A little learning is a dangerous thing;
Drink deep, or taste not the Pierian spring.
There shallow draughts intoxicate the brain,
And drinking largely sobers us again.
Think of the person who has read one book on a subject and suddenly thinks he knows everything. Or the beginning student of philosophy whose superficial encounter with skeptical arguments leads him to deny that we can know anything. A deeper inquiry, if only it were pursued, would in each case yield a more balanced judgement.
Similar
delusions of competence often afflict those who have studied a little
logic. Elsewhere
I’ve discussed the phony rigor often associated with the application of formal
methods. Here, however, what I have in
mind is the abuse of a more elementary part of logic – the study of fallacies
(that is to say, of common errors in reasoning).
The principle of charity
Beginning
students of logic, when they first learn the fallacies, often start thinking
they can see them everywhere – or more precisely, everywhere in the arguments
of people whose opinions on politics or religion they already disagree with,
though not so much in the arguments of people on their own side. (What are the odds?) A good teacher will inform them that
knowledge of the fallacies must be applied in conjunction with what is called
the “principle of charity.” This
principle tells us that, when an argument that could be read as committing a fallacy could also be plausibly
interpreted instead in a different way, we should presume that the latter interpretation
is the correct one.
The point of
this principle is not merely, or even primarily, to be nice. The point is rather that the study of logic
is ultimately about pursuing truth,
not about winning a debate. If we
dismiss some argument too quickly because we haven’t considered a more charitable
interpretation, then we might miss out on learning some important truth –
perhaps a truth that we are reluctant to learn, precisely because it comes from
someone we dislike.
But it’s not
just a failure to apply the principle of charity that can lead someone wrongly
to accuse another of committing a fallacy.
Sometimes people just don’t correctly understand the nature of some
particular fallacy.
Ad hominem?
Let’s
consider some common examples, beginning with the ad hominem fallacy. What
matters when evaluating an argument is whether its premises are true, and
whether the conclusion really follows from the premises, either with deductive
validity or at least with significant probability. And that’s all that matters, logically speaking. The character of the person giving the
argument is entirely irrelevant to that.
Ad hominem fallacies are
fallacies that neglect this fact – that pretend that by attacking a person in
some way, you’ve thereby cast doubt on the argument the person has given or the
truth of some claim he has made.
There are
different ways this might go. The
crudest way is the abusive ad hominem,
wherein, instead of addressing the merits of some argument the person has
given, you simply call him names – “racist,” “fascist,” “commie,” or whatever –
and pretend that sticking such a label on him casts doubt on what he said. Another common variation on the ad hominem fallacy is the circumstantial ad hominem or appeal to motive, wherein one attributes
a suspect motive to the person and pretends that doing so casts doubt on what
the person says. Of course, it does
not. A good argument remains a good
argument, however bad the motives (or alleged motives) of the person giving it,
and a bad argument remains a bad argument however good the motives of the
person giving it.
It is
crucial to emphasize, though, that calling someone a name, attributing bad
motives to him, or in some other way attacking a person or his character is not in itself a fallacy. It amounts to a fallacy only when what is at issue, specifically, is the
merits of some claim he made or some argument he gave, and instead of
addressing that, you change the subject and attack the
person.
But of
course, there are other contexts where the subject is the person or his character, rather than some argument he gave. For example, if a jury is trying to determine
whether a person’s eyewitness testimony is reliable, a lawyer is not committing
an ad hominem fallacy if he notes
that the witness has been caught in lies in the past, or is known to harbor a
personal grudge against the person he’s testifying against. Or, when you are deciding whether to believe
a used car salesman, you are not guilty of an ad hominem fallacy when considering that his motive to sell you a
car might bias the advice he gives you.
Again, in cases like these, what is at issue is not some argument the
person gave, which might be considered entirely apart from him. What is at issue is the credibility of the
person himself.
Or suppose
you call someone a “jerk” precisely because he is acting like a jerk. There is no fallacy in that. Indeed, there is no fallacy even if he is not acting like a jerk, but you’re just
in a bad mood. Name-calling may be
justified in the one case and unjustified in the other, but it is not a fallacy if the context isn’t one where
the cogency of some argument he gave is what at issue, and you’re distracting
attention from that.
People are
especially prone to make the mistake of confusing attacks on a person with the ad hominem fallacy when the context is a
debate or public exchange of some other kind – where, of course, one or both
sides may be making arguments. Suppose
Person A and Person B are engaged in some public dispute (on a blog, on
Twitter, or wherever). Suppose Person A
addresses the arguments of Person B, but Person B refuses to respond in kind,
resorting instead to ad hominem
attacks, or mockery, or changing the subject.
Suppose that Person A, appalled by this behavior, calls attention to
Person B’s personal failings – characterizing Person B as intellectually
dishonest, or as a sophist, or as a buffoon, or the like. And suppose that Person B then objects to
this and accuses Person A of committing
an ad hominem fallacy.
Is Person A
guilty of such a fallacy? Of course
not. He has not attacked Person B as a way of avoiding addressing Person
B’s claims or arguments. On the
contrary, he has addressed those claims
and arguments. His negative estimation of
Person B’s character is a separate
point, and a correct one. Person B – whether out of clueless
befuddlement or cynical calculation – makes of the false accusation that Person
A is guilty of an ad hominem fallacy
a smokescreen to hide the fact that it is really Person B himself who is guilty of this.
In the case
I just described, a person is accused of committing an ad hominem fallacy when he is not
in fact doing so. But it can also happen
that a person pretends (or maybe even sincerely believes) that he is not
committing an ad hominem fallacy when
he is fact doing so. To change my example a bit, suppose Person A
and Person B are engaged in some public dispute. Suppose Person B never addresses Person A’s
arguments, but simply and repeatedly flings terms of abuse, questions his
motives, and so on, with the aim of undermining Person A’s credibility with his
readers. Suppose Person A accuses Person
B of ad hominem fallacies, and Person
B responds: “I’ve committed no such fallacy!
After all, using such terms of abuse is not by itself fallacious. It’s only a fallacy when addressing an
argument, and I haven’t been
addressing your arguments. I’m just
telling people what a horrible person you are.”
Is Person B
thus innocent of an ad hominem
fallacy in this case? Not at all. He may not have committed this fallacy in a direct way, but he has still done so indirectly. True, he has avoided addressing any specific argument Person A has
given. Hence he has not in that way committed an ad hominem fallacy. At the same time, though, he has, through ad hominem abuse, tried to poison his
readers’ minds against taking seriously any
argument that Person A might happen to give.
Hence he has deployed a fallaciously ad
hominem tactic in a general way.
The bottom
line is this. Is a speaker resorting to ad hominem abuse as a way of trying to avoid having to address some claim or
argument another person has given? If
so, he is guilty of an ad hominem
fallacy. If not, then he is not guilty
of such a fallacy (whether or not his abusive language is unjustifiable for
some other reason – that’s a separate
question).
Appeal to emotion?
An appeal to emotion fallacy is committed
when, instead of trying to convince one’s listener of a certain conclusion by
offering reasons that provide actual logical support for that conclusion, one
plays on the listener’s emotions. The strength
of the emotional reaction makes the conclusion seem well-supported, when in fact the premises do not provide
strong grounds for believing it.
But here it
is important to emphasize that the presence of an emotional reaction does not
by itself make an argument
fallacious, not even if the speaker foresees such a reaction and indeed even if
he intends it. To take an artificial example
in order to illustrate the point, suppose some follower of Socrates, having
just heard the fatal verdict, wants desperately to believe that his hero
Socrates will somehow never die. You
hope to bring him back to reality, and present him with the following argument:
All men are mortal
Socrates is a man
Therefore, Socrates is mortal
He contemplates this reasoning, sighs heavily and resigns
himself to the cold, hard truth. The
argument raises profound emotions in him, as you knew it would. But have you committed a fallacy of appeal to
emotion? Obviously not. The argument is no less sound than it would
be if someone with no emotional reaction at all had heard it.
Still, you might think, the reason there is no fallacy here
is that the emotions in question are not such as to incline the person to want to believe the conclusion. Quite the opposite. But suppose the emotions in question were of that sort. For example, suppose one of Socrates’ enemies
feared that the hemlock would not kill him, and worried that perhaps Socrates
was immortal and could never be gotten rid of.
Suppose you present him with
the same argument just given. He is
reassured. But have you now, in this
case, committed a fallacy of appeal to emotion?
No. Here too, the
argument remains just as sound as it would be if some unemotional person who couldn’t
care one way or the other about Socrates had heard it. But what if you not only know that the person
will be pleased by the conclusion, but intend
for him to be pleased by it? What if
you hope that his positive emotional response to the argument will make him
more likely to accept it? Wouldn’t that make it a fallacious appeal to
emotion?
No, it would not. For
the bottom line is that the premises are clearly true and the conclusion
clearly follows validly from them. The
presence or absence of an emotional reaction, of whatever kind, does not change
that in the least. Hence there is no
fallacy of appeal to emotion. Such a
fallacy is committed only when there is some logical gap in the support the premises supply the conclusion,
which the emotional reaction is meant to fill.
But there is no such gap – and thus no fallacy.
Indeed, an emotional reaction can in some cases get a person
to be more rational, not less. In the second example, the person’s fear that
Socrates might be immortal is unreasonable.
He’s letting his fear of Socrates’ influence within Athens get the better
of him, and lead him to paranoid delusions.
The argument you give him, precisely
because it is pleasing to him, draws his attention away from these paranoid
feelings and back to reality.
Again, the example is admittedly artificial. But there are many topics that do
realistically carry heavy emotional baggage, yet where this does not entail
that arguments having to do with them must be guilty of the fallacy of appeal
to emotion. Matters of life and death –
war, abortion, capital punishment, and the like – are like that. No matter what conclusions you draw and what
premises you appeal to, they are bound to generate emotional reactions of some
kind in your listener. But that does not
entail that you are guilty of a fallacy of appeal to emotion.
The bottom line is this.
Are the premises of the argument true?
Do they in fact provide logical support for the conclusion (whether
deductive validity or inductive strength)?
Then the argument is not guilty of a fallacy of appeal to emotion,
whether or not it also happens to generate an emotional reaction in the
listener, and whatever that reaction happens to be.
Slippery slope?
A third fallacy that is widely misunderstood is the slippery slope fallacy. Someone commits this fallacy when he claims
that a certain view or policy will lead to disastrous consequences, but without
offering adequate support for this judgement.
It is an instance of the more general error of jumping to conclusions or
inferring well beyond what the evidence appealed to would support.
For example, suppose someone criticized a proposed small tax
hike by claiming that it would inevitably lead to a radically egalitarian redistribution
of wealth. It is hard to imagine how
this would fail to count as a slippery slope fallacy. Is there a logical connection between raising taxes slightly and radically
equalizing shares of wealth by way of redistribution? No, and it
is not hard to formulate principles that would both allow for some taxation while at the same time ruling
out radically redistributive taxation. Is
there nevertheless some strong causal connection
between raising taxes slightly and radically redistributing wealth? Obviously not, since there have as a matter
of historical fact been many cases where taxes were raised, but were never
followed by a radically egalitarian redistribution of wealth.
Notice that the problem here, though, is not that the argument claims that bad consequences would
follow. The problem is that the argument
did not back up this claim. This is often overlooked by people who accuse
others of the slippery slope fallacy.
They seem to think that any
claim that bad consequences will follow from a certain view or policy amounts
to a slippery slope fallacy.
In fact, there is no fallacy as long as someone explains exactly how the bad consequences are
supposed to follow. If you can show that
A logically entails Z, or that it
does so when conjoined with some other clearly true assumptions, then you have
not committed a slippery slope fallacy.
Or if you can identify some specific
causal mechanism by which A will lead to Z, then you have not committed a
slippery slope fallacy. You commit such a
fallacy only when you jump from A to Z without
filling in the gap between them.
What if you are wrong about the claim that A logically
entails Z, or wrong about the causal mechanism you claim links them? You are still not guilty of a slippery slope
fallacy. True, you are mistaken, and perhaps guilty of some
other logical error. But you haven’t
committed a slippery slope fallacy,
specifically, if you at least proposed some specific means by which A would
lead to Z.
There are other fallacies too that are often misunderstood,
but that suffices to make the point.
Knowledge of the fallacies is essential to reasoning well, but it is of
limited value if it is merely superficial
knowledge, and may in that case even impede careful reasoning. It can lead to seeing fallacies where they do
not exist, and thus lead away from truth rather than toward it. And if one’s knowledge of fallacies is
deployed merely as a further rhetorical
means of trying to make an opponent look bad, it constitutes sophistry rather than remedying sophistry.
Further
reading:
What
is an ad hominem fallacy?
The
ad hominem fallacy is a sin