The Cure for Copy-Paste Amnesia

‘Almost everything is imitation. The most original minds borrowed from one another.’ Voltaire, 1756.

Writing is hard. It can seem almost impossible when you read paper after paper of concise, carefully crafted words. It's easy to be tempted by the allure of cleverly written words. Once you read a concise sentence that says exactly what you want to say, it’s hard to un-see it. However, this temptation can sometimes lead to unintentional (or even intentional) plagiarism creeping into your work.

This article will share tips for avoiding plagiarism and review plagiarism detection tools for those without access to institutional software, including road-testing some available online tools. These tools may be useful for either the writer checking for unintentional plagiarism, or the mentor who wants to make sure there is no plagiarism before submitting.

What is plagiarism?

Plagiarism is considered a serious form of academic misconduct. The World Association of Medical Editors define plagiarism as, ‘the use of others' published and unpublished ideas or words (or other intellectual property) without attribution or permission, and presenting them as new and original rather than derived from an existing source’ (see their full Recommendations on Publication Ethics Policies for Medical Journals here). This definition applies to text or ideas in any electronic or printed format.

Self-plagiarism is where an author uses sections of text they have written in already-published material. This is also called text-recycling. Guidelines from the Committee on Publication Ethics (COPE) state that there are varying opinions on the acceptability of text recycling and that individual editors should consider how significant the overlap is, including the volume of text, the original source and whether it is referenced. This will determine whether the editor feels that the originality of the publication has become corrupted, and what action to take from there. Some forms of text recycling are considered generally acceptable within a field; such as proceedings or notes for conferences or courses. For further information on text recycling and guidance on when it may or may not be appropriate, follow this link.

Why would someone even consider it?

Unethical behaviour may be driven by three factors (coined the Fraud Triangle), including pressure (to get something done) or incentive (to reach a goal), opportunity (to create a shortcut) and rationalisation that it is okay to do so. These sociological effects on human behaviour probably influence 80% of the population (according to the concept of the 10:80:10 rule), who are at risk of behaving unethically if given the right conditions. This is how good people end up in sticky situations.

As an aside, according to this rule, 10% will never behave unethically and 10% will never hesitate to behave unethically!

So why might someone intentionally plagiarise in a scientific manuscript?

• Pressure to finish

  • degree deadline

  • manuscript deadline

  • credential application for specialist exams

• Lack of confidence or perfectionism

  • ‘I can’t word it better’

  • 'I'm stuck'

• Lack of training in scientific writing

  • feeling overwhelmed

How does someone unintentionally plagiarise?

·      Lack of understanding of what plagiarism is

·      Paraphasing instead of appropriately quoting

·      Forgetting to reference a source for a similar statement or idea

·      Making notes from papers that are ‘cut and paste’, then forgetting it is not their work

How to avoid plagiarism

Put some good habits in place as you read the literature;

• Keep original notes as you read

  • Summarise the findings of a paper in your own words

  • Never cut and paste a phrase or sentence into your notes – it is more likely to end up in your manuscript

• Write the first draft of a section (for example, introduction) entirely from scratch without looking at anyone else’s paper

  • It may not look great, but it will get better with each edit

• When you are writing a section that is more free-hand (Introduction or Discussion), do not look at anyone else’s Introduction or Discussion – only their Methods and Results section (when finding references to justify your statements)

• Do not read review articles until you are mostly happy with your own paper – then you may look at them to check that you didn’t miss any important original references

  • The caveat to this is referencing statements based on opinion or anecdotal evidence, which may require a review or textbook chapter reference

Once you are happy with your manuscript and close to the submission phase, you may like to do a plagiarism check using an online tool.

Why use plagiarism software?

The significance of plagiarism is often in the eye of the beholder. Views on this issue evolve over time. A journal editor back in 1881 did not seem to take too seriously the accusation from one author (in a Letter to the Editor) that a section of published text resembled his own work:

‘Plagiarism, as we take it, refers rather to ideas than to the mere mechanical arrangement referred to…for is not plagiarism the most subtle form of flattery? It is the tribute which mediocrity pays to genius.’ John J Mulheron, MD (Editor), Michigan Medical News, 1881.

Not too flattering for the author accused of plagiarism to be alluded to as mediocre. But in contemporary times, copying of text verbatim or even paraphrasing sentences can have serious consequences.

Many journals use plagiarism software with comprehensive databases that can flag an unacceptably high level of plagiarism. Reviewers, who are typically familiar with existing literature, may also detect plagiarism. If they recognise sentences similar to those in published papers, especially if they wrote those sentences, they might feel annoyed rather than flattered. This can negatively impact their perception of your manuscript. Given the subjective nature of the review process, it's crucial to uphold your integrity and avoid giving reviewers any reason to doubt it.

'I'll tell you one thing I am sure of. You're going to get caught. One way or another. It's a mathematical fact. It's like Vegas, the House always wins.' Carl Hanratty, 'Catch Me If You Can'

Alternatively, if you're mentoring or co-authoring with a junior researcher, you might want to ensure there's no plagiarism in the manuscript before it's finalised. This proactive approach can prevent embarrassment or reputational risk upon publication. It's important to discuss the benefits of plagiarism scanning with the first-drafter early on to avoid any misunderstandings. This discussion is also an opportunity to clarify what constitutes plagiarism and what is acceptable for the manuscript before submission. Submitting a manuscript containing plagiarism is not only unethical but a reputational risk.

What is available for individual use?

There are many providers designed for individual use for text similarity checks that are available online. It is not always clear, however, what sources they use to check for similarity. Many just scan webpages that are open access for the public. This is fine for students who may readily plagiarise from web pages, however, for a scientific manuscript, plagiarism is more likely to occur from published papers, which may or may not be behind a pay-wall.

TurnitinTM seems to dominate the market and has a large scholarly database for similarity checks. There are a few different products offered by TurnitinTM, such as iThenticateR and OuriginalR. These platforms are usually only accessible to publishing houses or universities, as they require large subscriptions. They are less accessible to the individual without institutional access.

There are many other plagiarism or similarity scanners that allow a pay-per-scan fee or are a part of a monthly subscription. Some are even free (but be aware of accuracy and document security!). The costs of scanning a paper can range from about USD$5 to USD$30. Most sites offer package deals for multiple documents or an ongoing subscription.

How do these platforms function for individual similarity checks?

I road-tested five ‘Plagiarism Checkers’ that were all under ~USD$25 per document, including two that are partnered with a major player in plagiarism checkers for universities. I chose them based on their description of databases used on their website and information gained from the literature. One of these was primarily a writing assistant (i.e. grammar checker) with an add-on plagiarism and AI-text scanner. Some of the scanners also offered bonus grammar or AI checks for an added fee.

I tested out several manuscript types that are already published in the veterinary medicine sphere (authored or co-authored by myself), including conference proceedings, textbook chapters and original papers. This means that I uploaded the text from documents that are already published to see if the software could detect the original document. Conference proceedings were not research abstracts but notes to accompany a presentation. I did not use all platforms for all documents due to the cost involved (it was a fun exercise, but I had to draw a line!).

Here are the results, with the range of percentage of similarity given by each individual platform.

I wish I could report a clear winner for accuracy but, disappointingly, I can’t. All platforms easily detected the two open-access pieces of writing. However, for content behind paywalls or in textbooks, none of the platforms could identify them. This suggests that either these 'single-serve' similarity scanners lack the power to identify material not readily available online, or the databases they use are not comprehensive enough for the veterinary literature.

Fortunately, I didn't find any evidence of unintentional plagiarism of earlier works in my own manuscripts. However, I did notice some sentences in papers published after mine that were similar, although this was sporadic and they weren't verbatim. There was some text recycling of sentences (see definition above) in conference proceedings. While it's common practice for speakers on the conference circuit to recycle text in conference notes or proceedings, I've personally made a conscious effort in recent years to avoid this. This can be challenging, especially when writing frequently on the same topic and liking your own sentences that you have used before! Broadly, most of the similarities detected at small percentage levels (i.e. <10%) by the platforms tested were common phrases, such as 'no significant findings'.

Funnily enough, there were some documents that flagged a decent percentage of ‘likely AI-generated’ text, when the papers were written before AI-generated text was even a thing! I can confidently say that I have never used AI to write or edit a paper. Are they suggesting that my writing is robotic? Or should I take it as a compliment? ;)

Overall, it seems that most of platforms that I tested were good at picking up on similar text in open-access documents but were unable to identify similar text in documents behind a paywall, at least for the small number of papers or book chapters that I tested.

In summary;

• Plagiarism constitutes academic misconduct, even if it is unintentional
  

• Plagiarism can interfere with your manuscript getting published and affect your reputation (as well as those of your co-authors)
  

• Consider using plagiarism checking software to screen for unintentional plagiarism, before submitting a manuscript

  • Be aware that using a plagiarism checker available online may not correctly identify all sources of plagiarism (based on the road test above)

  • If there is intentional plagiarism (and you are trying to get away with it), you can't be reassured by an online similarity checker not identifying it;

    • a journal editor may still identify it using a more-powerful plagiarism scanner

    • a reviewer may still identify plagiarism and recommend rejection

• It is highly important to avoid the temptation of plagiarism altogether by;

  • Knowing what it is

  • Having good writing habits from the start

  • Allowing adequate time for writing so you are not under pressure

  • Seeking help with writing when lacking confidence or ‘stuck’

'When you know better, you do better'. Maya Angelou (as quoted by Oprah Winfrey, 2014)

Disclaimer: in the spirit of declaring all sources, the images used in this article were generated by an Artificial Intelligence image creation program using key words and therefore any resemblance to published works is coincidental or unintentional