William J. Cronon, vice president of the AHA's Professional Division, writes: The AHA's Professional Division is commissioning a series of essays and advisory documents about common challenges historians face in their work. Although these essays will be reviewed and edited by members of the Professional Division, and although they will appear in Perspectives and on the AHA web site, they should not be regarded as official statements of either the Professional Division or the AHA. Instead, their goal is to offer wise counsel by thoughtful members of our guild in an effort to promote wide-ranging conversations among historians about our professional practice. Because plagiarism has generated so much public comment and controversy in recent years, we have focused some of our earliest efforts on this critical issue. We are most grateful to Peter Hoffer, an eminent legal historian at the University of Georgia and a member of the Professional Division, for producing the following "Reflections on Plagiarism" (the concluding part, "The Object of Trials," can be found in the March 2004 issue of Perspectives).

There are seven causes of inconsistencies and contradictions to be met with in a literary work. The first cause arises from the fact that the author collects the opinions of various men, each differing from the other, but neglects to mention the name of the author of any particular opinion.1
—Maimonides, A Guide for the Perplexed

For writers, readers, and teachers of history, as for Maimonides long ago, plagiarism is rightly both a mortifying and perplexing form of professional misconduct. It is mortifying because it is a species of crime—the theft of another person's contribution to knowledge—that educated, respectable people commit. It is perplexing, because, despite the public shame that invariably accompanies revelations of plagiarism, it continues to occur at every level of the profession, from prizewinning historians to students just beginning their careers. While many of these infractions have come to light because readers and writers of history are keen-eyed and implacable critics of the offense, additional cases may be avoided if authors and reviewers knew more about the offense. That is the purpose of this essay, the first installment of two on the subject of plagiarism.

Plagiarism is commonly defined as the appropriation of another's work as one's own.2 Some definitions add the purposive element of gaining an advantage of some kind.3 Others include the codicil, "with the intent to deceive."4 The historical profession has adopted a broad and stern definition of plagiarism, based upon ethical rather than purely legal conceptions. Its definition of plagiarism is the "expropriation of another author's text, and the presentation of it as one's own."5 It does not require that the act be intentional, nor that the offender gain some advantage from it.6 Nor for historians is the ultimate sanction against the offense a legal one, but instead the public infamy that accompanies egregious misconduct. As historians, we know, in the words of Lord Acton, the "undying penalty which history has the power to inflict on wrong."7

Plagiarism may include copyright violations, but the two are conceptually independent. Massive plagiarism may not involve a single instance of copyright infringement. Copyright is a property right defined by statute. In general, copyrighted materials can only be reproduced with permission of the copyright holder, but the "fair use exception" in the law permits quotation from most scholarly works. Plagiarism is first and foremost an ethical matter, and whether or not permission is required or obtained for use of another's work, the rules for source references and against impermissible copying or borrowing apply whether or not the source is under copyright protection.

The Statement on Standards of Professional Conduct prepared by the AHA reminds us that plagiarism "takes many forms." These may include "the limited borrowing, without attribution, of another person's distinctive and significant research findings . . . or an extended borrowing even with attribution."8 The bottom line is: work presented as original must be original; phrasings and research findings derived from others must be credited to others or the entire scholarly enterprise is undermined.

Historians adhere to these standards with the full knowledge that not everyone has the same attitude toward plagiarism as the historical profession. Some observers have noted that plagiarism may not only be common in painting, architecture, music, literature, and other forms of fine artistic expression; it is often regarded as a form of compliment. Uncredited borrowing occurs in popular art forms with disconcerting regularity. One best selling mystery-adventure novel based on a supposed code in the work of Leonardo Da Vinci relies upon and repeats the discoveries of many art historians, but neither mentions nor cites any of them.9 Folk music artists routinely rewrite and rearrange their predecessor's tunes in what Pete Seeger has called, according to Arlo Guthrie, "'the folk process.'"10 Legal writing is more accommodating to plagiarism than historians are. According to one leading legal scholar, "an individual act of plagiarism usually does little or no harm,"11 a perspective perhaps influenced by the fact that appellate judges' legal opinions are supposed to be derivative, based on lines of precedent elucidated in earlier judicial decisions.

At the same time, any realistic treatment of this highly complex and long-lived issue within the historical profession must include descriptive as well as prescriptive language. That is to say, one should not ignore the existence of long accepted usages, conventions, and occupationally mandated variations, nor the evolution of our standards in this area of professional ethics. 12

I. Avoiding Plagiarism

The first line of defense against plagiarism is the author. Even the most original historical scholarship rests in part upon earlier (secondary) studies. Historians should always give credit to those whose work they have consulted and to those who render assistance in the course of our work. Whether the form of circulation of historians' work is an article, book, museum exhibit, or other kind of publication, historians recognize scholarly debts in three ways: exact quotation, paraphrase, and general citations to works consulted. By their care and integrity in crediting these sources and by limiting the extent and monitoring the form of their copying or borrowing from these sources historians both avert the suspicion of plagiarism and avoid its commission.

All exact reproductions of another's words (direct quotations) should appear within quotation marks, or if in a block quotation, set off at the margins. All missing material from within the quotation must be indicated by ellipses. No words may be added except in square brackets. The order of the passages in the original may be altered by the author of the new work for literary or argumentative purposes, so long as the reference notes indicate the order of the passages in the original. The source of every direct quotation must either be cited in the text and fully described in a "works cited" section at the end of the piece (MLA style), or referenced with foot or end notes (Chicago Manual of Style). Publications without in-text reference apparatus (most textbooks, for example) should report all secondary sources in the text or a bibliographical essay. Failure to put the borrowing of exact words in quotation marks; failure to cite the source of the quotation in the reference notes with sufficient precision for a reader to check the quotation; and changing a few words in a nearly exact replication of another's text and then not giving any reference, whether inadvertently, through negligence, or intentionally, may be read as plagiarism. But even with full and correct references to the source, historians must take care not to borrow or copy excessively from any one source or group of sources.

A special case arises when an author quotes from a primary source quoted in part or fully in a secondary source. If the author relies on the secondary source for a portion or the entire text of the primary source, citation of the latter should take something like the form "A [the primary source], quoted in B [the secondary source]." This alerts the reader that the author has borrowed the quotation from the secondary source and has not consulted the original source. The author should not simply cite the primary source. If the author, however, guided by a secondary source, finds the entire primary source in the original, reads it, and then uses some portion of it, there is no need to cite the secondary source in which it was initially encountered. The purpose of scholarly citation of primary sources is to enable other readers to find and examine them for themselves. By contrast, in no case whatsoever should an author simply reproduce another author's documentary evidence with or without that author's reference notes, without fully crediting the author, giving the impression that the borrower had done the research. This is another form of plagiarism.

The second common form of indebtedness to another work is the paraphrase, the rephrasing of another's arguments or findings in one's own language. When in doubt, one should always prefer quotations to paraphrases, but there are reasons for preferring paraphrase to quotation including the inelegance of the prose in the original, the author's desire to avoid stringing together a series of long quotations, and the need to blend into a single paragraph the arguments of many secondary sources. Authors must paraphrase with great care if they are to avoid falling into plagiarism, for paraphrasing lends itself to a wide range of errors. In particular, a paraphrase, particularly after some time has passed in the course of research, may be mistaken by the author for his or her own idea or language and reappear in the author's piece without any attribution. Mosaic paraphrases patching together quotations from a variety of secondary sources, and close paraphrases, wherein the author changes a word or two and reuses a passage from another author without quotation marks, also constitute plagiarism.

In print, all paraphrases, no matter how long or how many works are paraphrased, must be followed by citations to the sources that are as clear and precise as those provided for a direct quotation. The citation should refer to the exact page(s) from which the material was taken, rather than a block of pages or a list of pages containing the material somewhere. If the material comes from a web site (for example another teacher's original lecture notes on an open web site), citation should include the entire web address and the date that it was accessed.

The third common manner of giving credit for a scholarly debt is the general citation to work in the field. Sometimes this will follow the author's summary of arguments or evidence from a number of works. In textbooks, a single paragraph may encapsulate three or four prior publications on the topic. All works an author consults should be either cited in the reference apparatus or in the bibliography. If particular pages were consulted, these should appear in references. By contrast, works not consulted by the author, even though they may be relevant to the topic, should not be cited. Such a citation would give the false impression that the author had used the work. By the same token, when an author makes a general citation to a work that contradicts the author's findings or conclusions, that fact should be noted in the citation.

If an author employs research assistants, their errors—for example the omission of quotation marks around a direct quotation or the omission of a reference at the end of a paraphrase—become the author's responsibility. The general rule that the supervisor is responsible for the acts of the employee applies here. What is more, the author had the chance, before publication, to review the entire text, and with that last clear chance goes the onus for all errors.

II. Conventions and Usages

Often it is hard to determine where plagiarism has occurred. Readers may disagree whether and how often an author has crossed the line between the permissible and the impermissible. Another way to formulate this general issue is that historians' use of others' work lies along a spectrum, a "continuum of intellectual indebtedness" in the words of William Cronon, in which possible misconduct in each work must be weighed on its own merits.

I would add to this another dimension ruled along an axis of long-established usages and conventions. As the AHA Statement on Standards reminds us, "historical knowledge is cumulative, and thus in some contexts—such as textbooks, encyclopedia articles, or broad syntheses—the form of attribution, and the permissible extent of dependence on prior scholarship, citation, and other forms of attribution, will differ from what is expected in more limited monographs."13

The advice on avoiding plagiarism in part I of the essay and the suggestions on dealing with plagiarism in part II apply with particular force to those works whose authors promulgate them as original contributions to scholarship, offering new findings, interpretations, and approaches.14 Conversely, personal letters, working documents, or in-house memos thus rarely exhibit the formalities of citation. Victoria Harden of the AHA Task Force on Public History suggests that this is particularly true when they are prepared as précis or summaries of existing scholarship by subordinates for their superiors, often on short notice. If at some time the author presents these reports or statements as original contributions to knowledge, or offers them as credentials for hiring, promotion, employment benefits, fellowships, or prizes, they must give credit to all sources consulted.

Certain kinds of historical writing or oral presentation of historical materials for general public consumption also commonly omit reference notes. Such materials may include guidebooks, captions at museum exhibitions, pamphlets distributed at historical sites, and talks or performances by re-enactors or historical interpreters. In the context in which these works are used or performed, their utility might be impaired if their authors or presenters were required to credit their scholarly debts. At the same time, it would be ideal if print or electronic versions of these materials include recognition, in some form, of the contributions of individuals to them and the scholarly sources on which they relied.

Lectures by history teachers to their classes and speeches at public meetings rarely include explicit references to the secondary sources on which the lecturer relies, particularly if the lecture is not presented as an original work of independent scholarship and the materials borrowed from others constitute only a small portion of the whole. The debt that teachers of history owe to their own teachers is pervasive and often results in lectures that borrow structure and theme from those mentors. While acknowledgment of this debt will never go out of fashion, it is commonly omitted.

Textbooks, like lectures to classes, are assumed to be cumulative and synthetic. In fact authors of textbooks rarely quote or cite precisely each secondary source they have used, and the topical structure and rhetorical formulae of new textbooks bear a remarkable similarity to older ones. It would be best if textbook authors limited their borrowing from any one secondary source and cited in a bibliography all the sources used. It is mandatory that any direct quotation from another work (excluding of course prior editions of the same textbook) be correctly identified.

What is generally termed popular history—journalistic accounts, memoirs and autobiographies, and articles by professional historians in general or popular journals of opinion, for example—rarely conforms to the same standards of citation as scholarly monographs and interpretive essays. Many popular histories, for example, have only a short list of works consulted. But wholesale borrowing from another work, even with attribution, is unacceptable. Ideas themselves cannot be plagiarized, but authors may not claim as their own the full-dress presentation, according to the AHA Statement on Standards, of "another person's distinctive and significant research findings, hypotheses, theories, rhetorical strategies, or interpretations."15

A particular case of book-length scholarship without footnotes or endnotes arises in some noteworthy series of books—for example, the Library of American Biography from Penguin/Viking, and the Landmark Law Cases and American Society from the University Press of Kansas. These are original contributions to knowledge by leading scholars designed primarily for classroom use. Series authors and editors are nevertheless very careful to observe the rules against plagiarism in these books to avert even the suspicion of surreptitious borrowing or copying from other works.

Common understandings, widely shared ideas, dates, names, places, and events in history do not need to be referenced, even if they were obtained from a particular source in print. For example, one does not need to cite a source to say that Washington was a Virginian, or that the Declaration of Independence was signed in 1776. Similarly, catch phrases, "conventional wisdom" (itself a phrase coined by the prizewinning economist, John Kenneth Galbraith), and even longer quotations borrowed from sources in common currency, like Shakespeare, the Bible, and Monty Python can be repeated without references.16 But if an argument or thesis is unique to a secondary source, and not a matter of general currency, the source should be cited.

The foregoing paragraphs in section II refer to secondary or scholarly sources. The appropriate use and citation of primary sources raises slightly different questions. It is plagiarism to take as one's own work portions of primary sources without citation, or through excessive borrowing even with citation; but when these are clearly indicated as the work of another, formal citation of the place where the author found the primary source (a printed documentary edition or the archive, for instance) is sometimes omitted. I believe that historians should treat primary sources with the same scholarly care as they apply to secondary sources, indicating exactly where they found a primary source (so that those who follow them can find it as well). Failure to fully credit a primary source (failing, for example, to give the title of the docket book or the file paper collection as well as the courthouse when citing a legal document) may not only lead to confusion among readers and suspicions of research misconduct, it lends itself to plagiarism.

A final, somewhat special case involves work for hire. If an author hires research assistants or "ghost writers," and by the terms of their contracts the latter agree that their names will not appear on the work as its author, they cannot argue that the final product plagiarizes their work. As Linda Shopes of the Task Force on Public History suggests, it is always good practice, however, for any supervisor who uses in his or her own work the research or writing of an employee to credit that employee by name. When an author relies upon the research of others not hired for that purpose—students in the author's class or individuals whose graduate studies the author is directing, for example—and those researchers' own language (as opposed to the documents or other evidence they find) is adopted or adapted by the author, it is unethical not to give credit to the researcher. This may be done in the acknowledgments section of the publication, in a note or notes, or in the text. If the author has depended upon the researcher to write up the results of the research and then uses these reports verbatim, the researcher should be given co-authorship.17

III. Detecting Plagiarism

Both academic and lay readers rely on the integrity of scholars. Authors owe it to their audiences as well as to themselves to avoid even a hint of plagiarism and are the best detectors of inadvertent mistakes in attribution or excessive copying even with references. This is true from the inception of the research to the closing stages of preparation of manuscripts for publication. Before any piece of scholarly research is presented orally, circulated, or submitted to a publisher, the author should review it carefully for plagiarism. Returning to the research notes and laying them against the text may reveal errors. The author should look for omissions and commissions that might have slipped into the successive drafts over time.

All scholarly journals and academic presses will send the work out to readers ("referees") to advise on publication, but referees cannot catch every instance of questionable use of secondary sources; nor should referees be held responsible if plagiarism slips past them. In particular, citation checking is not ordinarily part of their job. If the publisher (as is true of most trade houses) does not employ outside readers, the author has to be doubly careful. Book reviewers (or referees asked to help with hiring or tenure and promotion decisions) may uncover instances of plagiarism, but because that is not the primary reason for which they are reading the author's work, one cannot expect them to catch plagiarists in the act.

Despite all the reasons for which authors should and can avoid plagiarism, it occurs. The suspicion that a work contains plagiarism and its subsequent exposure are not pleasant occasions. Historical scholarship depends upon trust. Readers and publishers both rely on authors' claims of originality (indeed, book publishing contracts require authors to "warranty" that they have not plagiarized any other work).

In all cases of suspected plagiarism, the single most effective method of detection is the meticulous, side-by-side comparison of texts. This parallel reading of source (original) text and target (new) text will not absolutely prove plagiarism except in the most egregious cases, but it can raise or allay the level of suspicion. A reader comparing texts should not just look for similar words or phrases (for example, groups of three or four words) as these may in fact come from more than one author using the same primary sources or from the argot of a specific field. Instead, the reader should concentrate on unusual phrasing, for example uncommon verbs and unique combinations of modifiers. An example of parallel text comparison appears on the AHA web site.18

If the reader of parallel texts finds a few examples of questionable practices in a long work, they may, with the profession's accustomed charity, be attributed to mere coincidence. In the uncovering of plagiarism, as in all misconduct, one presumes innocence. But discovery of plagiarism throughout a manuscript or plagiarism in a series of publications suggests wanton and cynical disregard of ethical and professional standards, and will not be forgiven. The discovery may take years, but plagiarism is an offense that cannot be hidden forever.

—Peter Hoffer, Distinguished Research Professor at the University of Georgia, is a member of the Professional Division of the AHA.

Acknowledgments

The author is grateful to Maureen Murphy Nutting, Susan Mosher Stuard, and Denise Youngblood, members of the AHA's Professional Division, and William Cronon, its vice president; Arnita Jones, executive director of the AHA; Stanley N. Katz, chair of the AHA Task Force on Intellectual Property, and its members Michael Les Benedict and Michael Grossberg; the AHA Task Force on Public History, its chair, Linda Shopes, and its members Victoria A. Harden and Jamil Zainaldin; James Grossman, vice president for research and education at the Newberry Library; Nan McMurry, history and social science acquisitions librarian, University of Georgia Libraries; Lewis Bateman, senior acquisitions editor, Cambridge University Press; Fred Woodward, director, and Michael Briggs, editor in chief at the University Press of Kansas; Charles Grench, assistant director, and Amanda McMillan, assistant editor, University of North Carolina Press; Ashley Dodge, senior editor, Longman Publishing, College Division; Robert Brugger, senior editor, Johns Hopkins University Press; Williamjames Hoffer, Seton Hall University; and the members of the University of Georgia history colloquium for assistance in the preparation of this document.

Part 2 of this essay will be published in the March 2004 issue of Perspectives.

Notes

1. Maimonides [Moses Ben Maimon], "Introductory Remarks on Method," The Guide for the Perplexed trans. M. Friedlander 2nd rev. ed. ([1904] reprinted New York: Dover, 1956), 9.

2. Black's Law Dictionary, 7th ed., Bryan A. Garner, ed., 1170.

3. Joseph Gibaldi, MLA Handbook for Writers of Research Papers, 6th ed. (New York, 2003), 66.

4. Black's Law Dictionary, 1170, quoting Paul Goldstein, Copyright's Highway, 12.

5. The Statement on Standards of Professional Conduct (Washington, D.C.: American Historical Association, 2003), 10.

6. In this we are in accord with the Modern Language Association; see Gibaldi, MLA Handbook, 66.

7. John Edward Emerich Acton, A Lecture on the Study of History, Delivered at Cambridge, June 11, 1895 (London: Macmillan, 1895), 63.

8. Statement on Standards, 10. See below for examples of these.

9. Dan Brown, The Da Vinci Code: A Novel (New York: Doubleday, 2003). The book's "Acknowledgments" (n.p.) has only this: "My thanks also to Water Street Book Store for tracking down so many of my research books" and does not mention the individual titles and authors.

10. Arlo Guthrie quoted in Jon Pareles, "Critic's Notebook: Honoring Alan Lomax, Folk Music Crusader," the New York Times April 14, 2003, E3.

11. Richard A. Posner, "The Truth About Plagiarism," Newsday, May 18, 2003, reprinted at www.law.uchicago.edu/news/ posner-r-plagiarism.html (accessed May 1, 2003).

12. See, for example, Anthony Grafton, The Footnote: A Curious History (Cambridge, Mass.: Harvard University Press, 1997), 190–91.

13. AHA, Statement on Standards, 10. Older usages and conventions of citation were often not as precise or complete as those for citation in use today. For example, Oscar Handlin's Pulitzer Prize-winning The Uprooted (Boston: Little, Brown, 1937), Daniel Boorstin's Bancroft Prize-winning The Americans: The Colonial Experience (New York: Knopf, 1958) and his Parkman Prize-winning The Americans: The National Experience (New York: Knopf, 1965) did not have any notes. Neither did Perry Miller's much admired The New England Mind: From Colony to Province (Boston: Beacon, 1954). They merely had detailed bibliographies without page references to the quotations in the text.

14. This essay does not consider the question of falsification of research findings. A good survey of the issues raised in these cases appears in Ellen Altman and Peter Hernon, eds., Research Misconduct: Issues, Implications, and Strategies (Greenwich, Ct., 1997).

15. Statement on Standards, 10.

16. Maurice Isserman, "Plagiarism: A Lie of the Mind," The Chronicle Review: Chronicle of Higher Education, May 2, 2003, B12–B13. "No one expects the Spanish Inquisition."

17. In the area of electronic publishing, editors are taking an increasingly active role in areas that traditionally were categorized as "authorship." See Kate Wittenberg, "Scholarly Editing in the Digital Age," Chronicle of Higher Education, June 20, 2003, B12. It is not clear to what extent this development will continue, nor whether it will raise questions of proprietorship of electronically published scholarship.

18. See Susan Mosher Stuard and William Cronon, "How to Detect and Demonstrate Plagiarism."