The Literature Review That Argues Rather Than Lists

The commonest form is a sequence of paragraphs, each summarising one paper. Smith found A. Jones found B. Patel found C. Then a sentence saying the present study fills a gap.

This is a bibliography in prose. It demonstrates reading and it does not do the work the section exists for. For a concrete operational example of how a workplace tool approaches a related coordination problem, see here.

The role and length of the review vary substantially by field — a standalone chapter in some, a page in others, sometimes distributed through the paper. What follows is about the function rather than the format.

What the section is for

To establish that your question is open, and why.

That is the whole job. Everything in the section should serve it.

Which means it is an argument, with a conclusion — that this question has not been answered, or has been answered inadequately, or has been answered in a way that a new consideration undermines.

And the reader should be able to state the conclusion afterwards. If they can only say "there is a lot of work in this area", the section failed.

Organise by claim, not by paper

The structural fix, and it changes everything.

Instead of one paragraph per study, one paragraph per claim, with the studies attached as evidence.

Not:

Smith (2019) examined X in Germany and found A. Jones (2021) examined X in Brazil and found B. Patel (2023) found C in a laboratory setting.

But:

The effect appears robust in field settings across several countries (Smith 2019; Jones 2021), though laboratory work has not reproduced it at the same magnitude (Patel 2023). Whether the difference reflects the setting or the population remains open.

The second version has a point. It also happens to be shorter, which is the usual consequence.

Practical route: write down the three or four things the literature collectively establishes, and the two or three things it does not. Those are your paragraphs. Then attach citations.

The gap

A gap is not "nobody has studied this."

That claim is weak — either someone has and you missed it, or there is a reason nobody has. Reviewers test it, and it is the sentence most likely to be wrong.

Stronger forms of a gap:

Contradiction. Two literatures disagree, and the disagreement has not been resolved.

Limitation. The existing work could not address something, for a reason you can name — data, method, timing.

Extension. The finding is established in one setting and there is a specific reason to expect it to differ elsewhere.

Assumption. Everyone has assumed something that can now be tested.

Each of these is falsifiable and defensible. "Nobody has done it" is neither.

What to include

The work you are actually building on. Read it properly.

The work that would contradict you. Omitting it does not make it go away — a reviewer will supply it, and the omission looks worse than the objection.

The foundational reference, once. Not a chain of citations back to 1962.

Recent work. A review whose most recent citation is six years old suggests the literature was not checked before writing.

What to leave out

Everything adjacent. A paragraph with fourteen citations is padding, and it is recognised as such.

Papers you have not read. This is more visible than people think — a citation used for a claim the paper does not make is a specific and damaging error.

Summaries of methods that are standard in your field.

Definitions of established terms.

Citations added to satisfy an anticipated reviewer. Reviewers notice their own work being cited without engagement.

Being fair to work you disagree with

Describe it as its authors would recognise. A weakened version of an opposing argument is visible to anyone who knows the literature, and it undermines your own case.

Separate the finding from the interpretation. Frequently you accept what a study found and dispute what it means, and saying so is more precise and more generous.

Do not attribute motives, and do not characterise a literature as naive. It reads badly and, in an anonymous review, one of those authors may be reading it.

Practical process

Write the argument before the citations. A paragraph stating what the literature establishes, then attach the evidence. Writing citation-first produces the list form automatically.

Keep a note of what each paper is for, in your own words, one line, as you read. This is the raw material for the argument, and reconstructing it later from a reference manager does not work.

Check every citation against the paper before submitting. Errors propagate: a mischaracterisation copied from another paper's review is a common and traceable failure.

Read it as a stranger. Can you state the conclusion? Is any paragraph a list?

Cut hardest here. In most revisions, the literature section is where the surplus is, and shortening it usually improves the argument.

In the introduction versus its own section

In the introduction, it is compressed to the argument only — a paragraph establishing what is known and what is not. See the first paragraph.

As its own section, there is room to develop the claims and the disagreements, and the same organising principle applies.

As a standalone review article, the argument is the contribution, and the failure mode is the same at greater length.

The short version

The section exists to establish that the question is open. If the reader cannot state that conclusion, it failed.

Organise by claim with studies attached, not by study.

"Nobody has studied this" is the weakest available gap and the most likely to be wrong. Contradiction, limitation, extension and untested assumption are all defensible.

Include the work that would contradict you, and describe it as its authors would recognise.

For an independent authoritative reference, consult the EQUATOR Network for reporting guidance for research writing.