III.

Argumentation

"An argument is not the same as contradiction (...) it is a

connected series of statements to establish a definite proposition."

III.1. Deductive argumentation, validity and soundness

Logic is the business of evaluating arguments, sorting good ones from bad ones,
the science of proper reasoning. An argument in logic is not a violent quarrel, filled
with hostility and mutual resentment. An argument in logic is a reasoning structured to
persuade someone, to give her reasons to believe some statement. See for instance [1]:
(1) All men are mortal
(2) Socrates is a man
Socrates is mortal
The three dots "" on the third line of the argument mean Therefore and they
indicate that the final sentence is the conclusion of the argument. The other sentences
are premises of the argument. If you believe the premises, then the argument provides
you with a reason to believe the conclusion. This kind of argument is a deductive
argument (a syllogism in this particular case), in formal logic all arguments are
deductive arguments (inductive arguments, even good inductive arguments, are not
deductively valid, so we will not be interested in them).
We can define an argument as a series of sentences: the sentences at the
beginning of the series are premises and the final sentence in the series is the
conclusion. If the premises are true and the argument is a good one, then you have a
reason to accept the conclusion. When analyzing an argument, the first thing to do is to
separate the premises from the conclusion.
Premise indicators: since, because, given that, it is the case that.
Conclusion indicators: therefore, hence, thus, then, so.
In the former definition sentences is used in a technical way: in logic (unlike
in linguistics, for instance), only sentences (also called statements) that can figure as a
1

premise or conclusion of an argument are called sentences. So a sentence is something

that can be true or false. Questions, imperatives and exclamations are not sentences in
logic, only propositions which have a truth-value (which doesnt mean they are true
but that they could be true or false): Where is my umbrella? is not a sentence, it has
no truth-value.
Consider then the following argument [2]:
(1) There are only students in the classroom
(2) Students ought to listen in class
If you are inside the classroom, then you should be listening
There are two ways in which this argument could go wrong: 1/ if some of the
premises are false (maybe there is a teacher in the classroom [premise 1 is false] or
maybe a debate was taking place [premise 2 is false]); 2/ if even though the premises are
true the conclusion is false (maybe there is a recess or maybe youre not in the
classroom mentioned in premise 1). So, an argument could be weak because one of the
premises might be false (an argument gives you a reason to believe its conclusion only
if you believe its premises) or because the premises might fail to support the conclusion
(even if the premises were true, the form of the argument might be weak). The example
we just considered is weak in both ways, but logic just pays attention to the second one:
the truth of a premise has to be challenged with the facts (which are the domain of
observation or experimental sciences, not logic), logic challenges the relation of the
different parts of the arguments, their structure, the form of reasoning, not its
contents. When an argument is both valid and has true premises (a good argument, in
other words), it is called a sound argument. An argument is unsound (a bad deductive
argument) if one or more of its premises are false, or if it is invalid. Since logic alone
can only tell us something about the validity of arguments, nothing about the truth of the
premises, it is only one step towards the recognition of sound arguments. Consider the
following argument [3]:
Unicorns have one horn
Hobbits dont have any horns
No hobbit is a unicorn
Unicorns and hobbits certainly do not exist, nevertheless the argument is well
constructed: given those premises the conclusion follows necessarily, the form of the
argument is correct. Our senses, science, will tell us which sentences are true, logic will
tell us how sentences might be related, how to preserve the truth from the premises
in the conclusion. Logic builds arguments with a correct logical form, in which if its
premises were true, then its conclusion would necessarily be true. We call such an
argument deductively valid or just valid. Consider this other argument [4]:
You are reading these notes
These are Philosophy and Citizenship notes
You are a Philosophy and Citizenship student
Is this argument valid? It isnt. Certainly the premises are true, but it is possible
for the conclusion to be false (the teacher might be reading the notes, or an assistant

correcting them, or a parent thumbing through them). The truth of the premises does
not guarantee the truth of the conclusion, therefore the argument is not valid (notice that
the word "student" is present in the conclusion but not in the premises). The main task
of logic is to sort valid arguments from invalid arguments.
An argument is deductively valid if and only if it is impossible for the premises
to be true and the conclusion false. The crucial thing about a valid argument is that it is
impossible for the premises to be true at the same time that the conclusion is false.
Thats why argument [3] was valid, even though the talking of hobbits and unicorns
seems nonsense. Lets see a similar example [5]:
If you are reading these notes youre either a coconut or a palm tree
You are reading these notes and you are not a palm tree
You are a coconut
The conclusion of this argument is ridiculous. Nevertheless, it follows validly
from the premises. This is a valid argument. If both premises were true, then the
conclusion would necessarily be true. This shows that a deductively valid argument
does not need to have true premises or a true conclusion. Conversely, having true
premises and a true conclusion is not enough to make an argument valid. Consider this
example [6]:
(1) Real Madrid CF is a football team
(2) Atltico de Madrid is a football team
Rayo Vallecano is a football team
The premises and conclusion of this argument are, as a matter of fact, all true.
This is a terrible argument, however, because the premises have nothing to do with the
conclusion. Imagine what would happen if the Rayo Vallecano disappeared. Then the
conclusion would be false, even though the premises would both still be true. Thus, it is
logically possible for the premises of this argument to be true and the conclusion false.
The argument is invalid.
The important thing to remember is that validity is not about the actual truth
or falsity of the sentences in the argument, instead, it is about the form of the
argument: the truth of the premises is incompatible with the falsity of the conclusion.
Consider this new argument [7]:
In 2010 more than one third of the students failed the first exam of the year
In 2011 more than one third of the students failed the first exam of the year
In 2012 more than one third of the students failed the first exam of the year
In 2013 more than one third of the students failed the first exam of the year
One third of the students always fails the first exam of the year
This argument is an inductive argument, as an inductive argument it seems
more or less correct, but its not a valid one: it is possible for the conclusion to be false
even though the premises are true. "Inductive arguments do not guarantee the truth of
their conclusions, even if all the premises are agreed to be true. The most familiar form
of inductive argument is generalization from a number of particular cases as in
argument [7] or when noting, for example, that every animal we have seen with sharp
front teeth eats meat and concluding that all animals with sharp front teeth eat meat. But
3

notice that although we might be absolutely sure that we are correct about the particular
casesthat every such animal we have seen does in fact eat meatwe might still be
wrong in our generalization, our conclusion that all such animals are meat eaters. Thus,
it is essential in any inductive argument to begin with a well-chosen number of
particular cases and to make sure that they are as varied as possible. Inductive
arguments can be strong or weak, depending on the weight of the evidence for the
conclusion, the quality of the sample, and the plausibility of the generalization." (TBQ)
Inductive arguments are not evaluated as valid or invalid so in logic we will not be
interested in inductive arguments, logic is a deductive (formal) science, induction
belongs to empirical sciences which rely on evidence.
In considering arguments formally, we care about what would be true if the
premises were true. Generally, we are not concerned with the actual truth value of any
particular sentence (whether they are actually true or false). Yet there are some
sentences that must be true, just as a matter of logic. Consider these sentences:
1. It is raining.
2. Either it is raining, or it is not.
3. It is both raining and not raining.
In order to know if sentence 1 is true, you would need to look outside or check
the weather channel. Logically speaking, it might be either true or false. Sentences like
this are called contingent sentences.
Sentence 2 is different. You do not need to look outside to know that it is true.
Regardless of what the weather is like, it is either raining or not. This sentence is
logically true; it is true merely as a matter of logic, regardless of what the world is
actually like. A logically true sentence is called a tautology.
You do not need to check the weather to know about sentence 3, either. It must
be false, simply as a matter of logic. It might be raining here and not raining across
town, it might be raining now but stop raining even as you read this, but it is impossible
for it to be both raining and not raining here at this moment. The third sentence is
logically false; it is false regardless of what the world is like. A logically false sentence
is called a contradiction.
To be precise, we can define a contingent sentence as a sentence that is neither a
tautology nor a contradiction.
A sentence might always be true and still be contingent. For instance, if there
never was a time when the universe contained fewer than seven things, then the
sentence At least seven things exist would always be true. Yet the sentence is
contingent; its truth is not a matter of logic. There is no contradiction in considering a
possible world in which there are fewer than seven things. The important question is
whether the sentence must be true, just on account of logic. Logical truths are true in
any possible world.
We can also ask about the logical relations between two sentences. For example:
A1. John went to the store after he washed the dishes.
A2. John washed the dishes before he went to the store.

These two sentences are both contingent, since John might not have gone to the
store or washed dishes at all. Yet they must have the same truth-value. If either of the
sentences is true, then they both are; if either of the sentences is false, then they both
are. When two sentences necessarily have the same truth value, we say that they are
logically equivalent.
Finally consider these two sentences:
B1. My only brother is taller than I am.
B2. My only brother is shorter than I am.
Logic alone cannot tell us which, if either, of these sentences is true. Yet we can
say that if the first sentence (B1) is true, then the second sentence (B2) must be false.
And if B2 is true, then B1 must be false. It cannot be the case that both of these
sentences are true. If a set of sentences could not all be true at the same time they are
said to be inconsistent. Otherwise, they are consistent. Sometimes, people will say that
an inconsistent set of sentences contains a contradiction. By this, they mean that it
would be logically impossible for all of the sentences to be true at once. A set can be
inconsistent even when all of the sentences in it are either contingent or tautologous.
When a single sentence is a contradiction, then that sentence alone cannot be true.
Summary of logical notions:
- An argument is (deductively) valid if it is impossible for the premises to be
true and the conclusion false; it is invalid otherwise.
- A tautology is a sentence that must be true, as a matter of logic.
- A contradiction is a sentence that must be false, as a matter of logic.
- A contingent sentence is neither a tautology nor a contradiction.
- Two sentences are logically equivalent if they necessarily have the same truth
value.
- A set of sentences is consistent if it is logically possible for all the members of
the set to be true at the same time; it is inconsistent otherwise.

III.2. Formalization
Why do we speak of formal logic? Formal logic is opposed to informal logic, which
is the one we use in our everyday life, that is, with a natural language (English,
Spanish). The problem with natural languages is that they hide the logical form of the
arguments, but validity relies on the logical form. Natural languages are not precise
enough, theyre ambiguous and may distract us from the relevant items of the argument,
the relation of the sentences. See for instance those two arguments [1] and [8]:
(1) All men are mortal
(2) Socrates is a man
Socrates is mortal
(1) All which is mortal kills
(2) Socrates is mortal
Socrates kills
5

The second argument seems weird and in fact it is not valid because we are
using mortal in two different senses in the premises (mortal as having the capacity
to die as in "no mortal has ever challenged the power of Zeus", and mortal as having
the capacity to kill as in "this poison is mortal"), but in the conclusion we are treating
mortal as if in both cases it meant the same. See also [9]:
All men are palm trees
Socrates is a man
Socrates is a palm tree
This argument might be less interesting than [1], because the first premise is
obviously false: there is no clear sense in which all men are palm trees. Yet the
argument is valid. To see this, notice that both arguments have this form [F1]:
(1) All Ms are Ps.
(2) S is M.
S is P.
In both arguments [1] and [9] S stands for Socrates and M stands for man. In [1]
P stands for mortal; in [9], P stands for palm trees. Both arguments have this form, and
every argument of this form is valid (as we will prove later on). So both arguments are
valid.
What we did here was replace words like man or palm tree with symbols
like M or C so as to make the logical form explicit. This is the central idea behind
formal logic, and that's exactly what Aristotle, the founder of logical science, did with
syllogisms (but he went no further in formalization). We want to remove irrelevant or
distracting features of the argument to make the logical form more perspicuous. Starting
with an argument in a natural language like English, we translate the argument into a
formal language. Parts of the English sentences are replaced with letters and symbols.
The goal is to reveal the formal structure of the argument, as we did with these two.
Lets focus on [8]. When we are obliged to translate the argument to a formal
language, we realize that mortal does not mean the same thing in (1) and in (2), so we
could never use the same letter for both. The form of [8] is [F2]:
(1) All M are K.
(2) S is D.
S is K.
Argument [1] and [8] looked very much the same, a crucial difference of logical
form shows itself clearly only in formal language. In this case we translated mortal as
the capacity to kill with M and mortal as the capacity to die with D, the final
result is that clearly [F1] and [F2] are different, so the validity of one argument does not
imply the validity of the other one, they are not logically equivalent (and in fact [8] is
not a valid argument, because [F2] is an incorrect logical form).
When we translate an argument into a formal language, we hope to make
its logical structure clearer. We want to include enough of the structure of the English
language argument so that we can judge whether the argument is valid or invalid. If we
included every feature of the English language, all of the subtlety and nuance, then there
would be no advantage in translating to a formal language.
6

We are going to develop a logical language called SL. It is a version of

sentential logic, because the basic units of the language will represent entire sentences
(the smallest units are sentences themselves, simple sentences are represented as letters
and connected with logical connectives like `and' and `not' to make more complex
sentences). The arguments above [1], [8] and [9] and their formalizations [F1] and [F2]
do not belong to sentential logic but to quantified logic, in which the basic units are
objects, properties of objects, and relations between objects. Quantified logic is more
complicated than sentential logic, and we will not study it.
As any language, SL is made up of those elements: vocabulary and grammar
(rules of formation and rules of transformation). As its vocabulary SL possesses a
series of meaningless symbols instead of words, as its grammar certain rules of
formation specify which combinations of symbols are syntactically correct and
constitute an expression or a formula in SL, and finally some rules of transformation
(we will study them further down as rules of inference) indicate how can we convert
some combinations of symbols into different ones, that is, how can we deduce or infer
one expression from others expressions. In SL all those symbols and rules are perfectly
defined to avoid vagueness and ambiguity, thats what makes formal languages more
precise than natural languages. All formal systems, then, possess as basic constituents:
consistency (there are no contradictions within the system), completeness (all correct
expressions are deducible from previously defined rules, so the system contains all the
correct expressions or it is at least possible to derive them from it) and decidability (the
systems possess an effective [mechanical] method for deciding if an expression is part
of the system).
The basic element of SL are sentences, remember that a sentence is a expression
which can be true or false, which has two possible truth values: true or false, indicated
with T/F or 1/0. There are two kinds of sentences:
- Simple or atomic sentences: those which cannot be separated in other
sentences. For instance, even though Homer Simpson loves donuts can be split in
different parts (as subject and predicate), nevertheless it is not formed by smaller
sentences, its just one sentence. Homer Simpson is not a sentence, it has no truth
value, it asserts nothing.
- Complex or molecular sentences: those which can be broken down in atomic
sentences. For instance Homer Simpson loves donuts and hates vegetables can be
divided in two atomic sentences: Homer Simpson loves donuts and Homer Simpson
hates vegetables.
All sentences and combinations of sentences will be expressed through symbols
which can be sorted in two main types:
- Non logic symbols
o Sentence letters. Lower case letters (p, q, r, s, t) used to substitute
for sentences. For instance Bart Simpson is a bad student would
become p, and Bart Simpson is a bad student and a good skater
would become p and q. Those letters are called variables because
they do not stand for specific sentences, they can substitute different
sentences within different reasoning. Lisa Simpson will stay at home
and study might also become p and q because the formal structure
of this molecular sentence is the same as the previous one.
7

o Auxiliary symbols. Parentheses and square brackets used to simplify

the reading of complex sentences and reasoning. Thanks to them we
can know which relation is dominant between complex sentences
which can be interpreted in different ways.
-

Symbol

Logic symbols
o Monadic. Negation () is used to deny any sentence, it means no
or it is not the case that. It is monadic because it accompanies
one atomic sentence or one molecular sentence, but it cant be used to
build complex sentences, to relate or connect different sentences.
o Diadic. The connectives () are used in various ways to
relate sentences or groups of sentences, they are dyadic because the
necessarily connect two atomic sentences or two molecular
sentences.
Name
conjunction
disjunction
conditional
biconditional

Meaning
both and, and, but, also, although
either or or both, or
if then, therefore, then
"if and only if... then...",

So, finally, there are three kinds of symbols in SL:

Sentence letters
(with subscripts, as needed)
Connectives
Round and square brackets

a, b, c,z
a1, b1, c1, a2, b2, h37, n456,

( , ), [ , ]

With those symbols we can form any expression in SL. We define an expression
of SL as any string of symbols of SL. Take any of the symbols of SL and write them
down, in any order, and you have an expression. Not any expression is valid, since any
sequence of symbols is an expression, many expressions of SL will be meaningless. A
meaningful expression is called a well-formed formula. It is common to use the
acronym wff; the plural is wffs. The following rules of formation of formulas specify
for each possible expression in SL whether it is a well-formed formula or not, that is,
whether it is a syntactically correct expression in SL or not.
1. Every lower case letter is a wff.
2. If A is a wff, then A is a wff of SL.
3. If A and B are wffs, then (AB) is a wff.
4. If A and B are wffs, then (AB) is a wff.
5. If A and B are wffs, then (AB) is a wff.
6. If A and B are wffs, then (AB) is a wff.
7. All and only wffs of SL can be generated by applications of these rules.
Those rules define recursively all wffs in SL. The possible combinations of
symbols are endless since there are infinitely many sentence letters. So there is no point
in trying to list all the wffs, the rules describe the process by which wffs can be
8

constructed and thereby establish the limits of SL with just a list of six rules. It is
important here that A is not a sentence letter, it is a variable that stands in for any wff at
all. The capital letters A and B are used because they stand for any possible sentence
letter, they are an expression of the metalanguage that allows us to talk about infinitely
many expressions of SL (because A and B are part of the metalanguage, they are called
metavariables). It is important to distinguish between the logical language SL, which
we are developing, and the language that we use to talk about SL. When we talk about a
language, the language that we are talking about is called the object language. The
language that we use to talk about the object language is called the metalanguage. The
object language is SL, the metalanguage is English supplemented with some logical and
mathematical vocabulary.
With those elements (symbols and rules of formation) we are ready to formalize
any expression of a natural language as English, only a few points have to be clarified
concerning the use of parentheses. A wff like (qr) must be surrounded by parentheses,
because we might apply the definition again to use this as part of a more complicated
sentence. If we negate (qr), we get (qr), if we just had qr without the parentheses
and put a negation in front of it, we would have qr, which has a very different
meaning. The sentence (qr) means that it is not the case that both q and r are true; q
might be false or r might be false, but the sentence does not tell us which. The sentence
qr means specifically that q is false and that r is true. As such, parentheses are crucial
to the meaning of the sentence. So, strictly speaking, qr without parentheses is not a
sentence of SL.
Lets see some examples of formalization of reasoning.
(a) If you study and come to class, then youll pass.
p = you study
q = you come to class
r = youll pass
Formalization: (pq)r
This sentence states that in order to pass two conditions have to be met: coming
to class and studying. This means that it has both to be true that you come to class and
you study for you passing to be true also.
(b) It is not true that I abandoned my cat and that I hate onions
p = I abandoned my cat
q = I hate onions
Formalization: (pq)
This sentence states is that it is not true simultaneously that I abandoned my cat
and I hate onions, but one of those sentences could still be true alone, it is not each
atomic sentence which is denied but the conjunction of both.

(c) If you get caught copying the exam youll fail, and either youll be expelled
or punished with extra homework.
p = you get caught copying the exam
q = you pass
r = you will be expelled
s = you will be punished with extra homework
Formalization: pq(rs)]

III.3. Truth tables

Formalizing arguments is just a first step. In order to determine their validity, the
formulae have to be analyzed in connection with truth and falsity. The truth value of
molecular sentences depends on the truth value of its atomic sentences and the relation
established between them by the connectives. Any atomic sentence has two possible
truth values, it is either true (we will use T or 1 for truth) or false (we will use F
or 0 far falsity) but not both (in some systems of logic more truth values are possible,
but we are working with a bivalent logic). The truth value of an atomic sentence can be
specified by means of a table:
A
1
0
This table means that any atomic sentence (A) can be true (1) or false (0). We
dont know if it is one or the other (again, logic has no tools to determine the truth of a
sentence, only the transmission of truth between sentences), this depends on its meaning
(and we dont know it or we eliminated it through formalization). But we know for sure
that A has to be either 1 or 0.
The truth value of any atomic sentence is altered by connectives, and in fact we
can define the meaning of any connective by means of a truth table. For instance,
negation has the property of altering the truth value of any sentence: if one sentence is
true then its negation is false, if the sentence is false then its negation is true. Again we
can summarize this in a characteristic truth table for negation:
A A
1
0
0
1
With molecular sentences there are more combinations since the truth value of
the whole sentence depends on the truth values of its atomic parts. Each combination of
the truth values of the atomic sentences is tied to a specific truth value of the molecular
sentence, as in the truth table for conjunction:
A
1
1
0
0

B AB
1
1
0
0
1
0
0
0

10

This table means that a conjunction is true only when all of the conjuncts are
true, and false under any other circumstance. Each truth table works as a definition of a
connective, and we can build a characteristic truth table for any connective. Here is the
truth table for disjunction:
A
1
1
0
0

B AB
1
1
0
1
1
1
0
0

A disjunction is true when either one or both of the disjuncts are true. This truth
table allows us to see clearly that the disjunction we are using is an inclusive disjunction
(or means this or that or both) and not and exclusive disjunction (this or that but
not both), whose symbol is
The truth table for conditional:
A
1
1
0
0

B AB
1
1
0
0
1
1
0
1

This truth table is not obvious, but it is important to remember that when we say
If it rains, then the pavement will be wet we are only asserting that it is not possible
that it was the case that it rained and the pavement wouldnt be wet, the sentence affirms
that rain entails necessarily a wet pavement. Nevertheless the pavement could be wet for
any other reason (it was hosed down), so it is possible for the consequent (the pavement
is wet) to be true if the antecedent (it rains) is false, and also both could be false. What
the conditional establishes is that it is not possible for the antecedent to be true and its
consequent be false, so it is false when this happens and true under any other
circumstance.
The truth table for biconditional:
A
1
1
0
0

B AB
1
1
0
0
1
0
0
1

Because we could always write (AB)(BA) instead of AB, we do not

strictly speaking need to introduce a new symbol for the biconditional. Nevertheless,
logical languages usually have such a symbol. SL will have one, which makes it easier
to translate phrases like `if and only if.' AB is true if and only if A and B have the
same truth value.

11

How to build the truth table of a complex argument? The truth-value of

sentences that contain only one connective is given by the characteristic truth table for
that connective, but most arguments will have more than one connective. Consider the
sentence (pq)p . We consider all the possible combinations of true and false for p
and q, which gives us four rows.
p
1
1
0
0

q (pq)p
1
0
1
0

Now consider the subsentence (pq), this is a conjunction AB with p as A and

with q as B. p and q are both true on the first row. Since a conjunction is true when both
conjuncts are true, we write a 1 underneath the conjunction symbol. We continue for the
other three rows and get this:
p
1
1
0
0

q (pq)p
1
1
0
0
1
0
0
0

The entire sentence is a conditional AB with (pq) as A and with p as B. On

the second row, for example, (pq) is false and p is true. Since a conditional is true
when the antecedent is false, we write a 1 in the second row underneath the conditional
symbol. We continue for the other three rows and get this:
p
1
1
0
0

q (pq)p
1
1 1
0
0 1
1
0 1
0
0 1

The column of 1s underneath the conditional tells us that the sentence (pq)p
is true regardless of the truth-values of p and q. They can be true or false in any
combination, and the compound sentence still comes out true. It is crucial that we have
considered all of the possible combinations. If we only had a two line truth table, we
could not be sure that the sentence was not false for some other combination of truthvalues. The truth-value of the sentence on each row is just the column underneath the
main logical operator of the sentence; in this case, the column underneath the
conditional.
A complete truth table has a row for all the possible combinations of 1 and 0 for
all of the sentence letters. The size of the complete truth table depends on the number of
different sentence letters in the table. A sentence that contains only one sentence letter
requires only two rows, as in the characteristic truth table for negation. This is true even
if the same letter is repeated many times, the complete truth table requires only two
lines because there are only two possibilities. A sentence that contains two sentence
letters requires four lines for a complete truth table, as in the characteristic truth tables
12

and the table for (pq)p. A sentence that contains three sentence letters requires eight
lines. A complete truth table for a sentence that contains four different sentence letters
requires 16 lines. Five letters, 32 lines. Six letters, 64 lines. And so on.
In order to fill in the columns of a complete truth table, begin with the rightmost
sentence letter and alternate 1s and 0s. In the next column to the left, write two 1s, write
two 0s, and repeat. For the third sentence letter, write four 1s followed by four 0s. For a
16 line truth table, the next column of sentence letters should have eight 1s followed by
eight 0s. For a 32 line table, the next column would have 16 1s followed by 16 0s. And
so on.
The interest of truth tables is to find valid arguments, and more specifically
to recognise formal truths (sentences which are true regardless its content). With a
complete truth table, we consider all of the ways that the world might be. If the sentence
is true on every line of a complete truth table, then it is true as a matter of logic,
regardless of what the world is like. As was said before: logical truths are true in any
possible world.
So a sentence is a tautology in SL if the column under its main connective is 1
on every row of a complete truth table.
Conversely, a sentence is a contradiction in SL if the column under its main
connective is 0 on every row of a complete truth table.
A sentence is contingent in SL if it is neither a tautology nor a contradiction; i.e.
if it is 1 on at least one row and 0 on at least one row.
Also, we can use truth tables to find logically equivalent sentences or to prove
the validity of an argument.
Two sentences are logically equivalent in SL if they have the same truth-value
on every row of a complete truth table.
An argument is valid if it is logically impossible for the premises to be true and
for the conclusion to be false at the same time. An argument is valid in SL if there is no
row of a complete truth table on which the premises are all 1 and the conclusion is 0; an
argument is invalid in SL if there is such a row.

III.4 Proofs and laws of argumentation

Consider two arguments in SL:
Argument A

Argument B

pq
p
q

pq
p
q

13

Clearly, these are valid arguments. You can confirm that they are valid by
constructing four-line truth tables. Argument A is [(pq)p]q and Argument B is
[(pq)p]q. Argument A makes use of an inference form that is always valid: given
a disjunction and the negation of one of the disjuncts, the other disjunct follows as a
valid consequence. This rule is called disjunctive syllogism (DS). Argument B makes
use of a different valid form: given a conditional and its antecedent, the consequent
follows as a valid consequence. This is called modus ponens (MP).
When we construct truth tables, we do not need to give names to different
inference forms. There is no reason to distinguish modus ponens from a disjunctive
syllogism. For this same reason, however, the method of truth tables does not clearly
show why an argument is valid. If you were to do a 1028-line truth table for an
argument that contains ten sentence letters, then you could check to see if there were
any lines on which the premises were all true and the conclusion were false. If you did
not see such a line and provided you made no mistakes in constructing the table, then
you would know that the argument was valid. Yet you would not be able to say anything
further about why this particular argument was a valid argument form.
The aim of a proof system is to show that particular arguments are valid in a way
that allows us to understand the reasoning involved in the argument. We begin with
basic argument forms, like disjunctive syllogism and modus ponens. These forms can
then be combined to make more complicated arguments, like this one, Argument C:
[(l(jl))l]j
The proof system we will develop is called a natural deduction system.
Formally, a proof is a sequence of sentences. The first sentences of the sequence are
assumptions; these are the premises of the argument. Every sentence later in the
sequence follows from earlier sentences by one of the rules of proof. The final sentence
of the sequence is the conclusion of the argument. For natural deduction Argument C
would be presented in this way to be proved (were - marks the premises and
marks the conclusion):
-1 l(jl)
-2 l

And here is the proof by natural deduction of Argument C:

-1 l(jl)
-2 l
3 jl
j

(MP 1,2)
(DS 2,3)

By modus ponens (MP), 1 and 2 entail jl (we know 1 and 2 as premises, as

indicated by - before the numbers). This is an intermediate conclusion. It follows
logically from the premises, but it is not the conclusion we want. Now jl (3) and 2
entail j, by disjunctive syllogism (DS). We do not need a new rule for this argument.
The proof of the argument shows that it is really just a combination of rules we have
already introduced.
In a natural deduction system, there will be two rules for each logical operator:
an introduction rule that allows us to prove a sentence that has it as the main logical

14

operator and an elimination rule that allows us to prove something given a sentence that
has it as the main logical operator. When we add a line to a proof, we write the rule that
justifies that line and we also write the numbers of the lines to which the rule was
applied (as weve made with Argument C).
BASIC RULES
Conjunction Introduction (I or Conjunction (Conj.)

This rule says that if on one line we have written a truth,

and on another line we have another one, also true, then we can
write in just a line that both things are true (the conjunction of two
true sentences is itself true). Remark that picking the lines reversed,
you can obtain BA, and picking the same line you can achieve
AA and BB, which are also true (but trivial).

Conjunction Elimination (E) or Simplification (Simp.)

Since a conjunction is true if both conjuncts are true, knowing the truth of a conjunction we can
assert the truth of each of the conjuncts.

Conditional Introduction (I) or Conditional Proof (CP)

This proof allows doing something useful with hypothesis

(those sub-demonstrations which have a vertical bar to the left. It
means that if we supposed something (call it A), and we just
discovered (by using the rules) that supposing A made true B
(whatever it is), then we have something clear: we cant assure that
B always is true, but we can assure that A implies B, which is
written AB. This allows us to end the sub-demonstration and
continue working with what we were doing before. Remember that
you cant finish natural deduction inside a hypothesis.

Conditional Elimination (E) or Modus Ponens (MP)

Simply, if we are told that when A also happens B (thats

what it means AB), and they also tell us that now happens A,

then we can assure that B.

Biconditional Introduction (I)

15

In order to derive p
q, for instance, you must
be able to prove q by
assuming p and prove p by
assuming
q.
The
biconditional
introduction
rule (I) requires two
subproofs. The subproofs
can come in any order, and
the second subproof does not

need to come immediately after the first. The rule can be simplified
like this:

Biconditional Elimination (E)

Disjunction Introduction (I) or Addition (Ad.)

If we know that Its Thursday we also know that Its Thursday or cows can fly, Its
Thursday or Friday, or even Its Thursday... or not. All of them are true since we are using an inclusive
disjunction which is true as far a one of the disjuncts is true.

Disjunction Elimination (E) or proof by Cases (Cas.)

C
This is the most complicated rule, mainly because if we
are given a phrase with "or", like Its Thursday or Friday, what
can we deduce from it? That its Thursday? No, it may be Friday.
That its Friday? No, it may be Thursday. That its Thursday or
Friday? Well, yes, but we already knew that... We need more
information besides the AB. If, luckily, we happen to know AC,
and also BC, then we do know what happens when AB: both
one option and the other drive us to C, so C is true. This derivation
rule is called proof by cases, since we have to check each possible
case to see that they all involve the same conclusion.

Negation Elimination (E) or Double Negation (DN)

The negation changes the truth value of a sentence, so if

one negation makes A false, denying the negation of A will make it
true.

Negation Introduction (I) or Reductio ad Absurdum (Abs.)

16

that both B and B are true, youre not lost, since you just
discovered another truth: that its not possible for A to be true,
thats it, A its true. This procedure is called reduction to the
absurd (reductio ad absurdum): suppose something to achieve a
contradiction and be able to assert that what we supposed is false.
Its specially useful if you start supposing the contrary of what you
want to prove: if any contradiction can be discovered, then its
almost all done.

If after supposing A,
you achieved the conclusion

DERIVED RULES
Modus Tollens
(MT)

Hypothetical
Syllogism (HS)

Disjunctive
Syllogism (DS)

Ex
Contradictione
Quodlibet
(ECQ)
A

Unlike the method of truth tables, there is no (simple) method for constructing a
derivation of a valid sequent. Constructing derivations requires a certain amount of
imagination and creativity. (There is, in fact, a method, but it requires listing all possible
derivations in a mechanical way, and then selecting the one which works. This is very,
very inefficient, but possible.) However, Paul Tomassi in his textbook Logic presents the
following strategy, which he calls The Golden Rule:
You are presented with a sequent of the form:
A1, ., An B
where A1, , An are the premises and B is the conclusion.
Golden Rule:
1. Is the main connective of the conclusion the conditional ? If so, apply the strategy
for CP. I.e., assume the antecedent and try to derive the consequent. If not, ask:
2. Is the main connective of one of the premises the disjunction ? If so, apply the
strategy for Cas. I.e., assume each disjunct separately, and try to derive the conclusion
B. If not:
3. Try Abs. I.e., assume the negation of the conclusion and attempt to derive a
contradiction.
Tomassi adds never lose sight of the fact that each and all of the above
strategies can work together in a single proof. I.e., the pursuit of an overall strategy may
necessitate a sub-proof which itself requires a different strategy. Hence, apply the
Golden Rule at the outset to identify an overall strategy, and then reapply as necessary
throughout the process of proof construction.
17

III.5. Formal and informal fallacies

Fallacies are misleading arguments or, to be more precise, unsound arguments
which nevertheless look sound. Many of them are so tempting, and therefore so
common, they even have their own names. To call something a fallacy is usually only
another way of saying that it violates one of the rules for good arguments. To
understand fallacies, then, you need to understand what rules they break, and what their
defects are. If the defect is intentional usually the fallacy is also called a sophism.
III.5.1. Formal fallacies
Formal fallacies are those which involve an error in the form, that is, those
which are invalid arguments (the conclusion does not follow from the premises) no
matter the truth or falsity of the conclusion.
Affirming the consequent: a deductive fallacy of the form
pq
q
p

The second premise of a modus ponens asserts the antecedent. Affirming the
consequent, though, yields an invalid form. A true conclusion is not guaranteed even if
the premises are true. For example:
When the roads are icy, the mail is late.
The mail is late.
Therefore, the roads are icy.

Although the mail would be late if the roads were icy, it also may be late for
other reasons. This argument overlooks alternative explanations.
Denying the antecedent: a deductive fallacy of the form
pq
p
q

The second premise of a modus tollens denies the consequent. Denying the
antecedent, however, yields an invalid form. A true conclusion is not guaranteed even if
the premises are true. For example:
When the roads are icy, the mail is late.
The roads are not icy.
Therefore, the mail is not late.

Although the mail would be late if the roads were icy, it also may be late for
other reasons. This argument, again, overlooks alternative explanations.
III.5.2. Informal fallacies

18

Informal fallacies are a matter of unclear expression, they deal with the logic of
the meaning of language and cant be formalized properly. Informal fallacies involve
such things as the misuse of language, ambiguity, misstatements of fact or opinion,
misconceptions due to underlying presuppositions or just plain illogical sequences of
thought.
Ad hominem (against the man): attacking the persons authority rather than his or her
qualifications or reasons.
The philosophy teacher claims that its important to study philosophy and to live an examined
life because otherwise we would just follow previously set tracks without ultimate freedom, but what
would you expect a philosophy teacher to say?
St. Thomas defends that God exists because the Universe needs a first cause without which
there would be an infinite regress, but he was a monk.

Ad ignorantiam (appeal to ignorance): arguing that a claim is true just because it has
not been shown to be false.
Nobody has ever proved that there is no life outside the Earth, therefore aliens exist.

Ad misericordiam (appeal to pity): appealing to pity as an argument for special

treatment.
I know I flunked every exam, but if I don't pass this course, I'll have to retake it in summer
school. You have to let me pass!

Pity is not always a bad argument, but it is certainly inappropriate when

objective evaluation is called for.
Ad populum (appeal to the people): appealing to the emotions of a crowd, also,
appealing to a person to go along with the crowd.
Everyone's doing it!
People loves junk TV, so there ought to be more.
"I should avoid paying taxes since everybody does the same."

Ad populum is a good example of a bad argument from authority, no reasons are

offered to show that "everybody" is an informed or impartial source.
Begging the question / circular argument / petiquio principii: implicitly using your
conclusion as a premise.
God exists because it says so in the Bible, which I know is true because God wrote it, after all!

To write this argument in premise-and-conclusion form, you'd have to write

The Bible is true, because God wrote it.
The Bible says that God exists.
Therefore, God exists.

To defend the claim that the Bible is true, the arguer claims that God wrote it.
But, obviously, if God wrote the Bible, God exists. Thus the argument assumes just
what it is trying to prove.

19

Complex question: posing a question or issue in such a way that people cannot agree
or disagree with you without committing themselves to some other claim you wish to
promote. A simple example:
"Are you still as self-centered as you used to be?"

Answering either "yes" or "no" commits you to agreeing that you used to be
self-centered. More subtle example:
"Will you follow your conscience instead of your pocketbook and donate to the cause?"

Saying "no," regardless of their real reasons for not donating, makes people feel
guilty; saying "yes," regardless of their real reasons for donating, makes them noble.
Equivocation: using a key term or phrase in an argument in an ambiguous way, with
different meanings in different portions of the argument.
The laws imply lawgivers. There are laws in nature. Therefore there must be a cosmic
lawgiver.

False cause / incomplete generalization: generic term for a questionable conclusion

about cause and effect, usually generalizing from incomplete information. One of our
most common temptations is to draw conclusions from too little evidence. For example,
if the first Lithuanian I meet has a fiery temper, I might jump to the conclusion that all
Lithuanians have fiery tempers. If one ship disappears in the Bermuda Triangle, the
National Enquirer proclaims the Bermuda Triangle haunted.
False dilemma: reducing the options you consider to just two, often sharply opposed
and unfair to the people the dilemma is posed against. For example:
"Vote for me, or the country will be dragged into chaos."

Here is a more subtle example from a student paper:

"Since the universe could not have been created out of nothingness, it must have been created by
an intelligent life force. . . ."

Is creation by an intelligent life force the only other possibility? False dilemmas
often include loaded language; they also, obviously, overlook alternatives.
Loaded language: making an argument look good by mocking or distorting the other
side, using language whose only function is to sway the emotions (loaded language).
So, you defend eugenics so did the Nazis.
Pro-choicers defend assassination which is contrary to Human Rights.

Non sequitur: drawing a conclusion that "does not follow", that is, a conclusion that is
not a reasonable inference from the evidence. Very general term for a bad argument.
The "person who" fallacy: using reports of a few friends or one or two experiences of
our own to outweigh the careful summary and comparison of thousands records or
experimental research. To judge a set of examples, then, we often need to consider
background rates.

20

"I know a person who smoked three packs of cigarettes a day and lived to be 100..

Persuasive definition: defining a term in a way that appears to be straightforward but

that in fact is loaded. For example, Ambrose Bierce, in The Devil's Dictionary, defines
"faith" as "belief without evidence in what is told by one who speaks without
knowledge, of things without parallel." Persuasive definitions may be favorably loaded
too: for example, defining "conservative" as "someone with a realistic view of human
limits."
Poisoning the well: using loaded language to disparage an argument before even
mentioning it.
I'm confident you haven't been taken in by those few holdouts who still haven't outgrown the
superstition that. . .

More subtle:
No sensitive person thinks that

Post hoc, ergo propter hoc (literally, "after this, therefore because of this"): assuming
causation too readily on the basis of mere succession in time, establishing a necessary
connection where there is none.
I passed the exam wearing my red shoes, so they give me luck.

Red herring: introducing an irrelevant or secondary subject and thereby diverting

attention from the main subject.
"I think that we should make the academic requirements stricter for students. I recommend that
you support this because we are in a budget crisis and we do not want our salaries affected."

Usually the red herring is an issue about which people have strong opinions, so
that no one notices how their attention is being diverted. In a discussion of the relative
safety of different makes of cars, for instance, the issue of which cars are made in your
country is a red herring.
Straw man: caricaturing an opposing view so that it is easy to refute.
A. Sunny days are good.
B. If all days were sunny wed never have rain, and without rain wed have famine and death.

Or:
"Leftists want to steal all of your properties."

Weasel word: changing the meaning of a word in the middle of your argument so that
your conclusion can be maintained, though its meaning may have shifted radically.
Usually a maneuver performed under the pressure of a counterexample.
A. All studying is torture.
B. What about studying argument? You love that!
A. Well, that's not really studying.

21

Here "studying" is the weasel word. A's response to B's objection in effect
changes the meaning of "studying" to "studying that is torture." A's first statement
remains true, but only at the cost of becoming trivial ("All studying that is torture is
torture.").
III.5.3. The naturalistic fallacy
There are two fundamentally different types of statement: statements of fact
which describe the way that the world is, and statements of value which describe the
way that the world ought to be. The naturalistic fallacy is the alleged fallacy of inferring
a statement of the latter kind from a statement of the former kind.
Arguments cannot introduce completely new terms in their conclusions. The
argument, (1) All men are mortal, (2) Socrates is a man, therefore (3) Socrates is a
philosopher is clearly invalid; the conclusion obviously doesnt follow from the
premises. This is because the conclusion contains an ideathat of being a philosopher
that isnt contained in the premises; the premises say nothing about being a
philosopher, and so they cannot establish a conclusion about being a philosopher.
Arguments that commit the naturalistic fallacy are arguably flawed in exactly the
same way. An argument whose premises merely describe the way that the world is, but
whose conclusion describes the way that the world ought to be, introduce a new term in
the conclusion in just the same way as the above example. If the premises merely
describe the way that the world is then they say nothing about the way that the world
ought to be. Such factual premises cannot establish any value judgement; you cant get
an ought from an is as shown in this example:
Feeling envy is only natural.
Therefore theres nothing wrong with feeling envy.

This argument moves from a statement of fact to a value judgement, and

therefore commits the naturalistic fallacy. The arguments premise simply describes the
way that the world is, asserting that it is natural to feel envious. To describe the way that
the world is, though, is to say nothing of the way that it ought to be. The arguments
conclusion, then, which is value judgement, cannot be supported by its premises.
Bibliography:
Daniel Clemente Laboreo. Introduction to natural deduction. FDL, 2005.
Paul Tomassi. Logic. London: Routledge, 1999.
Anthony Weston. A rulebook for arguments. Indianapolis: Hacket Publishing, 2000. Section II.5 is mostly extracted
from here.
P. D. Magnus. An introduction to formal logic. CC, 2010. Most of the contents of this notes are extracted from here.

22