Thomas Aquinas, Summa Theologiae

Objection 1: It would seem that the evidence of two or three persons is not sufficient. For judgment requires certitude. Now certitude of the truth is not obtained by the assertions of two or three witnesses, for we read that Naboth was unjustly condemned on the evidence of two witnesses (3 Kgs. 21). Therefore the evidence of two or three witnesses does not suffice.

Objection 2: Further, in order for evidence to be credible it must agree. But frequently the evidence of two or three disagrees in some point. Therefore it is of no use for proving the truth in court.

Objection 3: Further, it is laid down (Decret. II, qu. iv, can. Praesul.): "A bishop shall not be condemned save on the evidence of seventy-two witnesses; nor a cardinal priest of the Roman Church, unless there be sixty-four witnesses. Nor a cardinal deacon of the Roman Church, unless there be twenty-seven witnesses; nor a subdeacon, an acolyte, an exorcist, a reader or a doorkeeper without seven witnesses." Now the sin of one who is of higher dignity is more grievous, and consequently should be treated more severely. Therefore neither is the evidence of two or three witnesses sufficient for the condemnation of other persons.

On the contrary, It is written (Dt. 17:6): "By the mouth of two or three witnesses shall he die that is to be slain," and further on (Dt. 19:15): "In the mouth of two or three witnesses every word shall stand."

I answer that, According to the Philosopher (Ethic. i, 3), "we must not expect to find certitude equally in every matter." For in human acts, on which judgments are passed and evidence required, it is impossible to have demonstrative certitude, because they a about things contingent and variable. Hence the certitude of probability suffices, such as may reach the truth in the greater number, cases, although it fail in the minority. No it is probable that the assertion of sever witnesses contains the truth rather than the assertion of one: and since the accused is the only one who denies, while several witness affirm the same as the prosecutor, it is reasonably established both by Divine and by human law, that the assertion of several witnesses should be upheld. Now all multitude is comprised of three elements, the beginning, the middle and the end. Wherefore, according to the Philosopher (De Coelo i, 1), "we reckon 'all' and 'whole' to consist of three parts." Now we have a triple voucher when two agree with the prosecutor: hence two witnesses are required; or for the sake of greater certitude three, which is the perfect number. Wherefore it is written (Eccles. 4:12): "A threefold cord is not easily broken": and Augustine, commenting on Jn. 8:17, "The testimony of two men is true," says (Tract. xxxvi) that "there is here a mystery by which we are given to understand that Trinity wherein is perpetual stability of truth."

Reply to Objection 1: No matter how great a number of witnesses may be determined, the evidence might sometimes be unjust, since is written (Ex. 23:2): "Thou shalt not follow the multitude to do evil." And yet the fact that in so many it is not possible to have certitude without fear of error, is no reason why we should reject the certitude which can probably be had through two or three witnesses, as stated above.

Reply to Objection 2: If the witnesses disagree certain principal circumstances which change the substance of the fact, for instance in time, place, or persons, which are chiefly in question, their evidence is of no weight, because if they disagree in such things, each one would seem to be giving distinct evidence and to be speaking of different facts. For instance, one say that a certain thing happened at such and such a time or place, while another says it happened at another time or place, they seem not to be speaking of the same event. The evidence is not weakened if one witness says that he does not remember, while the other attests to a determinate time or place And if on such points as these the witness for prosecution and defense disagree altogether, and if they be equal in number on either side, and of equal standing, the accused should have the benefit of the doubt, because the judge ought to be more inclined to acquit than to condemn, except perhaps in favorable suits, such as a pleading for liberty and the like. If, however, the witnesses for the same side disagree, the judge ought to use his own discretion in discerning which side to favor, by considering either the number of witnesses, or their standing, or the favorableness of the suit, or the nature of the business and of the evidence

Much more ought the evidence of one witness to be rejected if he contradict himself when questioned about what he has seen and about what he knows; not, however, if he contradict himself when questioned about matters of opinion and report, since he may be moved to answer differently according to the different things he has seen and heard.

On the other hand if there be discrepancy of evidence in circumstances not touching the substance of the fact, for instance, whether the weather were cloudy or fine, whether the house were painted or not, or such like matters, such discrepancy does not weaken the evidence, because men are not wont to take much notice of such things, wherefore they easily forget them. Indeed, a discrepancy of this kind renders the evidence more credible, as Chrysostom states (Hom. i in Matth.), because if the witnesses agreed in every point, even in the minutest of details, they would seem to have conspired together to say the same thing: but this must be left to the prudent discernment of the judge.

Reply to Objection 3: This passage refers specially to the bishops, priests, deacons and clerics of the Roman Church, on account of its dignity: and this for three reasons. First because in that Church those men ought to be promoted whose sanctity makes their evidence of more weight than that of many witnesses. Secondly, because those who have to judge other men, often have many opponents on account of their justice, wherefore those who give evidence against them should not be believed indiscriminately, unless they be very numerous. Thirdly, because the condemnation of any one of them would detract in public opinion from the dignity and authority of that Church, a result which would be more fraught with danger than if one were to tolerate a sinner in that same Church, unless he were very notorious and manifest, so that a grave scandal would arise if he were tolerated.

Articulus 2

Argumentum 1

Ad secundum sic proceditur. Videtur quod non sufficiat duorum vel trium testimonium. Iudicium enim certitudinem requirit.

Sed non habetur certitudo veritatis per dictum duorum testium, legitur enim III Reg. XXI quod Naboth ad dictum duorum testium falso condemnatus est. Ergo duorum vel trium testimonium non sufficit.

Argumentum 2

Praeterea, testimonium, ad hoc quod sit credibile, debet esse concors. Sed plerumque duorum vel trium testimonium in aliquo discordat. Ergo non est efficax ad veritatem in iudicio probandam.

Argumentum 3

Praeterea, II, qu. IV, dicitur, praesul non damnetur nisi in septuaginta duobus testibus.

Presbyter autem cardinalis nisi quadraginta quatuor testibus non deponatur. Diaconus cardinalis urbis Romae nisi in viginti octo testibus non condemnabitur. Subdiaconus, acolythus, exorcista, lector, ostiarius, nisi in septem testibus non condemnabitur.

Sed magis est periculosum peccatum eius qui in maiori dignitate constitutus est, et ita minus est tolerandum. Ergo nec in aliorum condemnatione sufficit duorum vel trium testimonium.

Sed Contra

Sed contra est quod dicitur Deut. XVII, in ore duorum vel trium testium peribit qui interficietur; et infra, XIX, in ore duorum vel trium testium stabit omne verbum.

Corpus

Respondeo dicendum quod, secundum philosophum, in I ethic., certitudo non est similiter quaerenda in omni materia.

In actibus enim humanis, super quibus constituuntur iudicia et exiguntur testimonia, non potest haberi certitudo demonstrativa, eo quod sunt circa contingentia et variabilia. Et ideo sufficit probabilis certitudo, quae ut in pluribus veritatem attingat, etsi in paucioribus a veritate deficiat. Est autem probabile quod magis veritatem contineat dictum multorum quam dictum unius. Et ideo, cum reus sit unus qui negat, sed multi testes asserunt idem cum actore, rationabiliter institutum est, iure divino et humano, quod dicto testium stetur.

Omnis autem multitudo in tribus comprehenditur, scilicet principio, medio et fine, unde secundum philosophum, in I de coelo, omne et totum in tribus ponimus.

Ternarius quidem constituitur asserentium, cum duo testes conveniunt cum actore. Et ideo requiritur binarius testium, vel, ad maiorem certitudinem, ut sit ternarius, qui est multitudo perfecta, in ipsis testibus.

Unde et Eccle. IV dicitur, funiculus triplex difficile rumpitur.

Augustinus autem, super illud Ioan. VIII, duorum hominum testimonium verum est, dicit quod in hoc est trinitas secundum mysterium commendata, in qua est perpetua firmitas veritatis.

Ad 1

Ad primum ergo dicendum quod, quantacumque multitudo testium determinaretur, posset quandoque testimonium esse iniquum, cum scriptum sit Exod. XXIII, ne sequaris turbam ad faciendum malum.

Nec tamen, quia non potest in talibus infallibilis certitudo haberi, debet negligi certitudo quae probabiliter haberi potest per duos vel tres testes, ut dictum est.

Ad 2

Ad secundum dicendum quod discordia testium in aliquibus principalibus circumstantiis, quae variant substantiam facti, puta in tempore vel loco vel in personis de quibus principaliter agitur, aufert efficaciam testimonii, quia si discordant in talibus, videntur singulares esse in suis testimoniis, et de diversis factis loqui; puta si unus dicat hoc factum esse tali tempore vel loco, alius alio tempore vel loco, non videntur de eodem facto loqui. Non tamen praeiudicatur testimonio si unus dicat se non recordari, et alius asserat determinatum tempus vel locum.

Et si in talibus omnino discordaverint testes actoris et rei, si sint aequales numero et pares dignitate, statur pro reo, quia facilior debet esse iudex ad absolvendum quam ad condemnandum; nisi forte in causis favorabilibus, sicut est causa libertatis et huiusmodi. Si vero testes eiusdem partis dissenserint, debet iudex ex motu sui animi percipere cui parti sit standum, vel ex numero testium, vel ex dignitate eorum, vel ex favorabilitate causae, vel ex conditione negotii et dictorum. Multo autem magis testimonium unius repellitur si sibi ipsi dissideat interrogatus de visu et scientia.

Non autem si dissideat interrogatus de opinione et fama, quia potest secundum diversa visa et audita diversimode motus esse ad respondendum.

Si vero sit discordia testimonii in aliquibus circumstantiis non pertinentibus ad substantiam facti, puta si tempus fuerit nubilosum vel serenum, vel si domus fuerit picta aut non, aut aliquid huiusmodi, talis discordia non praeiudicat testimonio, quia homines non consueverunt circa talia multum sollicitari, unde facile a memoria elabuntur. Quinimmo aliqua discordia in talibus facit testimonium credibilius, ut chrysostomus dicit, super Matth., quia si in omnibus concordarent, etiam in minimis, viderentur ex condicto eundem sermonem proferre.

Quod tamen prudentiae iudicis relinquitur discernendum.

Ad 3

Ad tertium dicendum quod illud locum habet specialiter in episcopis, presbyteris, diaconibus et clericis ecclesiae Romanae, propter eius dignitatem. Et hoc triplici ratione. Primo quidem, quia in ea tales institui debent quorum sanctitati plus credatur quam multis testibus. Secundo, quia homines qui habent de aliis iudicare, saepe, propter iustitiam, multos adversarios habent. Unde non est passim credendum testibus contra eos, nisi magna multitudo conveniat. Tertio, quia ex condemnatione alicuius eorum derogaretur in opinione hominum dignitati illius ecclesiae et auctoritati. Quod est periculosius quam in ea tolerare aliquem peccatorem, nisi valde publicum et manifestum, de quo grave scandalum oriretur.