I could understand that paraphrasing is important to avoid plagiarism. But the increasing rate of publishing is making paraphrasing more complex. Let me explain:
"Drug X has a cardio-protective effect when administered in small dose."
This is a core SENTENCE in any research discussing this drug.. Keep it in mind!
In 1999, there was only 10 researchers working on this drug. So, when each one of them was going to publish his article he would easily PARAPHRASE this sentence. The odds of changing the meaning while doing paraphrasing are unlikely.
Now, There are 1000 researchers trying to do the same! So, changing the meaning is LIKELY to happen because you have to write in different ways and utilize a wide range of vocabs which will affect the meaning. And worth yet, if you're citing a secondary article _ You're paraphrasing the paraphrased!!
And why all of this?
Just to get a paraphrasing report of
Was just trying paraphrasing the example you gave "Drug X has a cardio-protective effect when administered in small dose"
1) When given in a small dose, Drug X has a protective effect on the heart
2) One of the effects of Drug X, at small doses, is cardio-protection
3) A cardioprotective effect can be obtained by administering Drug X at small doses
4) Administration of Drug X results in cardio-protection at small doses
After above I've run out of options! :) Good point to think about!
It reminds me of a famous party game young people loved to play ages ago (I never was on such a party, so it's kind of thought experiment):
The game works best when there are about 10 guests and most of them have already drunken at least 3 glasses of wine (or whatever your favorite party drink happens to be). The host asks the guests to take a seat and arrange themselves in a circle (or ellipse). Then he takes out of his pocket a piece of paper on which a putative event has been written. The text should be long enough and with enough details so that the chance of memorizing it literally is near zero. Then he whispers the full text in the ear of the guest left of him (the other guests should not hear him whispering). The guest does the same, i.e. whispering the story to the guest next to him, but of course without reading from the slip of paper. At the end of this round robin, the guest to the right of the host is requested to speak out loud what he heard from the one before him. Finally, the host reads out loud, what he had written on the slip of paper. Laughter!
Of course, this experiment is more about the shortcomings of understanding spoken language and short-term verbal memory, but there will probably also be some typical other phenomena: filling up gaps in what one remembers, and replacing fading-away memories by words or phrases that one believes might have been the original words/phrases, in other words: paraphrases!
This (thought) experiment should be a reminder to all of us, that it makes sense to now and then go back to the original papers or books in which certain theories or methods or drugs have been described and explained. Instead of always/fully getting that information out of the textbooks or mouth of our lecturers, who got it themselves from their professors, who again ... yes, you know what I mean.
Great question Khalid S.A. Badran many appreciate it.
Paul Hubert Vossen, thank you so much for your inspirational insights.
So, what is the optimal solution for eradicating distortion in scientific research?
Should researchers continue paraphrasing despite their awareness of possible confusion for future generations?
Moohebat Shoyukhi : Regarding your two questions:
There is a deeper-going answer to the second question of Moohebat Shoyukhi 's. it is the answer to the more basic question: How shall we refer to the works and ideas/insights of those whom we build upon in our own texts?
Yes, paraphrasing is one way. However, there are multiple possible ways, some better, some worse, while the rest may be equally good. Moreover, different fields of science and research may have developed quite different favorite ways of referring.
But then, we must also consider what we are referring to: letters (the old days), articles (nowadays mostly called papers), monographs (specialized treatises), textbooks (for "students"), or even handbooks and encyclopedia.
In several branches of science, it is even customary to refer to single coherent pieces of knowledge (e.g., theories or laws) by type + originator + specifier, e.g.: the Theorem of Pythagoras about the squared lengths of the sides of a triangle. Publications in mathematics and physics are full of it, and there is IMHO no valid reason why this could not be done in (parts of) the empirical social and behavioral sciences, if the tenets of empirical science (e.g., producing testable and tested hypotheses, confirmed or falsified) are more worth than the paper on which they are written.
However, at some point, authors may come to a tricky question to decide: when may a particular claim or state of affairs be considered to be part of the common understanding between the author and his/her (intended) readership, in other words, part of the Body of Knowledge (BoK) for that scientific (sub)community, so that it doesn't need anymore an explicit reference, let alone a paraphrasing. For instance, in many textbooks in mathematics or the physical sciences, it is explicitly written in the preface to the book, that the text is intended for .
A partial misquoting could be as risky as a misleading paraphrase.
In some cases, lazy researchers or even busy health workers involved in patient care may catch sight of a source text and fail to look after drug side effects discussed in other parts of the source.
Here, once again, we review the discussion.
As Khalid S.A. Badran explained, this SENTENCE was part of a paper recommending a cardio-protective drug in small doses.
“Drug X has a cardio-protective effect when administered in small doses.”
Consider the paper also discussed:
“Despite having a protective effect, drug X might have a side effect on diabetic patients and increase their cardiac rate.”
Imagine, without studying the whole paper and scrutinizing drug X’s side effects and its possible interactions with other drugs, a researcher only quotes the first SENTENCE and acknowledges it.
“Drug X has a cardio-protective effect when administered in small doses” (Citation, Year).
Consider reviewers mistakenly accept the misquoted source, and a journal editor publishes it and approves it could be part of the Body of Knowledge (BoK).
Or
Another researcher misquotes the source SENTENCE in a way that misspells the drug name and deceives reviewers and editors to gain a publication opportunity; however, a drug company has already produced the misspelled drug for hair loss treatment.
Then, what will happen to the diabetic patients or even healthy people taking the misspelled medicine for hair loss treatment?
Measuring the degree of plagiarism for a given text T relative to another text T° is a typical case of disjunctive measurement on a bounded scale.
This is important to know, because you can't use the same formulas for disjunctive measurement as you would use for the other two types of bounded measurement: conjunctive measurement or bipolar measurement. Thus, let me explain in a few words, what disjunctive measurement is.
Disjunctive measurement can be characterized as follows, where we will use the letter P (see note below) for the degree of plagiarism on a scale from 0 (plagiarism-free) to 1 (full-plagiarism) [any other bounded scale may be used, too]:
This is not the right place to elaborate on the precise nature and form of the OR-operator. Suffice to say here that there exist various proposals for it. The simplest is arguably what is called "Einstein addition", because Einstein used it the define the addition of velocities on a bounded scale.
The nice thing about having the OR-operator is the following. If you have identified all relevant texts T° with which T should be compared, and calculated the degree of plagiarism for all those T°, then you can simply calculate the disjunctive sum of all those degrees of plagiarism to get the total degree of P for T. If, for some reason, you believe that the texts T° should be weighted according to some degree of relevance for text T, then you might prefer to calculate the weighted disjunctive sum of degrees of plagiarism w.r.t. all those T°. This is quite similar to the weighted average of an array of real numbers, but it takes into account that all P's are measured on a bounded scale instead of an infinite scale (real numbers don't have a lower or upper limit).
Hopefully, you will agree with me, that there is much more to say about measuring the degree of plagiarism as the major popular plagiarism tools want to make us believe.
_______________________________________
Note: P is in fact an abbreviation of P(T | T°), in words: the degree of plagiarism of text T relative to text T°. To calculate the total degree of plagiarism of text T we must compare it with all other relevant texts T°. We call a text T° relevant for T if T° has been published (in the same language) before T and if it has been made plausible that the author of T must have known about the existence of T°. Proving that, may require a lot of research, as you can find out in the many books about the origin of Darwin's theory about the origin of species, e.g., the book " Nullius in verba" by Mike Sutton, 2017.
- Misquoting and Distorting sources, as discussed above, are not detectable using the existing plagiarism tools.
- Manual detection or prevention of plagiarism, though the best way, is time-consuming, especially for health workers.
- With many plagiarized works added constantly to the Body of Knowledge, it seems pure knowledge is out of access.
- Indeed, the plagiarism calculators with the bounded scales developed by Paul Hubert Vossen could be valuable instruments for teaching academic writing and paraphrasing. This modern perspective on plagiarism detection and prevention sheds light on ethical paraphrasing.
- However, until the new generation of plagiarism detectors terminates academic misconduct, PARAPHRASING will remain an abiding concern for scientific writers influenced by erroneous clinical decisions. They may even prefer not to paraphrase or publish in medical fields lest they unintentionally slaughter or sacrifice many innocent people.
I fully agree with Moohebat Shoyukhi that "plagiarism calculators ... could be valuable instruments for teaching academic writing and paraphrasing". Do they play any other role in science and society? I think so.
Measuring something is never a goal in itself, except perhaps if you happen to be an expert in measurement techniques. In an article about plagiarism and intellectual property (in German language, see below) I read that concerns about violating "intellectual property" (in a broad sense) can be traced back to the beginning of Western civilization. Thus, the occurrence and detection of plagiarism is first and foremost a social and legal issue with a long history.
A measure of plagiarism cannot and will not of itself decide about the intention of the "perpetrator". I would even prefer to have a more neutral term than "plagiarism detector", perhaps something like "similarity detector". Beyond which value (threshold) of similarity a text starts to raise suspicion about the intentions and motives of the author should be left to the courts and committees, because it takes more than a significant measure of similarity between two objects, e.g., texts, to prove that it is "intentional stealing of intellectual property".
Lately, plagiarism software has become a tool in the hands of politically motivated investigative journalists (at least here in Germany where "law and order" are a prominent aspect of culture). This is unfortunate, because the goal of such "investigations" is not to clean up the depository of scholarly and scientific output, but rather to denunciate individuals (mostly) outside the scientific community who have allegedly got a false status in society, esp. in political establishment, by submitting PhD's apparently full of plagiarism. This use of plagiarism software is not what I am talking about, but of course once such tools are made public, it will be used by some people for whatever reason they believe it can be used for, even if the validity of the tool is still "under research".
_________
Malo, M. (2016). Plagiat und Zitat. Eine Skizze zur Entstehung des Begriffs geistiges Eigentum und seiner Nutzung in der Wissenschaft. In W. Sühl-Strohmenger (Ed.), Handbuch Informationskompetenz (pp. 323-334). Berlin, Boston: De Gruyter Saur. https://doi.org/10.1515/9783110403367-031
Whether intentional or unintentional, plagiarism is a non-ethical offense and the cornerstone of numerous plagiarism violations in the future.
Plagiarism is wider than the 10 or 12 plagiarism types proposed by commercial plagiarism tools. A comprehensive plagiarism assessment rubric, published in 2023, presents 120 plagiarism types resulting from seven plagiarism dimension composites.1 In this study, applying a five-point Likert Scale, the severity rank for each plagiarism dimension is calibrated between 60 and 84%.
Paul Hubert Vossen proposed a quasi-arithmetic plagiarism scoring formula for n plagiarism dimensions2, and his plagiarism calculator works based on the severity of the plagiarism dimensions (between 60 and 84%) and the occurrence of plagiarism cases in paraphrasing tasks and computes the plagiarism scores for 120 plagiarism composites3.
Notes:
1. Shoyukhi, M., Vossen, P.H., Ahmadi, A.H. et al. Developing a comprehensive plagiarism assessment rubric. Educ Inf Technol 28, 5893–5919 (2023). https://doi.org/10.1007/s10639-022-11365-1
2. Vossen, P. H. (2021a). A model of plagiarism based on corpus severities and probabilities. https://figshare.com/s/8455fab321a41e6a4de2
3. Vossen, P. H. (2021b). Plagiarism composite severity calculation. https://figshare.com/s/e4c80d0884c0cb5b7186
Of course, I agree with Moohebat Shoyukhi that plagiarism, intended or not, is a violation of the rules of ethical (academic) conduct.
I just wanted to point out that there are special cases of scholarly publications in which plagiarism tools will "detect" drastic occurrences of "plagiarism" that are not at all intended as plagiarism and therefore don't violate the rules of ethical academic conduct. This occurs when the plagiarism tools don't distinguish between "relevant" and "irrelevant" publications T° with which the "suspicious" publication T will be compared on signs of similarity.
I guess, you may already have guessed what kind of publications I mean: all scholarly works that take other scholarly or non-scholarly works as their object of study. In such cases, it is natural, even obligatory, that large parts (if not all) of the original text will be copied verbatim in the new text in order to analyze it or point out certain properties or hypotheses about it. This occurs in certain fields of the study of literature (part of linguistics or semiotics) or in theology (e.g., when a religious text is annotated piece by piece).
The irony: the study of plagiarism in science or fictional literature may lead itself to a lot of publications that will show a high (significant) level of plagiarism, IF the plagiarism tool used will not distinguish between relevant and irrelevant comparisons!
I cannot agree more! I wholeheartedly agree with you, Paul, that "the plagiarism tools don't distinguish between "relevant" and "irrelevant" publications."
Provided scholars have to incorporate a bulk of copied texts into scholarly work for research analysis purposes, Google detects verbatim copies as accurately as plagiarism tools. Journal editors have been Googling manuscripts for years to check for manual plagiarism detection inaccurately reported by plagiarism tools. Despite the capabilities, none of the two (i.e., Google and Plagiarism tools) discovers a secondary source paraphrased with falsified meaning drastically changed to show the writers' proficiency in paraphrasing. Nowadays, with the AI revolution, researchers may even have AI summarize a two-paragraph secondary source and report it as a paraphrased primary source with a false citation to speed up publications.
Publishing in the medical field is riskier than in mathematics or social sciences as the subject of medical studies is human resources voluntarily participating in clinical trials or experimental research who may not have full knowledge of drugs' adverse reactions.
- Badges
- Science topic
- Similar topics
- Chemistry
- Pharmacology
- Pharmaceuticals
- Drugs