Monday, December 05, 2011

Break It Up : An Ode to the Paragraph

Today I am thinking about : paragraphs.

You might not think that this topic has any hope of being interesting, and you are probably right, but I am thinking about paragraphs anyway. In particular, I have been wondering why I feel so wearied by long long long paragraphs in Science Papers. I can deal with them in Literature, but I am not so happy about them in Science Papers, especially ones I am reviewing, especially if the entire paper is really really long.

Assuming the content of a Science Paper is interesting and not enraging, it can be very pleasant to read a paper that contains paragraphs, each with a nice topical sentence followed by related text that flows in a logical way to a semi-stopping point, and then .. a break before the next paragraph, which continues the discussion or presentation of information. Reading text that has perfect paragraphs is like listening to beautiful music.

In one manuscript I was reading recently, the authors seemed to think that having a heading every couple of pages was sufficient for breaks. That is, within each heading, all the text was semi-related enough to go in one (pages-long) paragraph. I don't really know why they did this, but it made the paper more difficult (tiring) to read, at least for me. The writing is not bad; it's just not good.

But again, I don't know why long paragraphs wear me out. I don't have a problem with a short attention span, I don't have any particular problem with reading comprehension, and I found the overall topic of this particular paper moderately interesting. And yet, I kept putting the paper aside, to continue reading later. In fact, I have not yet finished reading it.

It seems strange to me that it would make that much difference to have a little indentation in the text now and then.

And of course having too many paragraphs is also annoying.

And single sentence paragraphs are also terrible in their own way.

And maybe I am extreme about this, but I think that the technical aspects of a paper -- even a 'dry' science article -- can have a big effect on how the paper is perceived, how much and how closely it is read, and how (much) it is enjoyed. Content is critical, but so is format and organization. 
Am I being shallow, focusing on the packaging and unduly enamored of a pretty text package? Is this mania for writing beauty related to the fact that I have synesthesia? Maybe, maybe not, but I think that paragraphs help a paper breathe, and that a big long chunk of text can suffocate a paper. (And maybe also a blog, but I think it's OK to hold blog-writing to a different standard than a science article).

Do such technical writing aspects affect how you review a manuscript or proposal? I don't mean that in the sense of writing quality, but in terms of the details of how the text is formatted -- paragraphs, headings, and such. Over-formatting is also annoying, but how much do such things really matter in how readers (including reviewers) perceive the quality of the overall document? And can such things affect how much a paper is cited?

22 comments:

Laura said...

Short answer: of course.

In behavioral economics there is a concept known as the fluency heuristic. When things are easier to understand or interact with we judge them as being better. If the technical aspects of writing make the reading more difficult than it likely will have an effect on how much the writing is cited.

"...people judge fluent statements to be more true, more likeable, more frequent, more famous, better category members, and to come from a more intelligent source than disfluent statements." (Oppenheimer, 2008)

http://web.princeton.edu/sites/opplab/papers/TICS686.pdf

Kevin said...

I suspect the lack of white space or clear paragraph breaks is a symptom of a deeper issue that causes your inattention, not the driving cause.

It seems to me that a lack of attention to the craft of wordsmithing probably correlated with a lack of attention to the craft of organizing one's thoughts. That is, folks who don't bother to edit their paper a few times with an eye toward readability probably haven't bothered to think carefully about what the right order for the concepts are, how to link them, and perhaps most importantly, how to present them succinctly.

I think that is probably why I would get tired from reading such a paper. Not because of the lack of indentation, but because the paper suffers from an overall lack of organization and/or clarity, and fighting through that is tough.

Anonymous said...

Paragraphs should reflect the structure of a paper and the organisation of the ideas. If there are no paragraph breaks, the paper is almostly certainly badly structured.
Of course, the best test would be to take a good paper with normal paragraph breaks, reformat it without the breaks and see if it is still readable.

Anonymous said...

Of course there is the other matter of HOW WE READ. I agree with the first three comments, but (in my experience - and I think there is research which backs it up) when you are reading you refer back and forward a little to keep your place and facilitate understanding. If you stop to think about a point (as I do), then finding your place again is much easier if there are paragraphs - and other punctuation. This, in turn, helps you "enjoy" a paper more because you feel less frustrated!

Anonymous said...

I like short paragraphs (about 10-15 lines max) in science papers. In the past few months, I have edited a lot of text from one of my PhD students who is defending soon. His first drafts were typically quite bad in terms of writing and organization. I proceeded to re-arrange the paragraphs and give them each a title for the student to realize the logic of the arguments and the organization of the paper. And I am pretty sure I told him what I was doing and why I was doing it.

In the next draft he kept all the titles, so now each paragraph is a section with a title. It is interesting he does not seem to mind that the paper looks really weird, with too much detail in those subheadings, which I put there just for the outline and need to be removed. But now anyone can follow the logic of the paper. I told him to try this technique on his next manuscript: think of a simple title for each paragraph; if you can't think of a single title, or if the title is too complex, the paragraph is probably not good.

Anonymous said...

I agree completely. I think how a paper "looks" has a huge effect on how it's perceived, though possibly an entirely subconscious effect for most people. I spend a lot of time fiddling with little details like spacing, figure placement (even in LaTeX, which is supposed to do a good job automatically), and even making sure section headers appear in aesthetically-pleasing places on the page.

I wish I could quantify the effect of all that fiddling, but the general amazingness of my research is a big confound.

Optixmom said...

I agree with all of the above especially Anon:4:53.

If I see a paper with very long paragraphs in a very long paper that I have to review my first response will be a negative one. I have a more positive response to clear ideas that are communicated in short sentences. It is easier for me to keep track of areas that I need to review more extensively because the ideas may be new to me.

Starting off by annoying the reviewer with bad paper formatting is not a good thing.

Anonymous said...

The length of paragraphs usually doesn't matter to me. However, the lack of any indentation or other spacing *between* paragraphs drives me absolutely around the bend (the visual equivalent of no paragraphs at all I suppose). I refuse to review any paper or proposal that doesn't have paragraphs delimited in some way.

Anonymous said...

An aside: I feel obliged to point out to Laura and other readers that perceptual fluency/fluency heuristic is not a behavioral economics discovery. It's from cognitive psychology, and dates back at least as far as the 1980s!

Stephanie said...

When I took scientific writing we were taught how to write paragraphs and sentences so that they work the best with how people's brains work, ie they are theoretically easier to read and understand. That is a very good idea. Yes, scientists are smart, but if even FSP has trouble reading a paper because of this, then it makes sense that we really need to think about this when we write. But, this requires admitting that we scientists are human. Some of us don't like that.

Anonymous said...

Definitely. Haven't formally reviewed a paper yet, but we *always* comment about the readability of papers in lab group meetings.

pyrope said...

I've just started reading a piece of fiction whose author is not a fan of periods. The sentences are many lines long and full of commas and parenthetical asides. This might make for an interesting written aesthetic, but it is exhausting to read. I need sentence and paragraph breaks to gather my thoughts and digest what I've just read. Readers of science expect to have to work through papers to a certain extent, but the harder we make our writing, the less likely our audience will succeed. And, the more likely they'll misinterpret what we were trying to say!

Anonymous said...

I routinely penalize student papers for having single sentence paragraphs (a bit of an oxymoron) and a lack of indentations, topic sentences, summary sentences, and transition statements. In my view, writing well is a part of the job!

I am often surprised to learn that I am the first professor or instructor to explain to students the basic outline of a paragraph - shouldn't this be taught in the 5th grade?

Notorious Ph.D. said...

In my (humanities) discipline, we teach students that a paper's big argument is supported by a number of smaller arguments, each of which gets their own paragraph. If you don't go through step by step, what you've got are a jumble of thematically linked passages, and your reader is left to figure out the structure of the argument on their own.

Shorter answer: not paying attention to paragraphs is not just bad writing; it's bad argumentation.

Anonymous said...

Obviously, paragraphs have a purpose, and readers make use of them accordingly. If an author is not constructing paragraphs, then the work is not consistent with the reader's expected conventions. Therefore, by definition, the reader must do a lot more work to construct the intended meaning.

Frankly, not constructing paragraphs is lazy, inattentive to good writing, and inconsiderate of the reader. Yes, bad writing does annoy me as a reviewer. There's only so much effort that should be asked of me as a reviewer, and when papers require a lot more than that, I'll make some general complaints and return for major revision, along with grumbling to to the editor.

Anonymous said...

In design we have a saying, "Amazing graphics will not make a bad design better, but poor graphics will ruin a good design." I think the concept applies here as well.

Arlenna said...

This is also why good writers are more successful at getting grants even if their science is supposedly "less meritorious" as so frequently cried about by people who are not successful at getting funded.

Anonymous said...

I totally agree with you, FSP. Poor writing is difficult to read. You are not being shallow. If the author cannot organize their thoughts well enough to form paragraphs, how is the reader supposed to understand the paper? It seems to me that the author is the one who should be doing the work, not the reader.

I once had a colleague tell me that I should make half of my talks impossible to follow so my audience would think my work was important. I didn't know what to say because when speakers (or writers) do that, I think they are just being selfish and lazy. However, poor writing and speaking is so pervasive, that maybe my colleague was correct. *sigh*

EcoGeoFemme said...

I think Kevin is right.

Also, I think too-short paragraphs are way more distracting than too-long ones.

Comrade PhysioProf said...

This is also why good writers are more successful at getting grants even if their science is supposedly "less meritorious" as so frequently cried about by people who are not successful at getting funded.

And if you go to the blogges where a lot of that crying is going on in the comments, what you will immediately notice is that the biggest criers are generally the least fluent, least literate.

Anonymous said...

Bad writing always affects me as a referee. I will reject a paper that I can't follow, and suggest major revision if I can only work out the meaning with difficulty.

Of course, I taught technical writing to engineers for more than a decade, so I'm particularly sensitive to bad writing. I'll give a little leeway to non-native speakers in matters of spelling and grammar, but not organization, and paragraphing is the most basic level of organization.

Laura said...

@ Anon 10:10

I won't deny that the fluency heuristic came out of cognitive psychology, but so did behavioral economics lead in large part by Tversky & Kahneman, two psychologists. The paper I linked to is authored by a psychologist. Behavioral economics is an interdisciplinary field of study that has a strong cognitive psych component. I'm not really sure why you feel so obliged as to point out what field has discovery claims on the effect as I did not in anyway denigrate cognitive psychology. I simply said that it is a concept 'in' behavioral economics.