Arguments 3: Fallacies
- Fallacies are forms of bad reasoning that occur so often
that we should take note of them and name them. Typically fallacies
are so common because they hold some sway over us: we find them
convincing steps in arguments when we should not.
- Most fallacies fall into one of three general forms:
misdirection and equivocation, non sequiturs, and confusion
- Most fallacies are bad reasoning because as steps in
arguments they do not ensure the preservation of truth. That
is, one can reason from true premises to a false conclusion
using these fallacies. This is true of non sequiturs and of
confusions with sourcing.
- Some fallacies, however, form good arguments but the
arguments are deceiving because they appear to be something
other than they are -- this is the case with misdirection and
- Non sequitur is a Latin phrase roughly meaning
"does not follow." These are the fallacies which appear to
lead to good steps in reasoning but in fact do not preserve
- Confusions with Sourcing occurs when people
confuse issues about the evaluation of sources with the
evaluation of claims.
- Misdirection and Equivocation occurs when
definitions are used which shift their meaning at some point
in the argument, or when there are premises which are false or
unwarranted but for reasons that are often overlooked.
- We will consider each of these kinds of fallacies in turn.
Misdirection and Equivocation
- Ad hominem Meaning literally "at the man," an ad
hominem is when one attacks a person instead of a premise or
inference. Attacks on a person may be relevant when we are
evaluating a source, but they are not relevant when we are
evaluating an argument (and, in fact, rarely is an ad hominem in
its more common instances relevant to source evaluation either,
since it is typically just an insult). Here is an example:
Smith: We need to oppose the ban on assualt weapons
because these weapons are not well defined and as a
result other kinds of guns could end up banned. This
would infringe on our second amendment rights.
The thing at issue here is whether we should oppose a
law. If Jones's response is meant to give us reason to
doubt Smith's conclusion, it has failed. It merely
insults Smith, and we have no additional reason to believe
or doubt Smith's conclusion or the validity of Smith's
Jones: That's just the kind of reasoning I expect from
a right wing gun nut.
- Tu quoque This "you too" fallacy is a kind of ad
hominem. It is to deny a conclusion based on the idea that the
speaker or some other relevant group is doing something
inconsistent with the claim.
Smith says that lawns do damage to the environment. But
I don't believe that. Look at Smith's house! He has a
huge green lawn.
If the implicit argument here is that it is wrong to conclude
that lawns are a source of environmental harm because the
person claiming this is inconsistent, then this claim is
clearly irrelevant. Smith may simply fail to follow his own
advice. We need then to see Smith's argument, and not this
tu quoque fallacy, to evaluate the conclusion.
- Red Herring This is to through something
completely irrelevant into the argument -- typically something
which generates heated emotions -- in order to lead the
argument astray. An example:
President Clinton says that we should pay off the
debt instead of reducing taxes, on the grounds that
this is better for the economy. I wonder what Monica
Lewinsky would have to say.
Ms. Lewinsky's opinion has no role in the argument that
is presumably being developed here, and is meant merely
to inflame and divert us.
- Appeal to Ignorance In this for of reasoning, one
assumes that lack of evidence is a kind of proof of a claim.
It is very common in conspiracy theories. Here, for example,
slightly emended and altered, is an argument for the existence
of extraterrestrial aliens hiding in our oceans for thousands
of years. I found this on the www:
If you had valued the peace and security of your private world,
then would you announce your presence to those war loving savages
upstairs? That's why millenniums upon millenniums those little
guys have no interests to come on land. Just as you would expect,
the extraterrestrials hide from us. They are here, in the oceans,
where we cannot see them.
In arguments like this, lack of evidence is treated as a kind of
positive evidence. The form of the implicit reasoning appears to
be a mistake like this: if the aliens were here and hiding, then
we would have no evidence; we have no evidence; the aliens are
here hiding. In propositional form, this looks like:
P --> Q
But here even if the premises are true, the conclusion could be
false (use a truth table to confirm this).
- Inappropriate Slippery Slope This is when we assume
without evidence that one kind of fact or event will lead to
another kind of fact or event. The idea of a slipper slope is
that you "slide" from one condition to the next. There can be
legitimate slippery slopes, but often it is not a valid bit of
Here is an example. Speaking on the television news program Wolf
Blizter's Late Edition, Jerry Falwell said of attempts to
legalize same-sex marriage:
[I]f, in fact, we legalize same-sex marriage, then why not
polygamy? why just two, why not a dozen? Why not...what's
wrong with bestiality?..."
Regardless of whether same-sex marriage is a good or bad thing,
Falwell's argument here appears to be that there is a slippery
slope from legalizing it to legalizing all these other things.
But he needs to provide some evidence of this. Since the very
thing at issue in this discussion is whether same sex marriage is
good, neutral, or bad, at best Falwell can be said to be begging
the question by claiming it will lead to or is as bad as these
other things. At worst, he is suggesting that legalizing same sex
marriage is bad because it will lead to these other things. But
without evidence of that, this is nothing but a fallacious slippery
Confusions with Sourcing Criteria
- Straw Man This is when someone constructs an
alternative conclusion to the issue at hand, or an alternative
argument than the argument that was made of a conclusion they
reject, and then attacks that conclusion or argument. This false
or irrelevant alternative is the straw man, and is set up to be
easily knocked down.
He is an example, from an interview on National Public Radio on 4
October 2005. Being interviewed is Senator Mary Landrieu (D-LA).
The interviewer is Steve Inskeep. The Senator is responding to
questions about her request for $250 billion for the state of
Inskeep: Are you sure you can justify all 250 billion dollars?
There are tens of millions of dollars for sugar cane research, for
Notice two fallacies here. First, we are told that FEMA money and
the money for Iraq may not be well managed. If this is meant to
say that we should not worry about whether the money for Louisiana
is well managed, then this is an instance of a form of "tu quoque"
reasoning: others are doing it, so why pick on me? Second, there
is a clear straw man here. The issues at hand here are (1)
whether all $250 billion that was requested is needed, since there
are things in the bill like money for sugar cane research, and not
just for home building and other reconstruction issues; and (2)
what kind of oversight should there be for this money? Note that
Landrieu ultimately attempts to deflect these questions by
constructing a straw man that we can think of as either an
alternative conclusion (you are saying I'm a corrupt Louisianan!)
or as an alternative argument (you are saying oversight is
required because you suppose we Louisianans are corrupt!) and then
Landrieu: We don't think every project in their will pass, but let
me ask you this: are people confidant that the money that they
have given to FEMA has been spent well? Why do they just jump to
the conclusion that any money that Louisiana would be spending
would not be spent well? We've not asked for a blank check, we've
not demanded this money, we've respectfully requested it, we have
no problem with transparency and accountability. So, we're open
to any system that we or others could suggest to make sure that
the money is spent well and wisely.
Inskeep: Including naming some federal overseer, some
L: Well, no, I don't support a czar, I would support a federal
facilitator, but we would not mind having any federal dollar
that's spent in any program should be accounted for. We're not
doing a really good job of accounting for our money in Iraq. But
just because a person from Louisiana asks for money the first
reaction is "Oh my God, we can't give them any because they are
corrupt" is obnoxious and insulting to every person in Louisiana.
- Inconsistent or Disputed Definition This is one form
of a straw man, but rather than changing the issue or argument,
one redefines some key term or concept. Then, typically, at some
point in the argument the person using the fallacy acts as if the
original definition or description were the relevant definition
for their point. This is a kind of sleight of hand, and is very
common. Here is an example, chosen at random; it comes from a
group called Summit Ministries:
Secular human is [in part the view that the correct psychological
theory of human beings is] Monistic Self-Actualization. This
psychology focuses on man's inherent goodness and predicts that
every individual can achieve mental health through the fulfillment
of physical or material needs. This is self-actualization. Monism
means that man is only body, that no soul, mind, or conscience
exist. If man is only matter, then his actions are simply the
result of mechanical impulses. This notion, called behaviorism, is
inconsistent because it directly contradicts the Humanist's
atheistic theology and naturalistic philosophy, which claim that
man, is the master of his fate.
Now, if we allow this to be a definition or necessary description
of something that this person is calling "secular humanism" --
call this sh1 -- then we might, for example, reject sh1 on the
grounds that it entails a very simplistic view of human beings.
But the thing being attacked in the implicit argument is not sh1
as defined or described here, but rather the beliefs of those
people who call themselves "secular humanists" (a review of the
surrounding material at www.summit.org/resource/worldview_chart/
makes this clear). But the most organized self-described
humanists in fact define their view in the following way (this is
from the Council for Secular Humanism,
Secular Humanism is a term which has come into use in the last
thirty years to describe a world view with the following elements
- A conviction that dogmas, ideologies and traditions,
whether religious, political or social, must be weighed and
tested by each individual and not simply accepted on faith.
- Commitment to the use of critical reason, factual
evidence, and scientific methods of inquiry, rather than faith
and mysticism, in seeking solutions to human problems and
answers to important human questions.
- A primary concern with fulfillment, growth, and
creativity for both the individual and humankind in general.
- A constant search for objective truth, with the
understanding that new knowledge and experience constantly
alter our imperfect perception of it.
- A concern for this life and a commitment to making it
meaningful through better understanding of ourselves, our
history, our intellectual and artistic achievements, and the
outlooks of those who differ from us.
- A search for viable individual, social and political
principles of ethical conduct, judging them on their ability
to enhance human well-being and individual responsibility.
- A conviction that with reason, an open marketplace of
ideas, good will, and tolerance, progress can be made in
building a better world for ourselves and our children.
Call this view sh2. The problem is that the implicit argument
against sh1 will not apply to sh2, even though it is clearly meant
to do so. There is here an attempt to attack sh2 by using the
definition of sh1, but sh1 is a different thing. It is like
proving that Tom Smith is a bad person, and then using the
observation that Tom is bad to conclude that Tom Jones is a bad
- Begging the Question This is when we assume in our
premises the conclusion, often in a disguised form. This makes
for a valid argument (trivially so!) but is usually then treated
as if something important has been shown when in fact a proof of
the following form has be provided:
The point being that there is not much of interest in an argument
that has this form, but if we somehow draw attention away from
the fact that we assumed the conclusion we might then make this
appear to be an interesting argument.
Here is a made up example of the kind of reasoning that would
exemplify begging the question:
Some have said that we should allow DNA evidence to be
used for new trials of death row inmates. But these are
the most dangerous criminals, and we should not be giving
them special consideration that might get them released.
Treating this as an enthymeme and using common sense to fill in
implicit premises, we can note that the use of DNA evidence in any
such trial would be to establish whether a prisoner were
potentially on falsely convicted. But if a prisoner were falsely
convicted, then we have no reason to believe she is a danger than
is anyone else. This argument assumes guilt to justify the idea
that these people are dangerous, in order to attack a process that
aims to establish potential innocence.
False Dilemma A false dilemma occurs when someone claims,
usually implicitly, that there are fewer alternatives of some
kind than there really are. For example, here is an example:
Smith: I'm opposed to the death penalty.
Here, Jones implicitly presumes that there are only two
alternatives: execution of criminals or letting them
go free, but of course there are other possibilities.
Jones: How can you want murderers to go free!
A false dilemma thus typically looks like an instance
of the rule (see notes 05) for disjunctive syllogism:
But if it is not the case that our only options are
P v Q, then the disjunctive syllogism cannot be used
to find a single alternative. That is, suppose it
is so that P or Q or R or S or T are true. Then the
most we can say if we rule out P is that Q or R or S
or T are true.
P v Q
P v Q v R v S v T
Q v R v S v T
- Ad populum This is to appeal to the idea that everyone
believes something as evidence that it must be true. An
observation that there is consensus about some claim might have
some benefit in evaluating a source, but it does not tell us
anything about whether an argument is valid. Here is a made up
example, not much unlike common discussions on cable TV:
Everyone knows that Hugo Chavez is a dictator. So, he is
not a democratic leader.
The conclusion here requires essentially that Chavez is a
dictator. That may be true, but appealing to what everyone
believes cannot be taken as evidence since it is the very thing
implicitly at issue. We need some other evidence in order to
construct a valid argument that does anything more than beg the
- Genetic Fallacy This is to attack a conclusion on the
grounds that we can explain some motivation for why it is being
asserted. Here's an example; suppose we hear the end of an
argument from Jones:
Jones: And for all of these reasons I have given, I conclude that
Lincoln was our greatest President.
What Smith says might be true, but it tells us nothing about
whether the conclusion is true, nor about whether the argument was
valid. (When we evaluate a source, we may care about the
motivation that leads the source to say certain things. But when
we are evaluating an argument, this is simply not relevant.)
Smith: You only say that because you too were born in