Thirty-eight dishonest tricks which are commonly used in argument, with the methods
of overcoming them
This is taken from "Straight and crooked thinking" by Robert H. Thouless,
Pan Books, ISBN 0 330 24127 3, copyright 1930, 1953 and 1974.
In most textbooks of logic there is to be found a list of "fallacies", classified in
accordance with the logical principles they violate. Such collections are interesting and
important, and it is to be hoped that any readers who wish to go more deeply into the
principles of logical thought will turn to these works. The present list is, however,
something quite different. Its aim is practical and not theoretical. It is intended to be a
list which can be conveniently used for detecting dishonest modes of thought which we
shall actually meet in arguments and speeches. Sometimes more than one of the tricks
mentioned would be classified by the logician under one heading, some he would omit
altogether, while others that he would put in are not to be found here. Practical
convenience and practical importance are the criteria I have used in this list.
If we have a plague of flies in the house we buy fly-papers and not a treatise on the
zoological classification of Musca domestica. This implies no sort of disrespect for
zoologists; or for the value of their work as a first step in the effective control of flies.
The present book bears to the treatises of logicians the relationship of fly-paper to
Other books have been concerned with the appraisal of the whole of an argumentative
passage without such analysis into sound and unsound parts as I have attempted.
Undoubtedly it is also important to be able to say of an argued case whether it has or
has not been established by the arguments brought forward. Mere detection of
crooked elements in the argument is not sufficient to settle this question since a good
argumentative case may be disfigured by crooked arguments. The study of crooked
thinking is, however, an essential preliminary to this problem of judging the soundness
of an argued case. It is only when we have cleared away the emotional thinking, the
selected instances, the inappropriate analogies, etc, that we can see clearly the
underlying case and make a sound judgement as to whether it is right or wrong.
The thirty-eight dishonest tricks of argument described in the present book are the
(1) The use of emotionally toned words (pp 10-25)
Dealt with by translating the statement into words emotionally neutral
(2) Making a statement in which "all" is implied but "some" is true (pp 27-38)
Dealt with by putting the word "all" into the statement and showing that it is then false.
(3) Proof by selected instances (pp 32-37)
Dealt with dishonestly by selecting instances opposing your opponent's contention or
honestly by pointing out the true form of the proof (as a statistical problem in
association) and either supplying the required numerical facts or pointing out that your
opponent has not got them.
(4) Extension of an opponent's proposition by contradiction or by misrepresentation of
it (pp 39-43)
Dealt with by stating again the more moderate position which is being defended.
(5) Evasion of a sound refutation of an argument by the use of a sophistical formula
Dealt with by analysis of the formula and demonstration of its unsoundness.
(6) Diversion to another question, to a side issue, or by irrelevant objection (pp 44-48)
Dealt with by refusing to be diverted from the original question, but stating again the
real question at issue.
(7) Proof by inconsequent argument (pp 49-50)
Dealt with by asking that the connection between the proposition and the alleged proof
may be explained, even though the request for explanation may be attributed to
ignorance or lack of logical insight on the part of the person making it.
(8) The argument that we should not make efforts against X which is admittedly evil
because there is a worse evil Y against which our efforts should be directed (pp 50-52)
Dealt with by pointing out that this is a reason for making efforts to abolish Y, but no
reason for not also making efforts to get rid of X.
(9) The recommendation of a position because it is a mean between two extremes (pp
Dealt with by denying the usefulness of the principle as a method of discovering the
truth. In practice, this can most easily be done by showing that our own view also can
be represented as a mean between two extremes.
(10) Pointing out the logical correctness of the form of an argument whose premisses
contain doubtful or untrue statements of fact (p 58)
Dealt with by refusing to discuss the logic of the argument but pointing out the defects
of its presentations of alleged fact.
(11) The use of an argument of logically unsound form (pp 58-64)
Since the unsoundness of such arguments can be easily seen when the form of the
argument is clearly displayed, an opponent who does this can be dealt with by making
such a simple statement of his argument that its unsoundness is apparent. For one's
own satisfaction when reading an argument of doubtful soundness, it will often be
found useful to make a diagram.
(12) Argument in a circle (p 64)
(13) Begging the question (pp 65-66)
Both 12 and 13 can be dealt with in the same way as 11; by restating your opponent's
argument in such a simple way that the nature of the device used must be clear to
(14) Discussing a verbal proposition as if it were a factual one, or failing to disentangle
the verbal and factual elements in a proposition that is partly both (pp 67-77)
This is really an incompetent rather than a dishonest way of arguing.
The remedy is to point out how much of the question at issue
is a difference in the use of words and how much (if at all) it
is a difference as to fact or values.
(15) Putting forward a tautology (such as that too much of the thing attacked is bad)
as if it were a factual judgement (pp 71-72)
Dealt with by pointing out that the statement is necessarily true from its verbal form.
(16) The use of a speculative argument (pp 78-83)
Rebutted by pointing out that what is cannot be inferred from what ought to be or
from what the speaker feels must be.
(17) Change in the meaning of a term during the course of an argument (pp 88-94)
Dealt with by getting the term defined or by substituting an equivalent form of words
at one of the points where the term in question is used and seeing whether the use of
this form of words will make true the other statements in which this term is used.
(18) The use of a dilemma which ignores a continuous series of possibilities between
the two extremes presented (pp 103-105)
Dealt with by refusing to accept either alternative, but pointing to the fact of the
continuity which the person using the argument has ignored. Since this is likely to
appear over-subtle to an opponent using the argument, it may be strengthened by
pointing out that the argument is the same as saying, "Is this paper black or white?"
when it is, in fact, a shade of grey.
(19) The use of the fact of continuity between them to throw doubt on a real difference
between two things (the "argument of the beard") (pp 105-108)
Dealt with by pointing out that the difference is nevertheless real. This again may be
made stronger by pointing out that application of the same method of argument would
deny the difference between "black" and "white" or between "hot" and "cold".
(20) Illegitimate use of or demand for definition (p 109)
If an opponent uses definitions to produce clear-cut conceptions for facts which are
not clear-cut, it is necessary to point out to him how much more complicated facts are
in reality than in his thought. If he tries to drive you to define for the same purpose,
the remedy is to refuse formal definition but to adopt some other method for making
your meaning clear.
(21) Suggestion by repeated affirmation (pp 111-114)
(22) Suggestion by use of a confident manner (pp 114-115)
(23) Suggestion by prestige (pp 115-118)
The best safeguard against all three of these tricks of suggestion is a theoretical
knowledge of suggestion, so that their use may be detected. All three devices lose
much of their effect if the audience see how the effect is being obtained, so merely
pointing out the fact that the speaker is trying to create conviction by repeated
assertion in a confident manner may be enough to make this device ineffective.
Ridicule is often used to undermine the confident manner, or any kind of criticism
which makes the speaker begin to grow angry or plaintive.
(24) Prestige by false credentials (pp 115-118)
The obvious remedy for this is, when practical, to expose the falsity of the titles,
degrees, etc, that are used. The prestige then collapses.
(25) Prestige by the use of pseudo-technical jargon (pp 116-118)
Best dealt with by asking in a modest manner that the speaker should explain himself
(26) Affectation of failure to understand backed by prestige (pp 118-119)
Dealt with by more than ample explanation.
(27) The use of questions drawing out damaging admissions (pp 199-120)
Dealt with by refusal to make the admissions. The difficulty of this refusal must be
overcome by any device reducing one's suggestibility to the questioner.
(28) The appeal to mere authority (pp 122-125)
Dealt with by considering whether the person supposed to have authority had a sound
reason for making the assertion which is attributed to him.
(29) Overcoming resistance to a doubtful proposition by a preliminary statement of a
few easily accepted ones (pp 128-130)
Knowledge of this trick and preparedness for it are the best safeguard against its
(30) Statement of a doubtful proposition in such a way that it fits in with the thought-
habits or the prejudices of the hearer (pp 133-135 and p 157)
A habit of questioning what appears obvious is the best safeguard against this trick. A
particular device of value against it is to restate a questionable proposition in a new
context in which one's thought-habits do not lead to its acceptance.
(31) The use of generally accepted formulae of predigested though as premisses in
argument (pp 161-166)
The best way of dealing with predigested thinking in argument is to point out good-
humouredly and with a backing of real evidence that matters are more complicated
than your opponent supposes.
(32) "There is much to be said on both sides, so no decision can be made either way",
or any other formula leading to the attitude of academic detachment (pp 166-167)
Dealt with by pointing out that taking no action has practical consequences no less real
than those which result from acting on either of the propositions in dispute, and that
this is no more likely than any other to be the right solution of the difficulty.
(33) Argument by mere analogy (pp 169-178)
Dealt with by examining the alleged analogy in detail and pointing out where it breaks
(34) Argument by forced analogy (pp 178-179)
The absurdity of a forced analogy can best be exposed by showing how many other
analogies supporting different conclusions might have been used.
(35) Angering an opponent in order that he may argue badly (pp 146-147)
Dealt with by refusing to get angry however annoying our opponent may be.
(36) Special pleading (pp 154-156)
Dealt with by applying one's opponent's special arguments to other propositions which
he is unwilling to admit.
(37) Commending or condemning a proposition because of its practical consequences
to the bearer (pp 157-158)
We can only become immune to the effect of this kind of appeal if we have formed a
habit of recognizing our own tendencies to be guided by our prejudices and by our
own self-interest, and of distrusting our judgement on questions in which we are
(38) Argument by attributing prejudices or motives to one's opponent (p 159)
Best dealt with by pointing out that other prejudices may equally well determine the
opposite view, and that, in any case, the question of why a person holds an opinion is
an entirely different question from that of whether the opinion is or is not true.
A similar list of graphical failures and how to correct them can be found
in Howard Wainer: Visual Revelations.
Rule 1: Show as little data as possible (minimize the data density)
Rule 2: Hide the data you do show (minimize the data/ink ratio)
Rule 3: Ignore the visual metaphor altogether
Rule 4: Only order matters
Rule 5: Graph data out of context
Rule 6: Change scales in mid-axis
Rule 7: Emphasize the trivial (ignore the important)
Rule 8: Jiggle the baseline
Rule 9: Alabama first!
Rule 10: Label: (a) illegibly, (b) incompletely, (c) incorrectly, and (d) ambiguously
Rule 11: More is murkier: (a) more decimal places and (b) more dimensions
Rule 12: If it has been done well in the past, think of a new way to do it