Quote mining is the alteration of the meaning of a person’s quote by taking it out of context or removing sections of the quote. It’s a disingenuous way to make it look like something someone said supports your position when, in fact, it does not.
Quote mining is easy to do if you’re creative about it. Consider the following passage:
“One day when we were kids, Charlie spent the afternoon stepping on ants in the driveway. He came into the house that evening with a guilty expression on his face. When his mother asked what was wrong, he admitted, ‘I’m a horrible person. I committed murder today! Poor ants.’”
In context, this passage is about an innocent child coming to terms with the world beyond him.
Suppose now that you wanted to make Charlie seem more nefarious. By cutting out just a few choice sections, you could twist the meaning:
"He [Charlie] came into the house that evening with a guilty expression on his face. When his mother asked what was wrong, he admitted, ‘I’m a horrible person. I committed murder today!'"
Technically, the new quote is correct - the words are the same as the original. But notice how choosing to remove certain contextualizing phrases has corrupted our understanding of Charlie as a "murderer."
Academic creationists have become notorious for quote mining - so much so that the term came into popular language among scientists in the 1990's to describe how their own quotes were being dishonestly used to suggest that they had doubts about the validity of evolutionary science.
In 1996, biochemist and high-profile creationist Michael Behe published a popular book titled Darwin's Black Box, which argued among other things that evolution could not account for certain biological structures, such as the bacterial flagellum. These parts, he claimed, were "irreducibly complex" - that is, if you removed any one piece, the whole thing would cease to function (a presumption promptly debunked by other biologists).
Behe tried to shore up his case by including numerous quotes in his book from evolutionary scientists that seemed to show how shaky the science of evolution was. On page 29, he quoted from a paper co-authored by Professor of Evolutionary Biology Jerry Coyne:
"We conclude--unexpectedly--that there is little evidence for the neo-Darwinian view: its theoretical foundations and the experimental evidence supporting it are weak."
Coyne responded quickly. In the February 1997 issue of the Boston Review, he wrote, "I went back to see exactly what Orr [Coyne's co-author] and I had written. It turns out that, in the middle of our sentence, Behe found a period that wasn't there."
What the paper had originally said was: "Although a few biologists have suggested an evolutionary role for mutations or large effect (Gould 1980; Maynard Smith 1983: Gottlieb, 1984; Turner, 1985), the neo-Darwinian view has largely triumphed, and the genetic basis of adaptation now receives little attention. Indeed, the question is considered so dead that few may know the evidence responsible for its demise.
"Here we review this evidence," the paper continued. "We conclude--unexpectedly--that there is little evidence for the neo-Darwinian view: its theoretical foundations and the experimental evidence supporting it are weak, and there is no doubt that mutations of large effect are sometimes important in adaptation."
Coyne added, "By inserting the period (and removing the sentence from its neighbors), Behe has twisted our meaning. Our discussion of one aspect of Darwinism--the relative size of adaptive mutations--has suddenly become a critique of the entire Darwinian enterprise. This is not sloppy scholarship, but deliberate distortion."
Quote mining is effective because few people ever go back to read the original sources. It's easy to do, since all quotes require selecting certain sentences to keep and others to omit. The goal, however, should always be accuracy. Mining for quotes does disservice to the reader, misrepresents the person being quoted, and makes the quote miner look more like a dishonest ditch digger.