An epiphany about science writing
Let me make one thing clear: this blog is about writing. Not writing about science -- there are plenty of science blogs out which are doing an excellent job of that, and I'm not sure how much I would have to contribute. Moreover, I have other avenues to make my technical viewpoints known, avenues more likely to attract an appropriate audience than a pseudonymous blog. No, I hope this blog to focus on, ... hm, I'm not quite sure how to put it yet, which is kind of the problem. But in any case I hope the net effect will be an avenue for me to practice, and improve the quality of my writing. A personal journal if you will, and with any luck, I'll have so few readers that I won't embarrass myself too much.
And my writing does very much need improvement. Not just the creative stuff, like the material that will end up here, but my technical, professional writing as well. That's what this post is about. I've always had trouble writing. Great difficulty just getting material out the door. I nearly didn't graduate, and in fact finished many months after I'd planned to, because my dissertation just wasn't done. It's not a recent problem either: I failed a lab course in college, in part due to inability to writeup the labs, and even as far back as the fifth grade, I was in trouble for not turning in even one of the half dozen or so writing assignments. I've always had trouble writing.
It's not just production, either. Some people write slowly, and well. They take time to craft wonderful prose, or text which is at least meaningful and comprehensible. Not me. I'd torture myself to do something, and the torture would increase tenfold upon reading it. It's even worse than I thought, would be my standard refrain, which is serious criticism indeed, considering how awful I thought it was while I was writing it. And in fact, this has always been the reason for the low productivity -- a combination of a small amount of taste and the ability (or disability) of applying it on the fly to the work I was producing. It was crap, I knew it was crap, and I found it difficult to accept how truly worthless I was. Then I'd get depressed, and wouldn't even be able to do anything, etc. Your standard positive-feedback loop, a vicious cycle.
But are there not classes, don't they teach you how to write in college? Sure. I even had or borrowed a copy of Strunk and White, and adopted a "Make every word tell" editing style -- which generally resulted in editing out every other word, delete the articles, the adjectives and qualifiers, and end up with taught, spare prose. Dry, a former colleague called a draft of mine. All of this is fine, a reasonable choice of style at the word-by-word or phrase-by-phrase level, but it cannot help if the sentence, paragraph, section, article levels are jumbled, chaotic, thoughtless. I could correct spelling, edit phrases, but could not put together a comprehensible story. Fine for a copyeditor, not so good if your living depends on writing intelligible prose.
Really, the unintelligible nature of my prose is not specific to writing. Often, I write the way I speak. And, as my wife will attest, my inability to construct an intelligible story extends to my oral utterances. I'm incomprehensible face-to-face too, it's not just in print. Put it down to the nature of my thought processes, if you want to -- I know many things, I can connect many things, but finding and telling one thread, with enough detail and glue so that the audience has any idea what I'm going on about, is hard for me.
Fine, so that's the background. Recently, I've been working on a review article. I'm coauthor number three, low guy on the totem pole, but it's also been some years since I've seen my name in print, so maybe I'm the hungriest (or the least busy) author. After much work, and only about a month after the deadline, we have a draft, about the right length, on the right subject, with the right number of references. A majority of the words are mine, and I put the sections together. Good, right? Unfortunately, not. I read it, and it's garbage. Incomprehensible. The boss is nicer, but still, consensus is that it requires a from scratch rewrite. (My wife is scathing too.)
I decide, from my perspective from a lifetime of failure in writing, to try again, all by myself, from scratch. The point is not whether what I write goes out unchanged, or even out at all, just about whether I can do it, produce something which at least approaches my own standards, regardless of how anyone else might see it.
The problem is, of course, that after so many decades of failure, I must be doing something seriously, consistently wrong. I have intimations of what it is, and can recognize the results, but don't know how to do it right. But seriously, how hard can writing be, compared to the technical issues I regularly tackle?
A google search leads me to several useful articles. Some were in the standard, "Identify your purpose, audience, and thesis first; then write your outline, be careful to keep your Methods separate from your Results and Discussion". It's the kind of useful, but generic advice I've probably read a hundred times before. The most important one is called, "The science of scientific writing", by Gopen and Swan. The authors are linguists and teachers of scientists, working at Princeton. They work in the context of scientific results about writing, and in particular about how readers interpret text. Thus they immediately put themselves into a worldview where their audience (professional scientists) can understand them, by discussing the results in terms scientific. The exact same points could have been made in literary criticism terms, the whole post-modern, texts-as-cocreations-between-the-readers-and-writers worldview, but it wouldn't resonate with scientists. English majors would need a rewrite with a different context. The other, truly brilliant thing they do is make their points and give their advice in the context of specific examples of real scientific prose, taken from articles either published or accepted. In other words, data. Scientists love data, it's the difference between bullshit and reality.
So what points do they make, what advice do they give? Well, read the whole thing. It's about subjects and verbs, stress positions, old information and new information, and where do readers expect these things to happen. Here's their summary:
Another excellent observation is about perceiving logical gaps. I think that that's my biggest problem, I consider some jumps obvious, which it's entirely opaque to my audience even what I'm talking about, much less why the connection is valid. A failure to articulate, in other words. (Not just limited to prose, either: I got docked points for failing to show work in algebra, too.) But how do you find such things? In truly opaque prose, with many structural problems of the kinds they outline, it doesn't matter where you start. Start anywhere, fix one kind of problem, then work on another, pretty soon it will become clear where the true issues lie. The closer a section of prose is to being coherent, the easier it is to identify and address the few remaining incoherencies.
So how did this work with my review article? Well, the rules given are a set of constraints on the writing. They aren't the only constraints, though: we also had a word limit, a subject area, a set of references to include, and a story to tell. So actually drafting the article became an exercise in solving a highly constrained system. Every phrase needs to be backed up with a citation, and needs to be accurate in the context of that citation. But the set of citations selected needs to make sense as a set, together. But every phrase and point needs to make sense in the context of the story of the section, and the sections into the piece as a whole. And every sentence and paragraph needs to obey the rules of reader expectations outlined by Gopen and Swan. It's a highly constrained exercise.
You might think that constraints would make writing more difficult. Oddly, that's not true, in poetry, or in prose. In this case, for me, it helped a lot. Because it turned a problem I couldn't solve ("Write beautiful, insightful prose") into one I could ("Identify a single satisfactory solution to the following set of constraints"). It was slow. It was very painful: I found myself frustrated, almost to the point of tears at points. (A wonderful indicator of forming new neural connections, if you can recognize it happening.) I spent days on the outline, something I've never done before, trying to figure out a way to make the points, satisfy the rules, cover the material, and tell the story, simultaneously. (Another observation which helped me feel comfortable blowing time on the outline appears in Stephen Covey's Seven Habits. It says, approximately, that everything built is built twice: first, in the mind, and only then in reality. The outline is about building the paper in my mind. Spending two full days outlining a 700 word section is not waste, but an investment, which will pay in the final product.) And the true beauty of the constraints-based approach to writing is that, if you identify the constraints correctly, and satisfy them, then the final result doesn't suck. It may not be beautiful, luminous prose that will inspire the reader to paroxysms of delight, but they will at least know what you're talking about. And in technical prose, that's what the point is. Communication with the reader.
So how's the result? Better than the first draft, says my wife, but still in need of serious editing. High praise, really, especially since she was able to figure out what the paper was trying to say. Maybe I'm learning something.
And my writing does very much need improvement. Not just the creative stuff, like the material that will end up here, but my technical, professional writing as well. That's what this post is about. I've always had trouble writing. Great difficulty just getting material out the door. I nearly didn't graduate, and in fact finished many months after I'd planned to, because my dissertation just wasn't done. It's not a recent problem either: I failed a lab course in college, in part due to inability to writeup the labs, and even as far back as the fifth grade, I was in trouble for not turning in even one of the half dozen or so writing assignments. I've always had trouble writing.
It's not just production, either. Some people write slowly, and well. They take time to craft wonderful prose, or text which is at least meaningful and comprehensible. Not me. I'd torture myself to do something, and the torture would increase tenfold upon reading it. It's even worse than I thought, would be my standard refrain, which is serious criticism indeed, considering how awful I thought it was while I was writing it. And in fact, this has always been the reason for the low productivity -- a combination of a small amount of taste and the ability (or disability) of applying it on the fly to the work I was producing. It was crap, I knew it was crap, and I found it difficult to accept how truly worthless I was. Then I'd get depressed, and wouldn't even be able to do anything, etc. Your standard positive-feedback loop, a vicious cycle.
But are there not classes, don't they teach you how to write in college? Sure. I even had or borrowed a copy of Strunk and White, and adopted a "Make every word tell" editing style -- which generally resulted in editing out every other word, delete the articles, the adjectives and qualifiers, and end up with taught, spare prose. Dry, a former colleague called a draft of mine. All of this is fine, a reasonable choice of style at the word-by-word or phrase-by-phrase level, but it cannot help if the sentence, paragraph, section, article levels are jumbled, chaotic, thoughtless. I could correct spelling, edit phrases, but could not put together a comprehensible story. Fine for a copyeditor, not so good if your living depends on writing intelligible prose.
Really, the unintelligible nature of my prose is not specific to writing. Often, I write the way I speak. And, as my wife will attest, my inability to construct an intelligible story extends to my oral utterances. I'm incomprehensible face-to-face too, it's not just in print. Put it down to the nature of my thought processes, if you want to -- I know many things, I can connect many things, but finding and telling one thread, with enough detail and glue so that the audience has any idea what I'm going on about, is hard for me.
Fine, so that's the background. Recently, I've been working on a review article. I'm coauthor number three, low guy on the totem pole, but it's also been some years since I've seen my name in print, so maybe I'm the hungriest (or the least busy) author. After much work, and only about a month after the deadline, we have a draft, about the right length, on the right subject, with the right number of references. A majority of the words are mine, and I put the sections together. Good, right? Unfortunately, not. I read it, and it's garbage. Incomprehensible. The boss is nicer, but still, consensus is that it requires a from scratch rewrite. (My wife is scathing too.)
I decide, from my perspective from a lifetime of failure in writing, to try again, all by myself, from scratch. The point is not whether what I write goes out unchanged, or even out at all, just about whether I can do it, produce something which at least approaches my own standards, regardless of how anyone else might see it.
The problem is, of course, that after so many decades of failure, I must be doing something seriously, consistently wrong. I have intimations of what it is, and can recognize the results, but don't know how to do it right. But seriously, how hard can writing be, compared to the technical issues I regularly tackle?
A google search leads me to several useful articles. Some were in the standard, "Identify your purpose, audience, and thesis first; then write your outline, be careful to keep your Methods separate from your Results and Discussion". It's the kind of useful, but generic advice I've probably read a hundred times before. The most important one is called, "The science of scientific writing", by Gopen and Swan. The authors are linguists and teachers of scientists, working at Princeton. They work in the context of scientific results about writing, and in particular about how readers interpret text. Thus they immediately put themselves into a worldview where their audience (professional scientists) can understand them, by discussing the results in terms scientific. The exact same points could have been made in literary criticism terms, the whole post-modern, texts-as-cocreations-between-the-readers-and-writers worldview, but it wouldn't resonate with scientists. English majors would need a rewrite with a different context. The other, truly brilliant thing they do is make their points and give their advice in the context of specific examples of real scientific prose, taken from articles either published or accepted. In other words, data. Scientists love data, it's the difference between bullshit and reality.
So what points do they make, what advice do they give? Well, read the whole thing. It's about subjects and verbs, stress positions, old information and new information, and where do readers expect these things to happen. Here's their summary:
1. Follow a grammatical subject as soon as possible with its verb.
2. Place in the stress position the "new information" you want the reader to emphasize.
3. Place the person or thing whose "story" a sentence is telling at the beginning of the sentence, in the
topic position.
4. Place appropriate "old information" (material already stated in the discourse) in the topic position
for linkage backward and contextualization forward.
5. Articulate the action of every clause or sentence in its verb.
6. In general, provide context for your reader before asking that reader to consider anything new.
7. In general, try to ensure that the relative emphases of the substance coincide with the relative
expectations for emphasis raised by the structure.
Another excellent observation is about perceiving logical gaps. I think that that's my biggest problem, I consider some jumps obvious, which it's entirely opaque to my audience even what I'm talking about, much less why the connection is valid. A failure to articulate, in other words. (Not just limited to prose, either: I got docked points for failing to show work in algebra, too.) But how do you find such things? In truly opaque prose, with many structural problems of the kinds they outline, it doesn't matter where you start. Start anywhere, fix one kind of problem, then work on another, pretty soon it will become clear where the true issues lie. The closer a section of prose is to being coherent, the easier it is to identify and address the few remaining incoherencies.
So how did this work with my review article? Well, the rules given are a set of constraints on the writing. They aren't the only constraints, though: we also had a word limit, a subject area, a set of references to include, and a story to tell. So actually drafting the article became an exercise in solving a highly constrained system. Every phrase needs to be backed up with a citation, and needs to be accurate in the context of that citation. But the set of citations selected needs to make sense as a set, together. But every phrase and point needs to make sense in the context of the story of the section, and the sections into the piece as a whole. And every sentence and paragraph needs to obey the rules of reader expectations outlined by Gopen and Swan. It's a highly constrained exercise.
You might think that constraints would make writing more difficult. Oddly, that's not true, in poetry, or in prose. In this case, for me, it helped a lot. Because it turned a problem I couldn't solve ("Write beautiful, insightful prose") into one I could ("Identify a single satisfactory solution to the following set of constraints"). It was slow. It was very painful: I found myself frustrated, almost to the point of tears at points. (A wonderful indicator of forming new neural connections, if you can recognize it happening.) I spent days on the outline, something I've never done before, trying to figure out a way to make the points, satisfy the rules, cover the material, and tell the story, simultaneously. (Another observation which helped me feel comfortable blowing time on the outline appears in Stephen Covey's Seven Habits. It says, approximately, that everything built is built twice: first, in the mind, and only then in reality. The outline is about building the paper in my mind. Spending two full days outlining a 700 word section is not waste, but an investment, which will pay in the final product.) And the true beauty of the constraints-based approach to writing is that, if you identify the constraints correctly, and satisfy them, then the final result doesn't suck. It may not be beautiful, luminous prose that will inspire the reader to paroxysms of delight, but they will at least know what you're talking about. And in technical prose, that's what the point is. Communication with the reader.
So how's the result? Better than the first draft, says my wife, but still in need of serious editing. High praise, really, especially since she was able to figure out what the paper was trying to say. Maybe I'm learning something.
posted by TFox at
2:36 PM
