With the recent rise of scientific fraud, researchers at Stanford University have begun looking at papers they know are fraudulent in a bid to figure out if they have a different vocabulary to clean ones.
In business there is an old saying, "If you don't understand what you're buying, check that you still have your wallet."
In short, fraud thrives in confusion.
According to the Journal of Language and Social Psychology researchers examined 253 papers that were retracted because they included fraudulent data, and then compared them with 253 papers that weren't.
They also looked at 62 papers that were retracted for other reasons (such as the research being unethical.)
What did they find?
Papers that include scientific fraud use more jargon, and are harder to understand than their clean counterparts.
It is unclear whether fraudsters do this on purpose, or if they are as bad at writing as they are at science.
That said scientific fraudsters also love using references - because if you stick enough in there tracking down what each reference says becomes a hassle.
Unfortunately this method of analysis is not suitable for finding fraudulent papers in general just yet.
According to Science Magazine David Markowitz, the lead author on the paper, told them they only managed to flag whether a paper was dishonest or not 57.2% of the time.
In short, more research needs to be done on fraudulent research in order to sharpen this method of winnowing out deliberate deception into something useable.
Just R20 for the first month. Support independent journalism by subscribing to our digital news package. 
