Tag Archives: Academic writing

Semicolons

As promised, today’s topic is the semicolon. I am always surprised by the numbers of writers who express genuine apprehension about semicolons. Somebody out there is giving people the idea that the semicolon is a risky bit of punctuation. There is, of course, the possibility for some initial confusion about the semicolon (is it more like a period? or a comma? or a colon?). But academic writers–people who study topics that are genuinely challenging (every semester I get at least one actual rocket scientist in a class)–can definitely handle the slight complexity of semicolon use.

The semicolon has two main uses. One, we can use a semicolon in the place of a comma in a list; in such cases, the semicolon does more than a comma, allowing us to include complex elements without worrying about unnecessary ambiguity. Two, we can use a semicolon in the place of a period between two complete sentences; in such cases, the semicolon does less than a period, allowing us to express a close topical connection between two independent sentences.

1. Semicolons in lists:

When we need to separate list items that are themselves complex or that have internal punctuation, semicolons work far better than commas.

Example:

The research on workplace equity confronts three main issues: the difficulty of finding an acceptable definition of workplace equity; the tension between workers, given that inequity will be perceived differently by different groups; and the tendency of managers to value business performance over working conditions.

Compare that to a version with commas between the list items:

The research on workplace equity confronts the difficulty of finding an acceptable definition of workplace equity, the tension between workers, given that inequity will be perceived differently by different groups, and the tendency of managers to value business performance over working conditions.

In the second example, the comma before given can create confusion. If we use semicolons, however, we avoid that ambiguity. (I will devote a future post to the topic of lists; we all use lists extensively in our academic writing, and there are things we can do to make sure we are using them effectively.)

2. Semicolons between sentences

When we need to divide two independent sentences while still maintaining a close thematic connection, semicolons work well.

Example:

Drying shrinkage can be eradicated by the application of the proper curing method; this reliance on curing means that we will need accurate measurements of the free water left in the concrete and of the relative humidity of the environment.

Compare that to a version in which both ideas are expressed in a single sentence:

Drying shrinkage can be eradicated by the application of the proper curing method, which means that we will need accurate measurements of the free water left in the concrete and the relative humidity of the environment.

The second example is less effective because of the potential ambiguity of the referent for which. You could, of course, solve this problem with a period instead of a semicolon. But those two sentences (having once been a single sentence) would still be closely connected. Showing that close connection with a semicolon can be a useful approach, especially for the academic writer who is looking to make complex connections among ideas without writing dense or ambiguous sentences.

The semicolon, as we just saw, can divide grammatically while uniting thematically. As readers, we are attuned to the end of sentences: when we reach the end of a sentence, we happily stop thinking about the grammatical relationship between the parts of the sentence. So ending sentences can be good. But short sentences–if they are too common or come in batches–can be bad.  A semicolon allows us to end a sentence while explicitly continuing our treatment of an idea. This benefit of semicolons is why we so often see them used with conjunctive adverbs and transitional expressions (e.g., however, instead, nevertheless, specifically, equally important, for example, in fact, on the contrary):

Ninety percent of Canadians recognize the components of a healthy diet; however, they fail to apply this knowledge when selecting foods.

These sentences could be separated by a period, but most of us will want to keep them closer than that. This is not to say that transitional words or phrases demand semicolons but rather that semicolons will often accurately reflect the relationship we are constructing when we use such expressions.

If you have questions about semicolons, please feel free to ask them in the comment section. And if you have no such questions, you should be happily using semicolons to great advantage in your writing. The next post will be an editing tip: identifying your own scaffolding phrases.

Links: National Inquirer meets Scientific Research, Social Media and Academics, DIY Academic Sentence Generator

This article from Inside Higher Ed describes an attempt by the Association of American Universities (AAU) to defend scientific research from mockery and poorly informed funding cuts. Make sure you follow the link that the article provides to the Scientific Inquirer: it will take you to the inaugural issue of a faux tabloid that the AAU has created to show how easy it is to make valuable peer-reviewed research sound daft and wasteful.

Here is something from The Chronicle of Higher Education on social media use among academics. The author summarizes a new report from the Centre for Information Behaviour and the Evaluation of Research, a report which found that academics are using social media for “collaborative writing, conferencing, sharing images, and other research-related activities”. Interestingly, the researchers found that the biggest users of social media were scholars from the social sciences and humanities; they suggest that the appeal of social media for these scholars may be the increased speed of dissemination. The Scholarly Kitchen provides a nuanced and critical response to the study, focusing in particular on the fact that ‘social media’ has been defined so broadly in this study as to be potentially meaningless.

Finally, if you have some extra time on your hands this weekend, The University of Chicago Writing Program has a great game for you: Write Your Own Academic Sentence. Here is my first attempt: ‘The linguistic construction of the gaze invests itself in the historicization of agency.’ To get the full effect, you need to use the ‘Edit It’ option on your sentence. Once you are satisfied, you can even get a second opinion from a virtual critic: ‘Your desultory treatment of the linguistic construction of the gaze deserves the obscurity into which it has fallen.’ Hopefully, once you are done playing, you’ll be able to get back to writing sentences that don’t sound like this!

P.S. I mentioned the Day in the Life of the Digital Humanities project last week. If you are interested in seeing how this project is shaping up, you can follow blogs from all the participants here.

How I Always Exaggerate Everything!

This blog has now been in existence for a few months, and I would like to pause for a moment to reflect. I have been pleasantly surprised by the degree of interest and would like to thank you all for reading and commenting. I particularly want to thank those of you who have blogged about my blog, added me to your own blogrolls, and tweeted my new posts. (Special thanks are due to The Thesis Whisperer and College Ready Writing.)

Although the actual experience of writing this blog has been very different than I expected (who knew it would be so engrossing!), the actual content has worked out more or less as planned. As was my intention, I started with key principles, sources, and strategies, all of which I hope will now act as a foundation for future posts. The completion of this initial phase means that I am now ready to tackle more specific writing issues. But I find myself unable to decide what to tackle first. I have given a ridiculous amount of thought to this–as those of you who know me will find easy to believe–without coming to a decision. What I did come to, however, was a realization about a quirk in my teaching style.

In trying to find a suitably significant topic for today’s post, I thought back to things that I present to my students as particularly significant. What I realized was that I use a lot of superlatives. In fact, you might say that I use superlatives in a way that suggests I don’t actually understand the concept of a superlative. ‘This is my favourite strategy’, I say about many different strategies. ‘This is the most important idea that you will learn in this class’, I say in reference to many different ideas. There is a pleasing simplicity about saying ‘If you learn one thing in this class, it should be this.’ But that simplicity is undermined by saying it about more than one thing. In course evaluations, students often mention my enthusiasm, although it sometimes sounds less like praise and more like pity (‘Rachael sure does care a lot about grammar.’) I am sure, overall, that my enthusiasm is a strength in the classroom. But the indiscriminate enthusiasm could stand to be replaced with something a bit more measured.

From now on, I will attend to the true meaning of ‘favourite’–and ‘best’, ‘most’, etc.–and exercise an appropriate degree of restraint. For instance, I will have only one favourite punctuation mark (although, at home, I will continue to have two favourite children because that is just sound domestic policy). Since that favourite punctuation mark is definitely the semicolon, I will make that the topic of my next post.

Links: Cascading Review, Digital Humanities, Research in Haiku

Here is a blog post from The Scholarly Kitchen that discusses some of the strengths and weaknesses of cascading peer review.  In cascading peer review, a journal will forward a rejected article (and its peer reviews) directly to another appropriate journal, thereby allowing the article to be considered by a new journal without the author having to repeat the submission process. This practice certainly isn’t standard but has been adopted by some larger publishers.

Friday, March 18 will be a Day in the Life of the Digital Humanities: “A Day in the Life of the Digital Humanities (Day of DH) is a community publication project that will bring together digital humanists from around the world to document what they do on one day, March 18th. The goal of the project is to create a web site that weaves together the journals of the participants into a picture that answers the question, ‘Just what do computing humanists really do?’ Participants will document their day through photographs and commentary in a blog-like journal. The collection of these journals with links, tags, and comments will make up the final work which will be published online.” Click here for an interesting collection of definitions of the digital humanities. I am planning to participate in this project, so I’ll discuss that experience in a future post.

Finally, here is a lovely blog devoted to dissertation research presented in the form of haiku. What makes this blog so enjoyable is that each contributor also provides a more conventional explanation of their research, allowing us to compare the poetry to the prose. I offer this link as more than just entertainment: I suggest trying it at home. Anything we can do to get at the essence of our research (even reconfiguring it as a 17-syllable poem) is likely to give us further insights into our own projects.

Links: Grammar Day, Fowler Revisited, Googlization

Happy National Grammar Day, everyone! Needless to say, it is always grammar day around here, but I am glad all of you get to join in the fun once a year. If you click the link, you’ll find grammar day e-cards, a theme song, a recipe for the grammartini, an exposé of common grammar myths, and even merchandise!

From The New Criterion, here is a review of Fowler’s A Dictionary of Modern English Usage: The Classic First Edition. In the review, Barton Swaim discusses a reprint of Fowler’s first edition as a way of revisiting the ongoing debate between descriptivism and prescriptivism. Swaim argues, in effect, that prescriptivism is both inevitable and way more fun. We will always, in his view, go looking for expert opinion about our writing decisions. And those expert opinions will be more stimulating than the bland descriptivist work of academic linguists. I have no trouble with the view that prescriptivism is more entertaining and often more immediately satisfying. However, what Swaim’s highly dismissive account of descriptivism fails to take into account is the possibility that persistent exposure to descriptivism might actually change the way we approach questions of usage. That is, a dominant descriptivist view might discourage our belief that all educated writers should use language in only one way and that all deviance from that way is deficiency. It may be unsatisfying to be told that a particular usage will be acceptable to some readers and unacceptable to others, but that may be all we, as writers, can hope for: a sound description of current practice to help us make up our own minds.

From Inside Higher Ed, here is a review of Siva Vaidhyanathan’s book The Googlization of Everything (And Why We Should Worry). The review includes an interview with Vaidhyanathan, who offers a clear-eyed case that Google’s ubiquity is significant. If everyone gets their information via Google, it matters how Google’s search standards are constructed. Despite its title, the book is clearly less a condemnation of Google, more a call for greater public conservancy of the wealth of human knowledge; Vaidhyanathan is urging us to have a serious discussion about the responsibilities and risks of the digital scholarship era.

Sentences

The most potent way to improve our academic writing is to think about meeting the needs of our readers. We are all generally aware that this imperative is true at the global level: we know that readers will be frustrated if our writing doesn’t meet their broad structural expectations. Sometimes those expectations will be based on their familiarity with the various genres of academic writing; sometimes those expectations will be based on promises we have made through our use of metadiscourse. We may also be familiar with the idea that readers need certain elements in order to understand our writing at the paragraph level. At the local level, however, writers are often unaware of what their readers will need for maximal enjoyment and comprehension. Indeed, readers themselves are often unaware of these needs; they simply know that some writing is easier to read than others. But if readers in fact have implicit expectations about writing at the sentence level, it seems plausible that writers could learn to anticipate and thus satisfy those expectations.

In this post, I am going to discuss three basic approaches to sentence clarity, each of which is designed to help us meet readers’ unconscious expectations. These strategies all derive from the helpful work of Joseph Williams. In addition of offering concrete advice for crafting better writing, Williams has given a superb articulation of why writing is so challenging to characterize: “But when we use clear for one [sentence] and turgid for the other, we do not describe sentences on the page; we describe how we are feeling about them. Neither awkward nor turgid are on the page. Turgid and awkward refer to a bad feeling behind my eyes” (Williams, Style, p. 17). This observation is crucial, especially for graduate students, who are often on the receiving end of feedback that they struggle to act upon. To be told, in so many words, that our writing is giving someone a headache is not without value. We need this knowledge because knowing the effect of our writing gives us the impetus to change; however, it is only the first thing we need to know. In most cases, such feedback gives us very little guidance about what to do next. We all need a subsequent step: what we will do after we learn that our writing is making our readers blink, shake their heads, and rub their temples.

That subsequent step can involve learning what readers generally need from sentences, so we have some strategies to work with when looking at our early drafts. Here are three such strategies to try during the revision process:

  • Place subjects and verbs together early in sentences
  • Choose stronger subjects and verbs
  • Employ the orienting–informing pattern

Subject-Verb Placement

The first strategy is the easiest: try to put your subject and verb close together and relatively early in the sentence. Before the verb arrives, the reader often feels a sense of incompletion. By delaying the verb, the writer is asking the reader to start processing the meaning of the sentence without its most crucial animating element, the verb. Consider this example:

Ontario’s education system, which requires further attention from policy makers to ensure coherence in educational policies, a need that has been extensively captured in a recent report published by the Canadian Education Association that evaluates the PCS policy initiative in its early years of implementation, is frequently cited for its attention to early years learning.

While we are working our way through all the things that the author wants to tell us about Ontario’s education system, we are likely slightly uneasy. The most common way of characterizing that uneasiness is to say that the sentence is ‘awkward’ or simply ‘too long’. Those judgments are the sorts that convey that there’s a problem but don’t tell us how to craft a solution. Observing that the verb comes very late in the sentence—42 words later—is a crucial step towards that solution. To rework this sentence, we could take all the things that the author wanted to say about Ontario’s education system and turn that into its own sentence, before turning to the ostensible main idea of the original:

Ontario’s education system requires further attention from policy makers to ensure coherence in educational policies, a need that has been extensively captured in a recent report published by the Canadian Education Association that evaluates the PCS policy initiative in its early years of implementation. Despite this lack of policy coherence, Ontario’s system is frequently cited for its attention to early years learning.

Both sentences now have a subject and verb that are close together, early in the sentence. The implication that the frequent citations happen despite a lack of policy coherence can be expressed in an opening clause before the second sentence.

Topic Subjects and Action Verbs

The second strategy asks us to look at what words we are choosing for our subjects and verbs. A grammatical subject that is also the topic of the sentence will give your writing more strength; similarly, a verb that is expressing the action of the sentence will make for writing that is more straightforward and energetic. Consider this example:

In the present study, focus is on the vibration of a clamped beam exposed to the recurring impact of the particles in a vibrationally fluidized bed.

Is ‘focus’ the true topic here? Does ‘is’ convey the action of the sentence?

The present study focuses on the vibration of a clamped beam exposed to the recurring impact of the particles in a vibrationally fluidized bed.

This revision makes the true topic (the present study) the grammatical subject and manifests the action with the verb (focuses).

Orienting–Informing Pattern

The final strategy is both simple and powerful: use the early parts of your sentences to orient the reader before providing novel information in later parts. This pattern is what Williams calls the old–new pattern, but I prefer to talk about the activities of orienting and informing. This terminology focuses on what the writer is doing for the reader: orienting and then informing. The concept of what is old and new often confuses writers with its hint of judgment; it is easy to think of new as good and old as bad. Talking instead about orienting and then informing puts the emphasis on the relationship being built between the writer and the reader who is being first oriented and then informed. Consider this example:

This slow reallocation is due to the high costs of job switching and migration. The Hukou registration system added to the migration costs. The Hukou system, which operated from 1958 until the late 1980s, was designed to restrict labour migration.

This slow reallocation is due to the high costs of job switching and migration. These migration costs were exacerbated by the Hukou registration system. The Hukou system, which operated from 1958 until the late 1980s, was designed to restrict labour migration.

This slow reallocation [orienting] is due to the high costs of job switching and migration [informing]. These migration costs [orienting] were exacerbated by the Hukou registration system [informing]. The Hukou system [orienting], which operated from 1958 until the late 1980s, was designed to restrict labour migration [informing].

Committing to this pattern is the best way I know to ensure that your reader will move easily from sentence to sentence.

You may have noticed that these three strategies for sound sentences don’t mention the use of active or passive constructions. This omission is purposeful: while there are many things to be said about active and passive, I find that these three sentence strategies are the best way to craft energetic writing. Most everyone is familiar with some degree of moral panic about the passive voice, either from an early encounter with Strunk and White (“The active voice is usually more direct and vigorous than the passive”) or your grammar checker’s relentless warnings about the dangers of passive voice. While there are absolutely cases of unhelpful passives, these three sentence strategies will generally root them out without limiting your ability to enjoy the affordances of appropriate passive voice. This topic is one for another post—since this one is already too long—but I didn’t want to end without acknowledging a topic that is often top of mind for writers eager to improve their sentences.

This post describes the fourth of five key strategies for strong academic writing; I have chosen these five simply because they are the ones that I most frequently turn to in my work with students. In the four other posts, I discuss reverse outlines, paragraphstransitions, and metadiscourse.

For more on sentences, you can consult this other post:

Links: English as a Lingua Franca, Commercial Manuscript Editing, Literature Reviews and Social Media

Here is an article from The Chronicle of Higher Education on the use of English as a lingua franca outside of English-speaking countries. It discusses many interesting issues–such as the cultural dimensions of communication–before concluding with a highly provocative point about the future of English. Although it is often assumed that being a lingua franca is a sign of linguistic vigour, some scholars suggest the opposite: that the English that is spoken as a global lingua franca may actually be a much reduced language and thus a vulnerable one.

This article from  Nature discusses the role of commercial manuscript editing. The author does an excellent job highlighting the various practical and ethical issues associated with commercial manuscript editing services. In doing so, she also tells us something about the priorities of scientific journal editors. These editors want manuscripts that they can send to reviewers without worrying that flaws in the writing will get in the way of assessment. But, interestingly, the flaws that worry them are not just grammatical; they are just as concerned about a poor mastery of the conventions of research articles. Novice academic writers need to understand what a manuscript must accomplish if it is to be considered for publication. The assessment ‘weak writing’ can be used to describe anything from punctuation difficulties to a poorly articulated methodology to improperly presented data; understanding the importance of all these elements will allow writers to see that the work of improving their writing will have to proceed on many fronts.  

Lastly, here is a blog post from The Thesis Whisperer about the similarity of literature reviews and social media. The specifics are Australian, but the general ideas about literature reviews are broadly applicable. And the social media analogy is clever. You really don’t have to be friends with everybody or every article!

Transitions

Learning how to make effective transitions is essential to strong academic writing. A lack of comfort with making transitions is one of the causes of the short paragraphs that so often afflict novice academic writing. When we do not know how to make smooth transitions, we are more likely to add in unnecessary paragraph breaks, imagining that starting a new paragraph will solve the problem. But creating short, choppy paragraphs only exacerbates the problem. Instead, we must focus on creating effective transitions between sentences, which we generally do in one of two ways: we use transition words or we use textual linkages. Both strategies have a role to play, but novice writers, unfortunately, often see transition words as their only way of moving from sentence to sentence. This over-reliance on transition words  is actually detrimental to our writing and blinds us to the possibility of using textual linkages to create more meaningful connections between sentences. Transition words are easy and thus allow us to avoid the hard work of grasping the actual connections in our texts. Indeed, texts full of transition words may actually feel choppy because unnecessary transition words can obscure the true nature of the relationship among sentences.

Here are a few key principles to help create clear transitions in your writing:

1. Avoid unclear reference. The single most important way of linking your sentences is through clear reference. Contrast these two simple examples: ‘A is connected to B. This is…’ and ‘A is connected to B. This connection is…’. Without the summary word (‘connection’), we cannot tell whether the ‘this’ in the first example refers to A, to B, or to the connection between them. We call this pattern ‘this + summary word’. There will be times, of course, when the reference is obvious, but generally the reader needs to have reference made explicit. So a simple principle: never leave a ‘this’ orphaned and alone.

2. Avoid unnecessary transition words. The transition words most likely to fall into this category are the additive ones: ‘in addition’, ‘also’, ‘moreover’, ‘furthermore’. (Both ‘moreover’ and ‘furthermore’ can be correctly used as intensifiers—where one sentence deepens the claim of the previous one—but they are so often used to indicate simple addition that I am including them here.) My first approach to a word like ‘also’ is to remove it; if you are using it to say ‘here comes another related point’, it is probably unnecessary. If you are instead trying to make a more complicated connection, removing ‘also’ and adding a more substantive indication of that link will be far more helpful to the reader.

3. Avoid the mere appearance of causality. When we overuse causal words, we often undermine the actual connection we could be making. When we say ‘A exists. Therefore, I am going to study A.’, we are missing a chance to give an actual rationale for our research. Look closely at your use of causal words (‘therefore’, ‘thus’, ‘hence’) and make sure that they accurately reflect the relationship you are trying to convey.

4. Use transition words to indicate a change of direction in your text. Whenever we are disagreeing with ourselves, it is essential that we indicate this to the reader. Consider these simple examples: ‘There is plentiful evidence for A. I think not-A.’ and ‘There is plentiful evidence for A. However, I think not-A.’ The first example sounds like you might be unintentionally contradicting yourself; emphasizing your intentions with a ‘but’ or ‘however’ lets the reader know what you are up to.

I will also make two quick points about other types of transitions.

Paragraph transitions generally need to be more robust than those between sentences. This need for more fulsome transitions can mean that ‘this + summary word’ becomes  ‘this + summary phrase’, where the phrase is a fuller indication of what was discussed in the previous paragraph. It also means that transition words are often out of place in paragraph transitions precisely because they create such a tight relationship. There are, of course, exceptions to this, but as a general rule words or phrases like ‘however’, ‘in other words’, or ‘furthermore’ may puzzle the reader when they appear at the start of the paragraph; at the very least, they may send the reader back to the previous paragraph and that is not the direction in which you want to be pointing your reader.

Transitions between sections are a different issue again. Transitions between sections can be made in several ways: at the end of one section, at the beginning of another, or at an earlier point at which an overall structure is created. (For instance, in a literature review, a writer may say that she is going to consider the literature on a certain topic from three different perspectives. The reader will then be fine with three independent sections without any explicit transitions between them.) One simple piece of advice for section transitions: do not rely on the section headings to accomplish the transition for you. As a rule of thumb, I suggest reading through section (and sub-section) headings as though they were not there. Not that they should actually be removed, but rather that the author should make sure that transitions are accomplished in the text, not through headings.

This post describes the third of five key strategies for strong academic writing; I have chosen these five simply because they are the ones that I most frequently turn to in my work with students. In the four other posts, I discuss reverse outlines; paragraphs; sentences; and metadiscourse.

For more on transitions, you can consult these other posts:

  • In Full Stop, I talk about the way we create flow across sentences.
  • Breaking Points looks at how we can signal the relationship we are trying to create between paragraphs.

Links: Job Interviews, Being Literal, Scholarly Reportage

Writing in Inside Higher Ed, this blogger suggests replacing face-to-face job interviews with video conferencing. I am not sure whether this is, in fact, a coming trend, but I did wonder what such a shift in practice might mean for how both sides of the equation assess the interaction. Communicative cues are so complex; what would we need to learn to present ourselves successfully via this new medium? Also writing in Inside Higher Ed, Dean Dad makes one particularly important suggestion: all interviews in a given round would have to be conducted via the same technology. It would not be fair for some to have face-to-face meetings while others had to engage in the more complex task of presenting themselves remotely.

This post from Motivated Grammar addresses the difference between prescriptivism and preference. In this case, the author dislikes ‘literally’ when it is used as a general intensifier (and thus not in the literal sense of literally) not because such usage deviates from some rule but because it is hyperbolic. Although most of us rarely need to take a position on the prescriptivism debate, we do need to think about how we make writing decisions. When we argue against a particular usage on the grounds of its actual strengths or weaknesses, we do more for the overall health of our writing than when we protest against its deviance from some imagined norm.

This article from Inside Higher Ed considers whether scholarly reportage–a method of inquiry that combines social theory with accessible narrative–is a passing trend or a credible option for scholarly work. I am curious what a method that blends ‘scholarship, memoir, and journalism’ might mean for writing. Working in multiple genres for a multifaceted audience would demand impressive writing skills. For more on this type of writing, here is an article from Dissent about Andrew Ross and the last twenty-odd years of cultural studies.

Paragraphs

A crucial strategy for improving academic writing is to pay attention to the importance of the paragraph as a unit of discourse. Novice writers tend to think of both full texts and sentences as areas for improvement, but they give less thought to the role of the paragraph. They recognize, of course, that a full text must possess a certain communicative goal, and they understand that sentences are the building blocks of the whole. But paragraphs? In my experience, these intermediate units are consistently neglected. This neglect greatly underestimates the important role that paragraphs play for the reader. A paragraph break means something to a reader; when we move from one paragraph to another, we imagine that we are leaving one thought (or issue or topic or argument or point or perspective or piece of evidence) and moving on to another. We attempt, in other words, to find some unity within a paragraph and to discern some diversity between paragraphs. When the writer has not managed paragraphs well, those attempts will lead us—consciously or not—to be disappointed. Most of us benefit from adding paragraphs to our list of things that must be effective if our writing is to succeed. To that end, here is my list of four things I wish every academic writer knew about paragraphs:

1. That they are very important. Simply stated, effort should be devoted to working on paragraphs, as well as on sentences and full papers.

2. That they usually need a topic sentence. The ‘usually’ is there to avoid the appearance of dogmatism, but I do in fact advise writers to start with the assumption that every paragraph will require a topic sentence. The main exceptions are introductory paragraphs (which often, in effect, act as a kind of topic paragraph for the whole text), transitional paragraphs (which exist to signal a significant shift in topic), and serial paragraphs (all of which refer back to a single topic).

3. That they should be thematically linked. The rest of the sentences should be recognizably about the theme announced in the topic sentence. These thematic linkages should also involve noticeable linguistic linkages, accomplished through strategic repetition and the use of key terms.

4. That their length is meaningful. The length of a paragraph should be determined by the demands of content, not by the number of sentences or space taken up on the page. When I ask students for the rationale behind a paragraph break, they frequently say something to the effect of ‘I thought it had gone long enough’. (The phrase ‘my high school English teacher always said …’ also comes up a lot in this regard, but the ongoing trauma of a high school English education lies well outside the scope of this post!)

Paying more attention to paragraphs can, needless to say, improve their internal cohesion. But this attention to paragraphs is also a key way to improve the overall coherence of a complete text. Our ability to engage in thorough structural revision can often be undermined by the difficulty of finding our way into our own text. Once that text is thought of as a series of paragraphs—each of which has an explicit role to play—we are better able to grasp the overall demands of structure.

This post describes the second of five key strategies for strong academic writing; I have chosen these five simply because they are the ones that I most frequently turn to in my work with students. In the four other posts, I discuss reverse outlinestransitions, sentences, and metadiscourse.

For more on paragraphs, you can consult these other posts: