Inference: Modus Ponens and Modus Tollens

Learn about inference, correct argument types, and deductive reasoning.

Inference

Let’s call a proposition with known truth value to be a fact. Establishing the truth value of some proposition from the given facts is called inference.

Examples

Let’s look at some examples to improve our understanding of inference.


Let’s assume the following facts:

  • F1F_1: Birds are of different colors.
  • F2F_2: A pigeon is a bird.
  • F3F_3: Pigeons are white or gray.
  • F4F_4: Siam is a pigeon.

From the facts stated above, we can infer the following conclusions:

  • C1C_1: Siam is a bird.

  • C2C_2: Siam is white or gray.

Siam

As F2F_2 and F4F_4, both are facts, hence C1C_1 is true. Siam is a pigeon, and a pigeon is a bird; therefore, we can conclude that Siam is a bird.

Similarly, from F3F_3 and F4F_4, we can infer C2C_2. Siam is a pigeon, and pigeons are white or gray; therefore, Siam is white or gray.

Here,

H={F1,F2,F3,F4}.H = \{F_1, F_2, F_3, F_4\}.

The conclusions C1C_1 and C2C_2 are inferred from HH. We can call HH the set of hypotheses. Further, note that;

C1HC_1 \notin H, and C2HC_2 \notin H.

Let’s look at another example:


Let’s assume the following facts:

  • F1F_1: Ruby and Julia are going together.
  • F2F_2: Ruby is going to campus.

From the facts stated above, we can infer the following conclusion:

  • C1C_1: Julia is going to campus.

As F1F_1 and F2F_2 are valid, C1C_1 should also be valid. Ruby and Julia are going together, and Ruby is going to campus. It means Julia is also going with Ruby to the campus.

Here,

H={F1,F2}.H = \{F_1, F_2\}.

The conclusion C1C_1 is inferred from HH. Further, C1HC_1 \notin H.

Inferences can be valid or invalid. Let’s look at an example of invalid inference.


Consider the following facts:

  • F1F_1: All healthy lions have four legs.
  • F2F_2: A healthy cow has four legs.

From the facts stated above, we can wrongly infer the following conclusion:

  • C1C_1: A healthy cow is a lion.
A lion
A lion

Given that F1F_1 and F2F_2 are valid, is C1C_1 also valid? From F1F_1, we can not infer that anything with four legs is a lion. Hence, we can not conclude C1C_1 from the given facts.

Only those inferences are valid in logic that are based on some established rules. We call templates of such established rules in logic as rules of inference. These rules of inference are building blocks for making sound arguments in mathematics.

Next, let’s look at the rules of inference and how each one is logically valid.

Modus ponens

Let’s assume that for arbitrary propositions, q1q_1 and q2q_2, we know that q1q2q_1 \Rightarrow q_2 is true. Further, we know that q1q_1 is true. Then, we can conclude that q2q_2 is true.

To understand it further, look at the following truth table:

Row Number q1q_1 q2q_2 q1q2q_1 \Rightarrow q_2
1 T T T
2 T F F
3 F T T
4 F F T

Given that q1q2q_1 \Rightarrow q_2 is not false means we are not talking about row number two of the truth table. Further, when we assume that q1q_1 is not false, this means we are not talking about row numbers three and four. The only possibility left is row number one, where q2q_2 is true.

Therefore, whenever we have the facts,

q1q2q_1 \Rightarrow q_2 is true, and q1q_1 is true, we can conclude that, q2q_2 is true.

This inference rule is called modus ponens, which is Latin for “method for affirming.” We also refer to it as the rule of detachment or the law of detachment.

Another way to look at the validity of modus ponens is by noticing that the following compound proposition is a tautology.

(q1(q1q2))q2.\left(q_1 \land \left(q_1 \Rightarrow q_2\right)\right)\Rightarrow q_2.

The above conditional proposition is true, and its hypothesis is true. Hence, its conclusion has to be true.

We can verify this by the following truth table:

q1q_1 q2q_2 q1q2q_1 \Rightarrow q_2 q1(q1q2)q_1 \land \left(q_1 \Rightarrow q_2\right) (q1(q1q2))q2\left(q_1 \land \left(q_1 \Rightarrow q_2\right) \right)\Rightarrow q_2
T T T T T
T F F F T
F T T F T
F F T F T

Examples

Let’s look at a few examples to understand modus ponens further:


Take the following propositions:

  • RR: It is raining.
  • GG: The ground is wet.
  • RGR \Rightarrow G: If it is raining, then the ground is wet.

Let’s assume that the following two propositions are true.

  • RR: It is raining.
  • RGR \Rightarrow G: If it is raining, then the ground is wet.
It is raining

Then, by modus ponens, we can conclude that the following proposition is true.

  • GG: The ground is wet.

For our next example, we take the following propositions.


Consider:

  • AEA_E: Alina is eating chocolate.
  • AJA_J: Alina feels joyful.

Assume that the following two propositions are facts.

  • AEAJA_E\Rightarrow A_J: If Alina eats chocolate, then she feels joyful.
  • AEA_E: Alina is eating chocolate.
Aline is eating chocolate

Then, by using modus ponens, we can conclude the following fact:

  • AJA_J: Alina is feeling joyful.

It is important to note that we can not conclude that AEA_E is true if we know that AEAJA_E\Rightarrow A_J and AJA_J are true.

Modus tollens

Let’s assume that for arbitrary propositions, q1q_1 and q2q_2, it is given that q1q2q_1 \Rightarrow q_2 is true. Further, we know that ¬q2\neg q_2 is true. Then, we can conclude that ¬q1\neg q_1 is true.

To understand it further, look at the following truth table.

Row Number q1q_1 q2q_2 ¬q1\neg q_1 ¬q2\neg q_2 q1q2q_1 \Rightarrow q_2
1 T T F F T
2 T F F T F
3 F T T F T
4 F F T T T

Given that q1q2q_1 \Rightarrow q_2 is not false means we are not talking about row number two of the truth table. Further, assuming ¬q2\neg q_2 true means we are talking about row number four. In row number four, ¬q1\neg q_1 is also true.

Therefore, whenever we have the facts;

q1q2q_1 \Rightarrow q_2 is true, and ¬q2\neg q_2 is true, we can conclude that, ¬q1\neg q_1 is also true.

This inference rule is called modus tollens, Latin for “method for denying.” Another way to look at the validity of modus tollens is by noticing that the following compound proposition is a tautology.

(¬q2(q1q2))¬q1.\left(\neg q_2 \land \left(q_1 \Rightarrow q_2\right)\right)\Rightarrow \neg q_1.

The above conditional proposition is true, and its hypothesis is true. Therefore, its conclusion has to be true.

We can verify this by the following truth table:

q1q_1 q2q_2 ¬q1\neg q_1 ¬q2\neg q_2 q1q2q_1 \Rightarrow q_2 ¬q2(q1q2)\neg q_2 \land \left(q_1 \Rightarrow q_2\right) ¬q2(q1q2)¬q1\neg q_2 \land \newline \left(q_1 \Rightarrow q_2\right) \Rightarrow \neg q_1
T T F F T F T
T F F T F F T
F T T F T F T
F F T T T T T

Examples

Let’s look at a few examples to understand modus tollens further.


Consider the following propositions:

  • NPN_P: Nemo is a pet fish.
  • NLN_L: Nemo lives in a fishbowl.

Let’s assume that the following two propositions are true:

  • NPNLN_P \Rightarrow N_L: If Nemo is a pet fish, then she lives in a fishbowl.
  • ¬NL\neg N_L: Nemo does not live in a fishbowl.

From these two facts, we can conclude ¬NP\neg N_P by applying modus tollens.

  • ¬NP\neg N_P: Nemo is not a pet fish.

If Nemo is a pet fish, then she lives in a fishbowl, and she does not live in a fishbowl, it means Nemo is not a pet fish.

Let’s look at another example:


Take the following propositions:

  • WW: It is winter.
  • WJW_{J}: Alex is wearing a jacket.

Consider the following statements as facts.

  • WWJW\Rightarrow W_{J}: If it is winter, then Alex will wear a jacket.

  • ¬WJ\neg W_{J}: Alex is not wearing a jacket.

Alex

By using modus tollens, we conclude the following fact.

  • ¬W\neg W: It is not winter.

We must be cautious that we can not conclude ¬WJ\neg W_{J} from the facts WWJW\Rightarrow W_{J}, and ¬W\neg W.

Quiz

Test your understanding of modus ponens and modus tollens.

1

Consider the following propositions:<br> SS: It is sunny. <br> HGH_G: Hazel is wearing sunglasses.<br> SHGS \Rightarrow H_G: If it is sunny, Hazel will wear sunglasses.

Assume that, SS and SHGS \Rightarrow H_G are true.

According to modus ponens, which proposition is true?

A)

It is not sunny.

B)

It is not sunny, and Hazel is wearing sunglasses.

C)

Hazel is wearing sunglasses.

D)

It is sunny, and Hazel is not wearing sunglasses.

Question 1 of 20 attempted