|
|
Table of Fallacies |
| Deductive Fallacies |
Inductive Fallacies |
Retroductive Fallacies |
Fallacies of Soundness
|
Fallacies that afflict the major premiss: |
Irrelevancies |
Errors in Observation In Induction the major premiss is a RESULT, i.e. a statement of the data or observations collected from a sample. If the result is false, then this statement does not accurately reflect what is actually true of the sample. Such errors are caused by experimental conditions that have a deleterious effect on the data collected, e.g. faulty instruments or faulty data-recording. |
False Cause In Retroduction the major premiss is a RULE, i.e. a statement of a known (or presumed) causal regularity. If the rule is false, then this regularity does not actually obtain. Hence, if the major premiss is false, the argument appeals to a causal mechanism that does not operate in reality (or, a least, is generally considered far-fetched and unbelievable.) |
|
Fallacies that afflict the minor premiss. |
Misrepresentations In Deduction the minor premiss is a CASE, i.e. an instance allegedly subsumed under a broader class. If the case is false, then the instance being considered does not really fall into the class it is alleged to fall into. That is, the facts of the matter have been misstated. Hence, a false minor premiss is a misrepresentation of the facts. |
Errors in Sampling. In Induction the minor premiss is a CASE, i.e. a sample chosen to represent a broader class. If the case is false, then the sample fails to represent the broader class that it was chosen to represent. Hence, if the minor premiss is false, however accurate or unbiased our observations may be, the data collected may not be reflected in the broader class. |
False Report In Retroduction the minor premiss is a RESULT, i.e. a statement of an observation of presumed fact. If the result is false, then the "fact" allegedly observed is bogus. Perhaps the observation was inaccurate, or perhaps it was misinterpreted. In any case, if the minor premiss is false, we find ourselves in the position of trying to "explain" something that never happened. |
| Fallacies that afflict either the major or the minor premiss, but we can't tell which. |
Ambiguities do not seem to take distinctive Deductive, Inductive, and Retroductive
forms. The fallacies can go by the same names, regardless of the kind of reasoning in
which they appear. |
Fallacies of Circularity
| Circular Justification Deductive circularities offer as justification precisely what we are trying to justify. |
Circular Evidence Inductive circularities involve the selection of certain evidence when we already know what the "evidence" will show. |
Circular Explanation Retroductive circularities appeal to a mechanism of explanation that is itself in need of explanation. |
ON THE NAMING OF FALLACIES
Several of the fallacies on this website do not appear on other lists for the simple reason that I named them myself. In some cases I was attempting to name a fallacy that had been described (but left unnamed) by other writers; in other cases I was attempting to describe and label a fallacy that I had observed in use, but could not fit into an existing slot.
I think people who study fallacies should be allowed to add new fallacies to our collective compendium, and to invent new names for the fallacies so introduced. After all, if no one took responsibility for observing, describing and labeling new fallacies, our knowledge of the subject could never progress. However, the listing of new fallacies and the creation of new fallacy names must be done cautiously and responsibly. I think at least three rules should be followed:
(1) Obviously, no one should invent a new name for a fallacy that already has a name. That just leads to confusion. (In several cases I have given a name to a fallacy only to discovered that it already had a more traditional name. In such cases I have generally adopted the older name as soon as I became aware of it. One exception was "secundum quid," which is clearly the oldest name for the fallacy that I call "destroying the rule." However, in this case the Latin phrase means nothing to modern students, and merely tends to confuse them. For purposes of teaching the subject, I prefer my own name for this fallacy.)
(2) New fallacies should be documented with actual instances. Like a lexicographer adding new words to a dictionary, a student of fallacies should attempt to describe current and past rhetorical practice, not merely construct artificial examples to fit the holes in an abstract conceptual scheme. Commercials, political speeches, bumper stickers, editorials, letters to editors and conversations with ones friends provide ample fodder for field research.
(3) Every effort should be made to fit examples from actual practice into the old categories, if this is possible. Rather than creating new fallacies, it may even be preferable to broaden or modify our understanding of old fallacies to accommodate the new examples. Only when such broadening would create serious confusion is it preferable to introduce a whole new fallacy category.
I have tried to follow these rules, but perhaps I have not always done so successfully. A rigorous application of these rules would require an intimate familiarity with all the past and present literature on fallacies. I have consulted a few sources, but my familiarity with the relevant literature is far from comprehensive. Much of what I now know was learned through email correspondence with people who have visited this page.