A quite loose, quite idiosyncratic foray into the nether murk of human behavior by a journalist and anthropologist who first, appropriately, ask the question, what is human nature? -- as opposed to, say, avocado nature or chimpanzee nature or computer nature. Guess again -- it's really not as easy as it sounds. Give up? All right, the distinguishing feature of human nature, according to Cannel and Macklin, is man's ability to organize observed human and natural data (""the world rubble"") into meaningful information which allows us ""to get through the day, the week, the year and the lifetime."" That is to say, ""It is only human nature that has to make information of its own nature, too, in order to keep chaos from encroaching."" The next question -- the central thrust of the inquiry -- is how is the world rubble processed? How does data become information? Give up? OK, according to Cannel and Macklin, there are several models of human nature through which the data -- related through ""stories"" -- are filtered; three such models -- the mechanistic (Edisonthropus Tinkerectus), the individualistic (Tarzanthropus Inherens), and the transactional (Morganthropus Exchangenesis) -- are hypothecated and explicated at length: the ""story"" of a woman walking down Broadway naked will be processed differently through each model. Depending on the model, we learn ""where everything belongs, and how and when and why it belongs there."" That is to say, ""Once you are into the story, you can tell what is order and what is chaos."" This sort of speculation is dandy -- big questions bounce around like variables at a sociology conference -- but The Human Nature Industry does indicate why they call the behavioral sciences soft.