I have read three books on academic writing back to back. Learn to write badly by Michael Billig, Writing for Social Scientists by Howard S. Becker, and Stylist Academic Writing by Helen Sword. You might ask: Isn’t writing dead? We have “AI”, right?
Yes, writing might be dead. But “good writing” is a skill that is now more valuable then it ever was. The same can be said for a lot of skills that AI allegedly could replace.1 If what you need is just writing yet another AI slop, that needs no thinking and no skill. But if what you need is communicating your idea to your readers and perhaps also changing their mind, then you need to write well.
The three books are all good and the authors convey their ideas differently. Billig writes in a critical manner. Becker (†) wrote in a more “back in my day” way. Sword backs her many ideas with an analysis of a corpus of research articles from different disciplines. I think in sum, I like Becker the most. That’s the third edition of his book, only a few years before he left us. In the book, I know that he kept updating the content to suit the new situation.
I can summarize the three books in an overarching way: Social scientists suck at writing the social sciences. I don’t want to point out any particular social scientist, I show the suckiness with my own old writing from my PhD research published in a decent communication journal.
Online activity is often cyberbalkanized, but it remains unclear whether this phenomenon leads to polarization of public opinion or if the relationship works in the reverse direction. This study tested the temporal association between cyberbalkanization and opinion polarization during the debate on political reform in Hong Kong. Online communities were constructed by a post-sharing network of 1,644 Facebook pages (101,410 shares); the differences between intra- and inter-community shares were derived, and a cyberbalkanization index was computed. A time-series analysis showed that the index temporally preceded the opinion polarization, i.e., most of the opinion poll’s respondents gave extreme ratings to government leaders, but not vice versa. The index was particularly predictive of polarization among youth.
The social sciences should be about people. However, in my abstract, there is almost no actor (except maybe the only common noun: youth). Billig calls this “unpopulated writing”. I threw in long jargons such as cyberbalkanization (and the adjective cyberbalkanized) without much thinking. Now rereading it, my logic was crazy. Can a study test a temporal association association? (Or rather We test a temporal association ?).2 Aren’t the differences between intra- and inter-community shares already the so-called cyberbalkanization index? The “i.e.” part is quite unclear whether it is talking about opinion polarization itself, or the temporal precedence of the so-called cyberbalkanization over opinion polarization.
In hindsight, it was bad. But it’s not super bad. The problem with the bad writing in the social sciences is what Becker’s student called “classy” writing: It looks impressive, pompous, but it is actually more like a pretentious word cloud. Usually, it contains a lot of nouns, a lot of passive constructs, a lot of clauses, and a lot of hedges. We usually learn to write like that because we do not take writing classes during our postgraduate training and we learn how to write either by emulating our supervisors or emulating how other scholars write in journal articles. Then, we write in those annoying styles. Becker suggested that we should not learn by just emulating any authors you read, instead select some role models you admire their writing. Sword also makes similar suggestions.
Becker also made clear that we should not expect to do right the first time. Instead, we should write a first draft quickly and then rewrite it from there a lot of times. I actually know this fact during my side career as a novelist. So, it is not so surprising.
In the end, I actually ask myself: Is it useful for me to write better? If we are talking about whether it is useful for my career, the definite answer is no. It is actually similar to my previous convictions on many academic activities: The academic system usually rewards people only for doing it and then putting it on their CVs. That’s already the reward. Doing it better usually does not improve the reward. There are more people writing than reading, so writing badly in most cases does not even get noticed. Besides, the publishing system (which includes also the writers, editors, reviewers etc.) has self-selected to accept the bad writing as the norm. So, those are not bad writings. Only “normal” writings. There are so many people who would want to write like Pierre Bourdieu, 3 for example.
Becker, in the last edition of his book, shared a story on his dealing with the “modern” publication business. He wanted to test how open the current publication environment is in accepting what he preached in his book. He submitted an article to the best sociology journal and the reviewers rejected his paper partly because of the writing style being perceived as “informal”. He concluded that:
My small experiment demonstrated to me the difficulty even a well-known scholar has trying to publish in the best journals these days without toeing the line on bibliography and stylistic matters. The organization of publication opportunities has changed. What used to be acceptable no longer is. And the new rules have taken on the inevitability of “what everybody knows.”
In the end of the end, writing better is only for my own sanity. In this insane world, being sane is a luxury.
As a tidbit, there is a retraction of a paper by a group of Hong Kong researchers, which leads to the associate dean of a department to step down. The retraction was because of AI-halluncinated references and human negligence. This news hit me a little bit because the protagonist was the PhD supervisor of my PhD supervisor. ↩
I must say this construct is actually quite common and kind of fine…-ish. The sentence fuctions as a report of what has bee done. A reader can probably know what the sentence probably means: In this study, we tested the association. But this formulation is just lazy and bad. The focus is now this study, is this information important? Grammatically speaking and in some contexts, this formulation is fine…-ish. For example, in an advertisement, an advertiser would say: This car sells well. The grammar jargon is “middle voice.” It does not mean that this car performs the selling action (which is not possible) in a good manner. Instead, This car sells well means that this car has the quality of being easy to sell a lot. Therefore, This study tested the temporal association could mean: This study had the ability to test the temporal association. But it is weird, especially when there is no adverb and it is not in a advertisement context. (I also would like to say that the distinction between active, passive, and “middle” voice is not that strong in my native language and that can be an excuse for me writing like that.) Another context this might work is the medical context: This excercise tests the strength of your muscle. But this is in present tense to describe the general quality of the exercise, not a reporting of what has been done. ↩
British sociologist Richard Jenkins says in his book on Bourdieu that Bourdieu and his followers engage in a collective bad faith (mauvaise foi) to write in a cryptic way by using complicated syntax and redefining common used words (such as “field” or the French word champ), because they adopt to the value that social scientists should write like that to distinguish themselves from the lay people. ↩