Scientists Don't Understand Each Other, Either
One of the hardest parts of science communication is translating jargon so that people outside the field can understand. Jargon is useful because it help scientists (and other experts) quickly and efficiently communicate complex ideas within our fields. But, of course, this can make us almost incomprehensible to anyone outside our field of expertise.
But one problem with jargon is that every field has it’s own jargon, developed in isolation from other fields, and without any thought of mutual intelligibility. And this has led to the creation of something I like to call “jargon homonyms”.
Jargon homonyms are words which look the same and sound the same across different fields but have completely different field-specific definitions.
Jargon homonyms can make inter-disciplinary communication not just difficult but almost impossible, because often none of the people trying to communicate will realize that they are talking about completely different things. It’s also not uncommon for scientists in one field to hear about something in another field and, without realizing that the topic is a jargon homonym, go off on how wrong the other scientists are. And that can fuel public mistrust in science.
Which is why we need a term like “jargon homonyms” because we can’t talk about problems we don’t have words for.1
Here are some examples of jargon homonyms.
Ecological:
In biology, this means: “pertaining to the study of ecology or ecosystems”
In epidemiology, it means: “processes which happen to or are measured in groups of people”
Bias:
In statistics, this means (roughly): the difference between the expected result of an estimation process and the desired target quantity
In epidemiology, this means (roughly): the difference between the number you calculated and the actual Truth.
In sociology, this means (roughly): discriminatory beliefs, actions, etc
In fashion, this means: cutting or sewing diagonally to the grain of the fabric
Fixed versus Random Effects
In epidemiology: fixed effects are relationships that are essentially constant at a group-level, and random effects are relationships that vary from individual to individual.
In economics: the meaning of these terms is essentially the opposite — fixed effects describe all the differences between individuals within a group, whereas random effects are essentially the opposite (don’t come at me, economists, I don’t understand your weird jargon!).2
This might seem like a niche issue. After all, the whole reason we have field-specific jargon is that scientists spend most of their time talking with others in their specific fields. Does it really matter if people outside our fields can’t understand us? In fact, jargon homonyms can have life-or-death importance.
Consider the term “airborne”.
In infectious disease epidemiology, this is an umbrella category that includes all the different ways that an infection can get from one person to another via the air. That means diseases which hitch a ride on big respiratory “droplets” but don’t stay in the air very long, and it means diseases which sneak into tiny respiratory aerosols (which epidemiologists often call “droplet nuclei”) and can float in the air for hours. While we do sometimes talk about droplets versus droplet nuclei, the category ‘airborne’ encompasses both. And that makes sense for epidemiologists because the statistical modeling approaches for these are the same (just with different parameters).
But in hospital infection prevention, the term specifically refers to infections that spread via those tiny respiratory aerosols. And this makes sense for infection prevention. Because the actions you need to take to prevent transmission from larger droplet-borne pathogens and smaller aerosol-borne pathogens are different. But there’s a problem here too, because particle sizes aren’t a nice dichotomy. Particle sizes are a continuum, and the choice of where to separate droplet from aerosol is (to a large extent) arbitrary.
And then there are the aerosol scientists. Since their whole speciality is tiny things floating around on particles in the air, they naturally have a MUCH more nuanced set of definitions for all the different types and sizes of things in the air. And so they have their own special definition of airborne.
So each field has it’s own special definition of airborne, which means that when an infection prevention specialist says an infection “isn’t airborne”3 it’s important that aerosol scientists and infectious disease epidemiologists don’t assume that means the infection “isn’t airborne” under their own definitions.
Worse, this means that the public has very little hope of understanding what “isn’t airborne” means.
But as you can imagine, scientists often don’t realize a term being used is a jargon homonym. And this can lead to heated arguments.
The solution is simple: when talking about science in public, including to scientists in other fields, we need to make sure we define our terms. This could mean providing a brief explanation of what our jargon means, or it could mean spending the time to explain ourselves in everyday non-jargon language.
What jargon homonyms have you gotten into trouble with? Share your examples in the comments!
E is for Epi is written by Dr. Ellie Murray. She has a masters in biostatistics and a doctorate in epidemiology from Harvard, has published 100+ scientific articles, and is the co-author of the textbook Epidemiology: An Introduction. Dr Murray works full-time on E is for Epi to bring you the highest-quality, up-to-date information on public health news. E is for Epi is entirely reader-supported and would not be possible without readers like you. Thank you to all the paid subscribers. If you find value in this work, help ensure it continues by becoming a paid subscriber!
Incidentally, this specific issue, not being able to talk about problems we don’t have words for, has it’s own jargon. When the problems that don’t have words are specifically problems faced by particular subgroups of society, this is called ‘epistemic injustice’. The idea is complicated, but basically it boils down to this: if you don’t have words to talk about injustices you face, then you can’t acknowledge those injustices… let alone fight them.
If you want to know how economists define fixed versus random effects, Nick Huntington-Klein has an online textbook: https://theeffectbook.net/ch-FixedEffects.html
Things can get even more complicated if it turns out that, even by their own discipline-specific definition, the infection prevention specialists were wrong.
The gulf in the public understanding of science jargon such as the word “theory” in immense. For example, creationist love to call creationism another theory, just like evolution is a theory. In reality creationism is at best an unproven hypothesis under scientific method jargon. Creationist never calls creationism a hypothesis. The general public is poorly educated in basic science and will see creationism as another theory to be a plausible theory evolution.
Here's a big example: Heritability. The scientific definition includes "Heritability, in a general sense, is the ratio of variation due to differences between genotypes to the total phenotypic variation for a character or trait in a population." (Briannica). And, in twin studies, heritability estimates rely on some key assumptions, including a lack of gene-environment interaction and a continuous, not dichotomous, phenotype. (Autism is dichotomous.) But many people, including scientists who should know the definition, and almost everyone else, interpret heritability estimates as the proportion (aka percentage) of cases of a condition, such as autism, that is simply *inherited*. Worse, that it's inherited without any interaction with the environment. That understanding is obviously wrong. And there does not appear to be any public effort to dispel this fundamental misunderstanding. The wrong interpretation of heritability estimates is having massive adverse effects on even the most basic autism epidemiology.
Scientists can and should do much better.