, Big Foot enthusiasts could cite ‘observations’ that would allow them to claim that Big Foot is a scientific fact.
I have given a lot of thought to my personal view on defining science since sometime way back in college (the 1990s for me). I make no claim to say that my definition is the right definition. I’m not sure any scientist would say I’ve got it exactly right and I’ve heard more than a couple others offer up their own definitions. But often the best learning is in the journey, not the destination. So even though I will put forward MY definition of science, please hear out my reasoning, as it may have more value than the definition itself.
So here is where I’m going:
Science (noun) is the building of knowledge that is documented and verifiable.
Now let’s dissect that…
Science (Noun). Science is a noun, not a verb. The notion that science is a process captures a lot about how science works, which I’ll get back to later. But it pushes science toward being a verb. I don’t science my studies; I do studies to contribute to science. There are a number of processes involved with making good contributions to science, and there is science in those processes, but science isn’t limited to that process.
Building of Knowledge. This does bring an active sense to the meaning of science. Science is not a static thing. It is increasing with knowledge. Expanding. Like carpenters building a house, sometimes something might get placed incorrectly – an error gets built-in. But other parts of how science functions do a good job of spotting these and ensuring that the building gets revised correctly. So science is ever increasing, expanding, adjusting, and revising knowledge.
Side-note: the error correction functions of science DO NOT mean that science will make arbitrary guesses and go down the wrong road willy-nilly. Through documentation, and by limiting to what can be verified, science proceeds carefully (perhaps even slowly).
Processes are used in the building of knowledge. But there is not just one single process. Sometimes you will hear people claim that science is limited to an ‘experimental process’ with a control and a test group. That is an extremely limited view of science. It may fit okay for a few scientific fields (e.g. chemistry), but is nearly impossible in other fields. For example, chemistry does experiments with quantities of molecules measured in ‘moles’. Before that could be applied to forest ecology one ought to realize that one mol of trees would occupy the entire land surface of billions of Earths (see my post: a mol of trees). That is not to say that one cannot do experiments in forest ecology; rather, those experiments necessarily use tiny numbers (relative to a mol of molecules) and they often have numerous cofactors from environmental gradients that chemists don’t need to worry about unless working with very dirty beakers. And for many ecology studies, correlative patterns are more useful than experimental tests.
Documented. For knowledge to be passed on accurately, it must be clearly recorded. This sounds simple, but there is a lot that goes into it! Science has some established patterns that have served it well. Yes, one might be able to clearly convey some new information in an audio or video recording. But one intent of documenting things is to preserve it for future centuries. Audio and video recording formats have changed a lot more than the written word. And yes, I have gone searching after scientific papers from the 1800’s.
The gold-standard is publication in a scientific journal. Here is the basic process that one or more scientists go through after completing a study:
- Write the study up as a manuscript.
- Submit the manuscript to a scientific journal (a periodical that publishes scientific papers, usually produced a scientific society or association).
- The journal editor finds two or three peer reviewers (other scientists with similar expertise).
- The peer reviewers go through the manuscript very carefully verifying that all methods and analyses are appropriate, that conclusions use sound logic, and checking all details throughout the manuscript. This is an arduous process and yet it is almost always done voluntarily (without pay). The reward is an early view on your colleagues work, and knowing that you give into a system that you will also use when you submit your own work. Peer reviews I have done typically take a full day of work (sometimes more). Reviewers recommend to the journal editor whether or not the manuscript is worth publishing and how much additional work is needed on the manuscript, or the study itself. Reviewers also submit all their comments, which typically run over one page per 10 pages of manuscript.
- The editor collects the reviews and passes judgement. They may reject the manuscript, accept it with major revisions, accept it with minor revisions, or accept it as-is (I don’t think I’ve ever seen one accepted as-is).
- If not rejected, the authors revise the paper as needed and resubmit. (If it is rejected the authors may try submitting to another journal – While rejection may be because the work was poorly done, it also happens if the study just isn’t a good match to the journal it was submitted to. If it was rejected because it was poorly done, the authors ought to make major changes to their study or drop it entirely*.
- Upon re-submission, the editor continues to pass judgement and decide when the manuscript is sufficient for publication. They may conduct further peer reviews. And they may make a lot of ‘editorial’ changes themselves (i.e. grammatical edits).
- Once accepted, the manuscript still goes through a publication process that includes proof-reading. Many journals have “page charges” that the authors must pay. [This has become increasingly common with open-source publication of PDFs on the internet. While I love the easy access, there are now some journals that are businesses*. Traditionally journals were run by scientific societies.]
- Finally, the manuscript is published as a paper in a scientific journal with clear details for others to cite the paper and for future scientists to find the paper according to the citation.
All that said, the internet has added one more detail: the “pre-print”. These are raw manuscripts that authors post at the same time as they submit to a journal. There are “pre-print servers” available to warehouse them and for people to browse. The idea (excuse?) was that authors would get even more feedback than they would from just the 2 or 3 peer reviewers through the normal peer review process. The reality is that an awful lot of poor information has hit the news during the COVID-19 pandemic before it was reviewed at all (e.g. ivermectin). I am not a fan of pre-prints.
Another way that science can be published is with what is commonly known as “gray-literature”. These are reports published in a whole variety of fashions. They may be simply ‘whitepapers’ that the authors release themselves. They may be reports from contractors to agencies that commissioned the work. They may be reports issued by agency staff either through a local office’s process, or through an agencies formal reporting system (e.g. the U.S. Geological Survey’s Open File Reports). The level of peer review in these varies. Some (especially whitepapers) may not be peer reviewed at all. Others may go through a peer review every bit as rigorous as a scientific journal. These do get cited sometimes within scientific journal papers, but scientists understand these are to ‘use with caution’.
Verifiable. Huh? What do I mean by ‘verifiable’? This is what ultimately distinguishes science from religion, mythology, the supernatural, and even out-right fiction. You saw a ghost once, so you believe in ghosts? Well if others don’t have a way to verify your ghost, it isn’t science. It must be testable, provable, evidenced. Old wood on a mountain top in the middle east? Without hard evidence it was floated by a man named Noah or trompled by exotic animals in pairs, then it is pretty hard to claim that it is scientific proof of a story in the bible. Science could be done on it to find out how old it is, what type of wood it is, even where it came from. But it is unlikely that there is a way to know the full story of how it got there in a verifiable way.
This opens a huge space for science and religion to co-exist. Science is about knowledge that is verifiable while religion covers beliefs about what cannot be verified. Concepts of souls and gods cannot be verified in any way that science can currently conceive. These are the realm of religion. Granted, we do things now with science that people from previous centuries would consider miraculous. Physics is closing in on the Big Bang and many fundamentals about reality. Biology has closed in on evolution and is beginning to close in on the origin of life. I admit that science is expanding while the ‘unverifiable’ is shrinking. I see why some people worry about the future of religion. But we are a long way from knowing what existed before the big bang or how it was triggered. There is still much room for ideas of ‘intelligent design’ within evolution. Religion and science can coexist… even be complimentary… for the foreseeable future.
*= I don’t claim that science is absolute perfection. Indeed I have a few critiques that will be the subject of another post. But I do believe it to have sufficient checks and balances as to avoid significant corruption or conspiracy, as I argue in another blog.