About a dozen years ago, I was excited to hear about a new book called Learn to write Badly: How to Succeed in the Social Sciences, Michael Billig (2013). I bought a copy and read it immediately, thinking it would give me some useful insights into why many social scientists write poorly.
To my annoyance, I discovered that the book’s 214 pages contained little worthwhile content. The author could have covered the things worth saying in 10 or 15 pages, perhaps 25 pages if he wanted to include a lot of examples.
Another thing annoyed me
My annoyance grew when I noticed a repeated stylistic feature that grated on me more and more each time I noticed it, which seemed to occur on almost every page. The author routinely included a comma that wasn’t needed. In some cases, the comma could even be positively misleading.
After a while, the author’s mis-punctuation started feeling familiar. It began to remind me of German. The author uses a comma where one is not needed (and may even be wrong) in English, but is always needed in German.
The problem relates to the distinction between restrictive relative clauses and non-restrictive relative clauses:
- a restrictive relative clause narrows down the set of entities discussed. For example:
‘authors who use commas incorrectly confuse their readers’.
The underlined relative clause narrows down the class of authors to include not just any author but only authors who use commas incorrectly. - A non-restrictive relative clause does not narrow down the set of entities discussed. Instead, it supplies extra information about the set of entities discussed. For instance:
‘authors, who use commas incorrectly, confuse their readers’.
The underlined relative clause comments on all ‘authors’ without limiting the class of authors. And it says that all authors use commas incorrectly.
Non-restrictive relative clauses are parenthetical. Accordingly:
- in spoken English, speakers typically pause before and after non-restrictive relative clauses; and
- in written English, those pauses are typically marked with a comma before and after non-restrictive relative clauses.
In contrast, there are usually no pauses (and no commas) before and after restrictive relative clauses.
Relative clauses in German
Unlike English, German punctuation requires a comma before all relative clauses, restrictive and non-restrictive.
Where does Billig’s punctuation go wrong?
Billig routinely uses commas with restrictive relative clauses. I went back through the first 2 chapters and I list below all relative clauses I found.
In the following tables, I present the relative pronoun in bold, along with the pronoun’s antecedent (and the comma, if present).
In table 1, I list 11 instances where Billig included a comma (incorrectly) to mark what appears intended to be a restrictive relative clause.
One of those instances is a sentence (from page 24) containing 2 restrictive clauses. Particularly strikingly, although the 2nd clause has no comma before it (correctly), the 1st has a comma before it (incorrectly).
Table 1. Restrictive relative clauses marked (wrongly) with a comma
| Page | |
| It would be strange were I arguing that we all need to be public intellectuals in a book, which is itself aimed at academics rather than at a wider audience. | 9 |
| It is certainly more manageable for the large number of academics, whose native language is not English but who find themselves being pressured to publish in English. | 9 |
| Also, there are two writers, whom I have already mentioned– William James and his namesake C.L.R. | 10 |
| … I am inevitably running the risk of providing a manual, which instructs readers how to write in the very style which I am criticizing. | 11 |
| The bad writing, which I am writing about, has not been produced by too little education. | 11 |
| In my view, the sorts of rhetorical faults, which I will examine later, are connected with the conditions under which we are working as academic hacks. | 12 |
| the intellectual circles, which specialist professors address in their writings, are becoming ever smaller. | 17 |
| The measures of citation analysis appear to identify which publications are most often cited in other articles and, thus, the auditors can claim to distinguish between academics, whose publications are being used by others, from academics whose publications are having little effect. | 24 |
| Institutions, keen on developing their research profile, might subsidize academic employees by paying the fee for them. | 25 |
| The sub-groups, about which Pepitone was talking, share similar basic principles, using similar experimental methods and adopting similar practices for writing their reports | 31 |
| It is hard to be confident that editors, who can write in this way, will act as guardians of good style. | 37 |
Table 2 lists 3 instances (in 2 sentences) where Billig correctly included no comma before a restrictive relative clause.
Table 2. Restrictive relative clauses marked (correctly) with no comma
| Page | |
| [Adam Smith] was not addressing a small group of specialists, but he was using non-technical words that he thought all educated readers would understand. | 15 |
| These figures might be somewhat inflated because they do not take account of the growing numbers of part-time teaching assistants that work in the research-based universities and whose employment is designed to ease the teaching burdens of the full-time, permanent academics | 23 |
Table 3 includes 3 instances where Billig included a comma before a relative clause that was probably intended to be restrictive, but where a non-restrictive reading is just about plausible as well.
Table 3. Relative clauses probably intended to be restrictive, but perhaps intended to be non-restrictive
| Page | |
| I feel confident that the overwhelming majority of specialist terms, which academic social scientists are producing in such profusion, are unlikely ever to trouble the eyes or ears of the general public. | 10 |
| Although the research university, which developed in the late nineteenth century, might resemble today’s universities, there were also important differences. | 17 |
| Tonette Rocco, writing in a book that aims to help and to encourage academics to publish, quotes an economist, who claims that a quality article in a good journal might be worth $200 for every year that the author works as an academic. | 23 |
Summary
In the 2 chapters I looked at (39 pages), I gathered 18 relative clauses. Of those relative clauses:
- 11 are restrictive and incorrectly punctuated, with a preceding comma;
- 4 are restrictive and correctly punctuated (with no comma);
- 3 clauses punctuated with a preceding comma are mostly naturally read as restrictive (making the comma incorrect) but might conceivably be read as non-restrictive (making the comma correct).
None of the incorrectly included commas made me misunderstand what was being said. But they all made me stop and think unnecessarily and I found their constant use unnecessarily irritating.
As for the book’s content, I felt that it was insufficient to justify its length. I will not read this book again.
“some useful insights into why many social scientists right poorly”. Oops! Right = write, nicht war? Interesting piece though!
I’d love to say it was conscious word play, or even that my subconscious was making a witty bid for glory. Sadly, it was just a typo. Thanks for letting me know, now fixed.
When I first started at IASC, Peter taught me a rule that has served me extremely well:
Descriptive* = expendable without changing meaning = offset by commas = begin with which.
Restrictive = essential to the meaning = not offset by commas = begin with that.
*Apparently “relative” is the more correct term and may well have been the word that Peter used. I find “descriptive” um, well, more descriptive.
1. Restrictive and non-restrictive are conventional labels for 2 classes of relative clauses. Huddlestone and Pullum’s A Student’s Introduction to English Grammar uses the labels integrated and supplementary.
2. I would often be happy using ‘which’ in restrictive relative clauses. The following is consistent with my attitude:
However, for “polished” prose, many American style guides, such as the 16th edition of The Chicago Manual of Style, recommend generally avoiding which in restrictive relative clauses. This prescriptive “rule” was proposed as early as 1851 by Goold Brown. It was championed in 1926 by H. W. Fowler, who said: “If writers would agree to regard that as the defining [restrictive] relative pronoun, and which as the non-defining, there would be much gain both in lucidity and in ease. There are some who follow this principle now, but it would be idle to pretend that it is the practice either of most or of the best writers.” English relative clauses – Wikipedia