The author’s words may either introduce direct speech or conclude it, sometimes they interrupt direct speech breaking the phrase into at least two intonation groups. The author's words preceding direct speech should be treated as a separate intonation group. So they are pronounced with almost any terminal tone (but mostly with the low falling tone), eg: And then he ˎsaid: →”Praps you are ˎright.” If the author's words follow direct speech they continue as an unstressed or partially-stressed tail of the preceding intonation group, eg: →‘What ˎis it for?’ he asked. →Is this for ˏme?’ he ,asked with sur,prise. If the author’s words form two or more intonation groups, the first of them doesn’t form a separate intonation group. The second and the third are always stressed and pronounced each on a lower pitch level. The nuclear tone of the final intonation group is usually that of the sentences in the direct speech. The non-final intonation groups may be pronounced either with the low rising tone or with the low falling tone according to their semantic importance. →“What a ˎpity!” was all I said | when he ,broke a ,glass. ↓“Do you ꞌthink that’s ˏfair?” she asked, | ,looking at me with sur,prise.
