Why summarising feels harder than it should

Reading a dense journal article can give you a comfortable sense that you have understood it. Then you sit down to write two hundred words about it and discover that the comfort was an illusion. Summarising is unforgiving in that way: it forces you to commit to what a text actually says rather than letting the general impression wash over you.

It helps to be clear about what a summary is not. It is not a paraphrase, which usually stays at sentence level, reworking one point in fresh language. It is not a description, which reports what a text does in order — "the author then turns to the question of funding." A summary states what the author argues, in compressed form, in your own words. That distinction alone will improve your academic writing considerably.

Start by pinning down the main claim

Before you write anything, answer one question: what does this author want me to believe by the end? Not what topic do they discuss — what do they claim about it? Write your provisional answer as a single sentence beginning "The author argues that…" It will probably be clumsy. That is fine; you are testing whether a claim exists at all.

The claim is usually easiest to find in three places: the abstract, the final paragraph of the introduction, and the conclusion. Introductions often signal it with phrases such as "this article contends" or "I will show that". Conclusions restate it, sometimes more boldly than the body of the text has earned. If the claim you find there differs from the one the evidence actually supports, note that gap — it is exactly the sort of thing your own critical writing should discuss.

Watch out for topic sentences that describe rather than argue. "This section examines the role of local authorities" tells you what is coming, not what the author thinks about it.

Separate the reasons from the evidence

Once you have the claim, identify the reasons offered in its support. A reason is an assertion that makes the claim more believable; evidence is the material that makes the reason credible — data, quotations, case studies, historical examples.

• Claim: Urban rewilding schemes deliver poor value for money.
• Reason: Their benefits are widely dispersed and hard to measure.
• Evidence: Cost-benefit analyses of three council projects, plus interview data from officers.

Sketching this structure as a short bullet list, sometimes called a skeleton outline, does two things. It shows you whether the argument is genuinely a chain of reasoning or merely a series of related observations. And it gives you a scaffold for your summary, because you will usually keep the claim and the reasons and compress or drop most of the evidence.

Decide what to leave out

Most of a source text will not survive into your summary, and that is correct. The skill lies in cutting deliberately rather than arbitrarily. Material you can normally drop includes definitions used only once, illustrative examples that repeat a point already made, methodological fine detail, previews of literature you are not discussing, and lengthy hedges about what future research might show.

The test for anything you are unsure about is simple: if you removed it, would the main claim start to wobble? If not, cut it. Keep the strongest single piece of evidence for each reason rather than listing everything the author cites. A summary that is proportionate — around a tenth of the original length for a substantial article — signals that you have judged what matters.

Write it in your own words without drifting

Read the original, then look away from it. Write from your notes or your skeleton outline, not from the open page. This single habit prevents most accidental plagiarism.

Be alert to patchwriting, where you keep the original's sentence structure and simply swap in synonyms. "The report asserts that provision is inadequate" is not a summary of "The paper argues that services are insufficient"; it is the same sentence wearing a coat. Change the architecture instead. Make a different element the subject, reverse the clause order, change the verb from passive to active. Technical terms — "monetary policy", "cognitive load" — should stay exactly as they are; the vocabulary of the discipline is not yours to invent.

Finish by comparing your version against the source. Check every fact, figure and nuance. If you have made the author sound more certain, or more critical, than they were, revise.

Test whether your summary does its job

A good summary stands alone. Someone who has never read the original should be able to follow it, and someone who has should recognise it as fair. Read yours aloud and ask whether the author would accept it as a reasonable account of their position, even if you plan to criticise it later.

Two quick checks are worth building into your routine. First, can you point to where in the original each sentence of your summary comes from, and note the page number for your citation? Second, does your summary explain why the claim matters, not just what it is? A summary that captures an argument's stakes is far more useful to your reader than one that merely lists its parts.