Tag: PEEL

  • Your Paragraphs Describe, They Do Not Argue: How to Rebuild Them (2026)

    Your Paragraphs Describe, They Do Not Argue: How to Rebuild Them (2026)

    Your chapter is the right length, every claim has a citation, and the feedback still says “descriptive”. The problem is almost never your research. It is that your paragraphs stop one sentence too early — at the evidence, instead of at what the evidence means. That missing sentence is where the marks are, and this is how to put it back.

    Start rebuilding your chapters in Tesify — free tier, your words, structure held in one place.

    What it costs to leave the paragraphs as they are

    This is not a stylistic preference. On the published undergraduate marking criteria at named UK universities, the word that appears in the 2:2 band again and again is descriptive, and the word in the band above it is critical. Nothing else separates them as consistently — we set out the actual band descriptors in how much of your dissertation mark is for writing, and the short version is that on some rubrics the prose requirement is literally identical for a 2:1 and a first, while the analysis requirement is not.

    Hull says where the marks sit, in a sentence that ought to be printed on every dissertation handbook. The interpretation step is “where you answer the question ‘So what?’”, and “These are the parts of your paragraphs that will get you the higher marks in any marking scheme.”

    So a chapter of well-referenced, well-spelled, entirely descriptive paragraphs is a chapter that has done the hard research work and then declined to collect on it.

    The four moves, and the one everybody skips

    Most UK universities teach the same shape under different initials. Oxford Brookes is relaxed about which you use: “There are a number of acronyms for remembering this, such as WEED, PEEL, SEE and TEEL. They all basically convey the same principle.”

    1. Point. The claim this paragraph exists to make. Newcastle: this is “usually the first (in some cases the second) sentence of the paragraph, commonly known as the topic sentence.” Sheffield adds what it should contain — a topic, “the general area under discussion”, plus a controlling idea, “the aspect or perspective on that topic that the paragraph will consider”.
    2. Evidence. The data, the citation, the quotation. This is the part you already do.
    3. Interpretation. Why the evidence supports the point, and what it means for your argument. Sunderland: “In order for your writing to be persuasive you will need to EXPLAIN to your reader how the evidence supports the point. This will involve analysing and evaluating the evidence.” This is the missing sentence.
    4. Link. The connection onward — optional, and contested (see below).

    Note that UK universities do not even agree what the letters stand for. Staffordshire teaches “PEEL stands for Point, Evidence, Explanation, Link. When writing a paragraph ask yourself, does it PEEL?” Newcastle teaches the same acronym as “Point, Evidence, Evaluation, Link”. That is not pedantry — explanation and evaluation are different jobs. Explaining says what the evidence shows; evaluating says how much weight it can bear. A strong paragraph does both, which is why the acronym you were taught is less important than knowing there are two moves hiding in that third letter.

    Newcastle is also honest about the model’s limits: PEEL “is a good starting point, but you may find it too rigid and restrictive for what you need to do, or too clumsy and repetitive in constantly linking back or forward.” Treat it as a diagnostic, not a template.

    A draft paragraph that ends on a quotation with no interpretation following it
    The commonest structural fault in an undergraduate chapter: the paragraph ends on the evidence.

    The published test for whether a paragraph describes or argues

    Hull publishes a comparison, adapted from Stella Cottrell, that turns this into something you can actually apply line by line:

    Descriptive writing does this Critical writing does this
    States what happened Identifies the significance of what happened
    Notes the method used Identifies whether a method was suitable or appropriate
    Gives evidence Argues a case according to the evidence
    Gives information Draws conclusions

    Leeds states the same rule as an instruction: “To write critically you will need to not only describe what happened, but also identify the significance of what happened.” And on method specifically: “Rather than simply noting the method used, which is the descriptive approach, a critical writer will show how appropriate that method was.”

    Run a paragraph against the left column. If every sentence lands there, you have found a descriptive paragraph, and you now know exactly which sentence is missing.

    Where paragraphs differ by chapter

    The four-move shape is not uniform across a dissertation, and applying it everywhere is its own mistake.

    Southampton draws the sharpest line: “The results chapter should provide a direct and factual account of the data collected without any interpretation or interrogation of the findings”, whereas “the discussion is rooted in analysis and interpretation, allowing you and your reader to delve beneath the surface.”

    So in a results chapter the interpretation move is deliberately withheld — it is not that your paragraphs should argue there, it is that they should not. The conventions for that chapter are different enough to need their own treatment, which we give in the results chapter guide. In a qualitative findings chapter the balance shifts again, and the ordering matters: our guide to thematic analysis covers how interpretation and quotation sit relative to one another there. And where you are writing up a technical decision, the paragraph has a fixed anatomy of its own — see the worked example in what to do when your statistical assumptions fail.

    The transferable rule is simply: know which job the current chapter gives your paragraphs, and do that job consistently.

    How long should a paragraph be? Nobody agrees

    Students lose hours to this, so it is worth stating plainly that the UK guidance does not converge.

    • The Royal Literary Fund gives a range of “150 – 250” — its published page omits the word “words” — or “between a third and two thirds of a double-spaced A4 page”, warning that a longer paragraph “gets harder and harder to follow” and a shorter one “looks scrappy”.
    • Hull gives “between 200 and 300 words long (they vary more than this but it is a useful guide)”.
    • Warwick counts sentences instead: “between 4 and 7 sentences long”, taking “about 2/5 of the page”, and splits at half a page.
    • Leeds, Staffordshire, Oxford Brookes and Newcastle all decline to give a number. Leeds: “There is no perfect length for a paragraph because this will vary between disciplines and types of assessments.” Newcastle notes that arts, humanities and social sciences “tend to use longer paragraphs than in the sciences”.

    Even the floor is contested — “too short” starts at one sentence for Warwick, one or two for Oxford Brookes, and fewer than three or four for Leeds. The useful conclusion is not a number. It is that a paragraph should be exactly as long as one point plus its evidence plus its interpretation, and if it is running past two thirds of a page you are probably making two points.

    A chapter checked by reading only the first sentence of every paragraph in sequence
    The five-minute audit: read only the first sentence of every paragraph, in order.

    The five-minute audit you can run tonight

    1. Read only the first sentence of every paragraph, in order. Newcastle sets the standard: “You should be able to get the overall gist and structure of an academic text by just reading the first line of each paragraph.” If what you hear is a list of facts rather than a developing argument, your topic sentences are doing the wrong job — Newcastle again: “If the first line of your paragraph is just a descriptive fact or quote from someone else, then they aren’t going to get a sense of your argument.”
    2. Write those first sentences out as a list. Hull recommends exactly this: “writing down a list of your topic sentences is an excellent way to check that your argument flows well from one point to the next.” Leeds suggests highlighting first and last sentences and asking “whether they represent a fair overview of the points that you want to make.”
    3. Find every paragraph that ends on a quotation or a number. Each one is missing its final sentence.
    4. Check your connectors mean what they say. Oxford Brookes: “A favourite linking word is ‘however’ which signals a contrast or contradiction, but if your next sentence doesn’t contain a contrast it will confuse your reader.” Where to source these honestly is covered in where to get your signposting phrases.
    5. Delete openings that look backwards. Leeds: “Generally, you should not start a paragraph by reiterating what you have just talked about. Your first sentence of a paragraph should alert the reader to a change of focus.”

    Doing this without rewriting the whole chapter twice

    The audit above finds the problem in an evening. Fixing it across 10,000 words is the part that defeats people, because every added interpretive sentence changes the paragraph’s length, its connection to the next one, and sometimes the order of the section.

    Rebuild your chapters in Tesify, concretely:

    1. Put your research questions beside the chapter, so every topic sentence can be checked against the question it is supposed to answer.
    2. Draft paragraph by paragraph in one workspace rather than across four documents, so you can see the sequence of points rather than only the paragraph in front of you.
    3. Add the interpretive sentence while the evidence is still on screen — that is the whole intervention, and it is far easier in place than in a later pass.
    4. Let the bibliography reformat itself while you spend your remaining time on the analysis rather than on punctuation in a reference list.

    The words stay 100% yours. That matters here more than anywhere: the interpretive sentence is precisely the sentence a marker is assessing you on, and it is the one you must write. Tesify holds the structure and absorbs the clerical layer so you have the hours to write it. There is a free tier — test the audit on one chapter before you pay for anything.

    Frequently asked questions

    What is the correct structure for a dissertation paragraph?

    One point, evidenced, interpreted, and connected onward. Most UK universities teach it as an acronym — Staffordshire uses PEEL for Point, Evidence, Explanation, Link — and Oxford Brookes notes that “There are a number of acronyms for remembering this, such as WEED, PEEL, SEE and TEEL. They all basically convey the same principle.”

    Which part of the paragraph earns the marks?

    The interpretation. Hull is explicit about it: the explanation step is “where you answer the question ‘So what?’”, and “These are the parts of your paragraphs that will get you the higher marks in any marking scheme.”

    Does the second E in PEEL mean Explanation or Evaluation?

    UK universities disagree. Staffordshire and Sunderland teach it as Explanation; Newcastle teaches the same acronym as Point, Evidence, Evaluation, Link. In practice you need both moves — say what the evidence shows, then judge how much weight it carries.

    How long should an academic paragraph be?

    There is no agreed figure. The Royal Literary Fund gives “150 – 250”, Hull gives “between 200 and 300 words”, and Warwick counts sentences instead: “between 4 and 7 sentences long”. Leeds, Staffordshire, Oxford Brookes and Newcastle all decline to give a number at all.

    Can a paragraph be one sentence?

    Not in academic writing, on the strictest published view. Warwick states it flatly: “One sentence is not a paragraph in academic writing. It is possible to find one sentence paragraphs in magazine articles and novels but good academic style does not have one sentence paragraphs.”

    How do I know if my paragraphs are only describing?

    Use the published test. Hull’s comparison sets “States what happened” against “Identifies the significance of what happened”, and “Gives evidence” against “Argues a case according to the evidence”. If every sentence in a paragraph falls in the left column, the paragraph describes.

    Is the topic sentence really always first?

    Usually, with one common exception. Newcastle puts it as “usually the first (in some cases the second) sentence of the paragraph”. The second-sentence version happens when the first sentence carries a short transition from the previous paragraph.

    What is the fastest way to check a whole chapter?

    Read only the first sentence of each paragraph in order. Newcastle sets the standard the test measures against: “You should be able to get the overall gist and structure of an academic text by just reading the first line of each paragraph.” Leeds suggests highlighting first and last sentences and checking they fairly summarise your points.

    Is a link to the next paragraph compulsory?

    UK guidance splits on this. Staffordshire treats the Link as a required element of PEEL, Hull calls links “optional but it will help your argument flow if you include them”, and Newcastle warns that constant linking can become “too clumsy and repetitive”. Use one when the connection is real.

    Why do my quotations weaken my paragraphs?

    Because a quotation is evidence, not argument, and a paragraph that ends on one has stopped before its most valuable sentence. Follow every quotation with your own reading of why it matters — that interpretive sentence is the one carrying the marks.

    Is it academic misconduct to use an AI tool while writing paragraphs?

    It depends entirely on what the tool does and what your university permits. A tool that rewrites your sentences for you is treated as misconduct at several UK universities. A drafting workspace where you write the words yourself is a different thing — but read your own regulations, because they vary, and disclose where your institution asks you to.

    Is Tesify free, and what happens to my unpublished work?

    There is a free tier you can test on a single chapter before paying anything. On confidentiality, the sensible standard for any tool is to check what its terms say about retention and training before you paste in unpublished data or participant material, and to anonymise transcripts first.