In scientific papers, the basic elements of any intellectual or research argument are paragraphs. Each paragraph should be a single unit of thought, a discrete set of ideas made up of closely related sentences.
The most generally applicable sequence to follow is — Topic, Body, Tokens, Wrap.
The opening “topic” sentence alerts readers to a change in topic and focus, and tells readers (in “panel” mode) what the paragraph covers. It should never be linked backward to material that preceded it (links are always made forward in "wraparound" sentences).
So beware of starting with linking words (such as 'However', 'Nevertheless', 'In addition') lest they lead you to look back. Instead, topic phrases should clearly signal a new focus.
Yet they also need to be carefully written, to give readers the impression of a smooth, "natural" progression of thought. Also, remember that a sign is nothing more than a brief scouting or naming prompt, not a mini-tour guide or preview of the argument to come.
The main sentences of the "body" give the central argument. In research work, they must clearly and carefully explain their reasoning, describe results, develop implications, elucidate formulas or elaborate and explain theoretical and thematic points. Body sentences are the mainstream of the paragraph, the core unit of thought.
Researchers normally have to offer tokens to back up and support their basic arguments. “Token” sentences can be sprinkled throughout a paragraph among the body sentences, at the appropriate points where they are most needed or useful.
Typically, symbolic phrases are examples, references, quotes other authors, supporting facts or an analysis of the "points of attention", accompanying exhibits, tables, charts or diagrams. To some extent, "symbolic" sentences are inherently digressive: they potentially deviate from the mainstream of the paragraph.
Therefore, they need careful management, especially when two or more symbolic sentences follow each other, without intervening "body" sentences.
Finally, the phrase “wrap” serves to pull the argument together, to let readers know that a building block has been put in place. It should be constructive and substantial, add value to the argument, and not simply repeat raw material. It should also handle any link to the next paragraph that is needed.
Rational and skimmed readers do not treat all parts equally. In search of the fastest possible appreciation of what is said, they pay particular attention to the beginning and end of paragraphs, the subject and the accompanying sentences - a technique commonly taught in "reading fast ".
When and if they take a closer look inside the body of the paragraph, readers may also initially skip symbolic sentences. And they will normally delay digging into “hard” formulas or difficult expository materials in search of a more intuitive (if rough) understanding gleaned from the sentences preceding or following them.
It follows that the beginning and end of paragraphs should always be written with the utmost care. Try separating these two sentences and look at them together. Check how they read, how substantial and informative they are, and how they could be improved.
Structure the paragraph well
Six things most often go wrong when writing paragraphs:
1. The author starts with a link to the previous paragraph, instead of a new topic sentence. Readers can conclude that it is simply "more or less the same thing" and skip to the next paragraph. Even those who persist can become confused — what are we going to talk about?
2. The paragraph begins with a sentence of formalism or another form of sentence without substance (or perhaps several such sentences). For example, authors may begin by discussing a caveat, definition, difficulty, or method problem that is part of the provenance of the argument to be presented.
If they persevere in their reading, they may not correctly identify the now submerged topic phrase, then find that the wrapper phrase seems unwarranted or biased because it does not match the apparent topic.
3. The author begins with the name and reference of another author, for example: “Harding (2007: 593) argues…” supports the work of others. VSSome students tend to write in this way, stretching over several pages, each beginning with the name of a different author, especially in the "literature review" sections.
But when the first words are someone else's name, the author inadvertently signals, "Here's a completely derived paragraph" — "or a section if this pattern repeats itself." The common reaction of critical readers is therefore to downgrade or skip the paragraph (or the entire section) and move on.
The easy solution to this problem begins by not thinking in terms of individual authors, but instead focusing on the schools of thought, or "sides" in an empirical controversy, that the authors to be cited represent. Write a clear, self-contained topic sentence.
Then explain the basic ideas or propositions of one or more schools of thought involved in the body phrases. Relegate author names to supporting references that come at the end of sentences, where they belong.
4. A paragraph ends abruptly, usually because the author realized it was too long. This usually happens because the token phrases have multiplied—perhaps because the intended brief exposition of an example or the analysis of an exhibit has become unwieldy.
Usually the writers here do a forced "emergency stop" and then usually write what should have been the final sentence at the beginning of the next one. The first then has a sequence of Subject, Body, Tokens but no looping phrase. And the next paragraph 2 starts with the wrap1 sentence moved around, and has a topic2 sentence buried in it (like that course you're reading, but hey SEO doesn't like big wrappers!).
5. Paragraphs become too long, extending beyond the acceptable search text range of 100-200 words to take up 300 words or more. This often happens because tokens have multiplied or bloated beyond limits that can be easily manipulated.
But because of their partly digressive character, the author hesitates to recognize the need to create to separate the tokens to treat them by thematic. Especially when discussing points of attention or complex expositions and not designed to be self-contained and easy to understand, body and symbolic sentences can become confused, creating a text where the main argument becomes difficult to distinguish.
The solution must be brutal. Once a paragraph exceeds 250 words, it should be partitioned, usually as evenly as possible, and separate topic and wrapper sentences are provided for each part.
If the problem stems from exposing a token or exposure too long, then the author needs to find a solution that can smoothly handle a partial digression. If a paragraph is between 200 and 250 words, that can be kept, as long as the looping sentence can still reconnect readers to the (now rather distant) topic sentence.
6. A paragraph is too short. For a search text, this happens if it is less than 100 words, and especially if it consists of a single sentence or less than 50 words. Short paragraphs occur because an author doesn't know what to say, or hasn't properly thought through how a point or set of points fit together or can be sequenced within the overall argument. Some reflect mixtures of points that the author did not recognize as such.
Other single-sentence paragraphs are "orphan" sentences that need to be incorporated into nearby longer paragraphs that weren't — for example, in seed lists or sequences of connected paragraphs. Orphan sentences should always be merged with their neighbors, so that they disappear.
Two groups of people need to be particularly careful to adapt to this search-level paragraph convention. First, Latin languages often write using several very short or single-sentence paragraphs, organized in subtle thematic ways that English-speaking readers have trouble following (so don't do like this course 🙂)
Such an audience will often only see a bewildering multiplicity of orphans interpreted as disorganized thought. Second, journalists, and now some college bloggers as well, use short paragraphs that look fine on newsprint or in narrow or spaced-out blog columns (that's what I say, like this course!).
All of these types of authors must aggregate orphans into longer lengths of at least 100-200 words if they wish to publish journal articles or research books in English. If you've ever seen a journalist's writing transposed into book form without this change, you'll appreciate that there are strong aesthetic reasons for making this change as well.
EN 
FR 
RU 
ES