Summary
This lesson opens the third unit of the course, devoted to semantics. It presents its six lessons and the thread that joins them, from the extension of a valuation to all formulas to the normal form algorithm and switching circuits; it places the unit on the path of the course, after deduction, which obtained formulas from others without attending to their meaning, and before metatheory, which will relate derivability to semantic consequence ; it explains why a single clause, that of the primitive connective, suffices to give meaning to the whole language, and why truth values are calculated with polynomials; it shows with a four-row table what semantics decides without searching for deductions or inventing markings; it specifies what is taken from the previous units and gives advice for studying it.
Learning objectives
- Describe the sequence of the six lessons of the unit and explain which results of each one the next one uses.
- Place semantics on the path of the course, distinguish semantic consequence from derivability and explain what soundness and completeness will assert.
- Explain why bivalence and truth-functionality make it possible to give meaning to all formulas with a single clause, and decide with a truth table whether an inference schema is valid.
- Recognize what the unit takes from the previous ones, distinguish the signs , , , and , and organize study starting from the entry assessment.
Entry assessment
Before working through the unit, it is worth measuring the starting point. This assessment is a self-assessment: it is not graded, and its only purpose is to guide your study. Its fifteen questions are of two kinds. The first four concern what the unit takes for granted from the first unit: definition by recursion and unique readability, the official abbreviations, the cases of an induction on complexity and the substitution of a subformula; anyone who misses one of them should review, before going on, the lesson that the question names. The remaining eleven cover, in order, the six lessons of this unit: two the first, two the second, one the third and two each of the last three. Anyone who has never studied the semantics of a formal language should not expect to get them right, and that does not matter: each answer, right or wrong, explains the point and names the lesson that deals with it. The same skills are measured again at the end, in the synthesis lesson.
Before starting the unit, measure your starting point. It is a short self-assessment: it is not graded and its result is stored only in your browser.
Question 1From earlier unitsA function on formulas is defined by giving its value at each variable and a rule that computes the value of from , from and from the values of both. What guarantees that exactly one function is thus defined?
That decides which values the function takes, not whether it is well defined. A rule with values and applied to a formula that could be read in two ways would give two results.This is precisely what recursion is for: the theorem guarantees once and for all, for the infinitely many formulas, that there exists one and only one function satisfying the clauses.They do not: the string is the same as . Recursion rests on the single formation rule, , which does have it.It is the theorem of definition by recursion on formulas, from the lesson on induction, and unique readability is its indispensable hypothesis: if a formula admitted two decompositions, the rule could assign it two values. The unit uses it from its first result, which extends a valuation of the variables to all formulas.Question 2From earlier unitsWith the official abbreviations of the course, which formula is ?
It confuses identity with equivalence. Idempotence will say that both always have the same truth value, but is a longer string than .That string is : the outer negation is missing, because disjunction is the negation of joint denial, not joint denial itself.The abbreviation admits equal components, like the formation rule, which from and forms .Disjunction abbreviates , and is, by definition, ; hence , letter by letter. The unit will return to this fact: one and the same string can have very different written forms, and be in normal form without looking like it.Question 3From earlier unitsTo prove by induction on complexity that all formulas have a certain property, which cases must be treated?
Negation, conjunction and disjunction are abbreviations, not formation rules. An induction that treated only them would leave out formulas such as , which is none of the three. It is a frequent error when proving the existence of normal forms.A variable is not a joint denial, and without the base case the induction does not get started.There are six cases too many and the only necessary one is missing: the six connectives are abbreviations of joint denials.Every formula is a variable or a joint denial of two formulas, and nothing else. This is how the unit proves the coincidence lemma, the replacement theorem and the existence of normal forms, always with a single inductive case.Question 4From earlier unitsWhen substituting the formula by the formula in , according to the definition of the lesson on induction, which occurrences of are changed?
That is replacing one occurrence, another operation, which the unit will also study; substitution changes all of them.Substitution acts on the string, where abbreviations multiply the occurrences: in , the formula occurs twice.What is put in is not substituted again: the definition runs through the initial formula, not the result.Substitution is defined by recursion, and its first clause changes the whole formula if it is ; the others apply it to the two components. The unit will prove that, if and are equivalent, the result is equivalent to , and will distinguish this operation from the replacement of a single occurrence.Question 5A valuation gives the value and the value . If has the value exactly when and have the value , what value does the formula take?
It applies the clause to the variables, not to the components. The components of the whole formula are and , and intervenes only through the second.The coincidence lemma of the first lesson ensures that the value depends only on the variables that occur in the formula, which here are and .The grouping is fixed by the string: by unique readability, the formula has a single pair of components, and its value is a single one.The clause is applied from the inside out: has the value , because has the value ; the whole formula is the joint denial of , which has the value , and of , which has the value , so it has the value . With numbers, .Review: Semantics of propositional logic
Question 6What does mean?
It is a frequent misreading. Denying that all valuations satisfy is not asserting that none satisfies it.That is unsatisfiability, which is much stronger: a single valuation that refutes suffices for .That is written and speaks of deductions; speaks of valuations. That both things coincide is what metatheory will prove.It is the negation of , which asserts that every valuation satisfies . For the variable , holds, and yet it is satisfiable. That is a contradiction is written .Review: Semantics of propositional logic
Question 7It is known that : every valuation that satisfies and satisfies . What follows?
Removing the premise changes what is asserted: , and is not valid. The premise does not disappear; it becomes the antecedent.It reverses the direction: the discharged premise is the antecedent, and the conclusion, the consequent.It also discharges the premises of , and that is legitimate only if is empty: , but is not valid.It is the semantic deduction theorem: a valuation that satisfied and refuted would give to and to , and would then satisfy without satisfying . Its proof reduces to the only false row of the implication.Question 8The formula is valid. Which of the following formulas is valid by the uniform substitution theorem?
It changes in some occurrences and not in others: it is not a substitution, and the result is not valid (the valuation that gives to and to refutes it).It puts different formulas in place of one and the same variable; it is not a substitution either, and the valuation that gives to and to refutes the result.It allows each variable to be changed into any formula whatever, provided this is done in all its occurrences and always with the same formula.It results from substituting by in all its occurrences; the uniform substitution of variables by arbitrary formulas preserves validity.Question 9By the convention of the course, which formula does the expression , written without parentheses, designate?
It associates to the right. It is equivalent to the designated formula, but it is not the same string, and the convention fixes only one.They are not: by unique readability, their first components would be equal, and they are and . They are different formulas with the same value in every valuation.Without a convention, indeed, it would designate nothing; that is why the third lesson fixes one, that of grouping to the left.Conjunctions of terms are defined by recursion associating to the left, and the convention abbreviates in this way that of the list . The other grouping is a different string, although equivalent by associativity.Question 10A truth function of two arguments assigns a value, or , to each of the four rows , , and . How many truth functions of two arguments are there?
Correct answer:
Each of the rows receives one of values, independently of the others: there are functions, one for each possible column of a four-row table. Six of them are those of the connectives of the course. Whoever answers has counted the rows and not the functions; whoever answers has multiplied by instead of raising to the power .Review: Functional completeness
Question 11Can a formula written only with and from the variables have the value in the valuation that gives to all the variables?
Length does not matter: the property is preserved at each step of the construction, so that no formula, however long, loses it.That one connective suffices says nothing about others. Joint denial suffices because negation can be written with it; with conjunction and disjunction, it cannot.There are infinitely many formulas, and no number of tables exhausts them; the answer is obtained once and for all with a property that all of them preserve.The variables have the value in it, and the conjunction and the disjunction of two formulas with the value have the value ; by the closure principle, the property passes to all formulas so written. It follows that negation cannot be written with and : the set is not adequate.Review: Functional completeness
Question 12A literal is a variable or the negation of a variable. Which of the following formulas is a literal?
It is equivalent to , which is a literal, but it is not one: it is the negation of a negation. When negating a normal form, the double negation of a literal must be reduced before going on.It is a conjunction of two literals, that is, an elementary conjunction, not a literal.It abbreviates , a compound formula that is neither a variable nor the negation of a variable.It is the negation of the variable . Literals are the pieces of normal forms: a clause is a disjunction of literals, and an elementary conjunction, a conjunction of literals.Review: Normal forms
Question 13Is the conjunctive normal form valid?
In a conjunction, a valid clause does not save the others: the conjunction has the value as soon as one clause has the value .Being satisfiable is not being valid: the valuation that gives to all the variables satisfies it, but another one refutes it.In a conjunctive form, validity can be read at a glance, by looking for complementary pairs clause by clause. It is one of the reasons why the unit studies normal forms.A conjunctive form is valid if and only if each of its clauses contains a literal and its opposite. The first clause contains one ( and ); the second does not, and a clause without a complementary pair has the value in the valuation that makes all its literals false.Review: Normal forms
Question 14To bring a formula to disjunctive normal form, conjunction is distributed over disjunction. Which is the correct rule?
It changes the second conjunction into a disjunction, a frequent error, and the rule so written is not correct: with true and , false, the left-hand side is false and the right-hand side true.The right-hand side is equivalent to , which is another formula: with false and , true, the left-hand side is false and the right-hand side true.It forgets to copy into the second term: with and false and true, the left-hand side is false and the right-hand side true.It is the distributivity of conjunction over disjunction: is copied into each term, joined to it by the same conjunction. Each application gives an equivalent formula, by the replacement theorem.Question 15A circuit has two direct switches, of the levers and , joined in series, and that block is in parallel with a direct switch of the lever . Which formula says when it conducts?
It swaps the roles: series is conjunction and parallel disjunction, not the other way round.It puts in series the switch of , which is in parallel: with on and off the circuit conducts, and this formula has the value .It puts in parallel the switches of and , which are in series: with only on the circuit does not conduct, and this formula has the value .The series lets the signal through if both switches let it through, and the parallel, if some branch lets it through: series corresponds to conjunction and parallel to disjunction.
Loading the content…
Completed
Sign in to save your progress