What’s Wrong With What’s Wrong?
Title: Fallacies
Author: Charles L. Hamblin
Publisher: Vale Press
Year: 1983 [first edition 1970]
Pages: 347
“What is a fallacy?” is one of those questions which history answered and … did not answer all at the same time. Two millennia of thought made them print thousands of dictionaries with a resolute definition of fallacy as follows (I picked up a random one):
A. an incorrect or misleading notion or opinion based on inaccurate facts or invalid reasoning [1]
All descriptions as such are a sort of tanaaa! when trying to come up with a definition of fallacy. [This definition in particular might just be the all-time champion of tanaaa!-ing, since it has more disjunctions than a restaurant menu. One might notice from the start that it embodies two seriously different notions: A1 inaccurate fact and A2 invalid reasoning]. Now, although I’m not saying that everyone who uses the term refers to this general meaning, it might be that, above all other definitions, laypersons make quick mental connection with this sort. I will not say upfront what is wrong with this definition, but you might have sensed it from the unveiled biased tone that some major miscues must have happened on the way to this generally accepted shot.
The purpose of the following set of texts is to arrive at some prevailing outline of theoretical apparatus with which to (1) refine the search for such “things” – I’ll stick to this naming until further elucidation–, (2) to appraise the major bugs regarding the import of the above mentioned definition into Argumentation Theory and (3) to follow some general rules of reorganizing the whole deal. In this post, we will be chiefly concerned with point 1.
N.B. I will henceforth refer to fallacies in hyperlink text when the general definition is concerned. These texts assume a common understanding of the subject. However, when you feel the meaning of the fallacy is somewhat slippery or uneasy, click on the link and spend some time together.
So,
Charles Leonard Hamblin (1922-1985) was a spooky Australian pioneer computer scientist and a professor at Technical University of New South Wales, Sydney. According to Wikipedia, he had some major contributions in that rigorously calculating … affair of his (the Reverse Polish Notation and the stack in computing are allegedly his contributions – have no idea what they mean). Oh, and another thing: he happen to write the most influential book on Fallacies in the 20th Century. Guess how he named it? He named it Fallacies – bet he had no copyright problems with this title[2] – and decided to make of it a ground-breaking, forward-looking intellectual endeavour. Well good for you, Mr. Hamblin, good for you!
Fallacies appeared in 1970 and came as a natural extension of what started with Stephen Toulmin (The Uses of Argument) and Chaïm Perelman (La Nouvelle Rhetorique) as a major reassessment of logic. Hamblin’s work resembles them in many respects (philosophical attitude towards symbolic logic and consistency on treating the field undertaken) and together they form the corpus of a later discipline called Informal Logic. We are not interested – at least for the moment – in these cultural issues, so moving on.
In this post, I will tackle the first chapter of the book (it’s a fairly big book, so I’m dividing it) called The Standard Treatment. Next posts will be significantly shorter, I promise.
“The Standard Treatment” (TST) as a concept is a rather disputed one (see Johnson 1990, 1991, Walton, 1991) but it is very useful for the first grasp on the subject of fallacies. It refers to the account on fallacies given at first by Aristotle on Sophistical Refutations, Topics, and Prior Analytics, account to which many scholars up to late 18th Century subscribed cheerfully and, most important, uncritically. The classifications and examples given in the textbooks following Aristotle might have varied sometimes from author to author, but in general “tradition has repeatedly proved too strong for its dissentients” (p.10). It wasn’t until the 16th century that some scholars felt the need to reject the foundations of TST and even the properness of such a subject in logic textbooks (e.g. Peter Ramus). And not until more recent times that authors formulated serious critical claims against it, like the one that out of fallaciousness one can not even make a theory at all, since it is impossible to mentally grasp erroneousness (e.g. De Morgan in the 19th century and Cohen and Nagel in the 20th). Be that as it may, Hamblin claims that we can bring together all these authors stemming from Aristotle into something we can shape as The Standard Treatment.
Even with its “dogmatic”, “worn-out”, and “debased”[3] manner, TST still has a more precise definition of fallacies than the one we first fished out of the dictionary:
B. A fallacious argument, as almost every account from Aristotle onwards tells you, is one that seems valid but is not (p.12)
Without any further explanation, the passage from A to B seems a big one and we could pin down major differences: (1) we are now taking into consideration arguments, which means that a false claim by itself cannot be a fallacy unless it is embedded into an argumentative context (note our subtle reference to gogreen18)[4], (2) we are now speaking about validity, and not inaccurate beliefs or facts, (3) a fallacy must bear some resemblance with what a valid argument actually is.
Hamblin seems to arrange TST into five (overlapping and imprecise – but hey, that’s the case to be made!) categories: fallacies inside language, fallacies outside language, ‘ad’ fallacies, formal fallacies and fallacies of scientific method. Since Hamblin’s main task is to criticize these categories, one should remember that this classification of TST is one that only serves the purpose of temporary organizing the history of fallacies.
We will now proceed in this manner: I will only write down the name of some fallacies and alongside the criticism that Hamblin made to this or that category. If you know what the general meaning of the name of the fallacy is, you can just read the criticism. If you have never heard of such puzzling items, you should follow the link attached to each one of them. It must be said that the link I’m providing may or may not correspond to what Hamblin gives as an example of TST, but I promise I will do my best so that the similarities will be greater than the differences. That being said, the main problems about the main fallacies are:
[just to make sure: what is after the ‘–’, is ‘Hamblin’s critical comment on’, not the ‘explanation of’ the fallacy; explanations come with examples in the link attached to the name]
A. Fallacies inside language (in dictione)
1. equivocation – Here, the problem is that “We are hardly capable of being deceived by any serious chain of reasoning exploiting the double-meaning in statements” (p.15), Hamblin was referring to examples of equivocation such as “Everything that runs has feet, The river runs, therefore, the river has feet”
2. amphiboly – How is it to be understood? Is ‘2x4+3’ an amphiboly (as De Morgan claimed)? Is “Save Soap and Waste Paper” (example cited by Irving Copi) even an argument? Since “puns and anecdotes do not take the place of logical analysis” (p.18), therefore, to function after the standard definition, such an argument should be that “truth in one sense makes us believe the truth in the other sense” and for Hamblin such an example is unlikely to be found.
3. composition and division – The problem here seems to be that, although the classic examples are conclusive and easy to spot, “in some cases, an argument of this form seems to be perfectly valid; thus ‘All the parts of this machine are made of metal, therefore this machine is made of metal’” (p. 21). In other words, there’s nothing wrong with the actual scheme, is the topic (the type of collection we are talking about) that renders the argument valid or not.
4. accent, figure of speech – these fallacies were lost along the way in the long history of TST. The fallacy of Accent first referred to being deceived by the misleading pronunciation of some words (Hamblin claims that this is unlikely to happen into an argumentative speech), and then to misleading emphasis inside a phrase (which makes it quite similar to amphiboly). Figure of speech referred to being deceived by the misleading etymology of a word (see endnote[5]).
B. Fallacies outside language (extra dictionem)
5. accident (or sweeping generalization) – The first definition of this fallacy was highly unusable (it was based on the elusive distinction between essential and accidental properties) but I. Copi (p.63) restated this fallacy as “applying a general rule to a particular case whose ‘accidental’ circumstances render de rule inapplicable” and this seemed decisive (see other examples here). Hamblin advises caution when using the word “general” here, since it is meant to refer only to something that “holds most of the time”, not always.
6. secundum quid (or hasty generalization) – It is a common mistake in argumentation that we move “invalidly” from a few samples to the whole class (without necessary qualifications), and it is just as common to be super-accurate and say that we do not actually know for sure the sun will rise tomorrow as it did yesterday. Nevertheless, before accusing someone of committing this fallacy, Hamblin’s criticism has to be solved: (1) if David Hume is right when saying that every argument from particular cases to a general rule must be fallacious, then inductive reasoning is synonym to fallacy (2a)“most or all generalizations are deductively unjustified; but we have a need for other criteria which will enable us to distinguish between the too-hasty ones and the relatively reliable” (p.29) (2b) “if an induction is based on ‘fewer than all instances’, it may lead us into error, but it also may not” (p.49) and later (2c) “we accept in principle that some inductive arguments are better than others, but by which cannons do we judge?” (p. 225)
7. ignoratio elenchi (or ignorance of refutation[6]) – There is nothing wrong with this fallacy in its meaning. It represents a legitimate accusation to put forward when someone is short of relevance. The problem with this legitimacy is that is too wide: “so described, this category can be stretched to cover virtually every kind of fallacy” (p. 31). Aristotle himself was ‘struck’ by this problem: “It strikes Aristotle, however, that if his definition of ‘refutation’ is really complete, he can regard all fallacies as due to failure, in some respect or other, to conform to it (167a 17-196a, 22)” (p. 87)
8. petitio principii (or begging the question) – I remember that the formula ‘begging the question’ was puzzling for me and that Hamblin’s ‘translation’ helped, so note that ‘begging the question’ should be read as ‘beg for that which is in the question-at-issue’ and even more clear “asking to be granted the question at issue, which one has set out to prove”. This is one of the major bones of contention between logic and … well, everyone else. We could extract the major points from Hamblin (1) “there is nothing wrong with inferring something from itself” (p.33), which means that what Formal Logic has to say about “p therefore p” or “p because p” is that they are both valid. (2) circular reasoning could come in many forms, none of which render the argument invalid (3) J. S. Mill claimed that all valid reasoning commits this fallacy, since, if we reason from ‘Socrates is a man’ to ‘Socrates is mortal’ using the premise ‘All men are mortal’, the conclusion – the fact that Socrates is mortal, is begged in the premise ‘All men are mortal’. I am afraid the issue is rather complex and even sketching it here would mean oversimplifying. We will return to it in due course.
9. many questions (or complex question) – “this having been said, it remains to ask what relevance these examples have in a list of fallacies. A fallacy, we must repeat, is an invalid argument; and a man who asks a misleading question can hardly be said to have argued, validly or invalidly, for anything at all” (p.39)
10. post hoc (or post hoc ergo propter hoc, or false cause) – This fallacy also is quite common especially when arguing about reasons (when we try to establish why someone did certain things or acted in a certain way), but consider Hamblin’s remark: (1) “Even so, this is puzzling. If we know that B always occurs after A we are well on the way to setting up a causal law” (p.37) (2) precise differences between necessary conditions and constant conjunctions is not actually an easy catch and (3) we already talked faulty, hasty and sweeping generalizations, are we now classifying ‘even hastier generalizations’?
C. ‘ad’ Fallacies
11. ad hominem, ad verecundiam, ad misericordiam, ad ignorantioam, ad populum, ad baculum, – you will find them all in English version everywhere as ‘appeal to’ except for the ad hominem, which is probably the only one that could not fit into this category (or could be crammed in artificially like ‘appeal to personal attack’). Although they sound so ‘latinish’, one should note that the genre was first invented by John Locke’s Essay Concerning Human Understanding in the 17th century – although ad hominem could be traced back to Aristotle (see Walton, 2004). The chief matters which might concern us are: (1) they are actually all peculiar versions of ignoratio elenchi – since they in some way evade the question at issue, (2) they change their ‘degree’ of fallaciousness based on context – this problem is close to the one regarding hasty generalization –, (3) turning them into deductive syllogisms comes in handy (X is an authority on facts of type T, X said S – which is a fact of type T, Therefore, S is true) but it is actually worse, since they are not meant to function as deductively valid. We will deal with these intimately in our following texts.
D. Formal Fallacies
12. The idea of classifying formal fallacies (you should check out this link with theoretical cautiousness) is also quite new, although affirming the consequent or the fallacy of undistributed middle stem from Aristotle’s Prior Analytics. Hamblin takes it that Whatley was the first to borrow the idea from Aldrich and propose this new group. Difficulties arising from the classification are numerous and of high theoretical importance (they virtually represent the rupture from symbolic logic). We will discuss them in detail when arriving at chapter 6 (‘Formal fallacies’) and 7 (‘The concept of Argument’). For now, note that this could be the most problematic chapters of Fallacies, as a field of study.
E. Miscellaneous
13. Oh, this is the best part even if Hamblin did not pay much attention to it. Today, Internet is full of crazy names which, although define fallacies that could easily fall under another category, refer closely to dubious nonachievements in argumentation. Sometimes, the actual problem is that the nomination has nothing to do with what is actually fallacious within such an argument. I am not saying that there is something entirely wrong with finding stage names for every conceivable rule breakage, it’s just that it seems something of an overdo. I have a personal top 5:
5. Appeal to Novelty
4. Reductio ad Hitlerum (argumentum ad nazium)
3. If-by-whiskey Fallacy
2. The Texas Sharpshooter Fallacy
1. Argument by Prestigeous Jargon
So,
as a temporary conclusion, what should remember is that:
(1) TST is flawed in many respects and should be dealt with carefully
(2) Logic has to have some uncovered fields on which, at least with classic tools, it cannot operate
(3) By means of logic or of substitutes, a distinction must be made between deductive and inductive criteria (resolve secundum quid problem)
(4) By means of logic or of substitutes, a distinction must be made between inferring and arguing (resolve petitio principii problem)
(5) Baptizing fallacies is not wrong in itself, it’s just that it is made with tools from TST and nobody seems to have a standard criterion[7]
(6) [this has not been argued here but we will refer to it in due course] – It might be the case that Hamblin’s treatment of The Standard Treatment is rather abusive, and that “we have reasons to think that the textbook tradition is not nearly so monolithic or dogmatic as Hamblin’s Coinage might suggest” (Johnson, 1990, p.165)
[1] Collins English Dictionary – Complete and Unabridged 6th Edition 2003. © William Collins Sons & Co. Ltd 1979, 1986 © HarperCollins Publishers 1991, 1994, 1998, 2000, 2003
[2] he insists on remarking that “Strangely, in a certain sense, there has never yet been a book on fallacies; never, that is, a book-length study of the subject”, p. 10
[3] the cataclysmic passage sounds like this: “And what we find in most cases, I think it should be admitted, is as debased, worn-out and dogmatic a treatment as could be imagined – incredibly tradition-bound, yet lacking in logic and in historical sense alike”, p. 12
[4] her actual introductory passage sounded like this: “the important thing to remember with claims is you wanna make sure that in order to be considered legitimate and valid, your claim has to be sound, it must be a sound claim; and the way that we judge that is by making sure that it doesn’t have any fallacies”. This passage is defective on so many levels that I cannot afford to explain it here. However, one should bear in mind that (1) validity and soundness normally have nothing to do with claims on their own and (2) fallaciousness is not the same thing as falsehood (3) it is actually the other way around, namely, “in order (for your argument) to be considered sound, it must be considered legitimate and valid”, although I am not very sure what does she mean by legitimate. I am guessing that gogreen18 means “argument” every time she uses the word “claim”, but this doesn’t account for the wide of the mark definition of fallaciousness.
[5] Hamblin quotes J. S. Mill, who fell into this trap glamorously: “The only proof capable of being given that an object is visible is that people actually see it. The only proof that a sound is audible is that people hear it: and so of the other sources of our experience. In like manner, I apprehend, the sole evidence it is possible to produce that anything is desirable, is that people do actually desire it”. Hamblin commented that “to say something is visible or audible, is to say that people can see or hear it, whereas to say that something is desirable is to say that it is worthy of desire or, plainly, a good thing”. (p.26)
[6] you may also find: misconception of refutation, red herring, smoke screen, wild goose chase, irrelevant thesis
[7] Hamblin notes that, since we have a fallacy of four terms, “there is nothing to stop us from having a fallacy of Five Terms, or even a Fallacy of Six terms” (p. 45) and later on that, if we take syllogistic laws as criteria for baptizing fallacies, we might end up with names such as Fallacy of Undistributed Middle with Illicit Process of the Minor Term and an Affirmative Conclusion with One Negative Premise: “a syllogism can brake more than one rule at once; and, with a little ingenuity, we can succeed in breaking all” (p. 200)
______________________________________________________
BIBLIOGRAPHY
HAMBLIN, Charles Leonard, “Fallacies”, Vale Press, 1983, Newport News, VA
JOHNSON, Ralph H., “Hamblin on The Standard Treatment”, in Philosophy and Rhetoric, Vol. 23, No. 3, 1990. Pennsylvania State University Press, University Park and London
JOHNSON, Ralph H., “In Response To Walton”, Philosophy and Rhetoric, Vol. 24, No. 4, 1991, The Pennsylvania State University, University Park PA
WALTON, Douglas, “Hamblin on The Standard Treatment of Fallacies”, Philosophy and Rhetoric, Vol. 24, No. 4, 1991, State University, University Park PA
WALTON, Douglas, “Argumentation Schemes and Historical Origins of the Circumstantial Ad Hominem Argument”, Argumentation, 18: 359–368, 2004, Kluwer Academic Publishers, Netherlands.
