those who testify of their own free will and those who are summoned to attend in the public courts of whom the former are available to either party, the latter solely to the accusers, we must distinguish between the duties of the advocate who produces witnesses and the advocate who refutes them.

He who produces a voluntary witness is in a position to know what he is likely to say: consequently the task of examining him would seem to be rendered easier. But even here such cases make a great demand on the acumen and watchfulness of the advocate, who must see that his witness is neither timid, inconsistent nor imprudent.

For the opposing counsel have a way of making a witness lose his head or of leading him into some trap; and once a witness trips, he does more harm to his own side than he would have done good, had he retained his composure and presence of mind. The advocate must therefore put his witnesses through their paces thoroughly in private before they appear in court and must test them by a variety of questions such as may well be put to them by his opponent. The result will be that they will not contradict themselves or, if they do make some slip, can be set upon their feet again by a timely question from the advocate who produces them.

Still, even in the case of witnesses whose evidence is consistent, we must be on our guard against treachery. For such witnesses are often put up by one's opponent and, after promising to say everything that will help our case, give answers of exactly the opposite character and carry more weight by the admission of facts which tell against us than they would have done had they disproved them.

We must therefore discover what motives they have for doing our opponent a hurt, and the fact that they were once his enemies will not suffice our purpose: we must find out whether they have ceased to be ill-disposed to him or whether they desire by means of their evidence to effect a reconciliation with him, in order to assure ourselves that they have not been bribed or repented of their previous attitude and changed their purpose. Such precautions are necessary even with witnesses who know that what they propose to say is true; but it is still more necessary with those who promise to give false evidence.

For experience shows that they are more likely to repent of their purpose, their promises are less to be relied on, and, if they do keep their promise, their evidence is easier to refute.

Witnesses appearing in answer to a subpoena may be divided into two classes: those who desire to harm the accused, and those who do not. The accuser sometimes is aware of their disposition, sometimes unaware. For the moment let us assume that he is aware of their disposition, although I must point out that in either case the utmost skill is required in their examination.

For if an advocate is producing a witness who is desirous of harming the accused, he must avoid letting this desire become apparent, and must not at once proceed to question him on the point at issue. On the contrary this point must be approached by a circuitous route in such a manner as to make it seem that the statement which the witness is really desirous of making has been forced from him. Again lie should not press the witness too much, for fear he should impair his credit by the glibness with which lie answers every question, but should draw from him just so much as may seem reasonable to elicit from a single witness.

On the other hand in the case of a witness who is reluctant to tell the truth, the essential for successful examination is to extort the truth against his will This can only be done by putting questions which have all the appearance of irrelevance. If this he done, he will give replies which he thinks can do no harm to the party which he favours, and subsequently will be led on from the admissions which he has made to a position which renders it impossible for him to deny the truth of the facts which he is reluctant to state.

For just as in a set speech we usually collect detached arguments which in themselves seem innocuous to the accused, but taken together prove the case against him, so we must ask the reluctant witness a number of questions relative to acts antecedent or subsequent to the case, places, dates, persons, etcetera, with a view to luring him into some reply which will force him to make the admissions which we desire or to contradict his previous evidence.

If this fails, we must content ourselves with making it clear that he is reluctant to tell what he knows, and lead him with a view to tripping him up on some point or other, even though it be irrelevant to the case; we must also keep him in the witness-box for an unusual length of time, so that by saying everything that can be said and more than is necessary on behalf of the accused, he may be rendered suspect to the judge. Thus he will do the accused no less harm than if he had told the truth against him.

But if (to proceed to our second supposition) the advocate does not know what the intentions of the witness may be, he must advance gradually inch by inch and sound him by examination and lead him step by step to the particular reply which it is desired to elicit.

But since these witnesses are sometimes so artful that their first replies are designed to meet the wishes of the questioner, in order to win all the greater credit when subsequently they answer in a very different way, it will be the duty of the advocate to dismiss a suspect witness while he can still do so with advantage.

In the case of advocates for the defence examination is in some respects easier, in some more difficult. It is more difficult because it is rarely possible for them to have any previous knowledge of what the witness is likely to say, and easier because, when they come to cross-examine, they know what he has already said.

Consequently in view of the uncertainty involved, there is need for careful inquiry with a view to discovering the character of the witness against the accused and what are his motives for hostility and what its extent: and all such points about the witness should be set forth in advance and disposed of, whether we desire to represent the evidence against the accused as instigated by hatred, envy, bribery or influence. Further, if our opponents bring forward only a small number of witnesses, we must attack them on that head; if on the other hand they produce an excessive number, we must accuse them of conspiracy: if the witnesses are persons of inconspicuous rank, we must minimise their importance, while if they are powerful, we shall accuse our adversaries of bringing undue influence to bear.

It will, however, be still more helpful if we expose the motives which they have for desiring to injure the accused, and these will vary according to the nature of the case and the parties concerned. For the other lines of argument mentioned above are often answered by the employment of commonplaces on similar lines, since the prosecutor, if he produce but few witnesses of inconspicuous rank, can parade the simple honesty of his methods on the ground that he has produced none save those who are in a position to know the real facts, while if he produce a number of distinguished witnesses, it is even easier to commend them to the court.

But at times, just as we have to praise individual witnesses, so we may have to demolish them, whether their evidence has been given in documentary form or they have been summoned to appear in person. This was easier and of more frequent occurrence in the days when the examination of the witnesses was not deferred till after the conclusion of the pleading. It is not clear to what Quintilian refers. There are, it is true, passages in Cicero where the orator speaks of evidence as already given, but the speeches where these references are found are all second pleadings. With regard to what we should say against individual witnesses, no general rules can be laid down: it will depend on the personality of the witness.

It remains to consider the technique to be followed in the examination of witnesses. The first essential is to know your witness. For a timid witness may be terrorised, a fool outwitted, an irascible man provoked, and vanity flattered. The shrewd and self-possessed witness, on the other hand, must be dismissed at once as being malicious and obstinate; or refuted, not by cross-examination, but by a brief speech from the counsel for the defence; or may be put out of countenance by some jest, if a favourable opportunity presents itself; or, if his past life admits of criticism, his credit may be overthrown by the scandalous charges which can be brought against him.

It has been found advantageous at times when confronted with an honest and respectable witness to refrain from pressing him hard, since it is often the case that those who would have defended themselves manfully against attack are mollified by courtesy. But every question is either concerned with the case itself or with something outside the case. As regards the first type of question counsel for the defence may, by adopting a method which I have already recommended for the prosecutor, Alive, § 17, 18. namely by commencing his examination with questions of an apparently irrelevant and innocent character and then by comparing previous with subsequent replies, frequently lead witnesses into such a position that it becomes possible to extort useful admissions from them against their will.

The schools, it is true, give no instruction either as to theory or practice in this subject, and skill in examination comes rather from natural talent or practice. If, however, I am asked to point out a model for imitation, I can recommend but one, namely that which may be found in the dialogues of the Socratics and more especially of Plato, in which the questions put are so shrewd that although individually as a rule the answers are perfectly satisfactory to the other side, yet the questioner reaches the conclusion at which he is aiming.