Skip to content

Science is science communication

This was a comment I made in response to a post from mathematician and science writer Jordan Ellenberg:

In many ways, science is science communication. Our first audience is ourselves. It’s no joke that “writing it down” is often a key step in understanding. And anyone who’s tried to write a textbook or expository article realizes that true clarity requires true understanding. I’ve done huge amounts of research in order to write a paragraph here or there in a book. Because I wanted to make it clear to others, which first meant making it clear to me.

P.S. Given the comments, I was perhaps not clear enough above. My point was not merely that scientists communicate with other scientists. My point was that, in doing science, I spend a lot of my time communicating to myself. The writeup is not merely what it takes to get published, it’s also often what it takes for me to formulate and understand what I myself am doing.


  1. George says:

    If we take science to be a synonym of knowledge (or accumulation of knowledge), understanding is just another word of it (science = episteme = standing over). I think that failure to communicating knowledge is due to limitations in language and not to the subject’s limitations in understanding (which are always present). This is one reason mathematics are so important.

    • Abhimanyu Arora says:

      I understand you mean science communication is science in itself…true especially in social sciences where policy recommendations are to be pithier than how we researchers tend to think.

      • dmk38 says:

        Andrew– I’m pretty sure you could have conveyed the same information by simply saying “Science *is*.” Of course, *doing* science involves communicating with other scientists.

        The problem with adding “science communication” is that it might lead someone to think you mean: “Making what’s known by scientists known to non-experts is the same thing as *doing* valid science.”

        That position is something a whole lot worse than a simple mistake. Proceeding as if that were so dissipates the value of what’s collectively known. The more scientific knowledge we accumulate, the more valuable it is to have a scientific understanding of how to be sure that that knowledge is recognized & its significance comprehended by those whose welfare it can improve.

        • Andrew says:


          I’ve added a P.S. above to clarify.

          • dmk38 says:

            I have a rhetorical advantage here. For if I fail to make *myself* clear, I can treat *that* as evidence that science communication is very tricky & demands professional training & expertise.

            Your making it clear to self is important but is still simply doing. Those who haven’t made validity of their inferences clear to themselves aren’t *making* valid inferences.

            But we risk getting bogged down in semantics. My point is simply that even after you have made the validity of your work clear to yourself & to others who share your professional knowledge & habits of mind, communicating the validity of what you have done to people who don’t will involve professional knowledge & habit of mind of a different sort. I gathered the debate you were commenting on was focused on that point, although it strikes me that both sides seemed to be reaching weird conclusions about it.

            In any event, I appreciate the clarification, & even more the benefits that I am able to get as you make yourself clear to yourself!

  2. Rahul says:

    As stated, that’s almost a tautology.

    “X is X communication” probably holds true for a huge number of human endeavours.

    OTOH, I agree with @dmk38 above: In typical usage “Science communication” is taken to mean communicating it to laymen. That’s a more debatable proposition.

    As an aside, I bet there were lots of first-rate scientists who’d write rotten textbooks or expository articles.

  3. Fernando says:

    Let k be a public language for communicating with self (e.g. an arrangement of twigs, scribblings on paper, math, dags, etc) and let K be the set of such languages

    Let s be probability of scientific discovery.

    Claim: ds/dk >= 0 for some ordering of K.

    (Proof is in the pudding)

  4. […] In many ways, science is science communication. Our first audience is ourselves. It's no joke that “writing it down” is often a key step in understanding. And anyone who's tried to write a textbook or expository article realizes …  […]

  5. […] Aqui uma citação interessante que vi no blog do Andrew Gelman aqui. […]

  6. […] Science is science communication. […]

  7. […] earlier written that science is science communication — that is, the act of communicating scientific ideas and findings to ourselves and others is […]