The scientific world is often rocked by cases of plagiarism. These are cases in which a researcher publishes something that turns out to contain material that he or she has taken from others without providing sufficient attribution.

Cases of plagiarism can have major and unfortunate consequences for those involved. Ultimately, as a researcher, you can be demoted or lose your job. It is therefore extremely important that researchers are aware of what plagiarism is, and where the line goes for what you can do with material created by others.

What is plagiarism?

The word plagiarism can be traced back to a poem by the Roman epigram poet, Martial (ca. 40-104 AD). Martial compared his epigrams to freed slaves, and the imitation of them to “plagium”: the crime of kidnapping people and selling them into slavery. You can read more about the concept of plagiarism (in Danish) in the information book published by the Committee for the Protection of Scientific and Scholarly Work (UBVA):  Ophavsret for begyndere – en bog til ikke-jurister, chapter 2.

The modern understanding of plagiarism is to take something that others have created and publish it as though you had created it yourself. It is for, example, plagiarism if a researcher publishes a book, and some of the pages later turn out to have been copied from another book without a clear reference to the source. Another example is to publish a scientific paper that appears to be based on your own research, but actually steals other people’s research results.

Plagiarism in research publications may relate to several different things.

Some plagiarism is committed when the author of a publication more or less word for word copies texts and phrases written by others.

In other forms of plagiarism, the author does not directly copy texts from other people’s publications, but rather appropriates their meaning content, conclusions, perceptions, research data, etc.

Finally, some forms of plagiarism relate to both text and meaning.

All these forms of plagiarism are addressed below.

Which rules and norms does plagiarism contravene?

Plagiarism may contravene a number of rules and norms.

Plagiarism may be contrary to responsible research practice

First and foremost, plagiarism violates scientific ethical norms for responsible research practice. This is, for example, stated in the Danish Code of Conduct for Research Integrity, under clause I:

“To ensure the reliability of the research, the researchers must be honest when reporting objectives, methods, data, analyses, results, conclusions, etc. This requires precise and balanced reporting by: […] acknowledging the work of other researchers.”

To this, clause 3.1.v. adds the following:

“When using one’s own work and the work of other researchers in a publication, researchers must provide sufficient and precise references to such other works.”

In serious cases, plagiarism may be deemed a case of research misconduct under the Act on Research Misconduct. Section 3(1) no. 4 of the Act defines plagiarism as follows:

Plagiarism: Purloining other people’s ideas, processes, results, text or special concepts of without providing proper attribution.

Section 3 (1), no. 1 of the Act on Research Misconduct defines misconduct in general terms as:

“Fabrication, falsification or plagiarism that has been committed intentionally or with gross negligence in the planning, execution or reporting of research.”

One might therefore logically suppose that all cases of plagiarism are thus by definition an expression of research misconduct, and that all such cases should therefore be dealt with by the Committee on Research Misconduct, but this is not the case in practice. In fact, in the area of plagiarism, there is a division of labour between the Committee on Research Misconduct and the research institutions, with serious cases of plagiarism being heard by the Committee, while the more trivial cases that cannot be said to involve research misconduct are dealt with internally at the institutions, including by the practice committees of the universities. This practice is now in accord with section 3 (2) no. 1 of the Act on Research Misconduct, which states:

“Research misconduct under subsection 1, no. 1, does not encompass instances of fabrication, falsification or plagiarism that have had little bearing on the planning, conduct or reporting of the research.”

In the executive order on the Danish Committees on Scientific Dishonesty (DCSD), it is similarly stated in section 15 (3) that the DCSD may set aside such cases. From practice, see, for example, the DCRM ruling of 14 March 2019: Here the Committee stated that plagiarism committed in the preface of books written by others does not in principle fall within the Committee’s remit unless scientific results are presented in the preface. See also the Committee’s rulings of 14 March 2019 (omitted reference to ministerial report) or 6 February 2020 (text coincidence in three sentences).

The mentioned division of labour between the Committee on Research Misconduct and the research institutions also follows from the fact that research misconduct presupposes gross negligence. Breaches of responsible research practice, do not, on the other hand, presuppose gross negligence.

In addition, it follows from section 4 of the Act on Research Misconduct, in conjunction with section 3 (1) no. 6, that the Committee on Research Misconduct can only deal with cases of plagiarism, etc., that are alleged to have been committed in scientific products. The research institutions, on the other hand, can also deal with plagiarism committed as part of teaching or popular communication.

A research institution that suspects plagiarism may have been committed in violation of the Act on Research Misconduct must refer the case to the Committee on Research Misconduct. See section 10 (3) of the Act on Research Misconduct. If the Committee on Research Misconduct does not believe that the plagiarism committed in the case is serious enough to be characterised as misconduct, the Committee will reject the case. The Committee may also refer the case to the relevant research institution, which will then take over. See section 17 of the At on Research Misconduct. A research institution may also on its own initiative take up a case that has been rejected by the Committee on Research Misconduct.

From practice, see, for example, the ruling of the Committee on Research Misconduct of 24 June 2018, in which the Committee stated:

“According to the allegation, which refers to the assessment committee’s statement on  the PhD dissertation concerned, the DCRM notes that the dissertation contains a coincidence of factual information from Article A and Article B, but that the degree of coincidence is of minor extent. In its assessment of the case, the DCRM emphasises that the assessment committee stated that the plagiarism was of a minor extent, but was not acceptable scientific practice. The DCRM also emphasises that the assessment committee admitted that plant taxonomy, which part of the dissertation addresses, contains an element of recycling from previously published works, but the Committee found that such permitted recycling could not be extended to apply to content and numerical data that ought to be based on the PhD student’s own observations. Finally, the DCRM emphasises that the dissertation does not contain a methodological section, and that neither the dissertation nor articles A and B refer to sample size or statistical information. The absence of a methodological section and the limited material and number of observations in the dissertation do not enable the DCRM to determine the extent to which there is an understandable coincidence of the numerical values in the dissertation and in articles A and B. Against this background, the Committee’s overall assessment is that the plagiarism has been of little significance in the planning, implementation and reporting of the research, and that it therefore does not constitute research misconduct, cf. section 3 (2) no. 1 of the Act. On the basis of the information presented in the case, the DCRM assesses that the case may contain instances of dubious research practice. The DCRM therefore refers the case to the research institution for further processing, cf. section 17 of the Act.”

Plagiarism may be contrary to the Copyright Act

Plagiarism may also contravene the Copyright Act. Briefly stated, this may be the case under the following conditions:

• Someone copies the text, composition, structure, etc., of a work, e.g. a book or an article, without citing the source.
• The material transferred must in itself reflect some originality on the part of the original author, i.e. creative aesthetic choices. The originality may lie either in the specific phrasing or in a creative way of compiling the material, e.g. the structure, composition, structure, etc.
• The material copied from a work is of reasonably concrete character; on the other hand, borrowing purely abstract elements of a work, such as its style, overall views, etc., is not an infringement of copyright.

You can read more about plagiarism and copyright (in Danish) in the information book published by the Committee for the Protection of Scientific and Scholarly Work, UBVA: Ophavsret for begyndere – en bog til ikke-jurister (Copyright for Beginners – a book for non-lawyers.) On the difference between copyright infringement and the scientific ethical charge of plagiarism, see chapter eight of the book and Morten Rosenmeier’s presentation at the UBVA plagiarism symposium in 2007.

The sanction for copyright infringement is typically the payment of a fee and possibly also reimbursement and compensation to the person whose copyright has been infringed. Copyright infringement may also be a criminal offence. Finally, the person whose copyright has been infringed may impose prohibitions, injunctions, etc., and oblige the plagiarist, or the plagiarist’s publisher, to withdraw the infringing publication. You can read more about the sanctions for infringement of copyright in chapter twelve of Ophavsret for begyndere – en bog til ikke-jurister (Copyright for Beginners – a book for non-lawyers).

Plagiarism can also lead to employment sanctions, etc.

Researchers are required by employment law to avoid conduct that is incompatible with their employment as researchers. Plagiarism committed by researchers can therefore also be deemed misconduct, which triggers disciplinary sanctions in the form of warnings, demotion, dismissal, etc. Plagiarism committed in a scientific dissertation that has led to an academic degree may also result in the degree being revoked.

In the remainder of this article, the focus is only on the extent to which plagiarism may conflict with the ethics of science, and unless otherwise stated, the focus is furthermore only on cases where the possible plagiarism has occurred in a research publication.

When is plagiarism deemed to have occurred in a scientific ethical sense?

The question of when plagiarism has been committed can be highly problematic. There are certainly some obvious cases, such as when an author publishes a book consisting largely of transcripts from another book. But there are also borderline cases, and these are the riskiest.  Many cases of plagiarism are simply due to the plagiarist’s failure to fully grasp where the line goes, and overstepped it. Below, we review some different cases.

Copying the original text of others

Many cases of plagiarism involve someone copying a text that others have written, without citing the source. This will often be deemed plagiarism. For example, it will usually be considered plagiarism if an article includes half a page from another article without citing the source, or if a book is published with sections that are in fact a kind of translation of material taken from a book in another language.

From practice relating to plagiarism in the form of text transfer, we can, for example, refer to:

• DCSD ruling of 2 December 2014 (PhD dissertation)
• DCSD ruling of 9 March 2015 (patent application)
• Committee on Research Misconduct ruling of 3 February 2020 (patent application)
• Committee on Research Misconduct ruling of 4 February 2020 (article)

Copying forms of text other than words, such as figures, graphs, equations, etc., can also be plagiarism – see, for example, the DCSD ruling of 24 August 2011.

However, there are grey areas. For example, doubt may arise as to whether plagiarism exists in cases where only very short passages have been copied, or possibly just individual sentences. This applies in particular if what has been copied is banal or meaningless, and does not express any linguistic independence. But it can be hard to say where the line goes.

You are, for example, currently reading an article about plagiarism from forskerportalen.dk. If someone posts the entire article on their own website and pretends to have written it themselves, then this would obviously be plagiarism. Posting an individual section, such as the introductory paragraph, without citing the source, would also be plagiarism.

“The scientific world is often rocked by cases of plagiarism. These are cases in which a researcher publishes something that turns out to contain material that he or she has taken from others without providing sufficient attribution.

Cases of plagiarism can have major and unfortunate consequences for those involved. Ultimately, as a researcher, you can be demoted or lose your job. It is therefore extremely important that researchers are aware of what plagiarism is, and where the line goes for what you can do with material created by others. “

But what if you just take part of the paragraph, and, for example, write:

“Cases of plagiarism can have major and unfortunate consequences for those involved. Ultimately, as a researcher, you can be demoted or lose your job. It is therefore extremely important that researchers are aware of what plagiarism is, and where the line goes for what you can do with material created by others.“

Or how about if you just take a single sentence, such as:

“Cases of plagiarism can have major and unfortunate consequences for those involved.“

It is difficult to set precise boundaries.

In its ruling of 2 December 2014, the DCSD stated, in relation to research misconduct:

“The Committee is of the opinion that there is no hard and fast line between plagiarism and non-plagiarism, but considers that the following remarks by S. Dutch, Professor of Natural and Applied Sciences at the University of Wisconsin, provide a constructive guideline: ‘Obviously copying 1,000 words verbatim without attribution is plagiarism… Is 100 words plagiarism? Almost certainly. 10? Perhaps.’ [Footnote omitted.]”

In a ruling of 3 February 2020, the Committee on Research Misconduct found that a PhD dissertation which included portions of text taken from works written by others was a case of plagiarism. In this connection, the Committee stated, amongst other things:

“In general, the Committee notes that, as a starting point, plagiarism must reasonably be suspected if a fragment of text of a certain length is copied from another text, typically one written by another author, without proper attribution, i.e. without a clear indication of the source. The reader must be in no doubt as to which pieces of text are the author’s own, and which are quoted or paraphrased. […] If the source is clearly indicated and emphasised in the reproduction (e.g. via italics or quotation marks), then this cannot be considered plagiarism.”

In the case in question, the respondent had copied a number of text fragments, each consisting of several hundred words. The Committee ruled that this was plagiarism. In this connection, the Committee stated that:

“The text concerned reproduces in a number of instances text taken from various sources without attribution […]. Taking the number of instances into particular account, the Committee finds that the respondent has thereby, through gross negligence, used the texts without providing proper attribution. In concrete terms, the fact that the respondent has at some points added individual words to the text that are not included in the cited source cannot lead to a different assessment. […] In assessing the significance of the plagiarism, the Committee places particular emphasis on the nature, scope and number of the copied texts. The Committee therefore assesses that the plagiarism has been of more than merely slight significance in the reporting of the research, cf. section 3(2), no. 1 of the Act [on Research Misconduct].”

The Practice Committee of the University of Copenhagen defined more precise guidelines in its ruling of 29 March 2016 on the PhD dissertation of former Minister of Education and Research Esben Lunde. In this case, the Practice Committee determined that it was not plagiarism that his PhD dissertation contained 19 instances of text coincidence with a number of existing works. In this connection, the Committee stated:

“On the basis of the rules governing research misconduct and good scientific practice, the Practice Committee finds that text taken from other people’s research works, and which is reproduced without citing a source, must in particular be assessed in relation to the following. First of all, the extent of the text: The longer the copied text, the more serious the plagiarism. Secondly, the originality of the content, assessed in relation to the field of research: The more original the content, the more serious the undisclosed reuse. Finally, the assessment should take account of whether the act was negligent or grossly negligent/intentional. […]

Consequently, the reproduction of very short passages of text that include general terms and commonly known content, and which are not essential to the research, is not considered contrary to the rules of good scientific practice. Reproduction of short passages of text which include the original research contributions of others, and where the reproduction may be attributed to negligence on the part of the person concerned, would be contrary to the rules of good scientific practice. Reproduction of longer text passages may also be in breach of the rules of good scientific practice, irrespective of whether the content of the text is generally known in the field of research.

If the reuse of a text is very extensive or relates to scientifically decisive content, and if this is deemed to be grossly negligent or intentional, then this will fall under the category of research misconduct.

Unacceptable scientific practice presupposes two conditions: First, an objective condition that a text passage of a certain length, or with sufficiently original content, must be reproduced, and second, a subjective condition concerning the degree of attentiveness. In relation to the above categorisation, breach of good scientific practice is distinguished from research misconduct through the degree of fulfilment of these two conditions.

In relation to the assessment of whether a breach of good scientific practice has taken place, as is relevant to the case in question, the objective condition regarding the extent of the text passage will depend partly on an assessment of the extent of the text and partly on the nature of the content. If the text passage describes general conditions, a greater extent of undisclosed reproduction may be acceptable than if the text passage is central to the research publication’s reasoning and thus characteristic of the publication’s research originality.

The qualitative assessment also includes a consideration of possible paraphrasing. Depending on the circumstances, extensive paraphrasing may in itself constitute a breach of good scientific practice, but if the author has, for example, performed further work on the issue and added new research value to the publication, this removes the paraphrase from the original text and makes it more acceptable.

The subjective condition for whether there has been a breach of good scientific practice relates to the degree of attentiveness. This condition will be met if the person concerned has failed to show the degree of attentiveness expected of a professional with a similar background in the field. This differs from research misconduct, in which the person must have acted intentionally or with gross negligence.”

In brief, the ruling determines that:

• The copying of short, banal sentences and fragments of text without attribution is not deemed to be plagiarism
• The copying without attribution of short sentences and fragments of text that reflect scientific originality and importance is deemed to be plagiarism
• Copying long fragments of text without attribution is deemed to be plagiarism. If the act is intentional or grossly negligent, it is deemed a case of actual misconduct

The ruling is not formally binding on universities other than the University of Copenhagen, nor is it formally binding on the Committee on Research Misconduct. However, the principles of the ruling appear reasonable and sensible, and it is therefore recommended that they are taken as a starting point, including beyond the University of Copenhagen.

In addition to the criteria set out in the Esben Lunde Larsen ruling, it must also be possible to take into account practice in the relevant research area. This is in accordance with clause 5 of the Danish Code of Conduct for Research Integrity, which states that:

“The Code embraces all research areas, recognising that standards for responsible scientific practice will be implemented in different ways within the various areas. This means that some recommendations may be more relevant to a particular research area and less applicable in others. The recommendations contained in the Code must always be understood against the background of established practice in the individual research areas.”

In some areas, it is, for example, not uncommon to find quite large text coincidences in the method descriptions of the methodological sections of publications, without this necessarily being considered a major problem. The cited part of the Code means that such cases are considered to be on the safe side of the rules on plagiarism. From practice, see the ruling of the Committee on Research Misconduct of 24 September 2019. Here, a majority of the Committee found that the plagiarism in the case was of little significance to the research reporting, and therefore fell outside the Committee’s remit, in part because the plagiarism occurred in the introductory and methodological sections.

Copying of non-original text, including quotations, references, etc.

In the cases mentioned in the previous section, a researcher had copied the original text of others into his own publication without sufficiently crediting the source.

However, sometimes the text a researcher has copied from someone else cannot really be said to have been written by that other person, but rather originates from a third source. It could be said that in such cases non-original text created by others has been copied. The question that arises in such cases is thus: Is it plagiarism if, in such instances, you merely attribute the original source without mentioning the intermediate work from which you copied the material?

The problem may be relevant in cases where a researcher borrows another researcher’s citations of other people, but refers directly to the cited source without mentioning the researcher from whose work he or she has taken the citation. Example: Politician A politician writes something in a private letter. Researcher B later finds the letter in an archive and quotes from it in a research publication. Another researcher, C, then reads B’s book, and on that basis repeats the quotation in an article. In the article, C refers directly to politician A, but does not mention that the text was copied from B’s book. In this situation, is C guilty of plagiarism in relation to B? He has referred to the original source, i.e. A. Should he have also referred to B?

Another related issue is whether copying the source references of others may be considered plagiarism. Example: A PhD dissertation refers to a scientific article that the author found in an old, exotic journal. A second PhD dissertation then refers to the same article. However, the author of PhD dissertation no. 2 only knew about the article because he read about it in PhD dissertation no. 1. Nonetheless, the author of PhD dissertation no. 2 refers directly to the article without mentioning that he only became aware of the article through PhD dissertation no. 1. Is this plagiarism in relation to the author of PhD dissertation no. 1?

The easy, uncomplicated answer is of course that we should refrain from using the quotations and source references of others, and that in examples such as those described, we should refer both to the original source and to the source that acquainted us with them. However, that answer is unrealistic, as there are certainly cases in which non-original text written by others is copied without source attribution.

Example: A researcher provides a scientific publication with the words Neil Armstrong famously said when he set foot on the moon: “This is one small step for man, one giant leap for mankind”. The researcher is unsure of the exact wording and so looks up the quotation in Wikipedia. In the final article, the researcher does not mention that he took the quotation from Wikipedia, but simply refers directly to Neil Armstrong. Here it is obvious that the researcher has not committed plagiarism in scientific ethical terms in relation to Wikipedia. We should of course consider whether it is a good idea to base research on Wikipedia articles, and whether if we do so we should refer to Wikipedia, so that readers can have the opportunity to check who is behind the Wikipedia article, cf. the DCSD ruling of 2 December 2014, p. 23 f. But this is a question of research quality, and has nothing to do with plagiarism as such; it is obvious that you are not a plagiarist if you simply look up a famous quote in Wikipedia without mentioning it.

Example 2: As the reader of this article, you have undoubtedly been seized with the urge to share with others the epoch-making insights you have gained from reading it. In this connection, you copy the reference to Ophavsret for begyndere (Copyright for beginners) above and write “You can read more about the concept of plagiarism in the information book published by the Committee for the Protection of Scientific and Scholarly Work, UBVA: Copyright for beginners – a book for non-lawyers, chapter 2”. Have you thereby plagiarised the article? Of course not.

No practice unambiguously determines how the line should be drawn.

The decisive criterion, as the examples above suggest, must be how well-known the copied non-original material is, including whether it is generally known or relatively unknown. It would therefore be contrary to responsible research practice, for example, to copy a work’s quotations from unknown sources, which the author has hunted down through archival studies, or to copy the references of others to little-known works. On the other hand, it cannot be plagiarism to copy well-known quotations such as the Armstrong quotation above, or to copy the references of others to well-known standard works.

This is in line with the practice of the Committee on Research Misconduct and the DCSD, which has stated in a number of rulings that common knowledge may be cited without a source being specified. See, for example, the ruling of the Committee on Research Misconduct of 3 February 2020 and the DCSD ruling of 12 December 2014.

Committee on Research Misconduct ruling of 3 February 2020:

“Common knowledge may be described without source attribution, without this being deemed plagiarism. Common knowledge may be defined as knowledge that may be assumed known by everyone within a particular group or regional, institutional, or professional community. This may include facts about, for example, geography, history, physics, language or literature. Word-for-word or close approximations of another author’s text on common knowledge must, however, be credited to that author.”

It can be difficult to know how to proceed in cases where the copied text is not exactly generally known, but on the other hand not entirely unknown, either. In such cases of doubt, the extent to which non-original material has been copied may perhaps play a role. For example, there may be a difference between whether an author who has copied the sources of others has done so many times or just a few times in the course of a publication.

If there are just a few instances, it may at least mean that the case does not fall under the remit of the Committee on Research Misconduct. See the Committee’s aforementioned ruling of 3 February 2020. Here the Committee found that a possible plagiarism was only of minor significance in the reporting of the research, taking into account the “nature, scope and number” of the copied text fragments.

Paraphrasing

As described above, copying text from others without attribution may be deemed plagiarism.

The question is whether this also applies if you “paraphrase”, i.e. rewrite the text, and reproduce its meaning in your own words.

If you rewrite the text thoroughly enough, you arrive at a point at which it can no longer be said that any text has been copied. However, the text must actually be genuinely rewritten;  it is not sufficient merely to replace certain words or swap some of the sentences around slightly.

As an illustration, see, for example, the DCSD ruling of 2 December 2014, p.15:

“Correct paraphrasing means that an author processes the thoughts and ideas of other authors and reformulates them using his or her own words and sentence structure […] as well as with source references to the works in which these thoughts and ideas are described. […]If the sentences have only been slightly restructured, , with certain words being replaced by synonyms, this will be deemed plagiarism if no source is stated.”

To this, one important thing must be added: Even in cases where you rewrite the text in such a way that it no longer contains any actually copied text, you may sometimes still have to refer to the source, because you have appropriated original opinions, conclusions, research data, etc.

Example: Scientist A makes an important discovery and describes it in an article. Another scientist, B, then publishes another article which recycles text from A’s article and gives the reader the impression that B is behind both the text and the important discovery to which it refers. This is of course plagiarism, in relation to both the text and its content.

Example 2: As above, but with the difference that B changes the text so that it is no longer identical to A’s, but the content of the article remains the same. This is plagiarism in relation to the text content, but not in relation to the text.

The question is: When does non-textual copying constitute plagiarism? This is addressed below.

Copying of material other than text, such as research data, understandings, information, etc.

So far, this article has dealt with whether it may be deemed plagiarism if you copy text from others without attributing the source.

However, cases of plagiarism can also occur in situations where someone copies material other than text without stating a source. Examples might include the research data and factual information of others, or their conclusions, opinions, information, etc.

Example: A scientist writes an article based on research data that has been collected by a colleague and described in an article written by that colleague. The scientist fails to mention that he has taken the relevant research data from his colleague’s article. He does not copy his colleague’s actual text, but bases his new article on the colleague’s research data without attribution. Is this plagiarism?

As stated above, the Code requires scientists to be honest in reporting “objectives, methods, data, analyses, results, conclusions, etc.”, and the Act on Research Misconduct defines plagiarism as “purloining the ideas, processes, results, text or special concepts of others”. On the face of it, one might therefore think that copying anything from others without citing the source, including their conclusions, opinions, research data, etc., is always by definition plagiarism.

But reality is more complex.

Legal research is, for example, based on studies of laws, judgments, administrative decisions, EU directives, etc. These constitute the legal research data on which the research is based. It is quite common for lawyers writing about the same topics to base their research on the same legal sources, and they do not usually state whether they are basing their research on laws and judgments, etc., that others may previously have written about. Example: A legal researcher reads an article referring to an interesting verdict handed down by the Eastern High Court on 4 January 2017. The researcher then writes an article about the verdict. The person in question feels no obligation under research ethics to mention the article in which he or she first saw the verdict; that is not how lawyers work. However, strictly speaking it is true that the researcher in question has appropriated research data from another person without attribution.

In “wet” areas of science, including medical science, the tradition is different. Here it would normally be viewed as extremely problematic if a medical scientist were to appropriate measurement results and the like from others without mentioning where they came from.

Other areas may also follow different practices. The plagiarism definitions contained in the Danish Code of Conduct for Research Integrity and the Act on Research Misconduct make it clear that appropriating other people’s “conclusions” is deemed plagiarism. However, it goes without saying that this depends on what type of conclusion is involved, including whether it is a special conclusion that can be traced back to a particular researcher, or just a conclusion reflecting familiar banalities and general knowledge. Moreover, the research ethics tradition in the field in question must also play a role. It would, for example, surely be considered deeply problematic if a scientist in the medical field were to appropriate the innovative conclusions of others about how a particular virus behaves under given circumstances. On the other hand, it would not be problematic in the same way if a legal researcher, without source attribution, were to conclude that case processing times are long in Danish courts, or if a social scientist were to state without source attribution that relatively few women are represented in Danish management boards. These are things that we all know.

The decisive aspect in cases like those mentioned here should be:

• How original and unusual is the material dealt with? Is it material of a special nature, and can it be traced back to specific researchers? Or is it rather material that merely expresses trivial knowledge?
• Is the copied material well-known, or relatively unknown?
• What does tradition dictate in the relevant field of research?

A number of rulings by the DCSD and the Committee on Research Misconduct have stated that common knowledge may be cited without attribution. See, for example, the Committee on Research Misconduct ruling of 3 February 2020, in which the Committee stated, inter alia:

“Common knowledge may be described without source attribution, without this being deemed plagiarism. Common knowledge may be defined as knowledge that may be assumed known by everyone within a particular group or regional, institutional, or professional community. This may include facts about, for example, geography, history, physics, language or literature.“

It also follows from the practice of the Committee on Research Misconduct that it is not necessarily plagiarism if two publications address the same general, previously known subject. See, for example, the Committee’s ruling of 4 February 2020:

“It follows from the practice of the Committee that the definitions of research misconduct and plagiarism, according to both the old rules for the Danish Committees on Scientific Dishonesty and the new rules for the Danish Committee on Research Misconduct, contain a condition that there must have been an actual unjustified acquisition of concrete results, etc., which can be attributed to a previous author, without this being duly credited in the later publication. It must thus be possible to demonstrate in concrete terms that a unique research idea or text has been copied and used without attribution, as opposed to, for example, situations in which scientists publish publications on the same subject in accordance with the usual methods in the field, cf. most recently the Committee’s ruling of 7 June 2018 in case no. 2017-01.

An allegation of plagiarised text, for example, should therefore indicate the precise location of the plagiarised text in the scientific product, as well as the precise scientific product from which the text was taken and where in that product it was taken from. If the allegation concerns several instances of plagiarism, it should contain precise information on all instances.

[…]

 To the extent that the products of the complainant and the respondent deal with the same subject, the Committee also finds that this is such a general idea that there can be no immediate question of plagiarism. In this ruling, the Committee has emphasised that it appears from both the complainant’s and the respondent’s product that the coincident topic has been dealt with several times in the literature prior to the publication of the complainant’s and the respondent’s product.”

The fact that assessments of a scientific ethical nature are generally influenced by traditions in the relevant research areas is, as previously mentioned, made explicit on p. 5 of the Code, which states:

“The Code embraces all research areas, recognising that standards for responsible scientific practice will be implemented in different ways within the various areas. This means that some recommendations may be more relevant to a particular research area, and less applicable in others. The recommendations contained in the Code must always be understood against the background of established practice in the individual research areas.”

What if you yourself have contributed to the source that is claimed to have been plagiarised?

Plagiarism means that you publish material that others have created, in such a way that it appears that you are responsible for it. It is therefore plagiarism if you, for example, copy a text written by another author in your own article.

If you reuse material that you have created yourself without attributing this to yourself, this is not plagiarism in the usual sense, but rather “self-plagiarism”. Whether self-plagiarism is considered remiss depends not on the principles described here in the article, but operates according to its own set of rules. You can read more about this in the article on double publication, overlapping publications and self-plagiarism here on the research portal.

On the other hand, it may be plagiarism in the traditional sense if you publish material that originates from publications to which both you and others have contributed, without providing sufficient attribution. Example: A, B and C write an article together. A then writes another article that reuses text from the first article, but fails to reference this. In addition to the fact that this may in some cases involve self-plagiarism, it may also involve traditional plagiarism in relation to B and C. This certainly applies in cases where A recycles text written by one of the others.

In several rulings, the Committee on Research Misconduct and the DCSD have indicated that these circumstances differ. See, for example, the DCSD ruling of 24 August 2011: Here, the DCSD determined that an author who, without attribution, had reused figures from an article he had written together with others, had not committed plagiarism, but at worst self-plagiarism. In its reasoning, the DCSD stated that: “the figures 5.18-5.20 in your thesis are identical with the figures in the article. Because you were a co-author of the article you may use these figures without a reference […]”. See correspondingly the ruling of the Committee on Research Misconduct of 12 November 2008, p. 4. Here the Committee stated:

“The Committee finds that, in principle, there can be no question of the acquisition of other people’s ideas, etc., without proper attribution if all the authors of the publication alleged to have been plagiarised are co-authors of the publication alleged to contain the plagiarism. Moreover, the Committee essentially finds that in general, an author’s reuse of text from an article with several authors, of which the author in question is one, is not plagiarism in the sense of the law.”

However, it goes without saying that this statement is too sweeping. If, for example, A, B and C publish a book together and each of them writes a chapter, it is clear that C cannot reuse text from A’s chapter in a new book without referring to A. See, in a similar context, ruling 27/2017 of the Practice Committee of the University of Copenhagen: A researcher who had used a figure created by three other researchers apologised, stating that he had been given permission by one of them, and that parts of the figure had been included in an article of which he was a co-author. The Practice Committee acquitted the person in question of plagiarism, stating that he had reasonable cause to believe that he had been given permission to use the figure. On the other hand, the reason was not that it had previously been included in the article of which the respondent was a co-author. See also the ruling of the Committee on Research Misconduct of 14 March 2019 (case no. 2018-07).

Here, several researchers were accused of plagiarising some ministerial publications of which they themselves were co-authors. The Committee acquitted them of plagiarism on the grounds that the possible lack of attribution was of little significance to the research reporting, and therefore fell outside the Committee’s remit.

How should attribution take place?

As described above, there are a number of cases in which the use of material created by others should be accompanied by source attribution. The question is what form this should take.

As previously mentioned, clause I and clause 3.1.v of the Danish Code of Conduct for Research Integrity state that researchers’ use of others’ work must be accompanied by “precise and balanced reporting”, and that “adequate and accurate references” are required.

What this precisely consists of is in principle influenced by the traditions of the various research areas, cf. the quotation from p. 5 of the Code, above. It is therefore possible that in some scientific fields, more restrictive attribution requirements may apply than in others.

In general, however, as a rule of thumb, source attributions in research publications must be clear and distinct enough to enable the reader to understand exactly what has been borrowed and where it has been borrowed from.

It is not, on the other hand, sufficient if:

• You write in the preface to a book that the book “builds upon” certain works created by others
• The works from which you have borrowed appear in your bibliography
• After copying ten lines of text from one source, you insert a source reference after the first line and leave it up to the reader to work out that this also refers to the other nine
• You rephrase the text slightly.

The requirements regarding the precision of the reference may depend on whether the text has been copied word-for-word or has been reproduced in some other form. If you copy text from others in a manner that requires a source reference, and if it is a word-for-word copy, it is most correct to mark this as an actual quotation, i.e. using italics or quotation marks. If, on the other hand, it is a paraphrase, i.e. you reproduce the meaning content of a text in your own words, it is sufficient to indicate clearly to the reader in some way that this has been borrowed, and what you are borrowing. See, for example, the Committee on Research Misconduct ruling of 3 February 2020:

“In general, the Committee notes that, as a starting point, plagiarism must reasonably be suspected if a piece of text of a certain length is copied from another text, typically one written by another author, without proper attribution, i.e. without a clear indication of the source. The reader must be in no doubt as to which pieces of text are the author’s own, and which are quoted or paraphrased. If the source is clearly indicated and emphasised in the reproduction (e.g. via italics or quotation marks), then this cannot be considered plagiarism.

Correct paraphrasing means that an author processes the thoughts and ideas of another author and formulates them using his or her own words and sentence structure, as well as attributes the source to the works in which these thoughts and ideas are described. If the words and sentences have only been slightly reordered or rephrased with  synonyms, then this is deemed to be plagiarism unless proper attribution is provided (source cited).

Common knowledge described without source attribution is not deemed plagiarism. […] Word-for-word or close approximations of another author’s text on common knowledge must, however, be credited to that author.”

Are there special requirements in connection with communication, etc.?

Thus far, this article has dealt with how sources should be referenced in research publications.

However, the question is whether what has been stated above also applies to texts that do not have the character of research, but rather of popular communication.

The question is relevant because what largely makes such communication popular is that the level of formal references is reduced to make the material more easily digestible. If, for example, a researcher writes a column or a newspaper article, he or she cannot expect it to be accepted for publication if it is full of scientific references. In the same way, books of popular science and encyclopaedias usually have fewer references than scientific literature. The question is therefore whether more lenient source attribution requirements apply to popular communication than to research? Can a researcher accused of having plagiarised someone else’s work in a popular article legally be excused on the grounds that this was a case of popular communication, and that he or she therefore did not have to cite the sources?

Cases of plagiarism committed as part of popular communication cannot be handled by the Committee on Research Misconduct, which, as previously mentioned, only deals with dishonesty committed in scientific products. See, for example, the Committee’s ruling of 13 September 2018, in which the Committee refused to handle a possible case of plagiarism in a self-published book that did not have the character of a scientific product. The above question may, on the other hand, be relevant for the practice committees of universities.

The Danish Code of Conduct for Research Integrity contains no precise criteria on this question, and it is therefore difficult to give a clear answer on this basis. Some in-depth discussions on the issue may be found in, for example, the Norwegian report God skikk – Om bruk av litteratur og kilder i allmenne, historiske framstillinger. Utredning fra et utvalg oppnevnt av Den norske Forleggerforening (DnF), Den norske historiske forening (HIFO) og Norsk faglitterær forfatter- og oversetterforening (NFF), februar 2006. (Good Practice – On the use of literature and sources in general, historical representations. Report from a committee appointed by the Norwegian Publishers’ Association (DnF), the Norwegian Historical Association (HIFO) and the Norwegian non-fiction writers’ and translators’ association (NFF), February 2006.)

In the assessment of the Committee for the Protection of Scientific and Scholarly Work (UBVA), the reference requirements for popular communication are probably not particularly lenient in connection with the copying of text. This applies, as far as we can see, not only if text is copied in a research publication, but also if it appears in, for example, a newspaper article or on a popular website.

On the other hand, more lenient reference requirements probably apply to popular communication when the copying deals with material other than text, such as research data, opinions, etc. If, for example, a doctor reproduces in a newspaper article the conclusions of a study published in The Lancet, it will surely be sufficient for the doctor to refer to the study as “a recent major study”, or “a recent study in the acclaimed journal The Lancet”, without having to refer to precise page numbers or the like.

See also the aforementioned Norwegian report.

Who is responsible for plagiarism? Plagiarism through negligence

If you overstep the limits described above, you may have committed plagiarism that is in breach of responsible scientific practice and in some cases the Act on Research Misconduct.

However, it is a condition in this connection that the plagiarism can be said to have taken place intentionally, i.e. on purpose, or through negligence, i.e. carelessness.

To be considered actual misconduct, the degree of negligence must be gross, i.e. reflect a very high degree of carelessness. Section 3 (1) no. 1 of the Act on Research Misconduct defines research misconduct as “fabrication, falsification or plagiarism which has been committed intentionally or with gross negligence in the planning, execution or reporting of research.”

With regard to plagiarism that cannot be characterised as misconduct, there are no clear rules on how much negligence is required. Therefore, there may in principle be cases where plagiarism is committed through negligence that cannot be said to be actually gross. See, for example, the ruling of the Practice Committee of the University of Copenhagen of 29 March 2016 (the Esben Lunde Larsen case), in which the Committee stated, inter alia:

“On the basis of the rules governing research misconduct and good scientific practice, the Practice Committee finds that text taken from other people’s research works, and which is reproduced without citing a source, must in particular be assessed in relation to the following. First of all, the extent of the text: The longer the copied text, the more serious the plagiarism. Secondly, the originality of the content, assessed in relation to the field of research: The more original the content, the more serious the undisclosed reuse. Finally, the assessment should take account of whether the act was negligent or grossly negligent/intentional. […]

If the reuse of a text is very extensive or relates to scientifically decisive content, and if this is deemed to be grossly negligent or intentional, this will fall under the category of research misconduct.

Unacceptable scientific practice presupposes two conditions: First, an objective condition that a text passage of a certain length, or with sufficiently original content, must be reproduced, and second, a subjective condition concerning the degree of attentiveness. In relation to the above categorisation, breach of good scientific practice is distinguished from research misconduct through the degree of fulfilment of these two conditions.

[…]

 The subjective condition for whether there has been a breach of good scientific practice relates to the degree of attentiveness. This condition will be met if the person concerned has failed to show the attentiveness that could be expected of a professional with a similar background in the field. This differs from research misconduct, in which the person must have acted intentionally or with gross negligence.”

The fact that plagiarism must have been committed intentionally or negligently is, for example, significant when an assessment is made of whether of a researcher may be considered co-responsible for plagiarism committed by co-authors. If a co-author plagiarises without having any reason to suspect that this is plagiarism, he or she cannot be held responsible for this. If, on the other hand, the co-author ought to have discovered the plagiarism, and the failure to spot the plagiarism indicates carelessness, then this is a different matter. See from practice, inter alia, the ruling of the Committee on Research Misconduct of 7 February 2020, in which a supervisor listed as a co-author of a PhD dissertation was found not to be co-responsible for a plagiarism committed in the dissertation.

Are there other rules for plagiarism besides those mentioned here?

What we have focused on in the above is the extent to which plagiarism may contravene responsible research practice under the rules of the Act on Research Misconduct, the Danish Code of Conduct for Research Integrity, or institutional rules based on these.

However, these are not the only rules that plagiarism may contravene. As previously mentioned, plagiarism may also contravene the rules of copyright law or labour law, for example. In relation to scientific ethics, one also sometimes comes across principles of plagiarism that differ from those described above. Private companies that conduct research, for example, may well have their own principles, and there are examples of universities and other public research institutions that follow their own principles of plagiarism. As far as is known, certain PhD schools have introduced the rule that compilation theses – i.e. theses consisting of a summary of a series of underlying articles – will not be accepted if the summary contains any text from the articles without precise attribution. Apparently, the reason is that these articles are sometimes written by someone other than the PhD student in question, and such rules are intended to ensure that no plagiarism occurs in the summary.

However, rules like this do not regulate what is considered to be responsible research practice, which is regulated by the Danish Code of Conduct for Research Integrity and the Act on Research Misconduct, together with the practice associated with these, including the practice of the Committee on Research Misconduct and the Practice Committees. Rules that PhD schools or others may impose do not affect the scientific ethical assessment of when something is plagiarism as defined by the Danish Code of Conduct for Research Integrity or the Act on Research Misconduct. If, for example, a PhD student submits a summary that repeats text from the underlying articles to a PhD school in violation of its rules, it does not automatically indicate that the PhD student has committed plagiarism in the scientific ethical sense.

A number of institutions also have rules on plagiarism committed by students. Such rules will normally be interpreted in reasonable accordance with the standards for responsible research practice as set out in this article.

The relationship between plagiarism rules and authorship rules

In accordance with the principles outlined above, publishing material created by others without crediting them sufficiently may be deemed plagiarism.

According to the rules on authorship described elsewhere on this research portal, in certain cases, researchers who have contributed to a publication have the right to be mentioned as co-authors.

In practice, the rules on plagiarism are primarily applied in cases where the plagiarist has plagiarised in a publication to which the plagiarised person has not contributed. On the other hand, cases in which a researcher who has contributed to a publication does not consider himself or herself to have been sufficiently credited are handled according to the norms for authorship, not the plagiarism principles. See, for example, the ruling of the Practice Committee of the University of Copenhagen no. 16/2016, in which a researcher accused some colleagues, whom he considered to have copied his idea, of plagiarism, but in which the Practice Committee handled the case under the authorship rules. In some cases, however, doubt may arise as to whether a case should be assessed according to the authorship principles, the plagiarism principles, or both.

Example: A young researcher is employed as a research assistant in a project led by a professor. The division of labour is that the professor lays down the overall framework, while the young researcher carries out the source studies and physical work. The young researcher records the results of the work in a written memorandum, which is intended to form the basis for a book to be published at the end of the project. It has not been agreed who is to be credited as the book’s author. At the end of the collaboration, the professor publishes a book that reproduces extensive fragments of text from the memorandum. The young researcher is not credited as a co-author, but merely receives thanks in the acknowledgements. Should the case be assessed according to the authorship norms, or the plagiarism norms, or both?

In our opinion, there may be cases in which the application of the plagiarism rules cannot be ruled out. There is some support for this in the legislative history of the Act on Research Misconduct, where, under the remarks to Section 3, it is stated:

 “It is expected […] that disputes of authorship will in principle be handled as cases of dubious research practice rather than as research misconduct in the form of plagiarism.”

That statement would be meaningless unless cases in which an author did not consider himself or herself appropriately credited could never by definition be considered plagiarism. See from practice the Committee on Research Misconduct ruling of 3 February 2020, p. 9. The case dealt inter alia with the fact that the respondent had used text from a scientific product that he had written together with others, without attribution. The Committee did not find that this was a case of plagiarism because there was an insufficient basis to establish that the respondent had acted intentionally or with gross negligence. However, the reason was not that cases of this nature could only be assessed according to the authorship norms. See also the DCRM ruling of 11 October 2019. Here the question was whether it could be deemed plagiarism that a researcher had been written out of a research application that he had originally co-initiated. The Committee found that, for various reasons, this was not the case – without referring to the fact that the rules for plagiarism could be displaced by the authorship norms.

For further reading

• Ophavsret for begyndere – en bog til ikke-jurister, chapter 8. The book may be downloaded free of charge as an e-book from the UBVA website, www.ubva.dk, under “Publications”.
• The Norwegian report: God skikk. Om bruk av litteratur og kilder i allmenne, historiske framstillinger. Utredning fra et utvalg oppnevnt av Den norske Forleggerforening (DnF), Den norske historiske forening (HIFO) og Norsk faglitterær forfatter- og oversetterforening (NFF), 2006. This is abbreviated “Kildebrukutredningen”.
• On plagiarism in student theses, at: http://www.stopplagiat.nu.
• You can see lectures, articles and filmed interviews about plagiarism from UBVA symposia on plagiarism. They are available as videos at www.ubva.dk. Topics covered include:
  • How much plagiarism is there?
  • The difference between the assessment of copyright plagiarism and ethical plagiarism
  • Plagiarism of patents, designs and trademarks
  • Why do people plagiarise?
  • How does it feel to be plagiarised?
  • The history of plagiarism