Scribd
Upload
95 views

B.A. Introduction To Logic 2012-13: Lecture 18: Quantificational Logic VI

This document provides a summary of quantificational logic rules and examples of proofs using these rules: 1) It reviews the existential quantifier rules of existential introduction (EI) and existential elimination (EE), and provides restrictions on EE. 2) It gives examples of proofs involving relational expressions and multiple quantifiers, demonstrating applying rules one quantifier at a time from wider to narrower scope. 3) It outlines a general strategy for quantificational logic proofs of eliminating quantifiers, deriving an unquantified conclusion, and introducing quantifiers.

Uploaded by

Ikki De Queero
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
95 views

B.A. Introduction To Logic 2012-13: Lecture 18: Quantificational Logic VI

This document provides a summary of quantificational logic rules and examples of proofs using these rules: 1) It reviews the existential quantifier rules of existential introduction (EI) and existential elimination (EE), and provides restrictions on EE. 2) It gives examples of proofs involving relational expressions and multiple quantifiers, demonstrating applying rules one quantifier at a time from wider to narrower scope. 3) It outlines a general strategy for quantificational logic proofs of eliminating quantifiers, deriving an unquantified conclusion, and introducing quantifiers.

Uploaded by

Ikki De Queero
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

[email protected].

uk

B.A. Introduction to Logic 2012-13


Lecture 18: Quantificational Logic VI
Recap
Rules of inference for the existential quantifier: EI: Given a formula containing a name on any line of proof, you may replace one or all occurrences of that name with a variable. Introduce the existential quantifier to the resulting matrix and write the formula on a new line. Annotate the new line EI together with the line number of the original line. The dependency-numbers of the new line are identical with those of the line of the original formula. EE: To infer a conclusion from an existentially quantified formula: first assume a genuinely typical disjunct and then derive the desired conclusion from that disjunct. Restate the conclusion on a new line of proof. Annotate the new line EE together with three numbers: (i) the line number of the original existential formula; (ii) the line number of the typical disjunct; (iii) the line number of the conclusion derived from the typical disjunct. The dependency-numbers of the new line consist of all the dependencies belonging to the derived conclusion except that you may discharge the dependency-number of the typical disjunct and replace it with the dependency-number of the original existentially quantified formula. Restrictions: Note carefully that EE cannot be legitimately applied if (i) the name in the typical disjunct is itself contained in the conclusion derived for EE, or (ii) the name in the typical disjunct is itself contained in any premise or assumption used to derive the conclusion for EE from the typical disjunct.

Relational expressions and multiple generality


We have now considered all four of the rules of inference specific to QL: UE, UI, EI, and EE. Our next task will be to look at the way in which those rules should be applied when dealing with proofs involving (i) relational expressions and (ii) multiply general formulas. Example 1: Suppose we want to prove the sequent: x [y [Rxy]] : y [x [Rxy]] Since both the premise and conclusion are universal formulas, the overall strategy will be to use UE and then UI (more on strategic matters later). However, both the premise and the conclusion contain two universal quantifiers. This means we will have to apply both UE and UI twice, as follows: {1} {1} {1} {1} 1. 2. 3. 4. x [y [Rxy]] y [Ray] Rab x [Rxb] y [x [Rxy]] Premise 1 UE 2 UE 3 UI 4 UI 1

{1} 5. Key points about this proof:

[email protected]

In applying UE and UI, we only ever eliminate or introduce one quantifier at a time. When we apply UE at lines 2 and 3, we start by eliminating the quantifier with the wider scope, before moving on to eliminate with the quantifier with narrower scope. (We unpeel the original formula, as it were, quantifier by quantifier.) When we use UI at lines 4 and 5, we start by introducing the quantifier with narrower scope, before moving on to introduce the quantifier with wider scope. Notice that in applying UI, there is no problem with the restriction on that rule: the only dependency in play here is the original universal formula, which contains no names.

Example 2: Suppose we want to prove the sequent: x [y [z [Rxyz]]] : z [y [x [Rxyz]]] Here were dealing with an extra universal quantifier in both the premise and the conclusion, but the same general strategy applies: {1} {1} {1} {1} {1} {1} {1} 1. 2. 3. 4. 5. 6. 7. x [y [z [Rxyz]]] y [z [Rayz]] z [Rabz] Rabc x [Rxbc] y [x [Rxyc]] z [y [x [Rxyz]]] Premise 1 UE 2 UE 3 UE 4 UI 5 UI 6 UI

Example 3: Suppose we want to prove the sequent: x [y [Rxy]] : y [x [Rxy]] This sequent involves both kinds of quantifiers. To complete the proof, well need to use both EE and EI, as well as UE and UI: {1} {2} {2} {2} {2} 1. 2. 3. 4. 5. x [y [Rxy]] y [Ray] Rab x [Rxb] y [x [Rxy]] y [x [Rxy]] Premise Assumption TD 2 UE 3 EI 4 UI 1,2,5 EE 2

{1} 6. Key points about this proof:

[email protected]

Notice again that we only ever eliminate or introduce one quantifier at a time. When we eliminate the quantifiers on lines 2 and 3, we start with the one which has the wider scope before moving on to eliminate the one with narrower scope. When we introduce the quantifiers on lines 4 and 5, we start with the one which has narrower scope, before moving on to introduce the one which has wider scope. There is no problem in satisfying the restriction on UI: the formula on line 4 which has been used for the purposes of UI contains the name b, but the dependency-number for that line doesnt take us to another line containing b. There is no problem in satisfying the restrictions on EE, since (i) the name in the TD doesnt appear in the conclusion derived on line 6, and (ii) the name in the TD doesnt appear in any premise or assumption used to derive the conclusion.

General strategy for proof construction in QL


Tomassi suggests a useful three-part strategy for proof construction in QL: First, apply the relevant elimination rules. Next, exploit the PL rules of inference to derive the desired conclusion in unquantified form. Complete the proof by application of the appropriate introduction rule.

As regards the first and third stages, the most important thing is to look very carefully at each formula youre dealing with, and ask yourself just what kind of formula it is: this will put you in a position to decide which elimination / introduction rules you should use. As regards the second stage, youll need to draw on your knowledge of proof construction in PL, remembering in particular the golden rule: Ask yourself: Is the main connective in the conclusion a conditional? If so, apply the strategy for CP, i.e. assume the antecedent and try to derive the consequent. If not, ask: Is the main connective of any member of the set of premises a disjunction? If so, apply the strategy for vE, i.e. assume the first disjunct and try to derive the conclusion, assume the second disjunct and try to derive the conclusion. Finally, draw that same conclusion from the original disjunctive premise by vE. If not: Try RAA. Remember: the trick is to assume the opposite of what you want and then try to derive a contradiction from that assumption together with any other formula or formulas already available in the proof. The double negation rules will allow you to finish things off to suit your purposes.

Example: Suppose you want to prove the sequent: 3

[email protected]

x [Fx Gx], x [Gx Hx] : x [Fx Hx] We start by taking a careful look at the premises: both of them are universal formulas. This means that the first step will be to use UE: {1} {2} {1} {2} 1. 2. 3. 4. x [Fx Gx] x [Gx Hx] Fa Ga Ga Ha Premise Premise 1 UE 2 UE

Now we look carefully at our conclusion: its also a universal formula. This means that the last stage will involve using UI to move from an unquantified formula to the conclusion. The unquantified formula well need for that last stage of the proof is Fa Ha, so now we have to work out how to derive that formula from what we already have. To do this, we apply the first part of the golden rule and we ask ourselves whether the main connective in Fa Ha is a conditional. The answer is yes, so we know that well need to do a conditional proof: {5} {1,5} {1,2,5} {1,2} 5. 6. 7. 8. Fa Ga Ha Fa Ha Assumption CP 3,5 MP 4,6 MP 5,7 CP

All that remains is to apply UI to complete the proof: {1,2} 9. x [Fx Hx] 8 UI

(As always when we use UI, we should check that theres no problem with the restriction on that rule. There is no problem here, because the dependency-numbers for line 8 refer us back to lines 1 and 2, which dont contain the name a.)

Reading
Tomassi, P. Logic. Chapter 6, VI.

Exercises
Exercise 6.6 http://fundraise.unicef.org.uk/MyPage/Charlie-KP-Marathon

You might also like