Every language has some way of reporting what someone else has said. To express what Jakobson [Jakobson, R., 1990. "Shifters, categories, and the Russian verb. Selected writings". "Word and Language". Mouton, The Hague, Paris, pp. 130-153] called "speech within speech", the speaker can use their own words, recasting the original text as their own, within an "indirect" speech construction. Or the other person may be quoted "directly", just as they said it, or more or less so. One major difference between direct and indirect speech lies in the way person specification within the speech report is cast. In direct speech, person reference is expressed exactly as it was in the original speech report. In indirect speech, the person reference is shifted to the perspective of the speaker. There is a third option--a "middle ground" situation known as "semi-direct" speech--with incomplete person shift. Semi-direct speech often involves coreferentiality between the current speaker--rather than the author of the speech report--and a participant within the speech report. In Manambu, a Ndu language spoken in the New Guinea area, semi-direct speech differs from both direct and indirect speech in a few interesting ways. Further examples of semi-direct speech and its various guises come from a number of African languages, other languages from New Guinea area, and Colloquial English. The existence of a semi-direct speech construction brings an additional dimension to the typology of speech reports: the necessity of including the perspective of current speaker in the overall picture. (Contains 1 figure and 5 tables.)