### Part (a)
We need to prove that for every solution (X,Y) of a simplified equation, both X and Y can be written in the form of a power of a word like Z (where Z can contain only one alphabetic letter).
1. Assume a simplified equation: Consider a simplified equation X=Y. Since X and Y are different words, we can write X and Y as sequences of letters. Let X=x1x2⋯xm and Y=y1y2⋯yn where xi and yj are letters of the alphabet.
2. Equal lengths: If m=n, then X=Y implies xi=yi for all i, which contradicts the assumption that X and Y are different words. Therefore, m=n.
3. **Assume m>n**: Without loss of generality, assume m>n. Then X can be written as X=Zk for some word Z and integer k such that Z is a prefix of X and Y. Similarly, Y can be written as Y=Zl for some integer l.
4. **Construct Z**: Since X and Y are different, k=l. Therefore, X=Zk and Y=Zl where Z is a word consisting of a single letter repeated multiple times.
5. Conclusion: Both X and Y can be written in the form of a power of a word Z where Z contains only one alphabetic letter.
■
### Part (b)
We need to prove that words W1,W2,⋯,Wn that are the answers of a simplified equation can be produced with fewer words.
1. Consider the rightmost words: Consider the rightmost words on either side of the equation, say Wi and Wj. If both have equal length, they are equal, and we are done.
2. Unequal lengths: Assume Wi is longer than Wj. We can re-express Wi as Wi′Wj and substitute this back into the equation, replacing all copies of Wi to yield …Wi′Wj=…WkWj.
3. **Cancel Wj**: Cancel Wj from each side to get another simplified equation in n words, …Wi′=…Wk.
4. Repeat the process: Repeat this process until the rightmost words on each side are of equal length, allowing us to equate them and express the substituted words in the form of n−1 words.
5. Conclusion: Reversing the substitutions lets us express the original n words in terms of these n−1 words.
■
### Part (c)
We need to prove that n words W1,W2,⋯,Wn that are the answers of a simplified system of equations can be produced with at most c words, where c is the number of connected components of the graph G.
1. **Construct graph G**: Define graph G with vertices {1,2,⋯,n} such that i and j are connected if in one of the equations, Wi and Wj appear on the right side of each side of the equation.
2. Spanning tree: Consider a spanning tree of each component of G. Delete all equations whose edges are not in a spanning tree. Now there are n−c edges in total, so n−c remaining equations.
3. Algorithm from part (b): Pick one of the equations and perform the algorithm from part (b) on it. This lets you express W1,…,Wk in terms of fewer words, say W1′,…,Wk−1′.
4. Group vertices: Group the vertices of G into k groups: one group for the vertices representing words which become expressed in the form …Wi′ for each 1≤i≤k−1, and one group for the remaining vertices.
5. **Construct new graph G′**: Construct a graph G′ that corresponds to the situation after all the W1,…,Wk have been replaced by all the W1′,…,Wk−1′. This graph will have k−1 vertices representing the words W1′,…,Wk−1′, as well as n−k vertices representing the unaffected words.
6. Number of components: Show that the number of components of G′ is at most c. Each group contains at least one vertex, and the new graph G′ is identical to the original G except that the two vertices representing Wi and Wj are merged into one new vertex.
7. Repeat the process: Repeat this process until we get c isolated vertices. Reversing the substitutions, we can express the n words in terms of c words.
■