Debunking Classical Pseudo-Paradoxes of Logic

Refuting the Liar's Paradox and Russell's Paradox: An Explanation for the General Reader

Introduction

There are constructions that, in courses on logic and philosophy, are presented as "profound paradoxes" capable of shaking the very foundations of mathematics and rational thought. A student encounters them and feels something like intellectual helplessness: the universe seems to crack at the seams, and a vague, inexplicable threat makes the hair stand on end.

The purpose of this article is to show that there is no paradox. There is a substitution of concepts - one that is easy to detect if you apply the most ordinary definitions of logic: the very ones introduced in the first lectures.

The explanation is deliberately structured to be accessible to someone without specialist training. Not because the subject is "too complex to formalize," but because the correct explanation is a simple one.

Part I. What Is a Proposition?

Before we can speak of paradoxes, we need to fix one definition - the central one for all classical logic.

A proposition is a sentence about which one can unambiguously say: it is either true or false. There is no third option. This is not an arbitrary restriction. It is the working definition on which the entire apparatus of logical inference rests: axioms, theorems, rules of proof. Only propositions participate in this apparatus. Only for them does it make sense to ask "is this true or not?"

It is important to understand, however, that there exists a vast number of sentences that are not propositions. Questions ("What time is it?"), commands ("Close the door"), exclamations, incomplete thoughts - all of these are sentences, but not propositions. Logic does not reject or ban them. It simply does not treat them as objects to which truth or falsehood can be assigned.

It is also possible, within a given formal system, to encounter a sentence that qualifies as a proposition in some other system - in a different, specific context - but does not qualify as one in the system at hand. In arithmetic, for instance, "a chair is less than 80" is not a proposition: arithmetic has no means of assigning it a truth value. Yet in some other system - say, one concerned with the price of furniture - that very same sentence could function as a perfectly valid proposition.

Moreover: one can formulate a sentence that is as loud and solemn as one likes - for instance, "I am a sentence that confirms all scientific laws." The laws themselves may well be true. And in some external system, this sentence might even carry genuine meaning. But within the formal system at hand, no theorem can be derived that would establish its truth or falsehood. Grandeur and even external plausibility do not make a sentence a proposition. Inside the system, it is simply a sentence - nothing more.

This distinction between a sentence and a proposition is precisely the key that dissolves all the "paradoxes" that follow.

Part II. The Liar's Paradox

How it is usually stated

A certain Liar is attributed a defining property: he always lies. He then utters the following:

"I am lying right now."

What follows is the standard analysis: if the sentence is true - then he is indeed lying, which means it is false. If it is false - then he is not lying, which means it is true. A contradiction is obtained, declared a "paradox," and presented as a "threat to the foundations of logic."

Where the bad substitution occurs

Let us check this sentence against the definition of a proposition.

The method is simple and standard - reductio ad absurdum. We temporarily accept the hypothesis that this sentence is a proposition, and observe what follows.

• Hypothesis 1: the sentence is true. Then the liar is telling the truth - that is, he is lying right now - contradiction.
• Hypothesis 2: the sentence is false. Then the liar is not lying - that is, the sentence is true - contradiction.

Both hypotheses lead to contradiction. This proves that the sentence can be neither true nor false. Therefore - by definition - it is not a proposition.

There is no paradox. What we have is a sentence that is tacitly and erroneously presented as a proposition. The moment we apply the correct definition, the "paradox" vanishes - not because we have "banned" it, but because it failed the elementary test for membership in the relevant class of objects (proposition).

How this looks intuitively

Imagine you open a box on which is written: "This box contains nothing." Inside is a slip of paper bearing the same inscription.

This is not a paradox of existence. It is an incorrect label. We do not rewrite the laws of physics - we simply throw the box away.

Or: you type the word "blue" into a field labeled "Age." The program returns an error. This does not mean that the mathematics of age has broken down. "Blue" is not a number. An input error is not a system paradox.

On the "eternal liar" as a system

The very idea of a "Liar who always lies" deserves separate attention. Let us take it seriously and try to construct it formally.

Suppose we are building a system - call it a "lie generator" - that is required to produce false propositions continuously and without exception. We can picture this concretely: take a formal system with axioms and theorems, then build a second system that takes each true proposition of the first and outputs its negation - that is, prepends "not" to every proven theorem. Such a system will generate reliably false propositions, and in that sense it genuinely "lies."

But such a system will of course never contain the sentence "I am lying right now" - because "it is not the case that I am lying right now" is not a true proposition. This is precisely what happens with the Liar's Paradox: what is offered to us as a supposedly false proposition turns out to be a sentence that is not a proposition at all. This is not "the system lying" - it is an input error.

Part III. Russell's Paradox (The Barber)

How it is usually stated

In a village, there lives a barber who shaves all those - and only those - who do not shave themselves. The question: does the barber shave himself?

• If he shaves himself - then he shaves himself, and therefore by definition he must not shave himself.
• If he does not shave himself - then he does not shave himself, and therefore by definition he must shave himself.

A "paradox" is declared.

Where the substitution occurs

Let us apply the same approach. The barber's job description contains a logical defect: the specification "shave everyone who does not shave themselves," when one attempts to apply it to the barber himself, generates an irresolvable contradiction.

Such a barber cannot be hired. Not because we "ban" thinking about him. But because the description does not define any real object, it is internally contradictory. It resembles a job posting that reads: "Wanted: an employee who works if and only if he is not working."

This is not a paradox of logic. It is an ill-formed definition of a pseudo-proposition, one that fails the test of "either true, or false, and nothing else."

This sentence, within logic as a formal system, is not merely neither true nor false - it is provably incapable of being either. And this is not just an observation: we demonstrated it directly, by deriving a contradiction from each of the two hypotheses.

In more precise terms: the predicate "the set of all sets that do not contain themselves" - from which the formal Russell's Paradox grows - likewise fails to define a coherent object. The attempt to construct such a set leads to contradiction, and this proves that no such set exists. This is not a crisis of set theory - it is an indication that not every grammatically well-formed description defines a real mathematical object.

There is another simple way to feel this intuitively. Take the imagined "set of all sets." Now simply add one more set to it - say, the set containing a single element, the number one. The result is "the set of all sets plus one more." But wait: if it contains more elements than the "set of all sets," then that first set was not, in fact, "of all." Which means the "set of all sets," the very moment it was fixed, was already incomplete. It cannot catch up with itself.

This is not a paradox - it is simply an ill-formed definition of an object. "The set of all sets" defines no stable mathematical object for the same reason that "the largest number" defines no number: you can always add one.

Conclusion

Both "paradoxes" operate by the same scheme:

1. A sentence (or definition of an object) is taken that contains a hidden internal contradiction.
2. It is tacitly accepted as a proposition (or as a valid definition).
3. A "paradox" is derived from this - a contradiction that is presented as a problem of logic or mathematics.

The correct answer in every case is the same: this is not a problem of logic. It is an input error - the submission of a pseudo-proposition, or the mistaken presentation of a plain sentence as if it were a proposition. The sentence failed the elementary test for membership in the class of propositions. The object failed the test for admissible values.

Logic does not ban such sentences or "expel them from the language." It simply does not treat them as propositions - just as a calculator does not treat an ellipsis, the word "...", as a number. This is not a limitation of the calculator. It is the correct behavior of a tool.

A student who encounters these "paradoxes" can breathe easy: the universe is not cracking. The hair can be put back down. This is simply an incorrect label on a box.