Sense and Nonsense about Plagiarism

Steven Dutch, Natural and Applied Sciences, Universityof Wisconsin - Green Bay
First-time Visitors: Please visit Site Map and Disclaimer. Use"Back" to return here.

It is certainly true that colleges and universities are seeing an epidemic of plagiarism, fueled by easy electronic access to resources, including “research papers.” Back in the days before the Internet, students at least had to put out the effort to type out their stolen work by hand; now they need merely cut and paste. It is equally true that we are seeing an epidemic of faulty definitions of plagiarism, including high-profile but inaccurate and unsupportable claims of plagiarism lodged against prominent authors and filmmakers. Bad definitions of plagiarism confuse students and simultaneously trivialize the problem. The examples cited in style manuals are commonly so pedantic that students might be pardoned if they conclude the whole issue is a matter of academic nit-picking. Some commentaries on plagiarism reinforce this notion. For example, Hunt (2002) said:

Scholars -- writers generally -- use citations for many things: they establish their own bona fides and currency, they advertise their alliances, they bring work to the attention of their reader, they assert ties of collegiality, they exemplify contending positions or define nuances of difference among competing theories or ideas. They do not use them to defend themselves against potential allegations of plagiarism.

Few of those are particularly respectable reasons for demanding citations. Most seem to be principally concerned with placing the author and her ideas in the academic pecking order so a reader can judge the work on superficial grounds rather than on its merits. Hunt’s passage is a celebration of form over substance. [In fact, the games I’ve seen reviewers play with references make a strong case that manuscripts at the review stage should not contain references so that reviewers are forced to base opinions solely on the substance of the paper.] Conspicuously missing from the list are the principal reasons for citations; they allow readers to check the accuracy of facts, gauge the credibility of the ideas being presented, know whether an idea is solidly established, controversial, or hypothetical, and find further information.

Peter Levin (2003) bluntly labeled the institutional hysteria over plagiarism a “witch-hunt.” Many statements about plagiarism point to a complete collapse of critical reasoning. For example, there are scattergun conceptions of plagiarism like that of Martin (1994), who broadens plagiarism to include citation of sources second hand, use of quotations from secondary sources even if the original source is cited correctly, use of ghost writers and professional speech writers, and corporate authorship of reports. Another site (University of Phoenix, 2002) actually claims that failing to include a page reference in a citation constitutes plagiarism. Charges of plagiarism are fast becoming the blood sport of choice among academic bottom-feeders.

Bad Definitions

The core definition of plagiarism is falsely citing someone else’s work as one’s own for the purpose of fraudulently gaining some advantage. Obviously copying 1000 words verbatim without attribution is plagiarism. It’s the less severe cases that cause problems. Is 100 words plagiarism? Almost certainly. 10? Maybe. How many words can be quoted, and how specific do they have to be, before quotation marks are needed? How extensively does a passage have to be reworked before it becomes original?

In a world as inundated by verbiage as ours, there is a very high probability that any author will duplicate someone else’s expression frequently without even knowing it. In fact, just now I went to Google, entered “plagiarism” and “without even knowing it” (as a complete phrase in quotation marks) and got 301 hits – 301 other people using that exact phrase in articles on plagiarism. In violation of some published canons of referencing, I will not cite them. Calling the use of a commonplace phrase “plagiarism” is so pedantic as to be beneath contempt. People who are completely unaware they need to cite references need to be educated; otherwise I submit that the idea of committing plagiarism without even knowing it is an oxymoron.

Many authors on plagiarism discuss “unintentional” or “unconscious” plagiarism. This term seems to be used in two ways. First, it sometimes is used to describe a student who generally knows the rules, but for some reason fails to cite a source, paraphrases without citing, or misjudges the boundary between common knowledge that doesn’t need citation versus longer or more esoteric material that does. This student is guilty of a breach of etiquette, not plagiarism, and needs to have her professional judgment fine-tuned. Calling something like that plagiarism creates an exaggerated idea of how serious the problem is. It’s also likely either to make students overly cautious to the point where they cite everything, no matter how trivial, or just give up and figure they cannot possibly avoid plagiarizing even if they try, so there’s no point in trying.

The other usage of unconscious plagiarism is to describe a student unconsciously duplicating or imitating the wording used in some source. We could imagine a student taking notes from a source, then writing the paper and unconsciously recreating or closely approximating some of the wording used in the source. Taken to its logical conclusion, this definition of plagiarism means any author with a style influenced by another author could be accused of plagiarism. Any of us who uses a turn of phrase or a distinctive method of organizing ideas that we learned from someone else is guilty of plagiarism. This is a definition of plagiarism so sweeping and so trivial as to be useless. Failure to distinguish between unintentional mimicry and putting one’s own name on a paper copied straight off the Internet waters down the definition of “plagiarism” to the point where it means nothing. I submit that “unintentional” or “unconscious” plagiarism is also an oxymoron.

Since outright copying of large excerpts without attribution is clearly plagiarism, and minor duplication of wording is not, the only significant issue involved in defining plagiarism is how extensive and how exact the copying has to be before it becomes plagiarism.

One obvious question I have never seen addressed in any discussion of plagiarism is this: in any copied text, is there any other viable way to express the idea? If there isn’t, then regardless of how exact the quote is, it can’t be plagiarism. Saying “The Civil War began when Confederate forces fired on Fort Sumter,” or “Plants obtain energy through photosynthesis” isn’t plagiarism for the simple reason those are about the simplest and most direct ways to express those ideas. Only contrived wording could rearrange those sentences significantly. (“When Confederate forces fired on Fort Sumter, the Civil War began,” or maybe Yoda-speak: “Began the Civil War did when Confederate forces on Fort Sumter fired.”) Therefore I propose:

Rule 1: If some particular wording is the best way to phrase an idea, or there are few other practical ways to phrase an idea, unattributed usage is not plagiarism. If all the best ways of expressing an idea have been used, a subsequent author has no choice but to use someone else’s words. If the idea has been covered that often, it has probably entered the domain of common knowledge anyway.

In fact, the very concept of “common knowledge” shows that the entire concept of plagiarism is riddled with logical fallacies, inconsistencies, and arbitrariness. Every single idea in the realm of “common knowledge” has a primary reference. The New York Times is an acceptable source for the dates of Pearl Harbor or the assassination of John F. Kennedy (the really primary source is the death certificate on file in Dallas). Every state has a statute defining the location of its state capital. The idea of “common knowledge” is an explicit admission that there are legitimate reasons not to cite sources and that they boil down to convenience and streamlining the narrative. If Martin (1994) can broaden the definition of plagiarism to include all unattributed usage, we could just as logically and legitimately broaden the concept of “common knowledge” to dispense with citations altogether. After all, the greatest works of literature were published in such an environment, and once something is published, it enters the common corpus of knowledge.

One of my principal complaints about style manuals is that, for reasons of economy, they use abbreviated examples, and therefore the examples of “plagiarism” they cite are largely cases where good alternative wordings just don’t exist. For example, a generally excellent style manual by Hacker (2000, p. 173-174) includes the following example. The underlined text is shown that way in Hacker’s book.


Half of the force holding Fort Pillow were Negroes, former slaves now enrolled in the Union Army. Toward them Forrest's troops had the fierce, bitter animosity of men who had been educated to regard the colored race as inferior and who for the first time had encountered that race armed and fighting against white men. The sight enraged and perhaps terrified many of the Confederates and aroused in them the ugly spirit of a lynching mob. -Albert Castel, "The Fort Pillow Massacre," pp. 46-47


Albert Castel suggests that much of the brutality at Fort Pillow can be traced to racial attitudes. Fifty percent of the troops holding Fort Pillow were Negroes, former slaves who had joined the Union Army. Toward them Forrest's soldiers displayed the savage hatred of men who had been taught the inferiority of blacks and who for the first time had confronted them armed and fighting against white men. The vision angered and perhaps frightened the Confederates and aroused in them the ugly spirit of a lynching mob.

Phrases like “men who had been” and “who for the first time had” are such commonplace usage that calling them plagiarism is just plain silly. “Toward them Forrest's” is inverted word order that is a bit florid for a typical student, but the only real place where the exactness of duplication and the specificity of the wording conclusively prove copying is the last underlined excerpt. So out of six alleged examples of plagiarism, two are commonplace wording, two others (“holding Fort Pillow…”, “armed and fighting…”) are wording that could naturally flow out of the specific situation being described, one is strongly suggestive of copying and only one is really clear-cut. The real problem in this passage comes from the close copying of sentence structure and the substitution of mere synonyms to break up the direct copying, something Hacker herself points out. Hacker does provide an example of acceptable paraphrase.


Albert Castel suggests that much of the brutality at Fort Pillow can be traced to racial attitudes. Nearly half of the Union troops were blacks, men whom the Confederates had been raised to consider their inferiors. The shock and perhaps fear of facing armed ex-slaves in battle for the first time may well have unleashed the fury that led to the massacre.

This doesn’t follow the wording of the original as closely as the preceding example, but is it really free of plagiarism? By many of the definitions being put forward today, including the criteria Hacker herself uses, no. “Nearly half of the Union troops were blacks” and the original “Half of the force holding Fort Pillow were Negroes” are similar enough to be open to the charge that the only difference is a little shuffling of words. The same is true of “raised to consider their inferiors” and “educated to regard the colored race as inferior.” “Shock and perhaps fear” are little more than synonyms for the original “enraged and perhaps terrified.” Ross (2004) takes the stance: “paraphrasing without acknowledgement is still plagiarism, whether it is superficial or radical.”

Hacker actually does an uncommonly good job of explaining what plagiarism is, giving a specific example, and providing a constructive alternative in a constricted space. Her manual is cited here to show how hard it can be to convey the real nature of plagiarism in the short confines of a writing manual or class handout, and how hard it can be to define when legitimate adaptation ends and copying begins. Unfortunately, such abbreviated examples probably serve to convince students that the whole issue amounts to nit-picking, that wholesale borrowing is legitimate if it’s sufficiently camouflaged, and that no amount of effort can avoid plagiarism anyway, so why bother?

Hacker’s example raises a much more fundamental issue than the mere mechanics of paraphrasing. By what bizarre logic can it be plagiarism to give open credit to an author for taking a certain position, and then use that author’s own words to describe that position? What possible purpose is served by forcing a writer to shuffle around an author’s words or substitute synonyms after plainly stating that another author’s ideas are being discussed? The whole concept is a yet another oxymoron.

In fact, these “paraphrases’ are not even paraphrases at all, but indirect discourse. No rational person would object to hearing “Jane said that she and Bill went to the movies” and finding out those were very close to the actual words of the speaker. We would expect it. We would hardly impute intent to steal the original speaker's ideas. If you paraphrased someone else's words in court and did not follow them closely you could be accused of perjury. The fact that so many "experts" on plagiarism can't tell the difference between indirect discourse and paraphrasing doesn't do much to lend credibility to the rest of their ideas. The objections to close paraphrasing simply do not stand up to analysis. I propose:

Rule 2: If an author is openly credited and a passage clearly implies paraphrasing or restatement of the original author's ideas, it is not plagiarism. Period. As Ross noted above, paraphrasing without attribution is plagiarism. But paraphrasing with attribution is never plagiarism.

Here’s an example from a university site on plagiarism (Frick, 2004) where respondents are asked to judge whether adaptations of an original text constitute plagiarism. The original text reads: “The first technology was the primitive modes of communication used by prehistoric people before the development of spoken language” (from Frick, 1991). One text variation reads “One of the earliest technologies was non-verbal signs that our prehistoric ancestors used before spoken language emerged.” Now I might buy this as plagiarism if the test answer were that the author’s ideas were cited without attribution. This is a made-up test, so the designer of the test is free to read the imaginary student’s mind, but in real life it wouldn’t be so clear. The test passage is such an intuitively obvious idea it would be perfectly possible for a student to come up with it independently.

But the “correct” answer, according to the Web site, is that the test passage is plagiarism because it’s a paraphrase of the original author’s words without attribution. How so? The common elements of both passages are “technology,” “prehistoric” “before” and “spoken language.” The idea that communication is technology might count as plagiarism, but it borders on common knowledge. “Prehistoric” is trivially obvious because communication obviously predates written history, and “prehistoric” pretty clearly implies “before.” “Spoken language?” Well, if the focus of the assignment is the evolution of communication, of course it will refer to spoken language.

This is a definition of plagiarism that, if it were to become general, would make it absolutely impossible for any scholar to generate any synthesis of existing ideas without being subject to the charge of plagiarism. The very aim of education is to give the student a web of interlocked ideas and to inspire synthesis. Every author who reads widely and spends a lot of time thinking about what she read would be vulnerable to accusations of plagiarism unless she immediately wrote down every idea when she got it and annotated the works that influenced it. And it wouldn’t be sufficient merely to cite the sources, her narrative would have to include a description of exactly how each source influenced every element of her writing. It would be squeaky-clean, just not very readable. So I propose:

Rule 3: If writing uses wording or ideas that an author could reasonably have composed on her own, a charge of plagiarism is not justified. There may have been plagiarism, but there is insufficient evidence to make the charge stick; therefore there’s no point in even raising the issue.

The Ambrose Affair

Stephen Ambrose has been loudly accused of plagiarism, so it’s instructive to look at a number of alleged examples, from Lewis (2002):

A passage in Undaunted Courage and one in History of the United States of America During the Administrations of Thomas Jefferson (Library of America, 1986) by Henry Adams (the Adams work was originally published in 1889 to 1891):

[Those words are quoted from Lewis’ site, which is why they’re indented, but if for some reason you preferred not to quote them exactly, say because it would clutter the narrative or break up the flow to observe all the technical niceties, how else could you possibly word it without running afoul of some definition of plagiarism?]

Adams, pg. 6: "The entire population, both free and slave, west of the mountains, reached not yet half a million; but already they were partly disposed to think themselves, and the old thirteen States were not altogether unwilling to consider them, the germ of an independent empire, which was to find its outlet, not through the Alleghenies to the seaboard, but by the Mississippi River to the Gulf."

Ambrose, pg. 52: "Fewer than one out of ten Americans, about half a million people, lived west of the Appalachian Mountains, but as the Whiskey Rebellion had shown, they were already disposed to think of themselves as the germ of an independent nation that would find its outlet to the world marketplace not across the mountains to the Atlantic Seaboard, but by the Ohio and Mississippi river system to the Gulf of Mexico."

First of all, to anyone aware of the Louisiana Purchase, whose entire purpose was to acquire an outlet to the Gulf, and Aaron Burr’s involvement in a scheme to create such an independent nation and subsequent treason trial, these passages are common knowledge. Apart from the mildly unusual words “germ,” “disposed” and “seaboard,” the two passages easily differ as much as Hacker’s original excerpt and “acceptable paraphrase” above. Note also that Ambrose introduced a new connection, the Whiskey Rebellion, that was not in Adams' passage. This is just not plagiarism by any sane definition of the term. Another example from Lewis (2002) is a little closer to the mark:

Passages in Stephen Ambrose's Nothing Like It in the World (Simon & Schuster, 2000) and their counterparts in an earlier work, The Great Persuader (Doubleday, 1970) by David Lavender:

Lavender, pg. 9: "Only three females were aboard - twenty-four-year-old Jessie Benton Fremont...Jesse was on her way to California, escorted by a young brother-in-law, to meet her explorer husband, John Charles Fremont, at the conclusion of his fourth expedition through the western reaches of the continent, this one in search of a usable railroad route to the Pacific."

Ambrose, pg. 49: "Twenty-four-year-old Jessie Benton Fremont...was on the ship, on her way to California to meet her explorer husband, John Charles Fremont. He had just completed his fourth expedition through the Western reaches of the continent, this one in search of a usable railroad route to the Pacific."

What are the salient facts of the passage?

And now, a challenge to readers. Weave those facts into a coherent paragraph without using so many of the same words that you put yourself at risk of a charge of plagiarism. One solution would be to put a period at the end of every bullet point and concatenate them into a classic example of eighth grade choppy writing. Or we could replace the periods with commas and get a classic eighth grade run-on sentence instead.

It’s obvious that Ambrose borrowed from Lavender. We know that because he cited Lavender. The issue with Ambrose is never that he failed to attribute, it’s that he copied wording too closely. This is an inconsequential enough passage that it’s unlikely that Ambrose would have included it if he hadn’t seen it in Lavender. But is it plagiarism? Those ellipses in both passages are troubling. The ellipsis in the Ambrose excerpt really inspires questions about how much original writing by Ambrose is being omitted to make the case for plagiarism stronger. There is one flagrantly duplicated line: “his fourth expedition through the western reaches of the continent, this one in search of a usable railroad route to the Pacific." Frankly, that's such clunky prose I'd be ashamed to admit I was the original author.

Lavender, pg. 16: "All contemporary accounts of the beach at Panama agree that it was appalling. Rain poured. Mud, mildew, and fungus oozed over everything. Sanitation in the improvised tent city was inadequate. Unwashed raw fruit caused epidemics of dysentery. Malaria swept through the weakened sufferers. There were outbreaks of cholera, threats of smallpox...Bayard Taylor, who also crossed the isthmus in 1849, was horrified by the vice, depravity and selfishness that he encountered on all sides."

Ambrose, pg. 50: "At the end of the trip, all were appalled by Panama City. It rained continually. Mud, mildew and fungus oozed everywhere. Sanitation in the tent city was lacking or completely absent. Unwashed raw fruit caused epidemics of dysentery. Malaria and cholera were common, as were threats of smallpox. Vice, depravity, and selfishness thrived."

This is a pretty clear case of tight paraphrasing that crosses the generally accepted line. The common use of “appall,” “mud, mildew, and fungus oozed,” “unwashed raw fruit caused epidemics of dysentery,” and “vice, depravity and selfishness” pretty much rule out chance duplication. And enough examples have been found in Ambrose’s writing to show a consistent pattern. By many published criteria, this is plagiarism.

But is it unethical? The fact that we can apply a label to something does not make the action itself wrong; it may mean the label is defective. The fundamental fallacy of Martin (1994) is that he correctly recognizes many degrees of appropriation of ideas, but instead of concluding that the term “plagiarism” needs better delineation, concludes that all appropriation is plagiarism. Although Lewis (2002) disparages Ambrose for copying Lavender’s “vibrant” prose, there’s nothing particularly vibrant about the original excerpts. The reason people know Ambrose instead of Lavender is that Ambrose was just plain a better writer. The principal objection to tight paraphrasing is that it misleads the reader into thinking the copied words are the writer’s own. In a student paper, where the goal is to tell how well the student herself writes, that’s an issue. If the prose were really dazzling and copied by a weak writer, it would be an issue. But any decent writer should be able to come up with a line like “mud, mildew, and fungus oozed.” It is very hard to make a plausible case that Ambrose harmed either the original writer or the scholarly community. I’m much less bothered by the copying than I am by the indication of hasty writing, sloppy proofreading and such unfamiliarity with the source material that Ambrose failed to spot a copied phrase when he saw it, in other words, poor craftsmanship.

It is impossible to use any source without incorporating vocabulary, logical structure, or concepts from the original source and thereby violating some definitions of plagiarism. The principal difference between “acceptable” paraphrasing and “plagiarism by paraphrasing” is cosmetic: does the user shuffle and reword enough to disguise the borrowing acceptably? Do we consider textbooks or popular expositions “plagiarism” since they rarely follow strict scholarly citation models? Since the concept of paraphrasing as plagiarism is logically indefensible, I propose:

Rule 4: If a source is cited in situ, or cited appropriately consistent with the format and style of the work, limited duplication of wording is not plagiarism. I am proposing we sharply curtail the definition of paraphrasing as plagiarism. I’m not arguing that we regard it as good practice, nor that lifting text without attribution should be acceptable, but there’s a difference between sloppiness on the one hand, and unethical behavior on the other. Call it the academic equivalent of decriminalizing pot. I propose scaling back this definition of plagiarism for the same reasons that many academics support decriminalizing pot: it’s unenforceable, attempts to enforce it waste time and resources, it does more harm than good, and it creates a pretext for busybodies intent on harassing others. The present definition is unworkable, has ragged and fuzzy edges, and leads to innumerable frivolous accusations. As a guideline, I suggest that anything longer than a line needs to be formally quoted. Highly original, distinctive, memorable or historic utterances also need to be quoted (“Fourscore and seven years ago,” "Don't have a cow, man," "Oh my God, they killed Kenny!"). Anything else is not plagiarism as long as the source is credited and there is significant original material added by the user. Excessive reliance on existing credited text is hack writing, nor plagiarism. Anyone literate enough to read a book with citations and be concerned with plagiarism should be literate enough to realize that a citation means someone else’s work is being melded with an author’s own. Anyone really concerned with telling which is which should take the time to check the original source. Anyone unwilling to expend the effort has no business getting involved.

Also, the “paraphrasing as plagiarism” concept simply doesn’t address the central problem of student plagiarism: the lack of independent thinking. It’s entirely possible for a student to shuffle words and ideas around enough to pass every imaginable test for plagiarism and still not have an original thought in the process. So defining close paraphrasing as plagiarism really doesn’t eliminate plagiarism; it merely makes more skillful and literate plagiarists.

I’d actually prefer to see some skillful plagiarism instead of the sloppy stream of consciousness writing I typically get. The worst problem I see with student writing is that so many students can’t even be bothered to look up sources to plagiarize. Plagiarism exists because it works, and it works because too many faculty cannot be bothered to read assigned papers or devise original writing assignments that are hard to plagiarize. Assigning five pages on generic topics like plate tectonics or the Franco-Prussian War is tantamount to saying “Here’s some busy work. Go copy something off the Internet.”

Plagiarism of Ideas

In addition to allegations of direct copying of text, there have been a number of high-profile cases alleging plagiarism of ideas by novelists and filmmakers. One was lodged by Steven Kessler, who claimed that Steven Spielberg and Michael Crichton had ripped off the idea for Twister from his screenplay Catch the Wind. He cited the following points of similarity:

  1. The protagonist was motivated to study tornadoes by the loss of a loved one.
  2. There were two rival teams of scientists each trying to develop a tornado probe.
  3. The probes in both cases were named after characters from The Wizard of Oz.

The first point is a stock literary plot device. We might as well charge Spielberg and Crichton, or for that matter Kessler, with stealing ideas from Moby Dick: the protagonist suffers a personal loss and spends the rest of his life at war with an impersonal force of nature. White whale, black funnel cloud; what’s the difference? The second point is also a standard plot device. Personally, I would gladly read a novel or watch a movie where completely rational people plan how to probe a tornado without getting killed, and without unraveling into petty personality conflicts. I see enough conflict and human irrationality in the news and my daily dealings with people, thank you, but most authors feel the need to include some sort of personal conflict. And who would ever dream of naming a tornado probe after a character from The Wizard of Oz, especially since one of the prototype devices used in real research was called TOTO? (Bluestein, 1983)

If such ultra-broad concepts of plagiarism gain ground, we can kiss all academic freedom good-bye. There is no idea that can’t be shown to incorporate elements of older ideas. The corporate attempts to tighten copyright control pale into insignificance compared with the danger that frivolous charges of plagiarism will do.

Finally, I can’t help but noting that the direction of most plagiarism charges outside the classroom is from the bottom upward. When Mark Svenvold (2005) published a book on chasing tornadoes, Michael Crichton and Steven Spielberg didn’t accuse him of plagiarism. Overly strict and broad definitions of plagiarism are creating a culture of academic mosquitoes who sustain themselves by swarming around more successful writers. I concur with Peter Levin (2003) that hypertrophied concern over plagiarism is fast becoming a witch-hunt.


Return to Pseudoscience Index
Return to Professor Dutch's Home Page

Created 1 August, 2005,  Last Update 24 May, 2020

Not an official UW Green Bay site