Abduction

Abduction is the least familiar mode of reasoning and the mode that was systematized most recently. In conjunction with deduction and induction, abduction is used to make logical inferences about the world. Furthermore, abduction offers great promise as a potential primary mode of reasoning for qualitative research.

The nature of abduction was first systematized during the late 19th century by the American philosopher and logician Charles Peirce. His form of an abductive inference is as follows:

• Some event, X, is surprising to us.
• But if some explanation, Y, were in place, then X would be ordinary.
• Therefore, it is plausible that X is actually a case of Y.

Another way to look at this inference is to suggest that it is, in fact, reasoning toward meaning. This means that abductive inferences are valid in different ways from the other two modes of reasoning. Whereas deductive inferences are certain (so long as their premises are true) and inductive inferences are probable, abductive inferences are merely plausible. Therefore, abductive inferences are weaker by nature than the other two sorts of inferences.

Even though abductive inferences are weaker, they can be extremely useful. When we reason to meaning, we are expanding the realm of plausible explanations. We are giving ourselves a chance to see things that we might otherwise miss by staying with tried-and-true explanations. This is what Russell Hanson called the “logic of discovery.”

Although there has been relatively little work done with using abduction, some of it has been quite fruitful. For instance, Gilbert Harman characterized abduction as “reasoning to the best explanation.” This notion of abduction has been incorporated into both expert systems research and artificial intelligence research. In addition, abduction has played an important role in semiotics.

Other work in abductive theory and practice has likened abductive researchers to detectives. In particular, Sherlock Holmes has been identified as an abductive thinker par excellence. That is, his so-called deductions are, in fact, abductions. In a famous case, Holmes infers, from the surprising fact that the watch-dog did not bark, the abductive conclusion that the dog knew the kidnapper. Umberto Eco has looked at this aspect of abduction in both his theoretical work and his novels.

Following the work of Peirce, Gary Shank has looked at the application of abductive reasoning directly to qualitative research. In this work, Shank argued that there are actually six modes of abductive inferences that all researchers use. These types of inferences lead to hunches, omens, clues, metaphors, patterns, and explanations. In fact, Shank went on to argue that abduction is actually the ground state, or default mode, of cognition in general. Furthermore, by using the formal structures of abduction per se, these six modes of inferences can be related to each other systematically. In this fashion, the power of abduction as a way to reason to meaning can be employed by [Page 2]qualitative research, which is the systematic empirical inquiry to meaning.