WHAT IS PLAGIARISM?
Plagiarism is one of the trickier violations of academic integrity. As a violation, it goes to the heart of the academic and scholarly enterprise. At most universities, academic conduct codes reflect this: one can get expelled for plagiarizing. At the same time, scholarship is, over time, a collective enterprise -- remember that Newton did what he did because he had stood "on the shoulders of giants" -- and scholars continuously need to draw from the works of others. It is therefore worthwhile to spend some time thinking about what counts as plagiarism.
Let's begin with a definition. Plagiarism in the context of a university course can be defined as submitting a piece of work which in part or in whole is not entirely the student's own without attributing those same portions to their correct source. As Gordon Harvey points out in his Writing with Sources (Indianapolis/Cambridge: Hackett, 1998), after offering a similar definition, this amounts to lying, cheating, and stealing (p. 22). It is lying because you are claiming to be the author of something you are not; it is cheating because you violate the rules of the game; and it is stealing because you take something from someone else without a permission.
Plagiarism can come in four main forms (the following categorization comes from Harvey and the explanations follow his examples, despite small differences, pp. 23-25):
I. As information or data which is not cited
Any time you present a piece of information or some amount of data which is not common knowledge, you must cite the source for that piece. Even if it is based on your own research, say, in a laboratory, you need to explain how you came about it. If it is from another source, you must cite the source. If you are wondering whether something is common knowledge or not, use your common sense, but also err on the side of caution. Everyone knows the earth is flat, and everyone should know Springfield is the state capitol of Illinois, but not everyone knows that Denmark has the highest alcohol consumption in Scandinavia; for that, you need to tell you readers where you learned it. (It might, for example, have been from the Finnish Research and Development Centre for National Welfare and Health website on alcohol consumption.) The same goes for precision: it is common knowledge there are about 270 million people in the United States, but if you say there are 275,773,000 people in the U.S., you'd better cite the U.S. Census Bureau or whatever source told you this.)
II. As an idea which is not cited
Let's say you read the following passage by Eugene Debs, from his "Revolutionary Unionism" (in Kenneth Dolbeare, ed., American Political Thought [Chatham, N.J.: Chatham House, 1998]):
The unity of labor, economic and political, upon the basis of the class struggle, is at this time the supreme need of the working class. The prevailing lack of unity implies lack of class consciousness; that is to say, enlightened self-interest; and this can, must and will be overcome by revolutionary education and organization. (p. 396)
You might write in your paper the following:
Debs argues, "The unity of labor, economic and political, upon the basis of the class struggle, is at this time the supreme need of the working class" (Debs, p. 396). The reason this unity is needed clearly suggests that the working class wasn't, at the time, collectively self-interested enough...
Now, although the terminology in the second sentence is different, the idea still clearly comes from Debs, and not citing him makes this a case of plagiarism.
III. A verbatim phrase or passage not quoted
This is the most obvious case of plagiarism, but even that is trickier than people think. You might, for example, think that it's not plagiarism if you borrow a nice verbal formulation simply because it's nice, and not because it has an immediate connection to the larger ideas. For example, you might be writing a paper in international relations and remember Thomas Hobbes's famous phrase about "convenient articles of peace" (Leviathan, ch. 13). You write:
Neo-liberal theory argues that supernational organizations should help antagonistic states seek convenient articles of peace...
Although you are not stealing any ideas from Hobbes, but merely a phrase, you are stealing it, and the passage is plagiarism. Again, the matter is slight tricky because there are many phrases that have become part of our common vocabulary; many Biblical phrases are like that, as are Shakespearean phrases. One might even argue that Hobbes's "solitary, poore, nasty, brutish and short" is one of those, even if "convenient articles of peace" isn't. When it doubt, use both caution and common sense. There is nothing wrong in crediting someone if he or she is the source from which you learned a phrase.
The point to remember about verbatim passages is that you don't get off the hook if you change a word or two, but otherwise quote a passage.
IV. A structure or organizing strategy which is not cited.
For example, if I had not said that the way the categories of plagiarism here are organized is Harvey's, I would have been plagiarizing, even if my explanations had been different. Once again, things are tricky, and sound judgment is needed. Some organizational strategies are commonsensically common: chronology, for example, is very common. If you write a paper on Plato, Aristotle, St. Augustine and Hobbes, in that order, it's perfectly fine not to cite all the other discussions that follow the same chronological order. However, if you could have written on any figures in the history of Western thought, and you pick both the order and the figures from someone, you'll want to cite the source: "Following Schmoe (1995), I will focus my discussion on Plato, Aristotle, St. Augustine and Hobbes. This is a useful selection because, as Schmoe argues,..."
There are some cases where uncited borrowing might not be plagiarism. For example, in my courses, I always declare my own lectures, for the purposes of the course, to be in public domain. They aren't really: my ideas are my ideas (or come from others who deserve credit). But it simplifies student writing if one doesn't constantly run into footnote clutter of the variety "Professor LaVaque-Manty, on May 24, 2000." Similarly, the contents of a data set shared by an entire course or some other source of information might not have to be cited. These are, however, special cases, and they depend on an explicit change of rules. Absent such changes, then, always assume that whenever you pass an idea that is not yours as yours, you are plagiarizing and violating academic integrity.
How one cites material depends on your discipline, your subject matter and on the possible special rules your instructor might have set up. It is every student's responsibility to be familiar with the prevailing standards for citing in his or her discipline. Manuals on these abound, both as books and online.