We did not go through 40,000 documents. What we did is we targeted particular kinds of documents and we set out the criteria for the test. NAMA extracted from the 40,000 population of documents, a smaller number, which I think was about 2,000 documents.
No comments