Scientific Writing = Thinking in Words

David Lindsay

• Editorial: CSIRO PUBLISHING
• Año: 2020
• ISBN:
  9781486311491
• Edición: 2
• Categoría: Pedagogía
• Categoría: Educación general

Telling people about research is just as important as doing it. But many competent researchers are wary of scientific writing, despite its importance for sharpening scientific thinking, advancing their career, obtaining funding for their work and growing the prestige of their institution.

This second edition of David Lindsay's popular book Scientific Writing = Thinking in Words presents a way of thinking about writing that builds on the way good scientists think about research. The simple principles in this book will help you to clarify the objectives of your work and present your results with impact. Fully updated throughout, with practical examples of good and bad writing, an expanded chapter on writing for non-scientists and a new chapter on writing grant applications, this book makes communicating research easier and encourages researchers to write confidently.

It is an ideal reference for researchers preparing journal articles, posters, conference presentations, reviews and popular articles; for students preparing theses; and for researchers whose first language is not English.

SECTION 1

THINKING ABOUT YOUR WRITING

1

A matter of attitude

Telling people about research is just as important as doing it. But many researchers, who, in all other respects, are competent scientists, are afraid of writing. They are wary of the unwritten rules, the unspoken dogma and the inexplicably complex style, all of which seem to pervade conventional thinking about scientific writing. This chapter brings these phantoms into the open, exposes them as largely smoke and mirrors, and replaces them with principles that make communicating research easier and encourages researchers to write confidently.

One of the greatest paradoxes in research is that, regardless of the field, work must be written and published before it can be considered complete, yet instruction in writing is rare in the training curriculum of budding scientists.

There is a common saying, ‘If you haven’t written it, you haven’t done it.’ A research project is not complete just because the last sample has been taken or the last set of data analysed. If you are in the world of research, it is of little value to have a colleague or two in the next office or laboratory know that you have discovered something. From the day that you completed undergraduate training and decided to become a researcher, your circle of colleagues or potential colleagues expanded from being a relatively few fellow students to an indefinite number of fellow researchers from all parts of the world. Communicating with them is a very different task from the one you were involved in as a student. In fact, you may have to spend as much time writing, reading or correcting manuscripts as you do on research itself. Even if you have told delegates at a large meeting or a convention what you have done, the proportion of scientists in your field that were there and listening to what you said is tiny and probably transient. ‘The spoken word evaporates but the written word stays on.’ The written word is permanent, all pervading and the best way to tell the world of research that you are a noteworthy part of it. In fact, one commentator described researchers as professional writers who do research to get material for their next research story. So, get used to the idea that you are a writer and work hard to become a good one.

If you haven’t written it, you haven’t done it.

Despite this, writing is one of the most inadequately developed of all the skills that scientists use in their research activities. Let us look briefly at the statistics:

• Probably 99% of scientists agree that writing is an integral part of their job as scientists.

• Fewer than 5% have ever had any formal instruction in scientific writing as part of their scientific training.

• For most, the only learning experience they have is the example they get from the scientific literature that they read.

• About 10% enjoy writing; the other 90% consider it a necessary chore.

These figures are, of course, approximate but they come from informal surveys conducted over many years in many countries and, I believe, are close to reality.

Beneath these statistics, it is easy to deduce a serious problem. For example, if 90% of scientists do not really enjoy writing then most of the scientific literature in front of us is written by people who did not enjoy writing it. The chances are that, regardless of the quality of the science, it has been cobbled together to get it published, reviewed by referees who have little more interest or knowledge about writing clearly than the authors and, finally, published in a style that has had little critical review. Thus, a big proportion of the literature on which developing scientists base their ideas of writing style and structure has been written and reviewed by people who knew little about style and structure, and probably didn’t enjoy writing anyway. That is not an effective model because it is highly variable and, on average, not very good.

We can develop the common saying further: ‘If you write it, but no one reads it, you still haven’t done it.’ The only reason for writing is to have what you have written read and understood by other scientists and this is often forgotten by scientists when they commit their work to paper. They believe – and are often encouraged to believe – that publication in a journal is the ultimate end point for a piece of research. It is not. Your paper must then be read and understood clearly by your scientific colleagues around the world in the relevant and related fields before the job can be deemed to have been completed successfully. So, we can extend the saying even further: ‘If you write it up and it is read but not understood you still haven’t done it.’

If you write it, but no one reads it, you still haven’t done it.

By contrast to the many bad models of writing that we come across, there are some beautifully written and structured papers that stand out like beacons because they are so clear to read and deliver their scientific message so forcefully. These are the models that we must attempt to follow. Unfortunately, they also stand out because they are much rarer than they should be.

The suggestions for better writing in this book draw directly and indirectly on these outstanding models and are usually presented as principles rather than rules. That is because rules imply strict adherence to a single way of writing and that would make scientific literature monotonous and dull.

But, applying well-thought-out principles to your own reporting style allows you to retain your personal imprint on your writing. It is up to you to decide if the principles make sense to you and, if they do, you can follow the further suggestions to modify the structure or the style of your writing to ensure that you are adhering to those principles.

Where does writing fit into the practice of research?

A successful piece of research yields results that are:

• defensible: the proper methodology and the model used to obtain them is appropriate and correctly implemented and the results are new

• robust: numbers are accurate and statistics are appropriate

• explainable: they can be put into perspective in relation to other people’s results and, possibly, can make a change in the ‘real world’.

Handling the first two of these is part of the research process with which most experimenters are familiar, but most are far less comfortable with the third, explaining in a research article what was done in a way that could make it stimulating for other researchers. This book concentrates almost entirely on this third component. But in doing so, it is important to remember that, in the world of research, bad information is even worse than no information. Bad information can mislead science for years. So, when embarking on your writing, it is vital to ensure that your results are defensible and robust and that you have systematically interpreted them thoroughly with colleagues, supervisors and anyone else who may be helpful. No amount of high-sounding prose can fix poor research. By contrast, good scientific writing can reinforce good research findings by making it easy to recognise that the work has been done well and readers can accept with confidence the new information in front of them.

Getting into the right attitude for writing

There seem to be two contrasting approaches to the recording and discussion of scientific results.

One is the positive attitude: ‘I have just been part of an adventure of discovery in science and I have found something that I want to share with you, the reader. In this article, I am going to take you on the same adventure and tell you what made me excited about it. In doing so I am confident that you will recognise and appreciate my scientific contribution.’

The other is far more passive and, regrettably, seems to be more common: ‘Research is the seeking and discovery of information that was not known previously. I am writing this for you, who have been trained a scientist to seek out information and make something of it. I am putting the data before you, together with some interpretation and I expect you to use your skills to work out much of what it means.’ This description of the second approach may seem harsh, and I am sure that no one thinks this way explicitly, but I believe that it is a fair interpretation of how many modern scientific articles are presented to readers.

If scientific articles are written to be read, then it is important for you as a writer to have a realistic impression of the sort of person who is likely to be a reader and how they go about reading. In reality, potential readers are not likely to be motivated much differently from you. That means that they are probably busy, lazy, impatient, grumpy and preoccupied with other things just as you are. They have other things than reading scientific articles on their daily agenda and they will be happy to convince themselves that they don’t need to read many of the articles in the journals that cross their desk or computer screen. They certainly will not be reading articles just in case they contain some unforeseen but useful material hidden in some obscure paragraph. So, your first task is to attract their attention and then try to hold that attention until the last full stop. That should be your goal but, even with a well-written article, it is unlikely that you will often achieve it. At least in the first instance, readers are selective until they get a feeling for the article and what it has to offer them. Then, if it really interests them, they will come back and scrutinise the whole article carefully and with scientific interest. The challenge is to make sure that, even if they spend just a few moments perusing your article, they will pick up the essentials of what it has to say. This means that they must find the most important parts clearly presented and in the places where they expect to find them. If they are forced to find your most interesting data buried in a heterogeneous mass of information in the Results or your most brilliant inspiration among a series of problematical comments in the Discussion, you will have little chance of having your work acknowledged or appreciated.

To write a paper successfully, you have to do more than commit your data and comments to paper: you must work hard to ensure that your data and comments are structured and presented so that the reader has easy access to them.

What is a ‘good’ style for scientific writing?

In writing scientific articles, many of us struggle to achieve a style of writing that does not come naturally to us. We imagine that we must follow a convoluted style based on vague impressions of what we have read in the scientific literature. Nothing could be further from the truth, and it is here that many of the models that we attempt to copy in the literature let us down.

There are just three immutable characteristics of good scientific writing that distinguish it from all other literature. It must always be precise, clear and brief, and in that order. If it is vague, it is not scientific writing; if it is unclear or ambiguous, it is not scientific writing; and if it is long winded and unnecessarily discursive, it is poor scientific writing. But do not sacrifice precision or clarity in order to be brief. So, for example, if it takes a few more words to make what you want to say crystal clear to as many readers as possible, then use those extra words.

The good news is that, if you are precise, clear and brief, then you do not have to conform to any other specific rules to be a good scientific writer. The style of good scientific writing is plain and simple English similar to that you would use in a conversation with a colleague. Or, as one author put it, ‘the best style is no style at all’. That is also good news because it is the style with which we are most familiar and most skilled. We use it every day, we get constant feedback on how successful we have been in getting across what we want to say and we are therefore confident with it. When writing about research, we often have to explain procedures and concepts that are complex, so it makes sense not to add further complexity by struggling with words and expressions that are unfamiliar to both the writer and the reader in order to conform to some imaginary style.

Of course, you may decide that you want to impress your readers with your knowledge and command of English. If so, think again. You should be writing to inform, not impress. Sometimes, when I say this to young scientists, they ask whether editors of scientific journals or reviewers might think them naive and unscientific if they use simple language. I can’t speak for all editors and reviewers, but I cannot imagine any of them complaining that authors were not obscure enough in explaining themselves. If you are a scientist and your ambition is to gain the Nobel Prize one day, then try to get it for your science and not for your literature.

There is another reason for writing in plain, simple English rather than using flowery, ornate or obscure prose. English is the de facto language of modern science; over 95% of more than 10 000 journals scanned by the analytical firm, Thomson Reuters, for calculating impact factors are in English. Yet, depending on the field, 50% or more of the researchers who may read a scientific article may not have English as their first, spoken language. If these people are discouraged by having to search for their dictionaries to understand what native English speakers have to say, the whole purpose of writing the article – to have it read and understood – will be lost. In fact, with the increasing spread of scientific expertise around the world, native English speakers have a serious obligation to their non-English speaking colleagues not to flaunt their good luck by inserting obscure words and expressions. Such words and expressions may be impressive, but for the wrong reasons.

If you are a scientist and your ambition is to gain the Nobel Prize one day, then try to get it for your science and not for your literature.

Remember, your primary aim when writing a scientific article should be to have as many people as possible read it, understand it and be influenced by it.

2

The fundamentals of building the scientific article

Most people associated with science and research agree that writing and publishing the article that describes their experiment is an integral part of the research process. Unfortunately, many think that this process is accomplished in three distinct phases: planning the work, doing the research and writing it up. That is a pity, because all three phases are so closely integrated that none can be completed successfully without involving the other two.

The relationship between good planning and the smooth execution of a research program is obvious, but the importance of thinking about writing the article during the planning is often overlooked. The title of this book Scientific Writing = Thinking in Words came from the conviction that thinking and reasoning at the planning stage facilitate both the experimental and writing stages and, if well thought through, writing an experiment can be as stimulating as doing the experiment itself and certainly not, as many people seem to feel, a necessary but unpleasant task.

Thinking and reasoning at the planning stage facilitate both the experimental and writing stages.

Broadly, the thinking process in writing a paper parallels that for desig...