I am not sure if I can be of real help. I wish just note that there is a close connection between epilepsy and language disorders (deeper perhaps than commonly assumed), that we are exploring in some detail. Well, not focusing on epilepsy indeed, but in disorders entailing abnormal profiles of brain rhythmicity. We do believe that basic brain oscillations can be the best route to achieve a good mapping of language to the brain. We are about publishing to papers in Frontiers examining the oscilopathic profile of language processing in autism spectrum disorders and schizophrenia. Both papers include references to our model of language processing in the brain from an oscillatory perspective (still in progress). Maybe this can illuminate your own research. Moreover, because the suite of brain rhythms is conserved across taxa, they allow to gain really interesting evolutionary insights, to the extent that we expect that human language resulted from a slight variation in the oscillatory behaviour of the primate/hominin brain. Let me know if I can be of further help.
I have an abstract devoting to this problem. I can't write a real article because I haven't got the catamnesis for patients in my sample. But may be it would be useful for you:)
Difficult question, I think for that you have to review the literature entirely. My experience in stroke patients suggests that about 1/3 of stroke patients carry language problems despite lesions far from typical structures.
A colleague of mine once pointed out that only about 12 % of the population have the textbook distribution of language functions, which means that the unexpected may be rather the rule.
Some files from Epilepsy which may be of indirect relevance to your questions are attached.
in people with brain tumours, the closest article I can think of to answer to your question is the meta-analysis by De Witt Hamer et al. 2012:
Hamer, P. C. D. W., Robles, S. G., Zwinderman, A. H., Duffau, H., & Berger, M. S. (2012). Impact of intraoperative stimulation brain mapping on glioma surgery outcome: a meta-analysis. Journal of Clinical Oncology, 30(20), 2559-2565.