CHAPTER 4-5
                PITFALLS THAT ENTRAP OUR THINKING

     When I was nine years old, we moved to a block with a tree 
nursery at the end of it.  The nursery seemed as huge as a jungle 
to us kids. (As an adult I am shocked at how small an area it 
covers -- about large enough for two sets of two houses back-to-
back.)  In our game, "tribe war", one of our favorite stratagems 
was to dig an "Ethiopian elephant pitfall trap" -- a pit in the 
path covered with slight sticks and grass for concealment -- and 
then lead the other tribe of two or three boys through the path 
in hopes that one would fall into the pit (all the way up to one 
knee.  We never worried about spraining joints or breaking bones, 
of course.) 
     Those elephant traps are my image of the pitfalls that we 
face in thinking clearly, except that the thinking traps are not 
created by malicious nature to trap any of us in particular.  
Rather, they lie in wait for all of us impartially, without 
design.  But if we do not learn to avoid them, we fall again and 
again, which not only prevents us from getting ahead, but also 
engenders such fear of falling that we stand immovable in one 
spot.  
     Two tactics help avoid pitfalls.  First, know the pitfalls 
so that you can spot their signs, the way a wary elephant somehow 
senses that something is amiss and steps around the trap, or 
tests it with a gingerly foot.  Second, go slowly, do not be 
induced into running heedlessly by others running before you, and 
test every abstract idea against the concrete experience of 
first-hand knowledge and scientific data.  Do not think that the 
path is safe just because others seem to run safely in it.  The 
saying "Make speed slowly" fits well here. 
     The profession of economics is currently graced by a great 
scholar who has an uncanny propensity for avoiding intellectual 
pitfalls.  For example, unlike almost all the rest of the 
profession, he avoided becoming a monotheistic devotee of the 
false god of post-World War II Keynsianism (a period which was 
different from the unusual circumstances of the 1930's depression 
when Keynes's doctrine may have been a useful trick.)  And I have 
noticed him avoiding at least two minor theoretical traps that 
almost every other textbook writer stepped into.  (There must be 
many more such cases that I did not notice.)  I have a hunch that 
he avoided these minor fallacious concepts not because he thought 
them through, but rather because he did not blindly plunge ahead 
and use the concepts just because everyone else did.  Instead, he 
seemed to draw back from them because he had not tested them 
thoroughly himself, and therefore just stepped around them 
"instinctively".  
     In contrast, his great rival (I eschew names to avoid 
unnecessary argument) who is considered the greatest theorist of 
his generation, and certainly is the deftest person 
mathematically who has ever plied the trade, has stepped into 
trap after trap over the years, plunging ahead with mathematical 
advances without testing the underlying reality of the abstract 
propositions.  This has led him to make some statements so 
foolish that most grade-school graduates would avoid them by 
testing them against their own daily experiences.  This Nobel 
prizewinner is so clever that he extricates himself from these 
traps well before they damage his reputation -- but not before 
they have damaged the thinking of his readers and followers who 
remain trapped in the fallacious concepts, unable to make 
progress.  
     It may be relevant that the former of those economists has 
worked as much with empirical data as with theory, always 
treating both sorts of knowledge as inseparable, whereas the 
latter economist has almost never worked with data.  This is 
powerful testimony in favor of William James' injunction to stick 
close to concrete experience, and never be so sure of your skill 
with abstractions that do not constantly test them against 
experience. 
     So...I advise you to make progress slowly, and to build the 
habit of testing your abstract ideas against concrete experience.  
     Now we proceed to the more technical task of analysing some 
of the main pitfalls to help you identify them before they trap 
you.  The more you read about the fallacies, the more sensitive 
to them you will become. 
     Fallacies are here classified as failures of logic, 
evidence, language, relevance, and overconfidence.  Most 
fallacies could easily be classified under a heading different 
than the heading under which I put it, however. 
     The same ideas that emerge in discussion of fallacies appear 
in at least three other contexts in this book, too:  Fallacies 
are used intentionally by propagandists as the principles for 
misleading people.  These same fallacies cause sadness, anger, 
and other painful feelings when we mis-evaluate our lives due to 
bad thinking.  And the principles of scientific thinking are 
devices for preventing fallacies from leading to unsound 
conclusions in research.  

                        LOGICAL PITFALLS1

     The study of logic, and of the logical fallacies, has 
traditionally been offered as a means of improving your 
purposeful thinking.  But deductive logic is only one part of 
reasoning, and reasoning is only one part of thinking.  The 
quality of the facts that one adduces obviously matters greatly; 
if you assume that energy is getting more scarce when it fact it 
has been becoming less scarce over the centuries, a logical 
argument based on a premise of increasing scarcity will 
necessarily be unsound even though its logic may be airtight.  
The quality of the theories you use matters, too.  If you invoke 
the simple Malthusian theory that the scarcity of oil must 
eventually increase because there is a limited amount of it and 
we use some of it up, but this theory is inappropriate because we 
create more oil by growing oil-bearing plants, then again your 
conclusion will be unsound though your deductive logic may be 
airtight. 
     The most serious drawback of logic is that the propositions 
it studies are misleading examples of the intellectual material 
that we must come to grips with.  Studies of logic specialize in 
isolated arguments like "Mammals breathe through lungs.  Whales 
are mammals.  Whales therefore breathe through lungs".  But most 
of our thinking concerns propositions that are interlocked with 
many other propositions, usually in a hierarchical fashion, and 
therefore cannot be evaluated in isolation.  It is the 
relationship of a proposition to the entire body of our thinking 
that creates the most subtle problems, but the most important 
ones.  For example, the Malthusian argument about the scarcity of 
oil is itself a branch of an entire tree of theory and data, and 
the statement about oil therefore cannot be sensibly evaluated 
without reference to the entire hierarchy of arguments; this is 
typical of scientific work.  And focusing on single isolated 
propositions distracts our attention from that larger task.  
     The classical logical fallacies are variations on a single 
theme:  A faulty or missing link in the argument.  The arguer 
gets away with this if the audience is not paying attention.  The 
obvious protection against logical fallacies is to check every 
link in the argument. 

Begging the Question, and Circular Reasoning
     These are two varieties of assuming something is so just 
because it is said to be so.  You beg the question (really, you 
beg that an assumption will be accepted) when you say "Will you 
grant that statement A is true?  If so, then we can deduce 
that..."  And you go on from there.  Or you just keep repeating A 
until it seems believable just because it has been heard so 
often.  
     The circular fallacy is proving A with B, and then proving B 
with A.  "Why should you believe in God?  Because the Bible says 
so.  Why should you believe what the Bible says?  Because it is 
the word of God." 

Non Sequitur

     In Latin "non sequitur" means "it does not follow".  That 
is, there is a break in the logic of the argument.  For example, 
in centuries past, in pogroms in Russia this cry was frequently 
heard:  "Save Mother Russia.  Beat the Jews".  There is no stated 
connection between the two ideas.  But amidst the passions of 
riots, the hearer does not pay attention to the break in the 
argument and simply acts as if one proposition implies the other.  

Post Hoc Ergo Propter Hoc
     The Latin title of this fallacy means "after it and 
therefore because of it" - that is, because B came after A, A is 
assumed to have caused B.  But do the birds which fly south cause 
winter to follow?  Precedence alone is a very weak test of 
causation.  (See Chapter 4-3.)  It is at the same level of 
thinking as the superstition that because every time you had corn 
flakes for breakfast you did well in the tennis match, corn 
flakes cause you to play well.  At best this is a clue to 
investigate dietary influences, rather than reasonable proof of a 
causal effect. 
     Indeed, no purely logical argument alone can establish 
causality.  In a complex situation, the decision to describe a 
relationship as causal is subtle and requires judgment, as 
described in Chapter 4-3.  

False Alternative

     "If you are against abortion, you must work for a law 
against it".  This fallacy is rather obvious:  You may think that 
abortion is a bad thing, but also think that a law against it is 
even worse.  Or you may be against abortion but judge that the 
effort to have it made illegal is even worse.  
     The rhetorical device known as "false alternative" attempts 
to box you into a corner from which you cannot escape except by 
accepting the speaker's desired alternative.  It does so by 
implicitly suggesting that there are only two alternatives, one 
that you do not accept and the other the speaker's desired 
alternative.  But usually there are many other alternatives.  To 
avoid being dragged into this fallacy, you need only use your 
imagination to postulate other alternatives, and have the force 
of personality to deny the speaker's false narrowing of the 
alternatives. 

Long Chains of Reasoning  
      The longer the chain of argument, the harder it is to check 
every link in the chain, and therefore the easier it is for an 
undetected fallacy to occur.  A computer program is a nice 
example of a chain of logical argument.  It is obviously easier 
for an error to slip by when a program contains hundreds or 
thousands of lines of programming code, all inter-related to 
other lines of code with a myriad of connections running every 
which way, than when a program contains just three or four lines 
of code where the logic is easy to grasp and an error therefore 
is obvious to your eye.  The same is true of a long argument in 
mathematics or symbolic logic.  (Chapter 4-2 discusses this 
matter further.)
     The longer and more complex the argument or computer 
program, therefore, the more important it is to check it not only 
by examining its logic but also by examining the output to see if 
it makes sense and squares with your everyday knowledge.  All 
things being equal, therefore, a policy of a government or an 
organization that hinges on a shorter line of argument is more 
attractive than a policy that hinges on a longer line of argument 
that ostensibly proves that the policy is sound.  (More 
discussion of this rationale in favor of muddling-through 
policies in preference to radical-change policies may be found in 
Chapter 3-1).  
              PITFALLS IN DEALING WITH INFORMATION          
Gathering and evaluating information is a skill.  If you do it 
poorly, your thinking suffers.  Lack of skill is a bit different 
than a fallacy, however, because the same actions that are 
unsound in one situation are sound in another situation.  The key 
questions about evidence are which, and how much -- how much 
evidence to gather, how much attention to give to particular 
pieces of evidence, which evidence you should consider relevant, 
and so on.  
     Let's briefly classify the places where error may occur:  
First, the aspects of gathering and evaluating evidence are:  the 
elements (observations), the dimensions (variables), the 
deductive logic used, the devices used to extend the data, 
predisposing abstractions (theories and world views), mechanisms 
used, and (perhaps surprisingly) the honesty of your character. 
Second, for each of these aspects of the information-developing 
process there are issues about what to include, what to emphasize 
and de-emphasize (the weights given to various elements), and 
what to exclude.  Space is lacking to cover this classification 
systematically, so I'll simply touch on some  high points.            
     Many fallacies result from the lack of scientific 
discipline.  Indeed, fallacy and scientific discipline are the 
names of the two opposite sides of the same coin.  Hence I'll 
refer frequently to Chapter 3-3 which discussed scientific 
methods for gathering information. 
     We tend to gather and process information in ways that are 
convenient for us.2  This implies that the information that is closest at hand, easiest to notice, and most 
familiar to us, is more likely to enter into our thinking than information that is harder to come by and less distinct.  Sometimes 
this economy of thought works well.  Sometimes, however, we risk falling into error by collecting information hit-or-miss, the 
easy way.  
     The fallacy of post hoc ergo propter hoc is an example of this kind of erroneous thinking.  The event that precedes another 
event comes to mind as the cause simply because it is the most available explanation of the occurrence.3
     We also use convenient similarities as a way of guessing the nature of new instances.  For example, Jacques Barzun tells us 
that many professors pigeon-hole a new student on the basis of her/his similarity to students they have taught earlier.  This 
practice certainly saves a lot of time in studying the new student.  (Bergun even recommends the practice.)  But it is easy to be 
fooled by surface characteristics and hence to put the student into the wrong pigeon-hole in your mind. 
     Even (or especially) scientists fall into this pitfall.  You have probably read and believed, as I have, that air travel is 
safer than auto travel per mile.  But driving 600 miles on a rural interstate highway by a 40-year-old seat-belted, alcohol-free 
driver in a car 700 pounds or more above average weight is slightly safer than an air trip of 600 miles, and 300 miles on the 
interstate are twice as safe as a 300 mile air trip.  Surprise.  And that's for the driver; a passenger sitting in back is even 
safer4.  But then again, though fatalities may be no greater in such a highway trip, the rate of injuries is about seventy times 
as great as for an air trip.  [get cite from Windle and Dresner]  And for a young, drunk, unseatbelted driver of a small car on a 
winding country road, the fatality rate is more than a thousand times higher than the safe-situation rate5.  Now how should we 
think about the matter?  (For some purposes one might wish to compute a measure of "total accident bodily damage" by adding an 
economic measure of the cost of injuries plus fatalities, using the sort of valuations of injuries and deaths that a court might 
arrive at -- very tricky business.  Clearly determining your purpose in advance might be the toughest part of this exercise.) 
     These are some pitfalls that scientific discipline helps you avoid: 
Insufficient Evidence, and Neglect of Variability
     A single swallow does not make a summer, and a single anecdote does not constitute proof of a general phenomenon.  The fact 
that Simha had a happy time in Salt Lake City does not imply that Salt Lake City is the best place to go to have a happy time.  
But a single demonstration that an artificial heart keeps a patient alive does indeed prove that an artificial heart can work.  
This error of neglecting variability is so important that it is discussed at further length in the next chapter. 
     The point which I want to hammer again (and again) is the importance of recognizing variability in a phenomenon, and 
responding to variability by collecting a sample of evidence large enough and representative enough to give you a reliable picture 
despite the variability, as in the example of the quality of taxi rides in various cities in Chapter 3-3.Improper Weighting of the 
Elements of Evidence
     A single vivid illustration can overwhelm a mass of other evidence.  A single picture of a brand-new immigrant receiving 
welfare can overpower careful statistical study which shows that immigrant families receive less in welfare than do average native 
families.  The impression of awesome power projected by a marching army on parade can lead to a decision to go to war, under the 
illusion that the enemy's power cannot be defeated.  Somewhere I have read Gamal Nasser saying that such an impression influenced 
him to plan war on Israel in 1967.  In the opposite direction, Charles DeGaulle wrote that Nazi parades of war equipment in 1937 
convinced French observers in Berlin that Germany was undefeatable, which led to French concessions with respect to Central 
Europe.6 
     Another illustration:  Many people who lived through the 
depression were so seared by the trauma of money worries that 
they saved obsessively in later years. 
     An important variant of improper weighting of the evidence 
arises from neglect of the proportions of the elements being 
considered.  We often give equal weight to all the elements when 
they differ greatly in their importance and ought to be weighted 
very differently.   This is like the story of the rabbit-and-
horse stew -- proportions 50-50, one rabbit, one horse.  This 
error is discussed at greater length in the next chapter. 
    I fell into this pitfall yesterday.  My friend Mort told me 
that he was worried about a brain tumor because he noticed that 
one eye was foggy, and he ran into a man who had had a foggy eye 
and it turned out to be a brain tumor.  I reassured him that the 
odds probably were 20 to 1 in his favor because there must be 20 
causes of a foggy eye other than a brain tumor.  Today he called 
and told me good news -- at worst it is a cataract, not a very 
serious matter nowadays.  (It turned out to be nothing at all).  
And I realized that there may be a thousand (I'm guessing) 
cataract sufferers for each occurrence of brain tumor, let alone 
other explanations of a foggy eye, and hence the odds I gave were 
way off the mark.  I had made the elementary error of ignoring 
the frequencies of the explanations, and counting each one 
equally. 

Unsound Biasing Preconceptions and Theories
     The pitfall of being biased by unsound prior expectations is 
as big as the world -- literally.  Indeed, an unsound conclusion 
often stems from an unsound world view.  For example, if you flip 
a penny ten times and get ten heads, you may expect a tail on the 
eleventh flip because you wrongly believe that there is a 
tendency for such things to "average out".  (This is factually 
wrong, of course; as it is said, the coin has no memory.)  
Indeed, you may even "see" a tail on the eleventh flip though it 
really is heads, because you are so sure that it must come out 
that way, as a result of the unsound theory of "the law of 
averages". 
     Or you may have acquired the view that Asians work harder 
than other groups, and you may therefore "observe" that an Asian 
in your office works harder than anyone else even though this is 
not so.  
     Yet you must make some generalizations -- about which type 
of dog is particularly friendly to children, and which plants 
make you sneeze -- or you could not get through the day.  Biasing 
Hopes
The Desire to Believe
     Too often we fool ourselves because we want to believe, 
rather than on the basis of evidence and sound arguments.  "I'll 
do it tomorrow," and allow ourselves to be convinced, though we 
put it off again and again with the same argument; this sort of 
thinking is a staple for all of us.  
     Our capacity to fool ourselves comes home to me every time I 
tiptoe so as not to wake someone up.  I can feel myself closing 
my ear passages to make it less likely that I will hear the 
sounds that I make.  In other words, I have an almost built-in 
mechanism to convince myself that I am am making less noise than 
I really am.  
              FALLACIES IN THE WAY WE USE LANGUAGE

     Philosophers have known for centuries how easy it is for 
others -- and for philosophers, too, or even especially 
philosophers -- to think badly due to pitfalls that are hidden in 
language.  Indeed, the entire body of philosophical pseudo-
knowledge called "metaphysics" is a collection of linguistic 
fallacies.  Some of these pitfalls are specific to certain 
languages, but the most important of them are the result of 
speaking any language.  
     The root of linguistic fallacies is confusion between words 
and things.  People often assume that because there is a word for 
something, the subject of the word is not just an illusion.  This 
is a flaw in metaphysics.  Just because philosophers have been 
talking about "the Good" for millennia does not imply that there 
is an meaningful entity corresponding to that term.  
Interestingly, in Hebrew the same word refers to "word" and 
"thing".
     Consider the confusion created by the great biologist Konrad 
Lorenz by the use of such terms as "married love" for some 
behavior of geese.  An unwary reader then asks (as one reviewer 
of a Lorenz book did):  "What is married love?"  and after 
associating a set of characteristics with the term -- for 
example, a "commitment" to each other --  goes on to say such 
things as "Would it not be nice if people could have the same 
commitment toward each other as geese do, instead of being 
fickle?"  (You laugh? People make such errors all the time.)  
(Stephen Jay Gould, review of Konrad Lorenz by Alec Nisbett, New 
York Times Book Review, Feb 27, 1979).  
     It helps if, instead of asking what a word" means", you ask 
what people mean when they use a word.  It may be that there is 
no agreement on what they mean; if so, the word is a hopeless 
mass of confusion, as is often the case in metaphysical 
discussions.  A way to work toward agreement among people about a 
word is to forge an "operational definition", or "working 
definition", which contains the operations to be used in 
identifying or measuring the phenomenon under discussion.  
Einstein's great breakthrough with the special theory of 
relativity was the substitution of an operational definition of 
time (time is what you read on a clock) for a definition in terms 
of its supposed properties. 
     Just because a writer's language seems to you to be nothing 
but a mass of confusion, however, to others it may convey useful 
meaning -- perhaps as poetry does, however, rather than as 
physics does.  This may be the situation with some theology 
(though some other theology may be simply muddle.) 
     No one can be considered safe from such confusions -- even 
the great Newton.  The basis of Einstein's great achievement of 
special relativity was demonstrating that a definition of time 
with respect to its supposed properties -- an idea that had been 
at the heart of Newton's thinking -- was shot full of confusion.  
Einstein's next step was to replace the ideas of absolute time 
and absolute space with definitions based on the operations used 
to measure time -- time is simply what one measured on a clock, 
and nothing more.       Even earlier, Ernst Mach -- from whose 
writings Einstein learned much in his youth  -- wrote about this 
"illusory notion" of absolute time, saying "It would appear as 
though Newton...still stood under the influence of the medieval 
philosophy...It is an idle metaphysical conception".  ("Newton's 
Views on Time, Space, and Motion", from The Science of Mechanics, 
Chicago, Open Court, 1902, excerpted in Arthur Danto and Sidney 
Morgenbesser, Philosophy of Science, Cleveland, Meridian, 1960, 
337-338).  
     If Newton could fall into such a linguistic trap, think how 
easy it is for the rest of us to be entrapped by language. 
     Perhaps the most troublesome linguistic fallacies stem from 
the various forms of the verb for existence, "is".  Especially in 
an ambiguous discussion, the term "is" (and its related terms 
such as "are" and "be") is (sic) one of the greatest sources of 
confusion in English.  This verb has two very different uses - 
first to indicate equality (as in "A car is an automobile" and 
"two and two are four"), and second, to serve as a joining word 
(a "copula", as in "Jack is skinny" and "Sereno is a fool").  
People think that by understanding the "is" they learn about the 
world.  I once met a person who intended to write a book about 
"What Is a Human?" It is unlikely that anything except confusion 
would come out of that endeavor. 
     An example:  People ask "Is a foetus a person? "  That is, 
they first decide whether a fetus "is" or "is not" a child, and 
then deduce from that definition a conclusion about what to be 
done.  If they answer "yes", they then conclude that "Destroying 
a foetus is murder"; if they answer "no", they conclude that the 
pregnant woman is not precluded from destroying it.  Everything 
here hinges on the work "is".  This is a typical example of why, 
in my view, this is (sic) the worst word in the English language 
-- but indispensable.  (Hebrew lacks such a word, but in the 
vernacular people use a substitute for it anyway.)  Perhaps the 
law requires such categorizations as whether a foetus "is" a 
person.  But apart from the law, I seldom find this usage 
necessary (though it is [sic] often convenient), and usually it 
is (sic) confusing when one is talking of morals. 
     The word "is" is at the root of the grave paradoxes in logic 
that required unraveling by the pathbreaking discoveries of 
Bertrand Russell and Alfred North Whitehead.  It is a fascinating 
and illuminating, though difficult, exercise to try to speak or 
write without using any version of "is," or related terms such as 
"exist".  The result is a purified English which its inventor, 
David Bourland, Jr., calls "E-prime".7 
     There are a zillion ways in which language can purposely be used to obscure meaning, all of them employed on occasion by 
politicians and advertisers (though this is not intended to suggest that most politicians and advertisers are not quite honest in 
their language; I believe them to be no less honest than people in other professions).  Someone joked that advertising is the only 
profession to have invented an entirely new linguistic device, the "indefinite comparative". I see this device now on the 
dishwashing liquid bottle.  "More Real Lemon Juice" says the label.  More than what?  Advertising can also claim to have removed 
one part of our anatomy, the armpit, and replaced it with another, the underarm.  Quite an achievement, and just with the simple 
technique of euphemism.     
     The power of language to mislead us derives in considerable part from its nature as communicator of abstractions through its 
easy-to-absorb and often emotive symbols.  In just three words, "Peace, Bread, and Jobs" conveys a great load of meaning to many 
people, enough to move people's minds and bodies.  Effort and precise skill are necessary to elucidate what is meant by such 
slogans.  But without such elucidation any reaction is just unthinking herd action. 
     Grand statements about any nation or culture are almost surely wrong.  And from-the-hip analysis of national character by an 
unlearned layperson is particularly obnoxious.  Nevertheless, I hazard this excursion into such statements:  Numerical slogans are 
peculiarly prominent in contemporary Chinese culture, perhaps as a nmemonic device.  (Was it so in earlier years, too?)  Any book 
about daily life in China is replete with allusions to (for example, in just two pages of a book about the Cultural Revolution; 
Bao Yuan, Born Red, Stanford U. P., 1987, pp. 82-83) the "Sixteen Points", the "Five Red Categories", "the Seven Black Categories" 
and the "Four Olds -- the old ideas, culture, customs, and habits of the exploiting classes".  Elsewhere in the book: "the army 
was undertaking a Three Supports and Two Militaries campaign" (p. 200), Liu Shaoqi was "pushing the capitalistic Three Frees and 
One Contract" (p. 175), the Central Committee's letters to the Soviet Central Committee containing "three peaces and one less" 
doctrine (p. 144), and much more.  There is a category of treasured possessions known as the "Five Wheels" -- bicycle, watch, tape 
recorder, and so on.  The menu of any Chinese restaurant offers numerical delicacies.  Can it be that the process of 
categorization, in which one of the elements takes on the evocative power of the others and all together have more evocative power 
than the elements when separated, is connected to the tendency in China for mass movements to arise so quickly and violently? 

    JUDGING BY THE SOURCE, AND OTHER FALLACIOUS IRRELEVANCIES     If you can ask "What's that got to do with the subject at hand?" 
or "So what?" (as Rudolph Flesch recommends we do frequently) and the answer is not satisfactory, then you have identified one of 
the irrelevancy fallacies, of which there are many interesting forms.  If you routinely apply this test to the arguments you hear 
-- and perhaps to your own, too -- you will be shocked how often you find irrelevancies. 
     There are many interesting irrelevancies that do not fall into the categories below, for example the advice to young lawyers:  
If you have the facts on your side, pound the facts.  If you have the law on your side, pound the law.  If you have neither going 
for you, pound the table.
     
Ad Hominem Attack
     Rabble-rousers surely throughout history have attacked an argument by attacking the source of the argument instead.  For 
twenty years biologists such as Paul Ehrlich and Garrett Hardin 
have attacked my studies of the economics of population by pointing out that once upon a time I was in the mail-order business, 
and my book on the subject is still selling well (fifth edition coming up!).  For those people to whom working in business is a 
sign of moral decadence -- even for two years, near the beginning of one's work life -- the implication is that one need not 
believe anything said by a person guilty of such a character flaw.  They also attack my intellectual credentials by pointing out 
that I do not have a PhD in economics, implying that without a PhD in a given field one cannot do credible work.  And they 
complete the triangle by attacking my motives, repeating from an introduction to one of my books that I was depressed for a 
period; they imply that my showing that the material life of humanity has been improving is simply a device to improve my own 
morale.  If this is not enough to impugn my motives, they suggest that I applaud more people being born because this is good for 
my supposed constituency in the business community (because I teach business, among other subjects, and have a PhD in that field.)  
They have not yet employed the standard ad hominem changes -- that I have an illegitimate child whom I refuse to acknowledge, and 
that I sneak money from my employer's or the public till.  Nor have they used on me the ad hominem device which they have used on 
another economist who did some similar research on the subject -- calling him a Catholic, and worse, a convert -- only because I 
am not Catholic.       This would be funnier if such attacks were not successful.  Perhaps the mechanism that makes ad hominem 
attacks successful is that the hearer does not want to associate her/himself with a given set of views lest that would seem or 
feel as if one is associating with a disreputable person. 
     The varieties of ad hominem attack are many and vivid.  Out-and-out insult is basic.  Also the "you are one, too" argument.  
Also the charge of inconsistency.  Discrediting the speaker by claiming that she/he has a vested interest in the argument is a 
staple of journalism, especially in Washington, D. C., where I now live.  (A typical example:  A biologist who has consulted to 
the chemical industry or has had a research grant from a chemical firm is dismissed as unable to render an objective opinion about 
the effects of a toxic substance, especially if the reporter considers her/himself to be an "environmentalist".) 
     Showing that the proposer of an argument is not credible is not always unreasonable, however.  When the evidence upon which 
the argument depends is unclear or mostly not available, your judgment must depend upon a complex of reasoning outside of simple 
deductive logic.  It is reasonable to place more trust arguers who have mostly been right in the past than in arguers who have 
mostly been wrong.  That is, the record of a person's forecasts should be taken into account.  It would seem that the track record 
of those who have forecast doom with respect to resources and the environment over the past two decades should cast into doubt 
these forecasts as they repeat them now and in the future; just about all their forecasts have been falsified by resources 
becoming more available and the air and water becoming cleaner rather than dirtier in the U. S.  (Curiously, however, this record 
of failure has not led to a loss of faith in these persons by the environmental organizations or by journalists.)

Argument to Emotion
     I remember as a boy of six in about 1938 hearing a speech by Hitler broadcast on U. S. radio.  I could not understand a word 
of German, but the frenzy with which Hitler delivered it, and the frenzy with which his audience in Germany cheered it, was 
unmistakable.  I cannot say for sure that his speech did not contain logical arguments, but it was obvious even to a little boy -- 
frightened both by the sound of the speech, and by the reaction of those in the group around me who understood German -- that 
emotion was at the heart of Hitler's appeal in that speech. 
     A prudent speaker enhances an argument with extra-logical elements such as vivid examples and emotional references.  (Indeed, 
that is what I have tried to do in the paragraph above in order to hold your interest and to make the point memorable.)  This is 
not a dirty trick unless the argument is unsound and the emotive elements are used to paper over the lack of logic. 

Argument to Authority
     The argument to authority is the other side of the coin from the argument "to the person" (argumentum ad hominem).  An 
argument to authority asks you to believe the argument because the proponent is an expert.  Again, though this is considered a 
logical fallacy, there are situations when the authority of the speaker is a reasonable consideration in your decision about 
whether or not to credit the argument. 
     Even within science, a person's record as a careful researcher and interpreter of evidence should sometimes be taken into 
account in judging the validity of a study.  Lack of space in a publication sometimes makes it impossible to describe in full 
detail how data are gathered.  If a piece of research reaches a conclusion that you find difficult to believe because it runs 
contrary to other studies, you are more likely to believe that the research is sound, rather than the result of a procedural 
error, if it was done by someone whose competence you have reason to trust.  
Appeal to Consensus
     "I have yet to meet anyone familiar with the situation who thinks India will be self-sufficient in food by 1971, if ever". 
(The Population Bomb, 1968, pp. 40-41).  This statement by biologist Ehrlich convinced many people that the Indian sub-continent 
was a "basket case" to be considered beyond saving by itself or by others.  In fact, there were at the time many agronomists and 
agricultural economists who felt that India had a very bright agricultural future if its farmers were unhindered by government 
regulation.  So the appeal to community belief was not even factually sound, as is frequently the case with arguments based upon 
community belief. 
     But even if there is great consensus on a belief, and even if the consensus includes those persons in the best positions to 
judge the matter, the consensus may be wrong.  And if the argument is not judged apart from the appeal to the general belief, the 
community may lose a sound idea. 
     "Fifty million Frenchmen can't be wrong."  But conventional beliefs are frequently wrong, perhaps more often than most 
members of the public thinks.  The presumption that the conventional belief is correct is too strong, in my experience.  This is 
certainly the case with economic issues.  The views stated in newspaper editorials about tariffs, trade gaps, budget deficits, 
currency controls, rent control, subsidies, and the like are more often wrong than right, as judged by the overwhelming and long-
standing consensus of the economic profession going back to Adam Smith.  (Is there a damaging contradiction in what I just said?) 

Appeal to Force
     One of the most famous lines from the movies comes from the mouth of a violent Mafia capo:  "I'm going to make you an offer 
you can't refuse".  Force is not exactly an argument, but it is included in the classic list of fallacies because it can win an 
argument.  You know how it goes:  The Mafioso implies that if you refuse the offer you will be dead.  
     More veiled threats are a more "civilized" form of this method.  Censorship is a related device. 






                             SUMMAR

                            FOOTNOTES 
     1The following section closely follows David Kelley, The Art of 
Reasoning, Chapter 6, pp. 123-132, including some examples. 
     2This section refers to the availability and 
representativeness heuristics of Kahneman and Twersky, as 
discussed in Nisbett and Ross, who review in a comprehensive and 
interesting fashion the evidence on reliability in human 
inference. 
     3Cite Kahnemann and Twersky on availability
     4Leonard Evans, Michael C. Frick, and Richard C. Schwing, 
"Is It Safer to Fly or Drive? -- A Problem in Risk 
Communication", xeros, no date.  William Freudenburg was kind 
enough to provide me with this information.  
     5 Discussion of Evans et al in Malcolm Gladwell, "Numbers Don't Lie - But They Can Mislead", Washington Post, August 13, 
1990, p. A3. 

     6"On May 1, 1937, a complete Panzer division, with hundreds of aircraft flying over it, marched through Berlin.  The 
impression produced on the spectators, and first and foremost on M. Francois-Poncet, the French ambassador, and on our military 
attaches, was of a force that nothing could stop -- except a similar force." The Complete War Memoirs, New York: DaCapo, 1984, p. 
26] 
     7  Albert Ellis and Robert A. Harper changed to E-prime when 
they revised their excellent self-help book, A New Guide to 
Rational Living (1961; 1975), and they assure us that it 
clarified their thinking greatly.  My knowledge of E-prime comes 
from the introduction to their book.  

     With respect to the observations (elements):  a) Inclusion: 
How far back should you collect data, if you are relying on 
historical experience?  How appropriate is the set of data (the 
universe) that you are working with?  Which theories are you 
relying upon in making such decisions?  b) Exclusion:  Which 
observations are you excluding as "outliers" because they do not 
fit your theoretical preconceptions?  Which observations are you 
excluding because they are not easily accessible -- looking for 
the watch under the streetlight because that is where the light 
is, rather than because you lost the watch there?  c) Weighting: 
How much weight has been given to observations because of their 
vividness and the clarity of par authority?  Which pre-disposing 
abstractions have influenced the weighting of the observations?
     With respect to the variables (dimensions):  a) Inclusion: 
Elements
     Inclusion
          Length of time series and appropriateness of universe, 
representative sampling, pda
     Exclusion
          Excluding as outoers because don't fit preconceptions, 
lamp-and watch, pda
     Weighting
          Vividness, clarity, known-ness, other availability; 
authority; predisposing abstractions (pda)
Dimensions
     Inclusion
          Important, and not unimportant; relevant (e. g. ad 
hominem), pda
     Exclusion
          Lamp-and-watch, political acceptability, pda
     Weighting
          Political acceptability, authority, pda
Logic
     Inclusions
          Premises
     Exclusions
          Premises, arguments
     Connectedness and validity
Mechanisms
     Inclusions
          pda
     Exclusions
     Correct understanding       

Page # thinking pitfa45# 3-3-4d