The Script

When A Certain Uncertainty Means Certainty: The Failure of Scientists to Communicate Effectively

Clouds Blue Question Mark Red Background Sky

Clouds Blue Question Mark Red Background Sky

Clouds Blue Question Mark Red Background Sky

Bernadette De Zoysa, Contributor

Hang on for a minute...we're trying to find some more stories you might like.

Email This Story

A physicist will be quick to point out that speed and velocity are not at all the same thing even though Merriam-Webster declares these words synonymous. When an ecologist uses the word cosmopolitan they are most likely referring to the broad distribution on an ecosystem and not a sophisticated global citizen or a vodka-based cocktail.  Experts in various fields of study tend to use jargon that leaves the rest of us quite confused and pondering: what are they on about? When one is immersed in a particular field of study, it is quite easy to forget that the rest of the world does not have any prior knowledge on what you study. This leads to miscommunication between scientist and the general public.

It’s not just in science. I experienced this the first time I looked at a menu in the US. I was baffled why anyone would want to eat biscuits with bacon and eggs sunny side up. Biscuits for me meant a chocolate covered sweet treat one had with their cup of tea. Why you would want it with gravy and sausages was beyond me. Luckily for me I had American company who patiently explained that biscuit in America referred to a savory buttery scone. My misunderstanding of American biscuits is unlikely to cause global social upheaval, but with today’s heated discussion on climate change the word “uncertainty” does play into the mass confusion.

What does the word uncertainty mean to you? In general use, to be uncertain is to be unsure, to not know. One might say “I am uncertain if it would rain today”. Indeed, the weather is quite unpredictable and using the word uncertain is quite appropriate in this case. One cannot say the same about climate: many confuse climate and weather, thinking the terms synonymous but in scientific terms weather refers to how the atmosphere behaves in the short term while climate describes the long term behaviour of the same.  But what does it mean when climatologists claim that “there is uncertainty in our climate change model” and why would they say such a thing if they were trying to convince people of the accuracy in of their models?

When a scientist talks of uncertainty they mean something different. Uncertainty is the measure of how well you know something. By stating the uncertainty of their research, scientists are able to verify the degree of dependability in their findings. Which leads the rest of us to wonder why on earth would they use the word uncertainty to mean the exact opposite of what it means in English? What’s next? Are we going to measure reliability with unreliability? Natural with unnatural? Imagine that! I can see it now on a supermarket shelf. A row of beautifully stacked cereal boxed with a label that would normally read “our cereal is made of 100% natural ingredients” instead they have the phrase “Our cereal is made of 0.1% unnatural ingredients”.  Sure! That stuff is going to sell like a wonder! While it may seem counter-intuitive to define uncertainty this way, turns out scientists had a very good reason for doing so.

To understanding scientific uncertainty we must first understand how scientific modeling works.Scientists seek to  model the natural word so that they can quantify and understand it better. They model the earth as spherical, atoms as a clump of protons and neutrons with electrons orbiting it like the planets orbit the sun (which is itself another model). Scientific modelling helps scientists understand the way things interact. But it has its limitations. Models are not perfect representations of what really is. Models tend to be a simplified version of reality.

One way to improve a model is to conduct experiments and measure if the model gives reasonable results when compared to measured results. This helps us improve the model to better represent the real world. The real earth, it turns out, is not a sphere, it is an oblate spheroid which is to say it is more like a squished rubber ball, squished in at the poles. More recent measurements using GPS satellites reveal that the earth’s shape is constantly changing. These results can be used to change and further improve the model. But scientists are also aware that GPS satellites are not perfect measuring tools and the results they give have uncertainties. Even the most precise measuring tools are not perfect. And the most complex of models are only models and they don’t represent the real world perfectly. This is why scientists refrain from using the word ‘certainty’.  There is always an element of uncertainty in human knowledge and scientists always acknowledge this and try to minimize the unknowns by measuring the uncertainty.

Pullquote Photo

If we had observations of the future we would trust them more than models… Unfortunately observations of the future are not available at this time. ”

— Thomas Knutson & Robert Tuleya

Similarly, a climatologist is unlikely to state that they are certain about their climate change model. They are more likely to acknowledge the elements that cause uncertainty in their findings and then state the level of confidence in the model. They would acknowledge these uncertainties because it is true in the scientific sense of the word uncertainty. Climate change models are not perfect. And by declaring the uncertainty in their models, they are able to verify the level of accuracy in their findings to other scientists.

However, if you read or hear about this out of context it would imply that climatologists don’t really know what they are doing. This unfortunate misunderstanding due to the way science is communicated to the public has led to many erroneous counter arguments against climate change and global warming. The admittance to uncertainty in climate change models by experts is taken as evidence to fuel the counter argument.  Even though those presenting their scientific conclusions on the matter only seek to convey the degree of dependability of their results.

Scientists admit that uncertainty in climate models arise from the nature of cloud formations and their composition, from measuring tools, from the inability to predict long-term patterns of solar radiation, from the inability to predict in detail how the biosphere interacts with the climate. Many skeptics of climate change models cite these as reasons to completely doubt the predicted results. This counter argument is being heard loud and clear, so much so that we have the chief of the EPA (Environmental Protection Agency) Scott Pruitt making statements like these: “”I would not agree that [CO2] is a primary contributor to the global warming that we see,” which he said in an interview with Joe Kernen on CNBC. Yet in the same interview, Pruitt also says that “I believe that measuring with precision human activity on the climate is something very challenging to do, and there’s tremendous disagreement about the degree of impact.”s

These skeptics are partially correct. It is fair to question the accuracy of climate change models based on factors of uncertainty. Scientists who support the reality of climate change ask these same questions in order to better their model. However, doubt is no reason to believe the opposite argument to being true. If Pruitt thinks that there is doubt and confusion in climate change research then he cannot conclude that CO2 is not a primary contributor to global warming. This is not logically possible.   

Erroneous arguments like Pruitt’s are being heard by the masses. Turns out that an argument based on a logical fallacy can still change minds if it is presented with effect, much like cereal boxes would sell based on its labeling and not the product enclosed.  The label of a cereal box may read “our cereal is made of 100% natural ingredients”. But can they really grantee that? 100% is a hard thing to achieve. 100% is perfection! As odd as it sounds wouldn’t it be more honest to admit the level of error? After all 0.1% unnatural is pretty close to saying 100% natural even though it’s not exactly the same. Of course it would sell badly for this reason and no sane marketing team will look at the experimental results and publish exactly that on a label. Unfortunately, scientists are not this clever about selling their results. They tend to be more transparent about the level of uncertainty.

Regardless, Sea levels are rising, ocean acidity is increasing, glaciers are melting and plant and animal geographical ranges are shifting. These findings all have a level of uncertainty associated with them. Does that mean we can ignore the evidence and carry on as always? Most definitely not! We cannot wait until we have a perfect model and perfect measuring system to make decisions. As climate experts Thomas Knutson and Robert Tuleya say “If we had observations of the future we would trust them more than models, … unfortunately observations of the future are not available at this time.”

If predictions prove correct or even approximately correct, a 2 Celsius degree warming will set in motion a series of events which human civilization is unlikely to survive. If the models predict such dire conclusions should we not set aside our need for perfect evidence and act before it is too late? Should we not err with caution?


Further Reading:


6 Responses to “When A Certain Uncertainty Means Certainty: The Failure of Scientists to Communicate Effectively”

  1. Anahita on June 17th, 2017 12:31 pm

    Love this article!! Great job Bernie–one of the few scientists who can also write 🙂


  2. Mami Horikawa on June 17th, 2017 5:07 pm

    This article reminded me of a talk at American Chemical Society national meeting. Yvette d’Entremont, AKA SciBabe is a analytical chemist, blogger, etc and complains about how people manipulate the meanings of science articles.
    Of course there is miscommunication between scientist and the general public, but I feel there is something else, not misunderstanding, but changing meaning of what data represent or something like that. Misunderstanding sounds so innocent, but actually, it is not just misunderstanding, but evil and mean science control. Scientists can take an advantage of innocence of the public.
    Your point is good, Bernie. Experimental science cannot avoid uncertainty and other errors. I guess there is miscommunication between science and “data”? or between non-science people and the real data? I don’t know. I think you are saying scientists are not doing well on translating science data into English in an easy way that everyone can interpret science data, there is always misunderstanding about the true science problem, I mean non-science people see whats going on properly.
    It was long since I just kept writing what came up in my mind.
    Anyways, what I’m trying to say is sweet biscuits with savory bacon sound excellent breakfast for me.


  3. Victor Venema (@VariabilityBlog) on June 18th, 2017 7:08 am

    Good article, which I could have written myself. Except for the wrong and alarmist last paragraph and except for the beating up of climate scientists for how they communicate (disclaimer: I am one of them).

    Scientists all over the world speak in the same way about uncertainty. Except for America (and partially the UK and Australia) the population gets the right message. The problem is not science communication, but deep systemic corruption in America and in large parts of the English language media.


  4. Jared Ellert on June 18th, 2017 11:29 pm

    So true. It is important that people understand how hard it is to convey these ideas so they are understood and can be talked about.
    And with all the passion about this topic, its hard to talk with the opposition (whatever side youre on), much less getting through the grit to why they are your opposition in the first place.


  5. Jared on June 18th, 2017 11:30 pm

    So true. It is important that people understand how hard it is to convey these ideas so they are understood and can be talked about.
    And with all the passion about this topic, its hard to talk with the opposition (whatever side youre on), much less getting through the grit to why they are your opposition in the first place.


  6. Ike Kiefer on June 19th, 2017 12:50 am

    Signal and noise are also essential concepts for science and data analysis. NASA GISS data sets of surface temperature have systemic corrections to the raw measurements that cool the past and warm the present. The magnitude of this imposed bias is nearly 2 deg C of artificial warming applied in a hockey-stick fashion. However, this large adjustment only results in less than 1 deg C of warming since 1880. In other words, the biased correction reverses the data’s natural trend from cooling to warming. When the supposed signal extracted from the processed data is smaller than the corrections applied, and the natural trend of the raw data is reversed in the processed data, true scientists would have to acknowledge one of two things: either that their correction methods were fatally flawed or that their error bars must be large enough to allow for either trend to be true. Either case means there is no scientific basis for global warming hysteria.


If you want a picture to show with your comment, go get a gravatar.

Navigate Right
Navigate Left
The Student News Site of Foothill College
When A Certain Uncertainty Means Certainty: The Failure of Scientists to Communicate Effectively