You've found the perfect source for your argument—but it's behind a paywall, only available as an abstract, or you're working from someone else's description of it. Can you still cite it? Sometimes. Here's how to navigate these tricky situations honestly.
Jan 3, 2026
By
Joe Pacal, MSc
Search citation styles
TL;DR
Only cite sources you've actually read. Try interlibrary loan, author emails, or preprint versions first. If you must cite something you only know secondhand, use a secondary citation format ("as cited in"). Never cite retracted papers as evidence—check Retraction Watch.
The Fundamental Principle
You should only cite sources you've actually read. This is a baseline expectation of academic integrity. Citing creates a chain of trust—you're vouching that the source says what you claim it says.
But "actually read" has some nuance, and genuine access problems are real. Let's work through the options.
Sources Behind Paywalls
Before you cite something you can't access, exhaust your options for getting access.
Try these first: Check if your library has access (many subscriptions aren't obvious from Google), use interlibrary loan (slower but usually free), check if the author posted a preprint or accepted manuscript, look for the paper on the author's personal or institutional website, email the author directly (researchers often share PDFs gladly), and check if your alumni library access still works.
If you genuinely cannot access the full text after trying these options, you have two choices: find a different source that makes a similar point, or use a secondary citation.
Secondary Citations (Citing Something You Read About)
When you're citing a source based on how another author described it—not because you read it yourself—you need a secondary citation.
This looks different across styles:
APA: In your text, name the original work and cite the source where you found it:
According to Smith (as cited in Jones, 2023), the effect was significant.
In your reference list, include only Jones (the source you actually read).
MLA: Similar approach—cite the secondary source with a note:
Smith argues that... (qtd. in Jones 42).
Chicago: Use "quoted in" or "cited in" in your note.
The key: You're being transparent that you didn't read Smith—you read Jones's description of Smith.
When Secondary Citations Are Acceptable
Secondary citations are fine for historical context or background claims that aren't central to your argument, sources that are genuinely inaccessible (out of print, untranslated, archive-only), and when the secondary source's interpretation is itself what you're discussing.
They're not ideal for key claims your argument depends on, recent sources you should be able to access, and situations where you could reasonably get the original.
Overusing secondary citations looks lazy at best and dishonest at worst. Readers notice.
Citing Abstracts Only
Can you cite a paper if you only read the abstract? Technically, no—abstracts are summaries written to attract readers, not comprehensive representations of the research.
Abstracts can omit important limitations, oversimplify methods, and misrepresent effect sizes or conclusions.
If you must work from an abstract, treat it as a secondary citation situation—you're citing the abstract's claims, not the full study. Better yet, find a way to access the full paper or find a different source.
What About Retracted Papers?
Citing retracted papers requires extreme caution. If a paper has been retracted, there was usually a serious problem—fraud, major errors, or ethical violations.
Generally, don't cite retracted papers as evidence for their claims. The retraction means those claims are no longer trustworthy.
Exceptions: If you're writing about the retraction itself, or if you're discussing the history of a scientific controversy, citing a retracted paper with clear notation is appropriate.
If you cite a retracted paper, include the retraction notice:
Author. (Year). Title. Journal, Volume, pages. Retracted. https://doi.org/xxxxx
Check Retraction Watch or the journal's website to see if papers you're citing have been retracted. This should be part of your pre-submission checklist.
Papers With Corrections or Errata
Unlike retractions, corrections don't invalidate a paper—they fix specific errors. If a paper has a published correction:
Check if the correction affects the parts you're citing. If so, cite the corrected information. Mention the correction if it's substantial and relevant to your discussion. Some styles prefer citing the correction notice; others just cite the corrected paper. Check your style guide.
The "I Can Only Find the Summary" Problem
Between AI summaries, abstract databases, and citation tools, it's easier than ever to encounter a source's claims without reading the source. This creates temptation.
Don't cite papers you haven't read just because you found a summary.
If you can't access a source, either find another source making a similar point that you can access, use a secondary citation properly, or leave it out.
The exception: systematic reviews and meta-analyses, where you may cite papers based on extracted data while acknowledging you're working from secondary analysis.
Practical Decision Tree
Can you access the full text through any legitimate means? → Do that, cite normally.
Have you exhausted all access options (library, ILL, author contact, preprint)? → If not, keep trying.
Is this source essential to your argument? → If yes, you need to find access somehow.
Can you find another source making the same point? → If yes, use that instead.
Must you cite this specific source? → Use secondary citation, clearly marked.
The exact formatting for secondary citations depends on your required citation style. Check the specific guide for your discipline below.
