Notes on Arguments, Truth and Validity ARTS2374 Ways of Reasoning

Recognizing Arguments
Just what exactly is the point of what we are doing here in this subject? One crucial thing is that we are
developing a model of language - a model which has very clear and well defined properties. The aim
of the model is to allow us to investigate the properties of the language we use which does not have
such clear and well defined properties. So our primary aim is to understand our own language use.
Logic, thought of in this light, is part of the endeavor of understanding ourselves and that endeavor is
what I think philosophy is in a nutshell.

There are many cases in which logical reasoning leads us to change our beliefs. Adding a piece of
information to our beliefs can lead to more commitments.

A newcomer to a village is speaking to the elderly village priest. “Well” says the priest, “this
may seem a sleepy village now, but I can tell you there were some happenings in the bad old
days. In fact, the very first confession I heard here was for a murderer!!!” Surprised and a little
shocked the newcomer makes his way to the local pub for some sustenance, to be welcomed by a
cheery rosy cheeked publican. “Oh” says he “it might interest you to know that I knew him as a
callow youth. When he first arrived as the new priest. I was the first one to take confession from
him.”

Now our newcomer is in a position to learn something that neither of his interlocutors told him.
His in a position to reason as follows.

The priest’s first confession was by a murderer.

The publican was the priest’s first confession.
So, The publican is a murderer.

Things we do with language

A very quick reflection about what we do with language results in distinguishing many different sorts
of activity:
Questions “When do tutorials begin?”
Proposals “Let’s not forget to study for exams this semester.”
Suggestions “Try doing a bit of study each week this time”
Commands “Jump three times then turn around and do it again.”
Exclamations “Boo Carlton!!!” or “Yeah Collingwood!” or “Ow!!”

Notice that none of these is true or false. They play a different role for us in our use of language than
do assertions. They do not function for us as descriptions of the world, as reports of how things are.

They do not assert anything. There are people who have worked on the logical properties of these
ways of using language. For us, we will stick with assertoric language, language which is used to
assert something.

Arguments are collections of assertoric sentences. (Declarative sentences)

It is important to note that we are sticking to these uses because modeling them is easier, not because it
is not possible to talk of the logic of questions, commands, or any of the other things we do with
language. It is possible to investigate the logic of these categories but more complicated.

Apart from arguments we can use declarative statements which can be true or false in other ways.
Among them are the following ways:

Loosely associated statements “Tuna are warm blooded. Some squid grow to immense size.”
Reports “A young male was seen leaving the scene. He was later arrested”
Exposition “When people talk of rights it is important to distinguish moral
and legal notions. The notion of a legal right is much more
straightforward than is the notion of a moral right.”
Explanation “I went to the bank to get some money. Finding I had none in
there, I proceeded to kick the ATM”
Illustrations, e.g.. “Not all fish are cold blooded. For example, tuna are warm
blooded.”
Conditional Statements “If cows flew, we’d all carry big umbrellas.”

Constructing an argument is just one of these but it is a fascinating part of the way we use language to
get ourselves around the world. Usually putting forward an argument is done in a social context. We
put forward arguments to produce certain effects in others. However, this is not the only way we use
arguments. I might be thinking hard about what course of action to follow and might find that
clarifying the considerations for each course of action the most sensible way to proceed. Or I might
wonder what to believe about a certain matter and find that the best way to proceed is to lay out clearly
the arguments in favour of the different positions.

In normal speech we use words like:

since in that seeing that
as indicated by as for
because given that owing to

to indicate what follows is a premise or is being offered as a reason to believe something else.

We also use other words to indicate conclusions:

therefore so entails that
thus hence consequently
for that reason it must be that as a result

So what do we do with language? The thing is we do many different sorts of things with language.
Here is a short and very incomplete list of things we do with language:
There are many things we do with language. I leave it to you to decide whether in performing one of
the following we use declarative sentences which are either true or false or not.
describe command request deny
warn accuse recommend answer a question
make a promise ponder lie to someone bullshit someone
explain order a pizza tell a joke give directions
advise compose a story tell a story say a poem
sing a song suggest a restaurant make a shopping list ask for help
Some Elementary Remarks About Truth and Validity
A prophet from their own people said of them “Cretans are always liars, wicked brutes, lazy
gluttons." This testimony is true.
St. Paul, Titus 1:12-14

Truth is a notion as tricky and as simple as any philosophical notion can be.

Truth and falsehood are properties of sentences. It is sentences which are true and which are false.

This is contentious. First notice that many sentences do not have a truth-value absolutely. Their truth-
value may change over time, or between contexts. So for example,
Michaelis is sitting.
may be currently true, was recently false, and will be false again. The sentence does not have a truth-
value for all time. Moreover, surprising enough there may be many people with that name. The
sentence does not even have a truth value as such, it depends also upon the person to whom the name
refers? So when we are talking about a sentence being true we are talking about the sentence fully
interpreted, with all the bits actually meaning something.

Some philosophers prefer to think of truth as indirectly possessed by sentences. These philosophers
think that truth is rather a direct possessed by what sentences express, what they mean: propositions.
Some others think it is the act of assertion, which uses a sentence, which is properly speaking true or
false. Right now we won’t have to adjudicate on this debate. It is a subtle and important debate with
very significant figures on both sides. But all of these philosophers will accept that it is appropriate to
talk of sentences being true. The open question is whether that is a property that sentences have
mediately or immediately.

For the moment we will treat fully interpreted unambiguous sentences as either true or false and just
one of these.

No sentence is both true and false. — The principle of non-contradiction.

Every sentence is either true or false. — The principle of bivalence.

These assumptions are going to be used for the first part of the course. It is important to realise that we
don’t really need to make these assumptions and for some philosophical reasons we might argue
against them.

Some philosophers have suggested that there really are true contradictions. Not cheap contradictions
like might be offered by a relativist, that abortion is both right and not right according to different
moral systems. That is no more a contradiction than is the pair of claims John is speaking to Jane and
Bill is not speaking to Jane. There is no threat of contradiction here. Different people stand in the
relation of speaking to Jane. There is no contradiction in that situation than there is when we say that it
is and isn’t raining in Australia right now, when it is raining in Sydney and brilliantly sunny in
Melbourne. There is not a contradiction there at all. Similarly with vagueness. Sometimes we say “It is
and isn’t raining right now” by which we mean that this is a vague matter. It is raining to some extent
but not enough to be really called ‘rain’.
A further principle to do with truth is an important one which takes us from mentions of a sentence to
the sentence itself.

For each sentence s, “s” is true if and only if s. – The principle of disquotation.

This principle says that there is no difference between asserting that snow is black and saying that the
sentence “snow is black” is true. We can go from one of these to the other with no change of
commitment. And yet notice, this is not trivial. The one is about a sentence having a property, the
second is not about language at all but about snow. For the principle to be plausible the sentences we
are talking about being true have to be sentences of our language. So for example the statement.

“Il pleut” is true if and only if il pleut.

is not itself grammatical. The phrase which follows the ‘if and only if’ is not part of our language. If I
replace that foreign phrase with its translation into English the biconditional becomes

“Il pleut” is true if and only if it is raining.

which is true but not the sort of a priori truth guaranteed by the principle of disquotation. It is not a
priori because it is not a priori that “Il pleut” translates to it is raining.

The thought of some philosophers is that there are some sentences which are both true and false in the
very same way. One example used to make this point is the famous liar paradox. The following
sentence states the paradox:

This sentence is false.

If this sentence is true then what is says must obtain so it must be false. But if it is false, then things
are as it says so it must be true. So this looks like a real contradiction. No prospect here of saying that
it is in different respects that the sentence is true and false, nor is there a prospect for saying that there
is vagueness involved.

St. Paul says of Epimenides that he speaks truly when he says that all Cretans are liars.

So, Epimenides, a Cretan, says “All Cretans are liars”.

But, if he is speaking truly, then all Cretans are liars, so he is a liar, so he is not speaking truly.

But if he is not speaking truly, then he is lying and he is speaking truly. (Well, not quite, he might be
lying, by saying all Cretans are liars, when in fact he is the only one!!)

This shows us that whichever way we take Epimenides’s claim it has to be false. But then St.Paul is
mistaken when he says “This testimony is true.”

This discussion has been called the “Liar Paradox”.

A simpler version of the Liar paradox
“This sentence is six words long.”
That sentence is true!!

“This sentence is seventeen words long.”

That sentence is false.

Nothing weird happening yet….

But now notice what happens when we look at the next example:

“This sentence is false.”

If it is true then it is false and if it is false then it is true. So it cannot have any truth value or it must
have both. If the world is as this sentence says, then it itself is false. But if the world is as it says, then
it is true, so if the world is as it says then it is both true and false. On the other hand there is an
intuition that says that the sentence doesn’t say how the world is at all. For that reason it can’t be true
or false.

So what should we say about truth and falsity? There is a great deal more to say. But not here. Perhaps
later. For now we rest with the idea of bivalence, and of non-contradiction and further a principle of
disquotation.

Strengthened Liar Paradox

A common thought is to think that the liar sentence “This sentence is false” is somehow incomplete:
that it is wrong to think that it is true and wrong to think that it is false. One suggestion is that it is
neither true nor false. On this suggestion there are some sentences which are true, some which are
false and some which because of various infirmities are neither true nor false.

Another sort of infirmity which leads to a sentence being neither true nor false is sometimes suggested
to be presupposition failure. For example, suppose there John has no children. The sentence:

All of John’s children are asleep.

seems to suffer from presupposition failure and has been suggested to be neither true nor false.

So what about saying that the same is true of the liar sentence? The idea is interesting. It means that
the sentence

This sentence is false.

is not false and not true. This maneuver blocks the argument above that makes the liar sentence both
true and false. Is this a solution to such paradoxes? No. It is clear that it is not a general solution. Now
consider the sentence:

This sentence is not true.

This sentence is known as the strengthened liar sentence. Suppose we go along with the suggestion
and consider it as neither true nor false. Notice then that the sentence is, as it says, not true. So if it is
neither true nor false, it is true. But if it is true, then it has to be false since it says it is not true. So if it
is neither true nor false then it is both true and false. This is a paradox.

Validity and Soundness

Validity is a property of arguments.

An argument is valid exactly when it is not possible for the premises of the argument to be true
and the conclusion false.

Alternatively, and equivalently, we can use Aristotle’s own formulation: an argument is valid just
when it is necessary that if the premises are true then the conclusion is true.

Colloquially we often say things like “What you say is valid” speaking of particular sentences people
have asserted. This is not the usage we have in mind. We are focussing on the notion of validity in
terms of what follows from what.

This is essentially a relational matter. As Aristotle points out (in the following excerpt from his Prior
Analytics) whether an argument is valid does not depend on whether the premises are actually true. It
depends on whether the conclusion would have to be true if the premises were true.

All Tuna are fish.

All fish are cold blooded.
So, All Tuna are cold blooded.

This argument is valid. It is just not possible for the premises to be true and the conclusion false. Now
you can know this argument is valid without knowing whether or not the premises are actually true. In
fact if the conclusion is discovered to be false. (Which it is!!) This leads us to see that at least one of
the premises must also be false.

As Aristotle discusses arguments which are valid deductions can have false premises and a true
conclusion. In that case we do not say that the deduction establishes the conclusion even though it
follows from the premises. It is not established because you are not forced to accept the premises so
are not forced to accept the conclusion.

All Japanese are Europeans.

Napoleon was Japanese.
So, Napoleon was European.

In wondering about invalidity, the question is whether the we would be inconsistent if we accepted the
premises and not the conclusion.

Sometimes a conclusion is so absurd that we cannot accept it. In that case the burden of proof is on us
to show that we don’t accept all the premises or that we can say why the premises don’t lead to the
absurd conclusion we can’t accept.

An argument is valid just when if the premises were true then the conclusion would be also.
Putting it another way, it is not possible for the premises to be true and the conclusion false.

But then, valid arguments might not have true premises?

That is right. Validity is about what would happen if the premises were true, not about their actual
truth value.

It is about the relationship between the premises and the conclusion not about actual truth.

A sound argument is a valid argument with true premises.

So a sound argument must have a true conclusion. Why?

Deduction and Induction

A deductive argument is one in which the arguer claims it is impossible for the premises to be true and
the conclusion false.

An inductive argument is one in which the arguer claims it is improbable for the premises to be true
and the conclusion false.

In a valid deductive argument the premises being true necessitate the truth of the conclusion, it is
necessary that if the premises are true then so is the conclusion. In this way in a valid deductive
argument the premises suffice for the truth of the conclusion.

In a successful inductive argument the premises being true make the truth of the conclusion probable.

Deciding whether an argument is put forward as a deductively valid argument or just as an inductive
argument is not always clear. Some bad deductive argument work well enough as inductive
arguments.
Consider this argument
If the patient has the disease then they will have a fear of water.
The patient has a fear of water.
Therefore, The patient has the disease.

This is a deductively invalid argument. To see that, you can well enough imagine cases where the
conditional is true, that anyone with the disease will have a fear of water. And where the patient has a
fear of water but this is because of an accident in a pool as a child and the patient does not have the
disease. So in fact the premises are true and the conclusion is false.

The fact that it is an invalid deductive argument does not mean that there is no role for this argument
to play in our reasoning. If this situation where the premises are true and the conclusion false are rare
enough, that is, if the premises are true, it makes the likelihood of the conclusion quite high. So it
might be that if the premises are true, this would provide good evidence to accept the conclusion, even
if it is not a conclusive reason.
Another argument of the same form makes a related point.
If it has been raining recently, then the road will be wet.
The road is wet.
Therefore, It has been raining recently.
This is also an invalid argument. We can readily enough imagine cases of people out with their hoses
wetting the road. But in circumstances with no hoses, or droughts , water restrictions and law abiding
citizens, we can see how the wetness of the road might indeed be very good evidence for the claim that
it has recently rained.
Often when we have conditionals such as that given in the last example and where the premise “The
road is wet” does work as evidence for the conclusion that it has been raining lately, we are committed
to more than the conditional given in the example. For example we might be committed to the claim
that the road’s being wet and its having rained lately march in step, when one is true so is the other.
This is the idea that the claim
The road is wet just when it has been raining recently.
or to put it in a slightly more formal way
The road is wet if and only if it has been raining recently.
Now notice that the argument
The road is wet if and only if it has been raining recently.
The road is wet.
Therefore, It has been raining recently.
is actually valid. We might have qualms about how true the first premise is but if we have evidence for
it and we can use it then the argument shows that it the premises are true then the conclusion will be
too.
The Language/Meta-language Distinction

A language used to discuss a language is a metalanguage with respect to that language, its object
language. We will use a metalanguage, ML, namely the language you are reading now, to talk about a
formal language L we will define for our propositional logical system and to introduce you to some of the
logics we can build around L. Some but not all metalanguages contain their own object languages as
sublanguages. When we talk about French in English, English is the metalanguage and French is the
object language. In that case the object language is not contained in the metalanguage, even though we can
have names for various expressions of the object language in the metalanguage, French sentences are not
English sentences.

“Il est Jules” is a French sentence, referred to by use of an English name of that sentence. I am using an
English name of a French sentence which is only mentioned in the previous sentence. ““Il est Jules”” is
the English name of a French sentence and not a French sentence. The French sentence is mentioned not
used. Being careful about distinguishing object and metalanguage very often is not important. For us it
will indeed be important because we will be interested in what can be said in a language.

We will abide by the convention that placing a bit of language inside a pair of quotation marks forms a
name of the original bit of language. So John is a person, “John” is the name of that person, ““John”” is
the name of the name of the person, etc. When we use “John” we are talking about John, When we
mention “John” we are talking about the name of John. Observing the use/mention distinction is very
important when talking about language.

Exercise (Do this in the First Tutorial)

Some of the following are false, or senseless. Make the smallest number of additions of quotation marks to
make them true (if possible). These questions are inspired by a similar set by Richard Cartwright.

1. Melbourne is south of Sydney, but Sydney is not north of Melbourne.

2. The last word of the answer to the previous question is Melbourne.

3. Quine’s friends called Quine Van.

4. The last word of #4 is obscene.

5. The last word of #4 is obscene.

6. Stretch was called Stretch because of his height.

7. John was called Stretch because of his height.

8. Stretch was so-called because of his height.

9. Galileo said the earth moves.

10. Galileo wrote the earth moves.

11. Galileo, who spoke no English, said that the earth moves.