73 thoughts on ““No one is going to force you to write badly. In the long run, you won’t even be rewarded for it. But, unfortunately, it is true that they’ll often let you get away with it.”

  1. It’s interesting how much difficulty people have in in articulating how to write well.

    I once had a student tell me that her professor told her that her writing needed to be more concise. So she asked her professor how to write more concisely and the professor said “Say the same thing but with fewer words.” Apparently the professor had no idea how completely seless that advice was. The question left unanswered was how to write the same thing but with fewer words.

    My guess is that pretty much every academic has had the experience of a student showing them (trying to use gender non-specific pronouns) a piece of writing and knowing that the writing wasn’t good but struggling how to convey clearly WHY the writing wasn’t good (or how to make it better). Even good writers find that difficult.

    I think that this gets to the heart of Pinker’s confusion. He doesn’t know why academics (supposedly) wrote poorly because he doesn’t know how to clearly describe the difference between good wiring and bad writing – except to say something like “good writing is more concise than bad writing.” If you don’t really know the mechanisms behind behind what makes poor writing than you can’t explain why being an academic is (suppose) associated with writing poorly. You don’t even know the causality behind the association between being an academic, and academic writing. Perhaps it is just that academics write about topics that are more difficult to write about. Or maybe it’s just that academics write about more things that Pinker doesn’t already understand well, so he finds their writing difficult to understand. Or maybe it’s that Pinker brings a different standard of evaluation when he reads an academic’s writing.

    In order to really show that academics write poorly, he’d have to lay out a matrix of criteria to be used to evaluate writing and then classify different types of writing (by prorn fifteenth professions) according to those standards. But that’s a problem because (1) it’s a largely subjective set of criteria and (2) it’s really hard to figure out what those criteria should be anyway.

    All of that said, some people have a better handle on this than others. I can recommend a very good book on writing; Style:Toward Clarity and Grace by Joseph M. Wliams.

    In it, he talks a bit about academic writing, and how to do it well (and yes, Andrew, it’s hard to do). He even provides a pretty good answer as to how to write concisely – where he breaks down attributes of concise writing (hint: it breaks down into coherence and cohesion) quite a bit better than just “Say the same thing with fewer words.”

    • “My guess is that pretty much every academic has had the experience of a student showing them (trying to use gender non-specific pronouns) a piece of writing and knowing that the writing wasn’t good but struggling how to convey clearly WHY the writing wasn’t good (or how to make it better). Even good writers find that difficult.”

      Alternatively:

      “Most academics struggle to explain to a student exactly what the student needs to do to improve their writing”.

      :)

    • Joshua,
      Before one judges a piece as good or bad, one has to ask themselves what the goal of the writing is. Allowing the reader to develop intuitive understanding about your work is very different than allowing the reader to exactly reproduce your work. I think it doesn’t need arguing that most academic writing fails in both these respects. I think E.T. Jaynes bit about “style of presentation” in Probability Theory: The Logic of Science is supremely relevant.

    • Pinker’s confusion. He doesn’t know why academics (supposedly) wrote poorly because he doesn’t know how to clearly describe the difference between good wiring and bad writing – except to say something like “good writing is more concise than bad writing.”

      This criticism is misguided. Pinker has devoted much of a book — The Sense of Style –– to clearly describing the differences between good writing and bad. See especially the first three chapters. And if you want a single passage, look to the paragraph on page 26 that starts with “The authors of the four passages share a number of practices…”

    • “It’s interesting how much difficulty people have in in articulating how to write well.”
      Not so surprising though. Even those of us who write reasonably well haven’t learnt to teach writing. I think I’m good at criticising students’ writing in order to help them improve, but “explain how to do it better” isn’t my strength either. The best I can do is what my PhD supervisor once did for me (and from which I learnt a great deal), namely to rewrite a page of my stuff with the comment “this is how this stuff can be understood – from how you wrote it it took me ages to figure it out”.

  2. Decades ago when word processing first came in, I suggested that anything could be downloaded but whenever a student(s) was/were using his/her/their own words, those words should be handwritten. The extra effort of pen to paper would encourage reflection and if nothing else, improve penmanship.

  3. I don’t agree with the general idea that academics write poorly, nor with the idea that writing well is hard.

    Writing well takes effort and time, yes, but it’s not “hard” in the sense that it’s impossible to figure out how to do it. It’s not a puzzle with the pieces missing. There *is* a path to clear writing, even if there isn’t a cookie cutter for it. If you have the gray matter to think a thought clearly, then you must have the gray matter to write it clearly. And if you’re not writing your thoughts clearly it’s because either a) you’re too lazy; or b) you don’t understand what you’re trying to write (which is why you can’t figure out how to write it); or c) for whatever reason you don’t want to express your thoughts clearly.

    Of all the kinds of writing, academic papers should be the easiest: it’s just a technical explanation. There’s nothing to figure out. Methods: “Data was collected from first year undergrads and entered into a spreadsheet. Data was uploaded to R and processed with Random Unverified Analysis Package. Results were exported to spreadsheet. Recent work by Partypooper et. al (2016) has shown that data checking and verification is likely to invalidate results. Thus, neither were performed.” Results: “The results demonstrate discrimination, inequity, ecological collapse and, ultimately, planetary implosion.” Solution: “Billions of dollars-worth of research is needed. Chickens must be de-breeded to their pre-agricultural genetic configuration.”

    So to whatever extent academic writing is bad, I don’t think it’s because writing is hard. It might be because *thinking* is hard. But not writing.

    • Jim –

      > There *is* a path to clear writing, even if there isn’t a cookie cutter for it. If you have the gray matter to think a thought clearly, then you must have the gray matter to write it clearly. And if you’re not writing your thoughts clearly it’s because either a) you’re too lazy; or b) you don’t understand what you’re trying to write (which is why you can’t figure out how to write it); or c) for whatever reason you don’t want to express your thoughts clearly.

      So do you disagree that poor academic writing is (fairly) ubiquitous? If not – then what do you think the reason is; widespread lack of gray matter, widespread laziness, many people not understanding the topics they’re writing on, or mamy people not wanting to express their thoughts clearly?

      Gotta say, each of those explanations seem implausible to me as a widespread phenomenon, to the point where even lumping them as a grab bag to choose from together doesn’t seem very explanatory to me.

      > It might be because *thinking* is hard. But not writing.

      Seems to me that thinking and writing closely related and interconnected, but not distinct phenomena. Iow, it isn’t an either/or. For example, being a fantastic thinker isn’t a guarantee of a great writer, and there are good writers who display facile thinking in their writing.

      • “So do you disagree that poor academic writing is (fairly) ubiquitous?”

        No. I don’t agree with that.

        “Gotta say, each of those explanations seem implausible to me as a widespread phenomenon”

        So you think there are lots of people who have great ideas but just can’t communicate them? To me that seems a lot more implausible than the causes I suggested, particularly given the close link you suggested between writing and thinking. I agree that there is a close link, and that’s why muddled writing is a reflection of muddled thinking.

        “and there are good writers who display facile thinking in their writing.”

        There aren’t any good writers who display “facile thinking”! You don’t have to be a good writer to write facile thoughts.

        • Following up: if a person is a whiz at adding single digit numbers in their head, would you say “s/he’s a great mathematician!” No. See?

  4. In my experience, good writing depends heavily on good thinking, and bad writing often results from fuzzy thinking. This is of course not the only factor, but clear thinking is essential. And clear thinking is often too hard to find.

    Also, this is a problem, at least in my field. I am astounded with how few researchers read the literature, depending instead on conference presentations and offline chats. With the covid effect on conferences, I’m not sure where researchers will get information.

    • Joe –

      > In my experience, good writing depends heavily on good thinking, and bad writing often results from fuzzy thinking. This is of course not the only factor, but clear thinking is essential. And clear thinking is often too hard to find.

      So do you think that poor writing is particularly common in academe? If so, do you think that’s because poor thinking is common in academe or do you think it’s more because of those unspecified other factors?

      • Yes: poor thinking is common in academia. Academic research also suffers from The Hero Complex, bias, monoculture, insularism, arrogance, relentless PR and money quest, among other problems. L

      • I like the good/fuzzy thinking separation here, but it might be less a quality of a person’s thinking per se but rather the thought process they present in their writing.

        I heard a seminar on writing by Larry McEnerny on YouTube that started out describing the different goals of academic writing. We initially write to clarify our own thinking about our research. However, that approach of “here are all of the steps I took and mistakes I made along the way and here is the clever result I took away from it” gets in the way of a reader understanding what you’re saying because they don’t care/aren’t in the same mindset that you were in when you were figuring this stuff out.

        I’d say poor academic writing is common in academia because our first drafts typically ARE records of our thought processes as we summarize an analysis. We’re focused on the topic while we write–causing our ideas to evolve over the course of writing out a draft–and it takes careful (and aggresive!) editing to ensure that the whole document reflects our latest ideas on the topic. Readers want to know your latest thoughts and their implications, not be led through your thinking process.

        Academic writing will stay largely low quality for the same reasons that low power studies will continue to be dominate many fields: it takes much more time and effort to make high quality stuff (experimental design, writing) and the advantages of high quality > low quality don’t outweigh the advantages of more stuff > less stuff. Paul Smaldino’s and Richard McElreath’s “The natural selection of bad science” illustrates this point with respect to research design/data analysis. I think a similar situation influences academic writing, aided no doubt by our lack of explicit training in writing.

  5. I think that one reason academic writing is often bad is that teachers often don’t stress good writing (or even describe what they mean by it). I have tried to in include a “Guidelines for written homework” section in the first day handout for all classes I have taught. Here is an example from an introductory statistics course:

    “Guidelines for written homework:

    1. Remember that one important purpose of written homework is to practice thinking statistically and to show me how well you have progressed in your thinking. Be sure to show your reasoning — I can’t evaluate it if you don’t show it. And keep in mind the following quote from the instructor’s manual for our textbook:

    If we could offer just one piece of advice to teachers using IPS, it would be this: A number or a graph, or a formula such as “Reject H0,” is not an adequate answer to a statistical problem. Insist that students state a brief conclusion in the context of the specific problem setting. We are dealing with data, not just with numbers.

    In particular, just handing in computer output is not satisfactory. You will often need to include part of the output with your solution, but you need to explain how it helps you solve the problem.

    2. Do not hand in a rough draft! Be sure to spend time organizing and writing your solution. Ask yourself if you would like to read your write-up. If not, rewrite it! Part of your grade will be based on clarity of organization and explanation. After all, communicating well is part of thinking well — and making the effort to communicate clearly is an important way to develop your thinking.
    Do not hand in extra computer output. Cut and paste (either by hand or on a word processor) so that figures and computer output come as close as possible to the point in your discussion where you refer to them. In some cases, writing on computer output (especially printouts of graphs) will work.
    Reminder: Answers in the back of the book are summaries, condensed to fit in as little space as possible. Do not use them as models for written homework.

    3. Write in complete sentences.

    4. Pay attention to correct use of vocabulary. You will be learning technical vocabulary in this course. Part of what you need to learn is to use it appropriately. Be especially careful of what in language learning are called “false friends:” words that are familiar, but have a technical meaning that is different from their common meaning. “Significant” is one example of such a word.
    Also be careful not to use mathematical vocabulary inappropriately in a statistical context. In mathematics, we can often prove an assertion. In statistics, we can usually only conclude that our result supports, suggests, or gives evidence in favor of a conclusion.

    5. Use symbols correctly. One symbol often misused is the equal sign. Do not use it except to mean that the two things it is between are equal!!

    If you introduce a symbol, be sure to define what it means. Common ways to do this include:

    Let µ be the mean of X.
    Denote the mean of X by µ.
    Let µ stand for the mean of X.

    Be careful not to let the same symbol stand for two different things in the same problem. Subscripts can often be used to avoid this confusion.

    • Thanks for this! You wrote:

      “Be especially careful of what in language learning are called “false friends:” words that are familiar, but have a technical meaning that is different from their common meaning.”

      Do you, or anyone else, have a concise way to “phrase for replacement”? So, instead of carefully explaining statistical significance every time, rather pointing to an alternative term? Same applies to other false friends from the technical literature.

      In addition to students “needing to learn” these terms, is there a way we can convey how they might start replacing this language?

      • I’m skeptical that “phrasing for replacement” is a viable solution to the problem. I think that a more pragmatic approach is to teach awareness of the problem. For example, instead of trying to “replace” the phrase “statistical significance”, emphasize that the word “significance” has different meanings in the phrases “practical significance” and “statistical significance” — in each case, the adjective truly (strongly) modifies the noun — in other words, the adjective *changes the meaning of the noun•. This, I believe, is the more general concept that students need to learn — that context may change the meaning of a word, sometimes very strongly.

        • I’m also also skeptical it would solve the problem in the short term, but perhaps it would help perturb the system a little. I have in mind the proposal to use s-values, as a measure of surprisal [1].

          I agree with you that it is good to teach awareness of this, but it seems that it is difficult enough to teach statistical reasoning as it stands. Why add the extra barrier of confusing or indirect terminology? Why not demonstrate to students and our colleagues that we (collectively) are working to make the communication of results easier, or at least less confusing?

          Not really sure what my semantics colleagues would say in general about adjectives changing the semantic meaning of a noun phrase (they tend to not always agree), but I believe in psycholinguistics we would say that the lexeme “significance” has at least two lexical-semantic senses [2], one roughly paraphrased as “importance”, and the other being the technical frequentist-statistical sense. The adjectives help select which sense is intended.

          1. Rafi Z, Greenland S. Semantic and cognitive tools to aid statistical science: Replace confidence and significance by compatibility and surprise. arXiv:190908579 [stat]. July 2020. https://arxiv.org/abs/1909.08579

          2. WordNet gives me three senses for “significance”, all unrelated to the statistical usage, but the differences are subtle, and probably don’t concern us here.

        • Thinking about “phrasing for replacement” a little more:

          I could accept replacing “statistically significant” with “statistically detectable” or “statistically discernible”, BUT with the proviso that the phrasing be “statistically detectable with the data and model used” or “statistically discernible with the data and model used”. I think the phrase “with the data and model used” is important, to emphasize that “statistically detectable/discernible” is NOT an inherent property of the question being studied, but that the results of the analysis depend on the data and model used to study the problem — different data and a different model might give different results. Leaving out the data and model suggests a false implication of certainty.

        • Totally agree. There is a real trick here to put the emphasis on the collection of data, model, method and assumptions without losing people. And I am pretty sure I don’t know the trick.

  6. Inger Mewburn: Hey, maybe we should make academy less elitist?

    Thomas Basbøll, exact quote: If you can’t find a way to like writing academically, I honestly think you’re better off finding something else to do with your life.

    Who else we should through away from academia? Obviously, people who don’t like doing statistics, they should be the first to go. People who dont like writing beautiful code? No place for them. People who dont like making informative graphs, presenting on the conferences? What about all these poor souls what dont like writing grant applications? Lets face it, uncovering the secrets of the universe is not for them.

    • I think it would be more fair if you either quoted us both or paraphrased us both.

      For example, you could quote her (as I did) as saying, “It’s highly likely you will graduate [from your PhD program] a worse writer than you started. This is because we spend a lot of time teaching you how to write in a particular ‘academic’ style that, not to put too fine a point on it: sucks.”

      In that context, my tone doesn’t come off quite so aggressive.

      Or you could paraphrase me: Hey, maybe spend some your time as a PhD student learning to enjoy writing.

      Seriously, writing is an essential part of academia. And, though it may sound odd, “uncovering the secrets of the universe” actually isn’t. You can be a really good academic and never really uncover a single “secret”. In some fields, statistics is an essential craft, as is making informative graphs. If you don’t enjoy them, those fields won’t make you happy. You’ll be doing a lot of it and it’ll make you miserable. Find a field where it’s not so important.

      Conference presentations and grant applications … interesting point. Some people never come to “like” these things. But I don’t think that should make them seek new careers. They are chores that can be managed — they can suck for a few hours a year — and you can still love your job.

      I (“honestly”) think writing is different. Along with reading, it occupies so much of your time, and you invest so much of your professional identity in it, that if you (honestly) think it “sucks” you may want to uncover the universes secrets in some other, non-academic field. Academia doesn’t have a monopoly on the context of discovery. In fact, it’s a bit elitist to assume it does!

      • Thomas,

        If I was paraphrasing you, I would have written something like “If you dont like writing under the current system, just leave”. So you actually ended up less aggressive in your own words =)

        I do agree with most of your points. I see that writing (and reading) is a essential part of the profession. If a tree falls in a forest and no one one writes a paper, it’s not science. I agree that academic writing style is about clarity and transparency, not about in-group signalling.

        You point is that academic writing, when done correctly, does not suck. But researches now, especially PhD students, are not required to write correctly, whey are tortured into producing a required amount of text to fit into a certain metrics. Being able to love wring is, unfortunately, a privilege, not a skill to be learned.

        Another point. Setting aside the problem with academia, science is a massive profession now. But we still expect every scientist to be excellent if everything: writing paper, teaching, leading research lab, applying for grants… These are all important task, and these are all hard tasks. Maybe you can still contribute to the community in your own way, and do academic writing (which is important!) with the help of you colleagues and lab mates? Not everyone born the same writer, some have dyslexia or dysgraphia, some may struggle with English.

        • That’s not really a very good paraphrase of what I said though, is it? Here it is with some context:

          “Writing is a big part of a research career. If you don’t enjoy it, you’re going to be spending a lot of time not liking your job as a researcher. If you can’t find a way to like writing academically, I honestly think you’re better off finding something else to do with your life. But don’t give up right away. During your graduate studies, and especially your PhD, spend some time learning how to enjoy writing. Insist on finding joy in the act of sitting down at the machine and addressing yourself to your peers.”

          We disagree about how torturous academia is and about whether enjoyment of writing can be learned. I will continue to tell people that if they can’t enjoy being PhD students they’re probably on their way into the wrong line of work (a lifetime of “torture”) or, more likely, the wrong field. My main advice is for them to find readers that don’t make them suck at writing. It’s hard and they should stick to it. I would never tell them to “just leave”.

        • There are two aspects to this I think:

          1) Communicating effectively and teaching people about your thoughts and how the world works

          2) Satisfying the gatekeepers of the publication world you work in.

          I think far too often people find these two at odds with each other, and that’s what’s considered the problem with academic writing.

        • I agree that many people consider this a big problem. And the tension between these two goals is of course sometimes real. But we have to insist on satisfying ourselves even as we satisfy our gatekeepers. That may be a bit harder. And, like I say, gatekeepers will often let you get away with writing less than well as long you’re doing what they say, but it’s worth the extra effort even if it’s not strictly necessary to get you published.

          Don’t let the gatekeepers ruin your relationship to your end reader. That’s what I was trying to say in my post. Don’t let them make you think your writing has to be bad. When it’s done, it has to look good to you. Letting them get you to write badly isn’t much better than letting them make you say something you don’t think is true.

        • If what is accepted as academic writing is pretense of causation where it has not been established, then the style of academic writing within that discipline is a problem that will not be solved by encouraging them to take their writing style and go somewhere else. It is one of the fundamental problems in a number of disciplines.

        • Yes, one the major causes of bad writing is trying to say something you don’t actually know is true. (Or, worse, knowingly saying something you think is false.) Bad writing is often “false” (dishonest) because the writer is being ambiguous where they could be clear, or clear where they should be ambiguous.

          If you feel you are being “forced” to do this — i.e., if you feel pressured into making stronger claims than your research justifies — then I do think you should switch specialty (or in some cases discipline). You’re not going to enjoy writing that way for a living.

  7. Here’s a philosophical point that I think is important. Many people think that science is about being right. That it is about thinking and making statements about reality in a way that they reflect correctly how reality really is. But I think science is something between people in the first place. It is about communication and convincing people. I believe that ultimately science aims at absolute agreement in free exchange, meaning that a statement that should be scientifically “true” can be accepted by everyone without being forced on anyone by power (this is obviously an idealisation; there are always people who refuse to be convinced – however science is about trying as hard as we can to convince – and being open to be convinced). There is no way to do this without caring for communication.

    I’m not sure how directly related this is to bad writing, however I come across many scientists who think they should be right in the first place and who communicate only because they have to. They may be willing to say that and why they are right, but are often much less willing to anticipate a critical or opposing point of view and to figure out what would be necessary to reach and convince somebody who takes such a point of view.

    Also, for example from what students ask me, I get the impression that many think scientific communication is about fulfilling some kind of ritual; saying things that have to be said because they have to be said as part of a ritual, rather than understanding how things need to be said in order to be able to convince somebody critical.

    • Well put. I think for me a big part of the appeal of Bayesian methods is that they provide a mechanical way to go from assumptions to (probabilistic) consequences. IF I accept the premises, then I must accept the conclusions, and also if I don’t accept the conclusions then I should look more carefully at the premises because there must be something I don’t actually accept.

      I think the backwards implication is the part Andrew was frustrated that the dogmatic Bayesian of 20 years ago wouldn’t accept

      • Yes, a major use of such models is that they help communicating a point of view, even if this point of view then turns out to be problematic or wrong.

    • > Many people think that science is about being right.
      Another perspective is that the assurance of science is simply that if research is adequately persisted in (one part of adequately being trying as hard as we can to convince widely – and being open to be convinced by others, without excluding participation) – the ways in which one has perceived reality incorrectly will eventually be corrected.

      Of course, that does not suggest one will ever get reality correct (and realize that) at least in finite time.

      • Yes, a scientific style isn’t just one with an air of rightness but one that is open to being wrong. By laying out clearly what you think and why you think so, you’re telling your reader all the things you might be wrong about. If your readers have reasons to think you are, in fact, wrong, they now have an opportunity to adduce those facts.

        If you write in the spirit of hiding your ignorance from your reader, or even in the spirit of just “sharing what you know” (as if you couldn’t be wrong), you’re not writing scientifically. It’s bad style.

        • > spirit of hiding your ignorance from your reader
          Gee, certainly how I started my career and I think it was a criticism David Cox had about Fisher( e.g. very convenient for Fisher to pick an example where its not clear regarding an possibly open question that more generally would need to be addressed).

          Writing scientifically requires maturity and being comfortable discovering how you are wrong. The value of that likely depends on your ability to redress it. Early career researchers likely don’t have that…

        • “Writing scientifically requires maturity and being comfortable discovering how you are wrong.”

          Is it that one needs to be comfortable discovering how they are wrong? Or is being willing to put up with the discomfort because the end is worth it? I think either would work. (e.g., women put up with a lot of discomfort during pregnancy and childbirth.)

  8. It strikes me that a basic knowledge of grammar, in a formal, old fashioned, sense, might also help. Understanding the elements of a sentence and how a sentence can be constructed might be useful.

    A friend of mine in history can be heard giving cries of anguish when marking papers. His undergraduates like to write long (20 words +) sentences but have never learned how to structure them.

    If one cannot construct the basic building blocks, it may be a bit difficult to assemble the final product.

    • I love this idea. It is exactly the same with students learning to write formulas – they tend to write long formulas that try to do everything, rather than breaking it into pieces (building blocks). So, good writing and good modeling involve the same kind of critical thinking skills. I never liked these dichotomies such as right brain/left brain, math vs. english, etc. It’s all really good critical thinking skills.

      • This reminds me of when I started as a freshman at The University of Michigan many years ago. Two semesters of Freshman Composition were required — but if you got an A or a “starred” B, you didn’t have to take the second semester. A friend a little older than me *whom I had met at an NSF summer program for high school students in math and science) told me that he was sure that I would not be required to take the second semester — because the good math majors always got excused from second semester freshman composition. (And I did get an A in the class, so was indeed excused from the second semester.)

    • Yes, I think grammatical correctness has been neglected in writing instruction. And by students themselves. Of course, that would quickly change if your historian friend graded according to his anguish. (Maybe he does, I don’t know, but many of his colleagues don’t.) Too many students who aren’t writing well are getting As and Bs. If they got the Cs and Ds they deserve both they and their English teachers would get their acts together in a hurry.

      (Aside: if teachers were less inclined to see a C as a poor reflection on their teaching and instead saw them as an accurate reflection of the student’s performance, none of this would be as painful for anyone. It would be easy to give a C if half the students got that grade or less.)

      In any case, along with sentences, we need to teach students how to compose coherent paragraphs, and how to arrange these intelligently into papers and essays. In an important sense, paragraphs, not sentences, are the buildings blocks of academic writing. (Writing them requires mastery of a range of sentence forms, of course.)

    • I read an intersting study – I can’t now come up with a link – that focused on instructional methodology for improving writing in non-native speakers of English.

      One group was given a course of instruction in grammar.

      Another was given a course of instruction in “meta-cognition” – in other words they focused on executive function as it relates to writing.

      In that study, a comparison of pre- to post-intervention assessments showed that the meta-cognition intervention resulted in more improvement in the students’ writing.

      > In any case, along with sentences, we need to teach students how to compose coherent paragraphs, and how to arrange these intelligently into papers and essays. In an important sense, paragraphs, not sentences, are the buildings blocks of academic writing. (Writing them requires mastery of a range of sentence forms, of course.)

      Seems to me that would align with the meta-cognition approach. Not that it needs to be an either/or zero sum choice – but I would suggest what you describe there is a “higher order” skill and as such is more likely to have a greater effext more of the time with more writers, as reflected in that study I described.

      • Joshua,

        I’d be interested to see that study if you remember it at some point.

        I’m wary about the whole apparatus of “meta-cognitive” writing instruction. I prefer to use the vocabulary of rhetoric (audience, purpose, etc.) but I think you’re right that thinking about the composition and arrangement of paragraphs moves us into the territory of “executive function”.

        • Thomas –

          > I’m wary about the whole apparatus of “meta-cognitive” writing instruction.

          I’d be curious to hear about why.

          > I prefer to use the vocabulary of rhetoric (audience, purpose, etc.) but I think you’re right that thinking about the composition and arrangement of paragraphs moves us into the territory of “executive function”.

          I also think that the focus on rhetoric, and standard rhetorical devices is also important, and I think that rhetoric is located along a spectrum…but it’s also, IMO, a sub-category of a metacognitive approach.

          Primarily, as an overview, I”m referring to such things as simple as understanding the importance of putting the writing down for a while and then coming back and re-reading with a cleared mind, or creating a good thesis, or the revising, or understanding who the audience is, etc. Process stuff. Understanding conventions of rhetoric is a piece of that.

          As another side-note, rhetorical conventions vary by culture. Some cultures have very different conventions about the function of writing. For some cultures, the hierarchical structures we prefer for academic writing, which stress clarity and being explicit, are considered insulting to the reader. The idea being that you should not make your point clear and even repeat it in different ways because it’s like saying that the reader is stupid and cant’ figure it out for themselves. For other cultures, creating a narrative have a higher level of importance in expository writing. For others, (such as the French), more explicit and clear writing might be considered “ugly.”

          One piece of the issues with academic writing, or writing in the academe, is that many times the people who are writing are coming from a different rhetorical culture and have never actually worked with understanding the rhetorical conventions of expository writing in English.

          Ok, and another side-note. There is a similar pattern with reading. Most academics I’ve talked to apply a “strategy” for reading (this goes back to the notion of metacognition). They don’t just read an article from the first word to the last, looking up all the unfamiliar vocabulary. They may briefly scan the abstract and the introduction, basically skip to the results and discussion, and then after that read the rest of the article selectively (also applying a practical algorithm) as needed only to the point where they can match their previous understanding of the topic with what they can learn from the article. They might do something like scan charts and tables, or read first and last sentences of select paragraphs, or scan through section headings.

          The reason why many academics read that way is because of the embedded organizational structure of (what is usually considered good) expository writing – where the intro serves a specific purpose (i.e., to lay out a “roadmap” for the organizational structure and introduce the thesis0 and the discussion and conclusion serve a specific purpose and the organization of paragraphs serves a particular purpose, and then the organization of individual paragraphs (like a fractal, reflecting the larger overall structure of the piece of writing) serves a purpose.

          But if you look at how many students read – even graduate students at prestigious universities – many don’t have a particular strategy for how they read. The don’t have a “metacognitive” concept about the act of reading. That is particularly true of non-Americans.

          Well, enough proving that I don’t know how to write concisely!

    • I mentioned at some point recently on this blog that my best writing instruction—better than anything I got in college, grad school, or professional life—was my high school English teacher Mrs. Fleischaker.

      While I’m sure she did a lot of subtle things that I still don’t understand, I can think of two big tactics she used:

      1) Frequent graded, timed writing exercises. At least once a week, the class period was given over to writing an essay (by hand, of course!) in the allotted time. She had us learn a set of shorthand notes that she would use to mark all of our essays individually and get us rapid feedback.

      2) Stylistic analysis. A focus of the class—and a frequent essay topic—was reading a passage and analyzing why the writer made the choices they did. Is there parallel structure, assonance/consonance, rearrangement of subject/verb, etc.? Why? Did it work?

      With regard to 1, writing is a skill and needs practice. But the rapid feedback, in combination with 2, meant that the practice was directed—in trying to understand why authors used certain writing devices, we were better able to use them ourselves to communicate concisely.

      That said, this kind of experience was very intense and I can’t see it being slotted into a Ph.D. program as an “add-on”—it really deserves its own focus. And with the current emphasis on multiple-choice assessments in grade schools, teachers are not supported for doing this kind of thing. But it definitely gave me the building blocks so that I could at least make the best of my subsequent sub-par writing instruction.

      Of course, I readily admit I don’t apply everything I learned from Mrs. Fleischaker as well as I could!

      • gec –

        > 1) Frequent graded, timed writing exercises. At least once a week, the class period was given over to writing an essay (by hand, of course!) in the allotted time. She had us learn a set of shorthand notes that she would use to mark all of our essays individually and get us rapid feedback.

        {> With regard to 1, writing is a skill and needs practice. But the rapid feedback, in combination with 2, meant that the practice was directed—in trying to understand why authors used certain writing devices, we were better able to use them ourselves to communicate concisely.

        While I don’t doubt what worked for you, and what you prescribed seem logical… there is a fairly extensive literature on whether/how much corrective feedback (from others) helps students (on average) learn to write better… and the results are pretty mixed.

        • there is a fairly extensive literature on whether/how much corrective feedback (from others) helps students (on average) learn to write better… and the results are pretty mixed

          I wasn’t aware, and that literature sounds important. I’ve found a bunch of citations, but the large majority are in the Journal of Second Language Writing. Can you recommend a study that involves students being taught in their first language?

        • John –

          Seeing as how you properly noted that I was being incautious above, I should take a step back here as well.

          It has been a while (say 15 years?) since I’ve looked at the literature – so I don’t have a set of citations to provide. Perhaps the consensus in the literature has changed since then, but my recollection from the time that I looked at this pretty closely is that there was far from a convincing consensus – even though some people felt strongly on either side of the debate.

          My own view is that results would necessarily depend on instructional methodology and the interpersonal nature of the interactions between individual students with different approaches to learning and different teachers with different approaches to teaching.

          That said, I gravitate towards s more “meta-cognitive” approach – and that applies to working with non-native English speakers as well as native English speakers. Anecdotally, I worked with many American students in graduate school who had never taken ownership over evaluating their own writing. They had handed in papers for years and gotten a lot of correctional feedback (largely through the messaging vehicle of grades) but still had little independent agency as the “executive” for improving their own writing output. Writing was largely a black box to them. It actually was a related sort of mirror effect from the same underlying problem that manifest as them having no meta-cognitive concept in approaching their reading.

          Of course, that could have been because they had been taught grammar and sentence-level writing poorly, or maybe even not at all. But I don’t think that was the entire explanation. For a lot of them it was also because they looked at writing as an exercise in getting it right or wrong and when they were told they had gotten it wrong (without having an autonomous sense of how to even evaluate their own writing) many were inclined to just decide that “I’m not good at writing.” This was particularly true for many of the students in the hard sciences that I encountered – in part because many of their professors in the hard science did not think it was within their area of responsibility to help students become better writers.

          It is particularly well-outlined with many non-native English speaking writers writing in English, who often feel that all they need to do to write well is to correct mechanical errors such all the grammar and usage errors (true of many Asian students in particular). In part that is a reflection of different cultural viewpoints on the purpose of writing; in some cultures the point is more to reflect back exactly that which was given to them in instruction (which would fit a right/wrong purpose) whereas in other cultures the emphasis is more on the student extending what material they were given to demonstrate an individual-level mastery.

          But I look at that as more of a distinction in an overall phenomenon rather than a completely different phenomenon all-together.

          I did just ask Mrs. Google, and she pointed me to this:

          https://www.tandfonline.com/doi/abs/10.1080/01411920500401997

          an abstract which I think fairly well reflects the understanding I gained with I was looking at this more closely. I will note that I consider sentence-level work on a scale that runs from grammar and usage, to sentence-level, to paragraph level, to larger organizational structures. That spectrum kind of parallels a spectrum that I see as running from a more didactic approach to teaching writing (such as teaching proper grammar) on one end, and an approach that emphasizes meta-cognition at the other end.

          But I’m not unaware that this debate is quite complicated and taking an either/or approach is probably mistaken.

          Sorry for the mini-novella. But be grateful it was only this long. I could go on for days on this topic.

        • The feedback Fleischaker gave was explicitly meta-cognitive in the way you describe, and was directed toward the idea of “ownership” that you describe. That said, I understand that the term “feedback” in learning sciences often refers to right/wrong or grading, so let me elaborate.

          First, the feedback itself was very concise. I mentioned we had to learn some shorthand for that to work, but I think that’s important because verbose feedback can overwhelm students and lead them to tune out. Too much commentary can also impair a student’s ability to map the feedback onto the elements that need work.

          Second, the feedback was about comprehension. Something would be flagged if it led to confusion. Sometimes this would catch typos/grammar errors, but that was not the purpose. Moreover, it did not enforce a style nor was it focused on “getting something right”. This invited you to think about how to improve the passage.

          A third tactic I forgot to mention, but which makes feedback particularly valuable, is that you could re-do assignments. That way you got to see whether your proposed resolution really worked.

        • gec –

          > The feedback Fleischaker gave was explicitly meta-cognitive in the way you describe, and was directed toward the idea of “ownership” that you describe.

          Yah. I actually noted that in your earlier comment and was remiss in not mentioning something about that.

          Your details in follow-on comments are very much in line with what I was going for.

          > Too much commentary can also impair a student’s ability to map the feedback onto the elements that need work.

          That, in particular!

          > A third tactic I forgot to mention, but which makes feedback particularly valuable, is that you could re-do assignments. That way you got to see whether your proposed resolution really worked.

          That also.

          A key piece of what I”m describing with a “meta-cognitive” approach is understanding how crucial a revision process is in producing effective writing. That might be the key stumbling block I came across with many students, who thought that they could produce effective writing without explicitly building in a revision process. I often found that students didn’t actually find out what their thesis was until the very end of their process of writing an assignment – after which what they needed to do was consider what they had thus far produced as a rough draft or even thinking through exercise, and then take the thesis they established at the end of the draft and revise with that thesis as the central organizing concept. But because they were on a deadline with a due date (and they were probably right up against that due date) the re-staring with a clear but new thesis was a non-starter.

          I think what set me off was your mention of “Frequent graded, timed writing exercises.” That makes a lot of sense within a larger frame where assignments could be re-done after incorporating specific and focused feedback.

        • Thank you, Joshua. I’ve had some similar experiences with my own students. And the article to which you linked seems interesting; I’ll check it out.

          They had handed in papers for years and gotten a lot of correctional feedback (largely through the messaging vehicle of grades) but still had little independent agency as the “executive” for improving their own writing output. Writing was largely a black box to them.

          Perhaps this passage deserves emphasis. Grades alone are nearly useless feedback in this context. And the sort of training that we’re discussing, whether “meta-cognitive” or narrow and concrete, takes a huge amount of time. Little wonder that we see so little of it at any level of education.

        • Joshua said,
          “Anecdotally, I worked with many American students in graduate school who had never taken ownership over evaluating their own writing. They had handed in papers for years and gotten a lot of correctional feedback (largely through the messaging vehicle of grades) but still had little independent agency as the “executive” for improving their own writing output. Writing was largely a black box to them.”

          Another anecdote: When I was in college, I signed up for a program to “tutor” kids from poor families. I was assigned a nine-year-old girl, from a family with eight kids (and no father present) who (all except one child who was institutionalized) lived in a garage apartment. We met weekly at the local library. One week she showed me something she had written in school. (The class had visited a zoo and the assignment was to write about their visit). I told her (in a matter-of-fact way) that when we write something, we read it over to see if has good spelling and grammar, and if it says what we want to say. So she started reading her essay, and very soon said something like, “Oh, this part doesn’t really say what I was trying to say. Let me see — Oh, I think this is a better way of saying what I wanted to say,” and proceeded to erase and rewrite. All she needed was the suggestion, and she really went with it.

          I wonder how many grade school teachers actually suggest that their students read and revise what they write.

        • Martha –

          Related to your anecdote. I have seen that when a teacher takes a piece of writing and gives feedback on it along with a grade, often the students will look at the grade as a kind of gatekeeper for what happens next. If they grade matches or exceeds their expectations, they’ll pretty much just stop with the grade. If it is lower than what they expected, they will then begin reading the feedback to see why it was lower. But often that effort is primarily so they can go back to the teacher to argue as to why the grade should be higher – not because they’re really focused on how they can use the feedback to improve their writing in the future.

          I think that probably the biggest issue here is what you’re describing – the lack of understanding how much revision is an inextricable part of the process – and relatedly how much thinking over time is an inextricable part of the process. I remember some anecdote about how someone went up to a famous writer (I think Baldwin) and complained about all the unproductive time just thinking about the writing but not writing, or writing something and revising it and then starting over, that was connected to their writing process. And the famous author responded that actually, all of that “unproductive” time IS a necessary part of producing writing.

  9. Actually in the first half of this year I had wanted to offer a workshop on academic writing for the (mostly Italian) PhD students in our department. Because of teaching reorganisations due to Covid, this hasn’t happened but I’d still be happy to do such a thing at some point in the future. One idea I have about it (which I feel is reinforced by the discussion here) is to let them write something or even to use something they have already written, and then to have the other students reading it. They should concentrate on understanding, not directly on assessing the writing, and should then tell each other what they understood, how clear it was to them, and what elements of the writing helped them or confused them. By this I’d like to emphasise that the whole thing is about communication with others really. The problem with teacher feedback, as valuable as it can be, is that this implicitly still communicates that writing is about satisfying authorities. But no, it’s about getting your point across to other people, and to anticipate what other people think when they read your stuff! Obviously I’ll still give feedback, but I’m really curious how such an “interactive” approach works as opposed to teaching some stuff as an instructor.

  10. Academic writing is hard, and I think some of it has to do with its style.
    Authors are supposed to be neutral, detached, so they force themselves
    to write in the passive voice. It’s hard to write well in a voice
    which isn’t dynamic, after a while it feels like a straitjacket.

    • Maxim:

      Sure, but it’s hard to write in a dynamic voice too! You can tell students and colleagues to avoid the passive voice, but it’s not like they’ll suddenly become clear writers. And, for that matter, non-academics also have difficulty with clear writing. I think that writing is hard, but it seems like it should be easy, hence the frustration. And training for writing is not focused in the same way that, say, sports training is. Students get lessons, they don’t usually get coaching. I think issues specific to academic writing such as the passive voice are the least of it.

      • Andrew –

        > I think that writing is hard, but it seems like it should be easy, hence the frustration.

        +1

        I agree that is a huge part of the issue. Writing well is often an inherently challenging process. It means committing yourself to something and then having to go back and change it or even throw it out. It means committing yourself to starting out on a journey when you really don’t know where you’re going to wind up or even if you’ll really get there. Maybe once you’ve committed a lot of time to going in a particular direction you’ll then uncover that direction won’t work and that you need to actually go off in a very different direction. Maybe you’ll even discover that your whole thesis actually doesn’t even work out!! The mistake is that people can tend to look at that as a failure rather than an important part of the process.

        It can be psychologically, emotionally, and intellectually very challenging. And who wants to go through that process if the function is primarily to publish an article describing work you’ve already done, or to get a grade from the teacher?

      • Andrew –

        Off topic (I like pretending that would ever stop me).

        I hope you’ll write something on this:

        –snip–
        Poverty persists in America even amidst the world’s richest nation. Attempts to attribute longterm poverty to social barriers, such as racial discrimination or lack of jobs, have failed. Some scholars now attribute poverty to culture in the sense that many poor become disillusioned and no longer seek to advance themselves. More plausible is cultural difference. The United States has an individualist culture, derived from Europe, where most people seek to achieve personal goals. Racial minorities, however, all come from non-Western cultures where most people seek to adjust to outside conditions rather than seeking change. Another difference is that Westerners are moralistic about social order, demanding that behavior respect universal principles, while in the non-West norms are less rigid and depend mostly on the expectations of others. These differences best explain why minorities—especially blacks and Hispanics—typically respond only weakly to chances to get ahead through education and work, and also why crime and other social problems run high in low-income areas. The black middle class has converted to an individualist style and thus advanced, but most blacks have not. Government has recently reduced crime and welfare in poor areas, but the ultimate solution to poverty is for the poor themselves to adopt the more inner-driven individualist style.
        –snip–

        https://link.springer.com/article/10.1007/s12115-020-00496-1

        • Joshua:

          Interesting to see this sort of reasoning escape its usual confines of country clubs, talk radio, and 4chan and make its way into the academic literature.

        • I followed the link and looked at the journal, which is called Society, more carefully. I’d say that it’s more like a political magazine along the lines of the New Republic or National Review than a traditional academic journal. This particular article reads like a high school or college term paper; for example:

          In the West most people are individualists. That is, they view life as a project. They focus on their own inner goals and values—such as achieving the “American dream”—and they seek to realize these out in the world. So their lives flow from the inside out. In the non-West, however, a more cautious and collective mindset usually prevails. Most people have little sense that they are separate from society, and they primarily seek to adjust to the demands made by the outside world, rather than seeking change. The chief goal is simply survival, not advancement. So life flows from the outside in. . . .

          The article is written by a college professor and it’s published by an academic press and has some of the trappings of academic writing (a website, a journal name with an official-looking citation, an abstract, and references), but I wouldn’t call it academic writing in any serious way.

        • Andrew –

          That forms an interesting link to the discussion of “academic writing,” and what inclusion/exclusion criteria we might use to determine what is writing that has the trappings of academic writing but isn’t academic writing – and how those criteria interact with the conclusion that “so much academic writing sucks.’

          If the standards are higher, then wouldn’t we necessarily expect a higher level of suckiness?

        • More here on this story from Adam Marcus at Retraction Watch. But I think that report misses some of the context, in that it seems to be treating that journal, Society, as a normal academic journal. As I wrote above, if you look at some of the articles they publish, it seems more like a political magazine than a traditional academic journal. The academicness of it is that the authors are college professors and it’s published in a format similar to that of other academic journals—I guess they do peer review, too—but in content these articles (not just the controversial one under discussion, but most of them) are basically opinion pieces, not research. Opinion pieces are fine for what they are—I write opinion pieces all the time—they’re just different from research.

  11. I taught writing for several years (first year students at very non-elite college). English 102 it was called. I had a background in journalism as well as academia. For my own purposes, and because I thought it might be an effective way to teach, I decided to teach by examples. I distributed two samples, written in the same style and on the same subject, and told the students the specimens had been rated by a professional (me, but I didn’t mention that) for writing quality on a scale ranging from 1 (the pits) to 10 (awesome). I told them the specimens had received ratings of either 1, or 10 (they might both be 1, or both 10, or one might be 1 and the other 10). In other words, they were extreme cases. I asked them to judge which was which. There were four possibilities. This was in order to have a starting point after which I planned to point out line by line, word by word. I would then compare the before-treatment (the one semester course itself) ability to distinguish 1s from 10s with the after treatment ability.

    The primary take-away for me was that almost no students cared one way or the other. Had no desire to write “well”, and didn’t want to think about what made good writing good versus bad writing bad, they just wanted to be told what to write that would be good enough to pass the class.

  12. I taught second-year “scholarly writing” at an okay law school and had a similar experience with 90% of the law students. (Being able to publish a law review article or even a good blog post can boost ones career as a young lawyer.)

Leave a Reply

Your email address will not be published. Required fields are marked *