Jan 28 2013

Up-Goer Five and Science Communication

Do you think you could communicate a scientific concept to a general audience using only the 1,000 most commonly used words? A thousand words sounds like a lot, but is it? Clearly this would not allow for the use of specialized scientific jargon, which is the point. A good science communicator should be able to translate complex science into everyday language, and use accessible analogies to make those concepts understandable.

This is something I do everyday, and not just on my blog and other social media. As a physician I have to communicate sometimes complex medical information to patients and their families. To make things more challenging my patients vary from being other physicians, health care workers, scientists or academics, to lacking a high school education or even not being a native English speaker.

Communicating to the public effectively means targeting a broad range of background knowledge. An effective science communicator should be interesting to experts while being understandable to a novice. Another challenge is to make scientific concepts simple without being oversimplified.

As a challenge to science communicators, geneticist Theo Sanderson created a website called Up-Goer Five – the concept and title was inspired by a comic by XKCD in which he explains the blueprints for a Saturn V rocket using only the 1000 most commonly used words.  The Up-Goer Five site has a text editor in which you can place text, and it will highlight every word that is not on the list of the 1000 most commonly used.

Of course I had to test it out, so I took the first two paragraphs from a recent blog post on confirmation bias. The results are below, with each violation of the 1000 rule in red and italics.

 It is my contention that scientific skepticism is an intellectual discipline and a cognitive skill set more than anything else. It is also a philosophy, a value system, and an approach to knowledge – but these are hollow without the knowledge and skills to apply that philosophy. This is especially true in our complex world, with sophisticated pseudoscience alongside mature and highly technical real science, ideologies of every stripe pushing their agenda, governments with power to protect, and markets and corporations with a profit motive to deceive. The internet is also drowning us in information, much of it dodgy. It is therefore not enough to have a generally skeptical outlook, or even to call oneself a skeptic. Skepticism is a journey of self-knowledge, exploration, and mastering the various skills that comprise so-called metacognition – the ability to think about thinking. 

Yikes. Can you imagine if I had to write this blog without using the words “science,” “knowledge,” or “information?”

This highlights what I think is the point of the humorous XKCD comic – it is, in fact, impossible to adequately explain some concepts using language that does not contain the necessary building blocks for those concepts. This leads to descriptions such as, “things holding that kind of air that makes your voice funny.”

Of course the whole point of learning new ideas is that you are building them on top of more basic ideas, so science education is always about explaining concepts with simpler language. But then, of course, you learn the language (jargon) of that level of complexity which allows you to learn the next level of complexity.

What makes Up-Goer Five humorous and not an effective guide to actual science communication is that it can lead to significant level jumping – applying grade-school language skills to college-level concepts. The inherent problem here is that language is intimately connected to conceptual understanding. If  you don’t have some notion of the concepts of “science” and “information” then it will be difficult to follow an article about science and information.

When a skilled science communicator crafts an essay for the general public they still have to decide what their target range is. You cannot write for a 5-year old and a post-graduate student at the same time. A good communicator should explicitly understand their audience – where is the lower and upper bar of the target range? If you set the lower bar too low then you will spend an incredible amount of time explaining every basic concept, and the average reader will be bored. If you set the upper bar too high then you will spend too much time on technical details. Having a feel for the sweet spot is key.

I also think that good science writing pushes the reader a little bit – enough that they are learning new concepts and words, but not so much that they feel left behind.

Obviously the 1000 most commonly used words is arbitrary and a gimmick. It is interesting, however. A more practical approach is to avoid the use of specialized jargon, but to assume a vocabulary appropriate to the target audience and the concepts being discussed.


27 responses so far

27 thoughts on “Up-Goer Five and Science Communication”

  1. dmaddock1 says:

    Well said. As an interested layman reader, good science writing should give me enough information and context to learn the key terms as I read. Your example paragraph (which is entirely italicized for me as a blockquote, btw) does this with the word “metacognition” for instance.

  2. I made the violating words red as well.

  3. ccbowers says:

    In a one-on-one or small group setting there is an initial period in which each party has to do quick assessments and adjustments in their communication that tailors the language and level of detail to the other person(s). To be honest I am not sure how good I am at doing this, but I do try.

    When I am on the other end of the communication (e.g. receiving healthcare information), I find it very frustrating when the other person cannot “adjust” the level of the conversation to one that is precise. Sometimes the person I am speaking to is not capable of doing this, but this is not always the problem. Removing so-called jargon often removes important information for those who understand it. I understand that it is a tough thing to judge as it is not just a matter of higher versus lower education, but there is also a large group of educated and/or intelligent people who may be just outside of their intellectual comfort zone. Being able to do this adjustment is an under-appreciated skill, and is vital to effective communication.

  4. There is wide gap between unusable (1,000 words) and no Up-Goer Five type filter at all. I’d love to see the same paragraph filtered with 2k, 3k, 4k and 5k most common words. An effective tool for actual science communication might be found at one of these levels.

  5. SARA says:

    I think this would be a more interesting, if you could change the range of to any number between 1000 and 50,000 of the most common words.

    What struck me at the time was that even just making it the most common 2000 words might make the rocket explanation understandable. I don’t know it that’s true, but his limits were pretty obvious in how awful it was to understand.

    On the other hand, I saw an editorial not long afterwards that said they loved how easy it was to understand. And I was wondering why it confused me and I had to think about each phrase to find a way to conceptualize it.

  6. ChrisH says:

    It would make playing for a very sucky Boggle game. I think of it because I did use a word that we commonly used at work: kip, which is a thousand pounds (I never got a chance to use slinch). None of my family believed it was an actual word, though it was in the dictionary with a different definition.

  7. I don’t think that “most commonly used” is a good criterion. There may be words that are understood by most but just not commonly used.

    I think “grade level,” or something like it, is probably a better criterion.

    For example, for the general public we might consider – high school graduate level vocabulary and without any specialized jargon. Also, you can have a higher tolerance for vocabulary if you also use a writing style that makes the meaning of words more implicit.

    So, I would say accessible writing depends mainly on those three variables – vocabulary level, use of jargon, and writing style.

  8. ChrisH says:

    On this part: “use of jargon”: it is often my habit to define the jargon at its first use (like kips and slinches). It is a habit from writing documents as an engineer where acronyms or abbreviations were defined, because PSD stands for more things than Power Spectral Density. 🙂

  9. I agree that much science writing–and writing in other specialized fields like law, politics, philosophy–is unnecessarily obscure. But to equate “simple” with “limited vocabulary” is a big mistake. Steve, you have it exactly right that learning new vocabulary is at the very core of education. Our brains build concepts the way an engineer builds machines: by grouping complex parts of ideas into assemblies that are themselves conceptually simple, and then combining these into bigger sub-systems, then combining these sub-systems into marvelously complex things that are still understandable when broken down the right way. The trick with communicating these complexities is not to make a futile attempt to wipe them away, but to understand what level your audience is at, and make sure you properly describe the path from their level to yours.

    “Jargon” gets a bad rap. I agree that some of the jargon necessary to experts is utterly useless in communicating to others. Doctors might have to use things like “dextral” or “dorsal” among themselves to avoid ambiguities; and changing these to “right” and “back” for a patient is fine. But I think it’s absolutely critical for a patient to understand, for example, the difference between a “symptom” and a “diagnosis” when they are thinking in ambiguous terms like “disease”. Hiding words like these from them hides the concepts, and lessens their ability to understand.

    One thing writers can do is simply include glossaries with their works. The web makes that simple for the reader. More tools should be available to make it simple for writers too. And yes, tools should look for unnecessary jargon as well, but they should be designed to be customizable by the writer for the particular needs of each paper.

    Another idea is to have policies in place that allow for multiple writers to adapt technical articles to different audiences. Treat such adapters like translators: here’s the original high-octane journal article, here’s a version in French, here it is at undergrad level, here it is at high-school level, here it is at politician-level.

  10. ccbowers says:

    I was surprised to see that “governments” was in red, and I thought that this word would be easily in the top 1000 for written english. It turns out that the list is based upon the usage of words in contemporary fiction, which obviously biases the words to those used in this type of story telling. The Oxford English Corpus has “government” as the 20th most commonly used noun in English writing around the world. Clearly, the list used matters a great deal.

    “Also, you can have a higher tolerance for vocabulary if you also use a writing style that makes the meaning of words more implicit.”

    Writing style is a very important component that is harder to quantify. Perhaps something like jargon density or difficult word density could be helpful, but even this does not adress the clarity of the writing. I agree that how common a word is is not a good criterion… many words are understood by children, but just don’t come up very often in conversation or writing.

  11. Aardwark says:

    I fully agree with Lee Daniel Crocker that the reductive approach to verbal communication (i.e. limitation of number of words) does not equal (or even correlate with) clarity of expression or ease of understanding.

    I am sure that what Theo Sanderson really meant to achieve with this is to give us a useful mental exercise and help us think about how we translate complex concepts into words. I do not think it should ever be used as a viable criterion of “simplicity” of content that we offer to our audience or of its “understandability” – much less as an editing tool (which is a truly horrible idea). Not even if we substitute it with “milder” versions with more words allowed.

    I do not think that any word should be excluded from use in advance or unconditionally, because it is impossible to say whether it really contributes to understanding or creates confusion, unless analyzed within the context it is being used in. Furthermore, it is quite possible to write a highly obscure text using only “simple” and “common” words, as it is to write a reasonably clear and concise one allowing some key notions and concepts to be described in less familiar terms, if these terms are the best to describe those notions and concepts – providing a proper definition is offered.

    The greatest merit of UpGoer Five, in my opinion, is that it reminds us to ask ourselves “What is jargon?” Can there ever be a list of words that are – as such – “jargon words”? Or is it closer to truth that only a particular use of a word in a particular context can be defined as ‘jargon’? In other words, same word may be “jargon” in one situation and not be jargon in another. I believe that it really depends on what we are actually trying to convey when we utter it or write it down.

    What follows is that it is impossible to build a vocabulary-based tool for “weeding out” jargon. We could try an algorithm-based one, though, but the algorithm would need to be about as complex as language itself – so I think there is no substitute for our own judgment in this matter, based on our own communication skill.

  12. BillyJoe7 says:

    Jre just used the word “revenant” on the Neanderthal cloning thread.
    (Even my iPad didn’t recognise it and tried to change it to “revue ant”)
    (But, actually, my son did recognise it – from a Internet game he used to play ?Diablo)

  13. Murmur says:

    This reminds me of a concept I think I first read in a Terry Pratchett book, but I might be mistaken on that. The name it was given was “Lies To Children” as it is a concept based on the idea that we tell children little lies about things in order to not confuse them (Q: “Where did i come from daddy?” A: “Mummy gave daddy a long hug and then 9 months later you came out of her stomach.” …. cue child not wanting to hug anything in case he makes a baby, as that is what the logic he has been given implies will happen). By oversimplifying the language we are forced to oversimplify concepts and in the end it results in oversimplified understanding that leads to oversimplified reasoning.

    I have just googled it and found out that it has another name: Wittgenstein’s ladder.

  14. BillyJoe7 says:

    “the same word may be “jargon” in one situation and not be jargon in another”

    This reminded me of the use of the word “anomaly” in climate science.
    Although, come to think of it, it’s probably jargon anyway.

  15. BillyJoe7 says:

    …I think I just echoed a murmur.

  16. Aardwark says:

    Douglas Adams nailed down a key concept regarding the “translation” of language (regardless of whether one translates between two different languages or two different “levels of complexity” within a single language) when he invented his ‘babel fish’, as a way to effectively write language problems out of the Hitchhiker’s Guide to the Universe.

    This key concept is, of course, that the babel fish is deaf to the words and unable to speak. Instead, it translates the actual meaning – at the level of the brain/mind. We may become able to create a ‘babel fish gadget’ one day, perhaps, after we learn enough about the way language-related processes are encoded in our cerebral cortex.

    But until then, I will keep taking my chances at communicating using full vocabulary. What really matters is how we use our words, not which words are ‘right’ or ‘wrong’ to use.

  17. eiskrystal says:

    In a way the list represents what we as a species think is important and at what level we communicate.

    So maybe “knowledge”, “ability” and “science” SHOULD be in the top 1000 words.

  18. ChrisH says:


    This reminded me of the use of the word “anomaly” in climate science.

    Also there is the infamous “hide the decline” and “using a trick” jargon, that was misinterpreted. Both “decline” and “trick” were not allowed at the Up-Goer-Five website. I did manage to substitute “fall” for decline, but I could not find anything to replace “trick.”

    Looking at the list, I find that it was just brute force without any editing of what 1000 words appear most often. This should explain why the are several contractions instead of just “do not”, “have not, etc.

    Some on the list are “can’t”, “don’t”, “he’d”, “he’s”, etc. (just find all apostrophes, and there are plenty).

  19. daedalus2u says:

    The reference to the babel fish points to a common error that people make about mental concepts, that any brain can hold any mental concept the way that any blank piece of paper can hold any written text.

    This is simply not the case. In order to instantiate a mental concept, a brain has to have a structure that is capable of instantiating that mental concept. Your brain has to have the “idea shaped hole” (analogous to the apocryphal “soul-mate shaped hole”, or the “God shaped hole”, or the “baby shaped hole”) so that the idea can “map” onto that structure so you are able to think about that idea.

    The brain has to self-modify its structure so that it can do the pattern recognition to recognize the mental concept that you are trying to think about. This is why it can take a long time to learn things, the brain has to remodel itself so the new concepts map onto them.

    Most of this neuronal remodeling occurs spontaneously, but it can also be consciously directed (to some extent). Usually this occurs when people refuse to allow their brains to remodel to allow the new concepts to map onto them. This is how denialists are able to persist in their denial of reality for so long. They are actively suppressing the normal brain remodeling that goes on in neurotypical individuals. People can also delude themselves and purposefully create a delusional world view. They deliberately generate a “false idea shaped hole”, so that they don’t have the capacity to perceive reality as it is. This is what YECs and AGW denialists do.

    When you accept something “on faith”, what you are in effect doing is making a “faith shaped hole” in your brain that can only be filled by whatever thing it is that you are accepting based on “faith”. Skeptics don’t do this, and people who do accept things “on faith” are simply unable to understand what not doing that is like.

  20. ccbowers says:

    “Skeptics don’t do this, and people who do accept things “on faith” are simply unable to understand what not doing that is like.”

    It seems like it should be harder for someone who doesn’t accept things “on faith” to understand what doing so is like (the flip-side of your statement). On the other hand a person who does accept things “on faith” most likely does not accept everything in this manner, so why can’t that skepticism be applied more broadly?

    I guess the ‘faith shaped hole’ you are talking about is really a broad type of ‘false idea shaped hole,’ and just like any committment to a false idea, it can be protected from scrutiny. Sorry, just talking out loud with my fingers and keyboard

  21. daedalus2u says:

    CC, I think that once you accept one thing on faith, you have to accept everything on faith, or your faith is brittle and will fail. There really isn’t a middle ground. Either everything is a miracle, or nothing is a miracle.

    Skeptics can tolerate being wrong, they just abandon those errors and move on, and know that now they are closer to what is real. What does someone with a faith-shaped hole do? Abandon faith? Which particular fath-derived idea? What do they trust instead? Their lying eyes? Then their whole faith-based edifice collapses.

  22. Aardwark says:


    “The reference to the babel fish points to a common error that people make about mental concepts, that any brain can hold any mental concept the way that any blank piece of paper can hold any written text.”

    I admit that this point initially escaped me. So, no babel fish, or at least not a universal one. But maybe a customizable model – adaptable to a particular set of mental concepts? Or a ‘mental concept esperanto’ babel fish – drawing only upon the ‘common ground’ in the mental-concept space? Or even a well-marketed BabelFish 2.0 – “Beyond Language”? Well, just amusing thoughts, really. I am actually not quite so fond of the whole ‘babel fish’ idea. After all, it is little but a narrative vehicle to justify vastly different beings from vastly different parts of the Galaxy conveniently being able to converse.

    As for myself, I much prefer to learn as many different (I dislike the word foreign) languages as I possibly can and thereby expand and enrich my neural network(s). If I had not eagerly done this since childhood, we would not be communicating in English right now, without any need for a babel fish.

    As for the notion that people who tend to accept things on faith must necessarily do so regarding any and all phenomena, this does not appear to always be so in reality. People can have multiple processing modes (or modules) and use them alternately. Voltaire made this point when he commented how amazing it is that people who solve complex practical problems in their lives every day, using highly objective reasoning, can still believe their local priest when he insists that Moon phases occur when God puts a part or whole of the Moon in His sleeve.

  23. ccbowers says:

    “CC, I think that once you accept one thing on faith, you have to accept everything on faith, or your faith is brittle and will fail. There really isn’t a middle ground. Either everything is a miracle, or nothing is a miracle. ”

    I think compartmentalization can allow for a middle ground. There do seem to be people who can protect certain ideas from scrutiny (e.g. accept things on faith), but, outside of these ideas, function in a totally different manner. I see a spectrum here, although I do think there is some clustering. There is an analogy to secular ideologies, in which skeptics are suceptible to. Although not necessarily about faith, there are emotional or intellectual commitments which can misform the “holes” in your analogy.

  24. Thadius says:

    As a student of Media and Journalism I learned that almost all journalism is written at a 5th to 8th grade level. I always found this practice annoying and detrimental to society as a whole. Most people read or consume news to learn about the world around them. Media producers on the other hand are interested in making money, and to do that they need to attract the largest audience. This means that language is geared toward the lowest common denominator. As a result, I think the general public misses out on a chance to learn more about any particular subject.

    I would argue that the “sweet spot” for communicating science is slightly above the average literacy level for that topic of the intended audience. Not so much jargon that the message is degraded for the average reader, but enough to challenge them.

    On the Babel fish, while perhaps not an accurate idea of how thoughts work, at least it gives us the most humorously concise refutation of God yet!

  25. Challenge accepted!

    Do you think you could tell about something hard to many people using only the 1,000 most often used words? A ten hundred words sounds like a lot, but is it? This would not allow for the use of hard words, which is the point. You should be able to explain hard ideas in easy words, and talk about other things that are like your things, to make your things easier to learn.

    This is something I do often, and not just on my computer place and other places. As a doctor I have to tell sometimes hard things to people and their families. To make things harder, people who come to me as a doctor are different from each other, from being other doctors, people who work with doctors, people who deal with hard ideas, to people who do not have high school or even not speaking very well.

    Telling hard ideas to people in a way that works means using the right kind of words for the right kind of people. A person who does that well should be interesting to people who know many words while being possible to listen to by people who don’t. Another hard thing is to make hard ideas simple without being too simple.

    As a way to try this out for people who explain hard ideas, Theo Sanderson made a computer place called Up-Goer Five – the idea and name comes from a drawing by XKCD in which he explains a hard drawing using only the 1000 most often used words. The Up-Goer Five computer place has a window in which you can place words, and it will show every word that is not one of the 1000 most often used ones.
    Of course I had to try it out, so I took a part of something I wrote on a hard idea about how people think. It is under here, with each bad word in red.

  26. Christopher_NW says:

    A very nice article with some good points. I remember the hardest talk I ever had to give was when I presented my research (bioengineering of entomopathogenic fungi with spider toxin peptides for controlling the spread of malaria; yes I know, a mouthfull) to a mixed audience at a research symposium last year. The audience consisted of members of various faculties including business and arts, other scientists and members of the public. The advice I was given was to make it understandable to the non-scientists, because the scientists (if interested) would come and chat to me after.

Leave a Reply