Good English and the scientific career: hurdles for non-native English speakers

13 02 2019

New post from Frédérik Saltré originally presented on the GE.blog.


It’s no secret that to be successful in academia, it’s not enough just to be a good scientist — being able to formulate and test hypotheses. You also need to be able to communicate that science effectively.

This implies a good command of the English language for anyone who wants a career in science. Mastering English (or not) will directly affect your work opportunities such as publishing, establishing networks at conferences, taking leadership of working groups, contributing to lab meetings (there is nothing worse than feeling left out of a conversation because of language limitations), and so forth.

But when it comes to language skills, not everyone is created equal because those skills mostly depend on a person’s background (e.g., learning English as a child or later in life), cultural reluctance, fear of making mistakes, lack of confidence, or simply brain design — this last component might offend some, but it appears that some people just happen to have the specific neuronal pathways to learn languages better than others. Whatever the reason, the process of becoming a good scientist is made more difficult if you happen not to have that specific set of neuronal pathways, even though not being a native English speaker does not prevent from being academically successful.

Read the rest of this entry »




Time to put significance out of its misery

28 07 2014

If you’ve been following this blog for a while, you’ll be no stranger to my views on what I believe is one of the most abused, and therefore now meaningless, words in scientific writing: ‘significance’ and her adjective sister, ‘significant’. I hold that it should be stricken entirely from the language of science writing.

Most science writing has become burdened with archaic language that perhaps at one time meant something, but now given the ubiquity of certain terms in most walks of life and their subsequent misapplication, many terms no longer have a precise meaning. Given that good scientific writing must ideally strive to employ the language of precision, transparency and simplicity, now-useless terminology should be completely expunged from our vocabulary.

‘Significance’ is just such a term.

Most interviews on radio or television, most lectures by politicians or business leaders, and nearly all presentations by academics at meetings of learned societies invoke ‘significant’ merely to add emphasis to the discourse. Usually it involves some sort of comparison – a ‘significant’ decline, a ‘significant’ change or a ‘significant’ number relative to some other number in the past or in some other place, and so on. Rarely is the word quantified: how much has the trend declined, how much did it change and how many is that ‘number’? What is ‘significant’ to a mouse is rather unimportant to an elephant, so most uses are as entirely subjective qualifiers employed to add some sort of ‘expert’ emphasis to the phenomenon under discussion. To most, ‘significant’ just sounds more authoritative, educated and erudite than ‘a lot’ or ‘big’. This is, of course, complete rubbish because it is the practice of using big words to hide the fact that the speaker isn’t quite as clever as he thinks he is.

While I could occasionally forgive non-scientists for not quantifying their use of ‘significance’ because they haven’t necessarily been trained to do so, I utterly condemn scientists who use the word that way. We are specifically trained to quantify, so throwing ‘significant’ around without a very clear quantification (it changed by x amount, it declined by 50 % in two years, etc.) runs counter to the very essence of our discipline. To make matters worse, you can often hear a vocal emphasis placed on the word when uttered, along with a patronising hand gesture, to make that subjectivity even more obvious.

If you are a scientist reading this, then you are surely waiting for my rationale as to why we should also ignore the word’s statistical meaning. While I’ve explained this before, it bears repeating. Read the rest of this entry »





Don’t torture your readers III

23 06 2014

TortureIt has been quite some time since I did one of these kinds of posts (see Don’t torture your readers and Don’t torture your readers II). However, given how popular they seem to be, I have decided to do a follow-up post on grammar problems that I tend to see far too often in science writing.

COMPOUND ADJECTIVES: This is a particularly abused component of scientific writing. Although it’s fairly straightforward, I’m amazed just how many people get it wrong. Most people appear to understand that when an adjective (that’s a qualifier for a noun, just in case you are a grammarling) is composed of more than one word, there is normally a hyphen that connects them:

  • e.g., ’10-m fence’, ‘high-ranking journal’, ‘population-level metric’, ‘cost-effective policy’

If two or more adjectives are given in a row, but none modifies the meaning of the others, then it is simply a case of separating them with commas:

  • e.g., ‘a long, high fence’, ‘an old, respected journal’, ‘an effective, enduring policy’

However, if the compound adjective is composed of a leading adverb (that’s a qualifier for a verb), then there is NO hyphenation:

  • e.g., ‘an extremely long fence’, ‘a closely associated phenomenon’, ‘a legally mandated policy’

There are other instances when no hyphenation is required, such as when the qualifiers are proper nouns (e.g., ‘a Shark Bay jetty’), from another language such as Latin (e.g., an ‘ab initio course’) or enclosed in quotation marks (e.g., ‘a “do it yourself” guide). Note in the last example, without the quotations, it would become ‘a do-it-yourself guide’).

A quick way to recognise whether a compound adjective should be hyphenated is to examine the terminal letters of the leading word; if the leading component ends in ‘ly’, then it is likely an adverb, and so the compound should not be hyphenated (although watch for sneaky exceptions like ‘early-career researcher’!). Read the rest of this entry »





Supercharge Your Science V.2

24 11 2011

I suspect a lot of ConservationBytes.com readers will be attending the imminent 25th International Congress for Conservation Biology to be held in Auckland from 5-9 December 2011 (it was to be held in Christchurch, but the venue was changed after that city fell down). I’ve now been to 3 previous ICCBs myself, and it should prove to be a good, informative (and fun) meeting.

I’ll be giving a talk or two, as will some of my students and postdocs, but I’m not spruiking those here (but you’re all invited, of course).

The main reason for this short post today is to advertise for Version 2 of our (i.e., Bill Laurance and me) popular ‘Supercharge Your Science‘ workshop. Yes, the organising committee of the ICCB decided it was a good idea to accept our application to repeat our previously successful series of presentations extolling the virtues of positive and controlled media interactions, social media and good writing techniques for ‘supercharging’ the impact of one’s science. You can read more about the content of this workshop here and here.

The description of the workshop (to be held from 19.00 – 21.00 on 6 December in the SkyCity venue) on the ICCB website is: Read the rest of this entry »





Don’t torture your readers II

22 02 2009

The second instalment of “Don’t torture your readers” (an attempt to stimulate better writing in conservation science) follows with some more mistakes, bad grammar and personal pet peeves.

  • DECIMATE (as in ‘… the population was decimated following…’) – I’ve seen this one used way too often. It is usually invoked by the author to imply some devastating reduction in population size (somehow it sounds bad); for this reason alone, the emotive language should be avoided. However, ‘decimate’ has a specific meaning: to reduce by every ‘one in 10’ (hence the ‘deci’ prefix). If you really mean the population was reduced by 10 %, use ‘decimate’. If you are just stating the population was reduced, state by how much and avoid emotive and incorrect terms.
  • DRAMATIC(ALLY) (as in ‘… we observed a dramatic decline in…’) – another meaningless, emotive word that belongs in the theatre, not in scientific writing. Quantify your meaning instead of relying on subjective terms.
  • CRITICAL(LY) (as in ‘… highlights the critical importance of…’ – This term is generally meant to communicate some urgent need or absolute necessity. While most authors would like to think their chosen topic is ‘critical’, many neither define to whom or what the results are ‘critical’, or even what the lack thereof would entail. In some circumstances it is used to infer some sort of threshold beyond which another state dominates, so I question the need for ‘critical’ at all in conservation writing. If you are trying to inflate the importance of your work, ‘critical’ is the word to use; if you mean a threshold, then simply state so.
  • FEW versus LESS – I’m amazed this still stumps so many people. ‘Few’ should be used to define a small number of countable (discrete) items (e.g., individuals, quadrats, plots). ‘Less’ should be applied to a measurable, continuous variable (covariate) that cannot be easily discretised (e.g., water, biomass, carbon). If you ever see someone write ‘less individuals’, get out the big red pen.
  • DATA – While on the subject of quantification, the word ‘data’ should always be followed by plural forms of the verbs (e.g., ‘… the data are…’; ‘… the data were…’). A singular ‘datum’ is one measurement and requires the singular form. A ‘dataset’ is a single group of data, so it too can use the singular form. If you want to communicate that your sample size was too small (for your intended purposes), you need to write ‘too few data’.
  • MIGHT/CAN versus MAY – I’ve often got this one wrong too. ‘May’ implies doubt or permission, so it is most often better to use ‘can’ or ‘might’ (where appropriate) when you expressly mean ‘under certain circumstances’.
  • THAT versus WHICH – This is not an easy one, and for a full discussion, visit this link. In the most basic description of the difference, ‘that’ usually introduces essential information in a restrictive clause, whereas ‘which’ introduces additional information in a non-restrictive clause. Quoting from the link given above provides some more clarity:

“What is FASCINATING to me is that . . . one way to determine . . . the correct word . . . is to ask the question, ‘Does the clause clarify which of several possibilities is being referred to?’ If the answer is yes, then the correct word to use is that. If the answer is no, the correct word to use is which.”

Seems somewhat counter-intuitive, but it’s correct (hence the confusion).

CJA Bradshaw

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to Furl





Don’t torture your readers

9 02 2009

This may seem a little off-topic for ConservationBytes.com, but I thought it pertinent to communicate how bad English hampers the understanding, popularity and implementation of good conservation science. I’ve started a list of common errors, unnecessary jargon, bad phrasing, archaic usage and overly complex constructions that I often see in conservation writing. Many of these are personal preferences, but I try to justify my suggested alternative in each case. Some of these apply to general English writing, others to science only, and others just to conservation/ecological fields. My hope is that students and young researchers can use my advice to improve the clarity of their writing. This first list is only preliminary – later posts in this theme will appear as I record more examples.

  • CONDUCT (as in ‘… we conducted the experiment…’) – What is wrong with ‘do/did’? I have never seen a scientist ‘conduct’ anything, but I have seen a few good operas.
  • PERFORM – See ‘conduct’. While some scientists would probably be more effective Thespians, let’s keep the theatre out of science.
  • VERY (as in ‘… there are very few species…’ – ‘Very’ has no place in scientific writing – I defy anyone to quantify what it means (i.e., it has an entirely subjective interpretation).
  • QUITE – See ‘very’.
  • SITUATED (as in ‘… our study area was situated in…’) – Simplify to ‘is/was’. Much easier, isn’t it?
  • SIGNIFICANT (as in ‘…this result has significant implications for…’; ‘… significant scientific advances…’; ‘… the functional significance of…’; ‘… can play a significant role…’ – This is probably the most abused word in science today. All the former examples mean nothing and are entirely dependent on the subjective position of the reader. Used without a statistical meaning per se (but more on the abuses of ‘significance’ as an arbitrary statistical paradigm in a later post), ‘significant’ and her sisters (e.g., ‘significantly’, ‘significance’) have no more place in scientific writing than ‘very’. Students often invoke this word simply to sound more scientific. Rubbish.
  • TO BOLDLY GO (i.e., any split infinitive; I couldn’t resist using one of the more infamous split infinitives) – I believe the jury is out really on the acceptable use of split infinitives, and I may be losing the battle, but an infinitive (for those of you who are grammatically challenged, an ‘infinitive’ is the base form of the verb prior to conjugation) can never be split by an adverb in English. How many times have you seen ‘… to significantly affect…’, ‘… to adequately measure…’ or ‘… to properly test…’. Sorry, all wrong (should be ‘… to affect significantly…’, etc.)
  • 10m (as in ‘… transects were set every 10m along…’) – You cannot write ’10metres’, so why, oh why, do people insist on sticking unit abbreviations next to the number? It should be ’10 m’!
  • i.e./e.g. – These abbreviations, id est and exempli gratia, literally mean ‘that is’ and ‘for the sake of example’, respectively. They are two words abbreviated each, so a full stop is required after each letter. Absolute correctness normally dictates the addition of a comma after the final full stop, but many journals drop the comma for whatever reason.
  • cf.confer (compare). It is one word, so its abbreviation requires a single full stop after the ‘f’.
  • its/it’s – Why is it so difficult for people to understand this one (especially in Australia)? In almost every other circumstance, an apostrophe followed by an ‘s’ indicates possession to a singular noun, as in ‘…the transect’s divisions’, ‘…the nearest neighbour’s value…’, etc. When the noun in question is plural, then the apostrophe sits nicely outside the terminal ‘s’ (e.g., ‘… the species’ attributes…’). This is a quasi-universal law EXCEPT for its/it’s. In this case ‘it’s’ is the contraction of ‘it is’, so ‘its’ becomes the possessive form. So, you can write ‘…its burrow…’, but ‘…it’s burrow…’ is incorrect. Still confused? There’s a simple way to remember – whenever you see ‘it’s’ in front of something, say ‘it is’ to yourself and see if the phrase makes sense. If it doesn’t, then it should be ‘its’.
  • CONTRACTIONS (e.g., ‘can’t’, ‘won’t’, ‘it’s’) – These are colloquial forms and should never be used in a scientific manuscript.
  • IN ORDER TO (as in ‘… in order to compare the plots…’) – What’s wrong with just ‘to’? I have rarely seen a situation requiring ‘in order to’. Unnecessary verbiage.
  • HAS BEEN SHOWN TO (as in ‘… is a species that has been shown to demonstrate a…’). There is simply no need for this verbiage. Simply state ‘…is a species that demonstrates a…’ and then reference the statement properly at the end of the sentence.
  • ABBREVIATIONS, ACRONYMS AND INITIALISMS – Use sparingly, if at all. They are often discipline-specific and have no meaning outside relatively small circles.
  • UTILISE – Just write ‘use’. For some reason people believe ‘utilise’ sounds more technical. Rubbish.

CJA Bradshaw

Add to FacebookAdd to NewsvineAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to Ma.gnoliaAdd to TechnoratiAdd to Furl