Atheist Talking Points:
Abductive reasoning is neither deductive nor inductive. It is most clearly illustrated by the nature of troubleshooting a problem, say a disease, or an electronic failure inside a computer. The conclusion is known. The cause is not known, but is inferred by the artificial creation of a “probable” series of events that, taken together as if they had occurred, could have produced such a result.
As can be seen, this would have considerable value in a trouble shooting situation, where a certain failure symptom can be known to be created by one of several possible previous chains of events. It might not be known which chain; but the narrowing of the unknowns to just a few chains of events is helpful in eliminating considerable effort by a technician. Such Abductive trouble shooting procedures are contained in many owners manuals that come with new products.
The same with medical trouble shooting, where certain disease symptoms are narrowed to a few possible chains of events or causes, by inferring causal properties based upon the final issue, the disease.
But how does Abductive logic fare in the determination of forensic and historical issues? What degree of accuracy and validation is to be expected?
In forensic analysis, much of the accuracy is obtained by the comparison of a questioned item to another item (or process) of a similar character. For example, the age of a log cabin can be accurately determined by the annular rings in the logs, by comparing them with comparable logs of known age, and counting the rings from the death of the cabin log, to the present time on the known log. The rings will show growth patterns that give era characteristics, such as wet years and dry years, allowing alignment of the two logs at the same time on the scale.
Such a forensic analysis is both common and acknowledged to be valid.
However, what of the following type of analysis, where no calibrated time scale is available? (This is an actual situation). A type of dinosaur remain is being sought along the bottom lands adjoining “ancient river flows” that still flow in the proximity. The group of analysts is brought to areas where the rock is the consistency of sand, being able to be brushed away without aid of any chipping, gouging or digging. When nests are found, the dinosaur eggs also can be brushed away without aid of chipping, gouging or digging. They are so fragile in fact that care must be taken not to destroy them with the soft brushes.
The senior analyst for the expedition insists that a) the rock is very hard because it is very old, 65 million years. Because the rock is 65 million years old, the dinosaur eggs must be fossilized, and also be very hard. [1]
Now in this case, the conclusion was determined well before any actual data was taken. The data to be taken did not include how hard either the rock or the eggs must be – they were assumed hard because the age was previously set.
So we now arrive at the embarrassing part. DNA was also found within the eggs that were supposedly fossilized. Still, the pre-established conclusion that the rocks and eggs were 55 mya was considered inviolable. No amount of valid information could derail the presupposition. Finally the condition of the rock and eggs was called a “new type of fossilization”. The results were withheld for a period of time, then released to publication in the journal, “Science”. [2]
The failure of this actual piece of scientific absurdity is couched in the Abductive reasoning that was applied to the rock before the expedition. This type of reasoning includes building a story to support the desired conclusion, and then to give more credence to the story than to the subsequent data that contradicted the story. Why would this happen? Certain fields of endeavor now have the firm belief that the established central story (meta-theory) is the foundation for rationality in the endeavor. Specifically, biology, at least paleobiology, requires that evolution be more than a theory – it must be True - in order for the observations of paleontologists to make sense. It is a requirement pure and simple. So any non-congruent input will not be allowed.
Abduction is so similar to Ad Hoc Ergo Propter Hoc as a fallacy of story telling, that it is difficult to separate a valid use of Abduction from a non-valid use. The difference is in the premises used within the story itself. In troubleshooting, the premises that are used to generate an inference of the cause are all true and valid premises.
In the Fallacy form of Abduction, the premises are fantasized, based on the desire for a chain of events to be true, regardless of the actuality. Fantasized premises are also the stuff of Ad Hoc Ergo Propter Hoc and “Just So Stories”, both of which are easy to rout.
Charles Darwin was the master Abductionist. His entire work was based on stories of adaptations and engineered breedings that seemed to him to be the same quality of evidence as that which is required for rational attainment of logical conclusions. It is still the case that paleobiology uses Abduction as justification for its extrapolated conclusions. Darwin built volumes of stories to support his theory. Biology still builds stories – but more sophisticated jargon based on the instantiation of DNA as a supporting truss. Such stories devolve to premises that are not, and likely cannot ever be, verified by any manner other than sub-stories and sub-sub-stories, in an infinite regression. Such a regression, not being based upon valid premises, necessarily cannot be halted at the baseline foundation of the First Principles.
The structure of a “science” based on layer after layer of fabricated stories is terrifyingly weak. Yet the structure persists, and so far is propped up against rupture by a seemingly never-ending supply of stories.
The historical and forensic sciences are vulnerable to the desire to build a case to support an observation. Historical and forensic endeavors cannot engage empirical verifications for their work. The observations are not necessarily repeatable or verifiable in any sense other that of a single instantiation: “an instance of (R) has been found”. The logically illegal extrapolation from the instantiation observation is (1) to create a story that imbues unobserved characteristics upon the observation; (2) to create a story that imbues unverifiable relationships between this instantiation and other instantiations; (3) to create a story that implies a coherence to a meta-theory that, in and of itself, is not verifiable.
What are we to say of specific science venues that proceed in the such a fashion? Where entire realms of endeavor are based on at least one of the Abduction Fallacies stated above? It can’t be too much to call it questionable. And it also can’t be too much to call it intellectually dishonest.
It certainly cannot be called truth. And it should not be called science.