Exercise 6.5.13

(a) α β + γ = α β α γ .

(b) ( α β ) γ = α β γ .

Answers

Lemma 1. Suppose that α is an ordinal, β 0 is a limit ordinal, and { γ δ } for δ < α + β is a non-decreasing transfinite sequence of ordinals. Then

sup δ < α + β γ δ = sup δ < β γ α + δ .

Proof. First we note that α + β is a nonzero limit ordinal by Lemma ?? Let A = { γ δ δ < α + β } and

γ = sup δ < β γ α + δ .

We show that γ is the least upper bound of A , which shows the result by the least upper bound property (Lemma ??). First consider any γ δ in A so that δ < α + β .

If δ < α then γ δ γ α = γ α + 0 γ since the sequence is non-decreasing and 0 < β . On the other hand if δ α then by Lemma 6.5.5 there is a unique ordinal 𝜀 such that α + 𝜀 = δ . Then, since δ = α + 𝜀 < α + β it follows from Lemma 6.5.4a that 𝜀 < β . Hence by the definition of γ we have that γ δ = γ α + 𝜀 γ . Thus in all cases γ δ γ , which shows that γ is an upper bound of A since γ δ was arbitrary.

Now consider any 𝜀 < γ . Then by the definition of γ it follows that there is a δ < β such that 𝜀 < γ α + δ since 𝜀 is not an upper bound { γ α + ζ ζ < β } . Thus again by Lemma 6.5.4a we have that α + δ < α + β so that γ α + δ A . Hence since 𝜀 < γ α + δ it cannot be that 𝜀 is an upper bound of A . Since 𝜀 < γ was arbitrary this completes the proof that γ = sup A as desired. □

Lemma 2. If α is an ordinal such that α > 0 then 0 α = 0 . Otherwise if α = 0 then 0 α = 0 0 = 1 .

Proof. First if α = 0 then clearly 0 α = 0 0 = 1 by Definition 6.5.9a. Then if α > 0 we show the result by transfinite induction on α . First if α = 1 then we have

0 α = 0 1 = 0 0 + 1 = 0 0 0 = 1 0 = 0 ,

where we have used Definitions 6.5.9b and 6.5.6a. Now suppose that 0 α = 0 so that we have

0 α + 1 = 0 α 0 = 0 ,

where we again have used the same two definitions. Lastly suppose that α is a nonzero limit ordinal and that 0 β = 0 for all β < α . Then we have by Definition 6.5.9c that

0 α = sup β < α 0 β = sup β < α 0 = 0 ,

where we have used the induction hypothesis. □

Lemma 3. If α is an ordinal then 1 α = 1 .

Proof. We show this by transfinite induction on α . First for α = 0 we have 1 α = 1 0 = 1 by Definition 6.5.9a. Now assume that 1 α = 1 so that 1 α + 1 = 1 α 1 = 1 α = 1 by Definition 6.5.9b and the induction hypothesis.

Lastly suppose that α is a nonzero limit ordinal and that 1 β = 1 for all β < α . We then clearly have by Definition 6.5.9c that

1 α = sup β < α 1 β = sup β < α 1 = 1 .

This completes the inductive proof. □

Lemma 4. If α , β , and γ are ordinals where α > 0 and β γ then α β α γ .

Proof. Clearly if β = γ then α β = α γ so that the conclusion holds. So assume that β < γ .

Case: α = 1 . Then by Lemma 3 we have

α β = 1 β = 1 = 1 γ = α γ .

Case: α > 1 . Then it follows from Exercise 6.5.14b that α β < α γ .

Hence in either case α β α γ is true. □

(a)

Proof. We show this by transfinite induction on γ . First for γ = 0 we have

α β + γ = α β + 0 = α β (by Definition 6.5.1a) = α β 1 (by Example 6.5.7a ) = α β α 0 (by Definition 6.5.9a) = α β α γ .

Now assume that α β + γ = α β α γ so that we have

α β + ( γ + 1 ) = α ( β + γ ) + 1 (by Definition 6.5.1b) = α β + γ α (by Definition 6.5.9b) = ( α β α γ ) α (by the induction hypothesis) = α β ( α γ α ) (by the associativity of multiplication) = α β α γ + 1 . (by Definition 6.5.9b)

Lastly suppose that γ is a nonzero limit ordinal and that α β + δ = α β α δ for all δ < γ . First if α = 0 then α β + γ = 0 β + γ = 0 by Lemma 2 since β + γ > 0 since γ > 0 . Hence we have

α β + γ = 0 = 0 β 0 = 0 β 0 γ = α β α γ ,

where we have used Definition 6.5.6a and the fact that 0 γ = 0 by Lemma 2 since γ > 0 .

On the other hand if α > 0 then we have by Definition 6.5.9c that

α β + γ = sup δ < β + γ α δ .

It then follows from Lemma 4 that { α δ } for δ < β + γ is a non-decreasing sequence since α > 0 . Thus we can apply Lemma 1 so that

α β + γ = sup δ < β + γ α δ = sup δ < γ α β + δ = sup δ < γ ( α β α δ )

by the induction hypothesis. It then follows from statement (6.6.1) in the text that

α β + γ = sup δ < γ ( α β α δ ) = α β sup δ < γ α δ = α β α γ

by Definition 6.5.9c. This completes the transfinite induction. □

(b)

Proof. We show this by transfinite induction on γ as well. First for γ = 0 we have

( α β ) γ = ( α β ) 0 = 1 (by Definition 6.5.9a) = α 0 (again by Definition 6.5.9a) = α β 0 = α β γ . (by Definition 6.5.6a)

Next suppose that ( α β ) γ = α β γ so that we have

( α β ) γ + 1 = ( α β ) γ α β (by Definition 6.5.9b) = α β γ α β (by the induction hypothesis) = α β γ + β (by part a) = α β ( γ + 1 ) . (by Definition 6.5.6b)

Lastly, suppose that γ is a nonzero limit ordinal and that ( α β ) δ = α β δ for all δ < γ .

Case: α = 0 . Then if β = 0 we have

( α β ) γ = ( 0 0 ) γ = 1 γ (by Definition 6.5.9a) = 1 (by Lemma  3 ) = 0 0 (by Definition 6.5.9a) = 0 0 γ (by Lemma ??) = α β γ .

On the other hand if β > 0 then first we note that 0 = β 0 < β γ by Definition 6.5.6a and Exercise 6.5.7a since both β 0 and 0 < γ . We then have

( α β ) γ = ( 0 β ) γ = 0 γ (by Lemma  2  since  β > 0 ) = 0 (by Lemma  2  again since  γ > 0 ) = 0 β γ (by Lemma  2  yet again since  β γ > 0  as shown above) = α β γ .

Case: α > 0 . Then we have

( α β ) γ = sup δ < γ ( α β ) δ (by Definition 6.5.9c) = sup δ < γ α β δ (by the induction hypothesis) = α sup δ < γ β δ (by statement (6.6.1) in the text since  α > 0 ) = α β γ (by Definition 6.5.6c)

Since these cases are exhaustive this completes the inductive proof. □

User profile picture
2024-07-15 11:42
Comments