Y'know those lists you catch in the Grauniad periodically "Ten Top Tips to Typers" where the first rule is avoid alliteration? That link has solicited Top Ten lists from a string of living authors. One common theme in many of those lists is "write early and write often". Here's a list culled from the works of Mark Twain, which ends with "Write without pay until somebody offers pay. If nobody offers within three years, the candidate may look upon this circumstance with the most implicit confidence as the sign that sawing wood is what he was intended for." Which is a great winnow; if you write because, like Luther, "Ich kann nicht anders", then you may not be happy starving in your garret but it will be clear why you were put on the planet. A couple of weeks ago, my old boss asked for comments on the subject of science writing. That was a little odd because, my primary fault as a scientist was/is a failure of finish: the ability to write it all down and submit it to peer-review. It's much easier to write the Blob - they accept anything. So I dashed off a few comments and suddenly it was TenTopTip list:
- Don't try to read hundreds of novels now, it's too late for that. My reading daughter and I both did this http://testyourvocab.com/ and we both had about the same size of vocabulary but she is 40 years younger than me. Improving your wordpower in adulthood is going to be a struggle similar to that of learning a foreign language - easy when you're three, a slog when you're thirty.
- Read The Blob! One of the long words that I knew for the vocab test was hypnopompic which I needed to write a post about narcolepsy. This is my 500th post so there's plenty of other material to divert you from doing some work. 282,000 words is more than Moby Dick but less that Anna Karenina.
- My Gaffer in Grad School chid me thus "Don't write run-on sentences" because I often erred in being too clever rather than being too simple. Go with Twain: "Use plain, simple language, short words and brief sentences." I only listened to that advice with half an ear and you'll find lots of examples on The Blob where I disappear up my own clever orifice.
- A later boss used to say that, after you've landed the last result, you should write the title first because that should articulate the take-home message. Then write the Abstract which should make clear the structure for how you get from start to take-home. The paper should then write itself.
- It also helps a lot to use a Table of Contents structure of headings, sub-heads and sub-sub-heads. If the headings do not have equal 'value' then your structure needs adjustment, For a paper, you can delete the 1.2.1; 1.2.2; 220.127.116.11; 18.104.22.168 structure when the hierarchy is populated with words. For a thesis it's usual to leave these structural signposts in.
- Don't sweat too much about apostrophes - it's just a convention and is not consistent: we use a possessive apostrophe in "John's hat" but not in "its hat" (because we've allocated it's to the contraction of "it is". That's how is is but it didn't have to work out that way. Tell annoying apostrophe-nazis to do something about deforestation in Borneo. Try to use correct spelling and grammar (and apostrophes!) in letters of application or grant application: there's no advantage in distracting/annoying your readers with obvious errors - it will make them think you frankly don't give a damn.
- When we were young (before the First War) one of the standard exercises in English was to do a Précis which required you to read, say, 1500 words and summarise the piece in, say, 250 words. We used to joke that the first step was (of course) to delete all the adjectives. The Art of Summary is not formally taught any more and more's the pity because to do a Précis properly you have to read with care and attention and abstract the key points into your summary. Very useful skill that, and 'abstract' is one of the tasks you need for every paper and grant application you write.
- Read a (science) book and write a book review. It's a good discipline, for the same reason writing a Précis is. A Book Review should give the title, cost and page count, the author and publisher and ISBN. Then it should summarise the book, and maybe recommend whether readers should read (all of) it (or not) and buy it (or not). THEN it should add something that the reviewer brings to the table laid out by the book. That makes the review valuable in and of itself. All in 500 words! You may be able to get the review published if you're quick off the mark with a new book and have something interesting to say about it. In The Examiner, say, or The Scientist.
- Write early and write often: you get better at anything if you practice, for the last 11 months I've written something (usually about 500 words) on The Blob every day usually before breakfast. That amounts now to much more than a quarter million words.
- Get someone to read a chunk of your stuff and identify your verbal tics, then work to eliminate them. Far too many of my sentences begin with 'So'. Consequently I must do better.