How can the channel capacity be larger than one, in the noisy channel coding theorem?

Question

The Noisy channel coding theorem stated that:

For every discrete memoryless channel, the channel capacity $C=\sup\{{I(X,Y)}\}$ has the following property: For any $\epsilon>0$ and $R<C$, for large enough N, there exists a code of length N and rate ≥ R and a decoding algorithm, such that the maximal probability of block error is less than $\epsilon$.

We know that chances are that the capacity C is larger than one. For example, if C=3 and we choose R=1.5, how can the rate of a code be 1.5?