On Academic Writing

Choose your audience

Think who you are writing for. In academia that often translates into what journal are you submitting your paper to? or if it is a research proposal, what panel is going to read it? If it is your master’s or PhD thesis, who are the readers? who is the opponent, the examining committee? Each of these formats have bounds and expectations on length, order of sections, or a minimum set of features that need to be part of your product.

Thinking on your audience makes your writing more empathetic. It helps you tune the language to the expectations and the format targeted. For example, a journal like Ecology and Society expect research articles of 5000 words, Environmental Research Letters 4000 words, Science 3000 words, Nature 2500 words. Some social science journals are more generous, Global Environmental Change tops at 8000 words for research articles, while Annual Reviews offers something like 9000 for literature reviews.

Journals like Nature, Science or PNAS are interdisciplinary journals at heart. They expect as little jargon as possible while still being rigorous. Often their format follows the Introduction, Results, Discussion within their word limits, leaving the methods at the end for the interested reader (and with different word count, ~3000 words). That means you need to tailor the message in a way that your average reader gets what you did and how you find your conclusions without diving deep into the how — the methods.

Disciplinary journals (like E&S, GEC, ERL) are more niche. Your audience are people within the same disciplinary interests as yours. On one hand it means you are free to use more standard jargon that is common in your discipline without having to define it (e.g. niche or biodiversity in ecology, or inequality, discount rates, or opportunity cost in economics). They also follow the more tranditional Intro - Methods - Results - Discussion - Conclusion structure. Be aware of different practices in referring to literature cited. Make sure to read their information to authors page where you often find additional formating information on figures, tables, measuring standards, or guidelines for supplementary materials.

Make an outline

Once you have chosen your target journal and your audience, it is useful to think on the best way of delivering your message. Here I assume you’ve already done your research, you already have results, you already know what do you want to write about. At this point my advisors often told me to chose what are the top (3) take home messages I want my readers to remember from my work. These messages should match my number of figures and their content, each figure helping me support my key points. It is a good time to sit with your co-authors and decide on the outline, this is what messages need to be deliver and when.

Let’s assume you’re targeting E&S, a 5000 word outline. It could look something like:

• Introduction [1000w]

• Methods [1000w]

• Results [1500w]

• Discussion [1000w]

• Conclusions [500w]

Now, if an average paragraph is 250 words, that means you only have four paragraphs in your intro. You can further refine as:

• Introduction [1000w]

  • What is the problem [250w]

  • What have others done to solve it and fall short [250w]

  • What is the research problem this articles aims to solve [250w]

  • What is the question and how is it going to be answered [250w]

These questions are just prompts. Actually your concise and clear answer to each of these questions should be the lead sentences of each paragraphs. The sentence tells the reader the red thread, the flow of the story, and the most important information. If you were to present your paper in 5 minutes with 5 slides, and 3-5 bullet points per slides, these points are the key points you want the reader to remember. Agree on the take home messages early with your co-authors, it is easier to edit and refine text once everyone is on the same page on what the structure of the manuscript should reflect.

Keep balance: the hourglass model

The journal Nature has a very useful guide on how to write good abstracts. That is a 250 word paragraph that summarizes your paper. It follows the hourglass model. It starts with a set of general statements that introduces the reader to your problem at hand. Then it narrows down the general problem to a specific one that you’re set to solve. These two components often reflect the content of the Introduction, where the last paragraph reflects more concretely the aim of your paper and your research question. The next sentences reflect the methods describing what you did to answer you questions, and the results followed quickly on what you found. The rest of the abstract explains why that result is important, trying to position it back to the general context. Here is where you help the reader interpreting the break through but also the limitations of your work.

The same structure is expected in your paper. The Intro and Discussion mirrors each other in bringing the reader from the realm of the general problem to the specific question you’re attacking, and back from your results to why they matter. The Methods and Results section mirrors each other as well. Here you are more specific on your language to help the interested reader replicate what you did, or be precise on what you found. You want to be concise here, don’t lose your readers on the details. In some journals, methods go at the end precisely to avoid distractions, and many results end up in supplementary material to show your reviewers all the robustness tests you made to make sure your claims are correct, but without distracting the non-expert reader from the take home messages of your paper.

The hourglass model helps you keep balance. The length of Intro and Discussion as well as content and level of generality should be matched. A paper with a very long Introduction but relatively short Discussion seems off, it’s unbalanced. The same is true for the Methods and Results section: if you applied two different methods, one expect at least two key results coming from each approach. Common mistakes that we all do is mixing parts of the method in the introduction, or mixing material for the discussion in the results section. It feels that we can’t wait to tell the reader why some results looks the way it does. While there is no right or wrong ways, it helps to keep the results as factual as possible, and condense the interpretation in the discussion. Finally, the abstract is a summary of the whole paper, and it mirrors in (more or less) length and importance the conclusion paragraph. Some journals do not have conclusion section and dislike paragraphs summarizing what has already being said. In that case, the last paragraph of the discussion section is your conclusion. Bear that in mind to make sure your reader gets a happy forward looking ending.

Keep it simple

Keeping it simple is probably the hardest part. The simpler it is, the easier your reader will get what you mean. But what does it mean to keep it simple? In writing, let’s start by getting rid of jargon. Jargon are overly technical wording that you might think makes the text sound elegant but in reality might be confusing more than helping. If you have to use jargon, always define it on its most simple way, without introducing more jargon. Another way to keep it simple is by using short sentences. Remember, a sentence starts with a subject that performs a verb with some complementary information at the end. Follow that recipe. Avoid complicated constructed sentences that span more than one or one and a half lines of text. If it get’s long, it is always possible to split in two.

The most important sentence are the first sentences of your paragraphs. They are called lead sentences. They should be affirmative, short, and contain the take home messages you want to communicate (e.g. the bullet points of your slides). In some languages like Spanish or German, we expect the pun to come at the end. English speakers, however, expect that piece of information early, at the start of the paragraph. Avoid using the first sentence to describe a figure for example, instead of writing:

• Figure 3 illustrates on the x-axis the level of hunger in cats while the y-axis shows their happiness, as you can see, cats when they are hungry can be more friendly and needy

You can write: “Cats become friendly when hungry”. Within a sentence, the most important information also comes first. In this case, the important actors are cats, not the figure really. Another example is:

• As the prominent scholar XYZ (year) said in their seminal paper: “Me, myself and I”, the Sun is the middle while the Earth goes around it.

Avoid praising unnecessarily people, no one cares really who says what or when. That is what the reference is for and typically goes at the end of the sentence. Unless the central topic of your writing is that particular person, or the context of writing becomes relevant to understand the meaning (e.g. it was the year of the war), then it is just noise, distractions to the reader. Instead you can rewrite as: “Earth rotates around the Sun”. The object performing the main action is the Earth, so it should come first, not last in the sentence.

Designing good lead sentences will help your reader finding key information. Nowadays most of us do not have time to engage fully with a text. Think of reading the news on a newspaper. If you have limited time to get the most of the content, you can read the title, sub-headings and lead sentences to find the info you were interested on, get the high level structure of the text. A good outline performs that function, lays out the structure of your argument in a logical sequence. Lead sentences make it clear and memorable, they can be the slides of your presentation.

Train for the long run

Writing is like running marathons. You need to train frequently, some do it every day. It does not have to be long, perhaps 15 minutes a day will do. But getting into “writing mode” gets harder if you have not done it in a while, as it is getting into running. It does not have to be a book project. It can be journaling, blogging, simply putting down some impressions of your day.

For people whose livelihood depends on writing successfully, it is good to remember that the quality of your text is independent of your emotional state. As a writer you will have good and bad days. Sometimes you will feel great and believe your text is amazing, and viceversa. But a week or month later when you revise and edit, you will realize that it wasn’t as great (or crap) as you thought, some edits are in order.

If you’re a student starting your career, remember quantity is seldom a proxy of quality. Ask yourself what was the last time you read a master thesis that was longer than 80 or 100 pages (believe me I’ve seen those). Or if you are writing the first paper of your PhD, when was the last time you read a 25k words paper? and was it your favourite? Actually if you look around scientific journals, the more interdisciplinary and wider the readership is, the shorter is the format they expect us to write.

Don’t let perfection be the enemy of the good. You can edit forever, but you are doing a disfavor to your current work and your next one in procrastinating too much with perfection. If you’re a PhD student, every extra day you spent in Paper 1 is a day less for Paper 4 or the kappa of your thesis. Let me put it clear: it will never be perfect. But you still need to graduate. If you aim to become a professional writer, you still need to sell some books to pay the rent. Put yourself some deadlines and keep them.

Share with your friends and colleagues your drafts. Ask for feedback and provide constructive criticism when asked for it. Often a fresh pair of eyes on a text can easily spot the structural weakness of our arguments that are blind to us seeing it every day. At the end is up to you to make the changes or not. But it is often useful to hear other people’s perspectives on your ideas. Taking the time to make other people’s work better is a gift, cultivate such friendships and be kind and constructive to others.

Spice it up

Scientific writing can be boring and dry, but it does not have to be. You can use analogies, metaphors, creative examples, even jokes. Some authors like to make titles inspired on Bob Dylan songs. Keep track of the articles you really like, not only because of the ideas presented, but for the way the authors captured your attention and made things fun to engage with. When I’m stuck with little inspiration I often go back to some of these favourites. To find some of these treasures, dare to read outside your disciplinary comfort zone. When you read other disciplines it is harder to engage with the technical content, which means your radar is more sensitive to the tricks authors use to keep you on the hook.

Writing is fundamentally a creative process. Nurture your creativity through experiences, read widely not just science. The more you read, the wider the repertoire of tricks you have been exposed to that you can use on your own writing. But inspiration also comes from unexpected places. A beautiful sunset, and art exhibition, an engaging conversation with a friend on the pub after work. Consider those moments of enrichment part of the exercise of writing, and just practice, practice, practice.