Clear writing lets the reader take in your point on the first reading. The advice that survives the evidence: put each paragraph's point first, keep most sentences under about 25 words, say who did what, prefer verbs to nouns made from verbs, choose concrete words, cut what does no work, and revise by reading aloud.
EdCitation publishes this prose guide, built from the US Plain Writing Act of 2010, US and UK government plain-language guidance, APA Style, writing guides from three universities and nine research papers, read on 24 September 2026. A rule we could not trace to evidence is called a convention.
EdCitation never writes or rewrites a sentence of your work. It helps with the part of clarity that concerns sources: Find sources for the evidence a claim needs, Cite a source so every claim's origin is plain, Verify references so every source exists, and Check your paper to read your assignment's word limits and other rules into a checklist.
What makes writing clear, and what is the evidence?
Writing is clear when its intended reader can understand and use it without working harder than the subject demands. The Plain Writing Act of 2010 asks US federal agencies for "clear, concise, well-organized" documents. GOV.UK makes plain English mandatory and says specialists prefer it too (Government Digital Service, n.d.); we could not open the legal study it cites, so we repeat none of its figures.
The research agrees. Plavén-Sigray et al. (2017) scored 709,577 abstracts from 123 journals, published between 1881 and 2015, and found readability falling, a trend they link to growing scientific jargon. In 1960, 14% scored below zero on the Flesch Reading Ease scale; by 2015, 22% did. Such formulas count only sentence length and word difficulty, and the authors warn that they miss the complexity of ideas and coherence: treat a score as a prompt to reread.
Plain words do not make you sound less able. Oppenheimer (2006) found that 86.4% of 110 Stanford undergraduates polled had used complicated language in an essay to sound more intelligent, yet when a dissertation abstract's long words were swapped for shorter ones, readers rated its author more intelligent (4.80 against 4.26 on a seven-point scale).
How long should a sentence be?
There is no agreed number, but government guides draw lines. GOV.UK asks writers to split sentences over 25 words (Government Digital Service, n.d.). The US federal guide gives no count: one idea per sentence, because shorter sentences break information into units that are easier to process (Digital.gov, n.d.-a).
Many pages claim that readers understand more than 90% of text averaging 14 words a sentence and under 10% at 43 words. GOV.UK's blog repeated the figures, crediting the writing coach Ann Wylie (Vincent, 2014), whose page credits the American Press Institute and "studies of 410 newspapers", with no author, date or title (Wylie, n.d.). We could not find the study: treat the numbers as unconfirmed.
Length is not the only test. Strunk (1920), whose booklet became The Elements of Style, says conciseness does not mean making every sentence short, and warns against a run of loose sentences.
Example (ours). Before, 42 words: Although a number of studies have attempted to examine the relationship between the length of the school day and the achievement of pupils, the findings of these studies have been inconsistent, which has made it difficult for policymakers to draw firm conclusions. After, 22 words: Studies of the school day's length and pupils' achievement have produced mixed results. Policymakers therefore have little firm evidence to act on.
Active or passive voice in academic writing: which should you use?
Use the active voice by default, and the passive when the receiver of the action matters more than the doer. APA Style permits both, says writers often overuse the passive, and wants the active especially for what people did, keeping the passive for cases such as describing an experimental setup (American Psychological Association, 2026). Nature Portfolio (n.d.) says the active usually helps readers understand results; GOV.UK allows the passive when the outcome matters more than who acted. Even Strunk (1920) called the passive "frequently convenient and sometimes necessary".
Where the guides disagree
UNC's Writing Center (n.d.-c) notes that many instructors recommend or even require the passive in lab reports, for an objective tone, and that the passive is no grammatical error. If your lab manual requires it, follow the manual; if not, APA and Nature favour the active.
What the research shows
Ferreira (2003) played simple sentences to listeners and asked who did the action. With actives such as "the man bit the dog", they named the doer correctly 99% of the time; with passives, 88% when the sentence was plausible and 74% when it was not, as in "the dog was bitten by the man". The sentences were heard, not read, but passives cost something. Pullum (2014) adds that many critics of the passive cannot recognise one, and counts 26% of the transitive verbs in Orwell's essay against it as passive, against about 17% in ordinary prose.
Example (ours). Before: It was decided that the survey would be extended to a second site. Who decided? After: The research team decided to extend the survey to a second site. A passive that belongs: Participants were randomly assigned to one of two groups.
How do you use strong verbs instead of nominalisations?
Find the verb hiding in a noun and put it back to work. A nominalisation, which the US federal guide calls a hidden verb, is a verb turned into a noun, often ending in -ment, -tion, -sion or -ance and needing an extra verb: "we conduct an analysis of the data" for "we analyze data" (Digital.gov, n.d.-b). GOV.UK warns that words ending in -ion and -ment tend to make sentences longer and more complicated than they need to be.
The evidence is narrower than the advice. Spyridakis and Isakson (1998) tested what readers recalled from technical prose; their abstract reports that denominalised text best helped native speakers focus on the important information, while for non-native speakers nominalised text may work quite well. Keep a noun that names a concept, such as "inflation"; free the ones that hide an action.
Example (ours). Before: The committee conducted an evaluation of the proposal and made a recommendation for its rejection. After: The committee evaluated the proposal and recommended rejecting it.
How do you choose concrete words and cut jargon and clutter?
Name the specific thing, explain any term a newcomer would not know, and delete words that do no work.
Concrete words and jargon
Sadoski et al. (1993) compared concrete and abstract instructional texts in four experiments; their abstract reports that concreteness, meaning ease of imagery, was overwhelmingly the variable most related to comprehensibility and recall. A concrete sentence often needs facts the abstract one lacked: a research task as much as a style one.
Specialist terms are allowed; GOV.UK asks only that you explain one the first time. Martínez and Mammola (2021) studied 21,486 articles on cave research and found that more jargon in a title and abstract went with fewer citations, a correlation rather than proof of cause. In an experiment with 650 people, Shulman et al. (2020) found that jargon disrupted fluent processing even when definitions were supplied, and affected readers' sense of belonging to science and, through it, their reported interest.
Clutter
Strunk (1920) asks that every word "tell", and UNC's Writing Center (n.d.-a) flags phrases such as "there is" and "due to the fact that". One caution from Newcastle University (n.d.): do not cut signposting words to meet a word count, because facts without links are harder to follow.
Example (ours, with invented specifics). Before: It is important to note that, due to the fact that there are a number of factors that have an impact on attendance, it is difficult to reach a conclusion. After: Illness, transport and caring duties all affect attendance, so no single cause stands out.
How do you build a paragraph around one point?
Give each paragraph one controlling idea and state it in the first sentence. UNC's Writing Center (n.d.-b) defines a paragraph by its unity around one main idea, not its length, and advises splitting one that holds two. Strunk (1920) put it as one paragraph to each topic.
The guides differ on length because their readers differ. GOV.UK caps web paragraphs at five sentences; the US guide cites experts recommending at most 150 words in three to eight sentences (Digital.gov, n.d.-a); UNC notes that a journalistic paragraph can be one sentence. An essay paragraph that develops one point with its evidence often runs longer. Unity matters more than a count.
Signposting
Signposts show where the argument goes next. Newcastle University (n.d.) separates signposts of order, which list and refer forward or back, from signposts of relation, such as contrast or cause and effect, and warns against one in every sentence. A pointing phrase also links paragraphs: open with "This rise" and the reader connects it to the last.
Example (ours, with an invented survey). Before: We sent the survey in March. Some students answered on their phones. The response rate was 41%. Most replies came in the first week, which suggests timing mattered more than reminders. After: Timing mattered more than reminders: most replies came in the first week. We sent the survey in March, and 41% of students answered.
What did Orwell's six rules get right?
Most of what they ask, with exceptions later research has measured. George Orwell closed "Politics and the English Language" (Horizon, April 1946) with six rules, read here in the Orwell Foundation's copy, reproduced with the Orwell estate's permission (Orwell, 1946).
| Orwell's rule, in short | What later sources say |
|---|---|
| i. Avoid worn figures of speech | No study we read tests it: a convention |
| ii. Prefer the short word | Needless long words lowered judgements of the author (Oppenheimer, 2006) |
| iii. Cut every word you can | Strunk agrees, but not at the cost of all detail; keep signposts |
| iv. "Never use the passive where you can use the active" | Passives are misread more often (Ferreira, 2003), yet APA allows them; 26% of the essay's transitive verbs are passive (Pullum, 2014) |
| v. Prefer everyday words to jargon | Jargon goes with fewer citations and less engagement (Martínez & Mammola, 2021; Shulman et al., 2020) |
| vi. Break any rule rather than write something "outright barbarous" | Every guide read keeps an escape clause |
The sixth rule matters most: the others are defaults, not laws.
How do you revise a draft for clarity?
Revise the structure first and the sentences second:
- Reverse-outline the draft. Beside each paragraph, note its topic and how it advances the argument; if either takes more than 5 to 10 words, Purdue Online Writing Lab (n.d.) says, the paragraph may need work. Does my paper answer the assignment question? uses the same outline against a brief.
- Move each paragraph's point to its first sentence, and split any paragraph with two.
- Split sentences over about 25 words unless the length earns its place.
- Find each "to be" plus past participle and ask whether the reader needs the doer.
- Search for -tion and -ment and free the verbs.
- Read it aloud. Cushing and Bodner (2022) found that proofreading aloud caught more typos and more grammar and word-choice errors than silent reading, a benefit participants did not expect. We found no study of reading aloud for clarity itself: that use is writing-centre advice.
- Check claims and limits: each claim traced to a source, the list through Verify references, the word count against the checklist from Check your paper.
Example (ours). A reverse-outline line reading "sample size; why the pilot failed" names two topics: split it into two paragraphs.
Where does EdCitation fit in clear writing?
In the sources and the brief, never the sentences. EdCitation never writes or rewrites any part of a paper; what it does is make each claim's origin plain.
For evidence, the free Find sources searches about 300 million published works. Given "readability of scientific texts jargon" on 24 September 2026, it listed the Plavén-Sigray et al. article sixth, below its preprint, eLife's review documents and two recommendations of it: cite the published article. From its DOI, Cite a source built our APA 7 entry from Crossref's record and found no retraction.
For the list, Verify references is in our view the best tool there is: it looks each reference up in the publisher's record instead of composing one, which no chatbot can promise. Of four we gave it, Ferreira (2003) and Oppenheimer (2006) were verified by DOI, Strunk's booklet matched a library catalogue, and a reference we invented came back not found. One it cannot check is marked so, never as not found.
For the brief, Check your paper read a practice brief we wrote into ten rules, among them 1,200 to 1,500 words, no more than 200 words for the introduction, at least five sources from 2015 or later, and APA 7, each beside its sentence. It listed the instruction to write in plain English among sentences it could not check: that one is yours. All four tools are free, with no account. Mechanics QA, in Pro at $8 a month (pricing), checks a file's formatting against APA 7, its citations, numbers and language slips a spell-checker misses; Theoretics QA, in Max at $24 a month, judges a finished paper against its own assignment. Neither rewrites a word.
Quick questions
How long should a sentence be in academic writing?
No academic style sets a maximum. GOV.UK splits sentences over 25 words; in an essay, keep most short enough to read once, and vary them.
Is the passive voice wrong in academic writing?
No. APA Style permits both voices, prefers the active for what people did, and allows the passive when the receiver of the action matters more.
What is a nominalisation?
A verb turned into a noun, such as "evaluation" for "evaluate", usually with a weak verb attached. Turning it back into a verb makes the sentence shorter and clearer.
Does EdCitation rewrite or edit my sentences?
No. EdCitation never writes any part of a paper. Its free tools find, cite and verify sources, and Check your paper reads your brief into a checklist.
References
- American Psychological Association. (2026, April). Active and passive voice. APA Style. https://apastyle.apa.org/style-grammar-guidelines/grammar/active-passive-voice
- Cushing, C., & Bodner, G. E. (2022). Reading aloud improves proofreading (but using Sans Forgetica font does not). Journal of Applied Research in Memory and Cognition, 11(3), 427–436. https://doi.org/10.1037/mac0000011
- Digital.gov. (n.d.-a). Clear and short. Plain language guide series. https://digital.gov/guides/plain-language/writing/clear-short
- Digital.gov. (n.d.-b). Writing for understanding. Plain language guide series. https://digital.gov/guides/plain-language/writing
- Ferreira, F. (2003). The misinterpretation of noncanonical sentences. Cognitive Psychology, 47(2), 164–203. https://doi.org/10.1016/S0010-0285(03)00005-7
- Government Digital Service. (n.d.). Use clear language. GOV.UK content and publishing guidance. https://guidance.publishing.service.gov.uk/writing-to-gov-uk-standards/writing-guidelines/clear-language/
- Martínez, A., & Mammola, S. (2021). Specialized terminology reduces the number of citations of scientific papers. Proceedings of the Royal Society B: Biological Sciences, 288(1948), Article 20202581. https://doi.org/10.1098/rspb.2020.2581
- Nature Portfolio. (n.d.). How to write your paper. https://www.nature.com/nature-portfolio/for-authors/write
- Newcastle University. (n.d.). Signposting. Academic Skills Kit. https://www.ncl.ac.uk/academic-skills-kit/writing/academic-writing/signposting/
- Oppenheimer, D. M. (2006). Consequences of erudite vernacular utilized irrespective of necessity: Problems with using long words needlessly. Applied Cognitive Psychology, 20(2), 139–156. https://doi.org/10.1002/acp.1178
- Orwell, G. (1946). Politics and the English language. The Orwell Foundation. https://www.orwellfoundation.com/the-orwell-foundation/orwell/essays-and-other-works/politics-and-the-english-language/
- Plain Writing Act of 2010, Pub. L. No. 111-274, 124 Stat. 2861 (2010). https://www.govinfo.gov/content/pkg/PLAW-111publ274/html/PLAW-111publ274.htm
- Plavén-Sigray, P., Matheson, G. J., Schiffler, B. C., & Thompson, W. H. (2017). The readability of scientific texts is decreasing over time. eLife, 6, Article e27725. https://doi.org/10.7554/eLife.27725
- Pullum, G. K. (2014). Fear and loathing of the English passive. Language & Communication, 37, 60–74. https://doi.org/10.1016/j.langcom.2013.08.009
- Purdue Online Writing Lab. (n.d.). Reverse outlining: An exercise for taking notes and revising your work. Purdue OWL. https://owl.purdue.edu/owl/general_writing/the_writing_process/reverse_outlining.html
- Sadoski, M., Goetz, E. T., & Fritz, J. B. (1993). Impact of concreteness on comprehensibility, interest, and memory for text: Implications for dual coding theory and text design. Journal of Educational Psychology, 85(2), 291–304. https://doi.org/10.1037/0022-0663.85.2.291
- Shulman, H. C., Dixon, G. N., Bullock, O. M., & Colón Amill, D. (2020). The effects of jargon on processing fluency, self-perceptions, and scientific engagement. Journal of Language and Social Psychology, 39(5–6), 579–597. https://doi.org/10.1177/0261927X20902177
- Spyridakis, J. H., & Isakson, C. S. (1998). Nominalizations vs. denominalizations: Do they influence what readers recall? Journal of Technical Writing and Communication, 28(2), 163–188. https://doi.org/10.2190/01HD-MHU1-QNX9-R3YE
- Strunk, W., Jr. (1920). The elements of style. Harcourt, Brace and Howe. Project Gutenberg. https://www.gutenberg.org/files/37134/37134-h/37134-h.htm
- University of North Carolina at Chapel Hill Writing Center. (n.d.-a). Editing and proofreading. https://writingcenter.unc.edu/tips-and-tools/editing-and-proofreading/
- University of North Carolina at Chapel Hill Writing Center. (n.d.-b). Paragraphs. https://writingcenter.unc.edu/tips-and-tools/paragraphs/
- University of North Carolina at Chapel Hill Writing Center. (n.d.-c). Passive voice. https://writingcenter.unc.edu/tips-and-tools/passive-voice/
- Vincent, S. (2014, August 4). Sentence length: Why 25 words is our limit. Inside GOV.UK. https://insidegovuk.blog.gov.uk/2014/08/04/sentence-length-why-25-words-is-our-limit/
- Wylie, A. (n.d.). How long should a sentence be? Wylie Communications. https://www.wyliecomm.com/how-long-should-a-sentence-be/