Scholars cite their sources of information for three chief reasons: 1) intellectual honesty; 2) as a guide for their readers and colleagues to the full context of the material consulted; and 3) to be sure that the information conveyed is accurate.

Intellectual honesty is the admission that humanity is linked together in a kind of collective learning process. Very little is discovered "de novo," that is, without a solid foundation in other researchers' previous exploration and understanding. Citation is an act of humility and an act of appreciation for what other scholars have pieced together about the nature of a particular problem or an aspect of some phenomenon.

Citation can also take the interested reader to the sources consulted, so that s/he might "check" the researcher's use and interpretation of the evidence already acquired. For example, a quotation by Abraham Lincoln takes on a very different meaning when read in full context. "Fellow citizens, we cannot escape history." Often Lincoln captured the essence of a thought in very few words. Aaron Copland used these very words in his hagiographic musical composition "A Lincoln Portrait" (1942). They resound in our own hearing -- but the careful researcher would discover that these words, penned by Lincoln in his 1862 annual message to Congress, were written in the context of the president's support for gradual abolition of slavery (by 1900), monetary compensation for slaveholders thus deprived of their "property," and the possible relocation of the ex-slaves outside of the United States. The context reveals a very different portrait of Lincoln than does Copland's "portrait".

Connected to the idea of context is the idea of accuracy. Putting aside the issues of editions with errors and fraudulent documents, literary scholars and textual analysts have long been aware of the legitimate differences (intended and unintended) that occur between different versions or editions of a written or a printed work. In the bibliographic world authors often add or delete material from a book, correct errors, and refine arguments. The subsequent work is largely like its first version, but sufficiently different to have the phrase "second edition" or "revised edition" attached to it. In computer programs differing versions are often given progressively higher numbers, for example WordPerfect 5.1, WordPerfect 5.2, Wordperfect 6.0, etc. Both systems notify readers and users saying, essentially, "hey, something is different here."

But the issue becomes more difficult with e-documents. The problem of what version to cite is augmented by the Internet's capacity to simultaneously maintain differing editions or versions of the same e-document. Often times the version found and cited by a scholar is dependent upon the server where the particular e-document is mounted. Cite what you find. It is all the more important, therefore, that the citation clearly point the reader to the intended version (see the links to URNs -- universal resource names -- for some possibilities about identifying specific versions of e-documents).

One word of warning: the most recent version of an e-document is not necessarily the most accurate or the most desirable version of a document. What version a scholar cites depends upon the questions that are being asked -- and answered.

Citation is necessary for an author's intellectual honesty and for a reader's safety -- "caveat lector" (let the reader beware) is an apt motto -- and an apt warning.