This is the publication of a Freiburg (Breisgau) dissertation of 2007. There is an introduction on Roman legal procedure (1–17), followed by a brief excursion into Athenian law concerning witnesses (18–25). About 200 pages on the main topics follow. Almost every piece of relevant Ciceronian text is examined and a considerable amount of external material introduced — though not, as we shall see, enough. The topics covered are: summons; timing and form of testimony (written and/or verbal); oaths; questioning; sealing of testimony; embassies; laudatores; qualification for and exclusion from testimony, including women, slaves and minors as witnesses; reliability of witnesses; offences of witnesses and their punishment. The investigation of the texts adduced is generally careful and the conclusions sane. Regret arises, however, because certain evidence is not brought to bear. First and foremost, the tablets from the Casa del Bicentenario at Herculaneum containing testimony in the Petronia Iusta case (TH 16–29 = Pugliese Carratelli, PP 3 (1948), 165–83). Here we have written evidence in a civil case, in the form ‘scripsi et iuravi’, duly witnessed and sealed by seven witnesses — the same number as used in wills — thus solving the author's problem in 83 ff. The oath, in this case by the Genius of the emperor and his children, seems routine and probably prescribed by Augustus' legislation, but is likely to have been the rule earlier in another form. The evidence is hearsay and is related to a conclusion about the issue in the case: ‘hence I know that the woman in question was born of Petronia Vitalis free’ (or ‘… a freedwoman of Calatoris Themis’), ‘which is the issue in question’. The witnesses are thus not only allowed to argue but seem to be expected to argue. Is this relevant to Cicero? Yes: Asconius (Corn. 60–1C), describing the testimony of five consulars against the tribune Cornelius, says that they declared both that they had seen the tribune reading the text of his bill himself and that this was relevant to the charge of damaging the majesty of a tribune. It follows that witnesses were not expected to be objective (contra Steck 120).

It would also have been worth pointing out that the cross-examination of witnesses (80 ff.) usually led to an altercatio between the advocates over the validity of the testimony (Quint. 6.4.1 ff.), a point that tends to be neglected in discussions of Roman legal procedure. The epigraphic lex repetundarum (line 33) regarding interrogation by the praetor may indeed refer to witnesses during the trial itself, but reflect an earlier stage in the development of the quaestio — I miss here reference to the first volume of Roman Statutes (ed. Crawford). This could have been profitably adduced also in a closer investigation about the evidence in the initial lines of the Lex Latina Tabulae Bantinae for disqualification of men as, inter alia, witnesses (144 ff.). There is much useful material in this work: it is a pity, therefore, that there are a few important lacunae.