The existence of a phrase in a large monolingual corpus is very useful information, and so is its frequency. The authors introduce an alternative approach to automatic translation of phrases/sentences that operationalizes this observation. The authors use a statistical machine translation system to produce alternative translations and a large monolingual corpus to (re)rank these translations. The results show that this combination yields better translations, especially when translating out-of-domain phrases/sentences. This approach can also be used to automatically construct parallel corpora from monolingual resources.