to go outside these physical speculations, whether the Trojan or the Peloponnesian war was the greatest, what was the nature of the shield of Achilles, or whether there was more than one Hercules. In forensic cases, however, which consist of accusation and defence, there is one kind of conjecture by which we enquire both about an act and about its author. This sometimes treats the two questions together, as, for example, when both the act and the identity of the author are denied, and sometimes separately, as when the first enquiry, whether the act was committed, is followed by a second, where, the act being admitted, the question is by whom it was committed.

The act itself again sometimes involves a single question, as, for example, whether a man is dead, and sometimes two, as, for instance, whether he died of poison or of some internal disease. Another form of conjecture is concerned with the act alone, it being admitted that if the act was really committed, there can be no doubt as to its author. A third form is concerned solely with the author, the act being admitted and the dispute turning on the question as to who committed it. This third form is complex.

For the accused either confines himself to denying that he did it or accuses another of having done it. Further, there is more than one way of transferring the charge to another. At times this results in mutual accusation, which the Greeks call ἀντικατηγορία, and some of our own authors concertative accusation. i.e. mutual or reciprocal accusation, see VII. i. 3. At times, on the other hand, the charge is transferred to some person who cannot be brought to trial, and may be either known or unknown: again, if the person is known, he may be someone outside the case or the victim himself, who may be alleged to have committed suicide.

In such cases we compare characters, motives and other circumstances in the same way as in eases of mutual accusation. Cicero, for instance, in the pro Vareno diverts the charge from the accused to the slaves of Ancharius and in the pro Scauro throws the suspicion of Bostar's murder upon his mother.

There is also a different form of comparison, which comes into play when both parties claim the credit of some act, and yet another kind, when the question is not as between two persons, but as between two acts; that is to say, the question is not which of the two committed an act, but which of two acts was committed. Finally, if the act and the identity of the author are both admitted, we may still raise the question of his intention. I shall now proceed to detail. As an example of joint denial covering both the act and the identity of the author we may take the following statements, I have not committed adultery, I have not sought to establish myself as tyrant. In cases of murder or poisoning the denial is often divided as follows:

The act was not committed, and, if it was committed, it was not by me. But if the defence say, Prove that the man was killed, the burden falls solely on the accuser, for the accused can say nothing more against the charge except perhaps in the way of casting certain suspicions, which he should throw out in the vaguest terms, since if you make one definite assertion, you will have to prove it or run the risk of losing your case. For when the question lies between our statement and that of our opponent, one or other will be regarded as true. Thus when the point on which we relied for our defence is overthrown, there is nothing left but the points that tell against us.

For example, when the question turns on symptoms, which may point either to poisoning or internal disease, there is no third course left open and consequently each party sticks to his statement. At times the question turns on the nature of the fact, whether, for instance, death was due to poisoning or internal disease, and arguments are introduced which are drawn from the circumstances alone without any reference to the person concerned.

For example, it makes a difference whether the deceased was cheerful or depressed, had been working or taking his ease, had been awake or sleeping previous to the festive gathering that was followed by his death. The age of the deceased is also an important factor, and it is desirable to know whether he died suddenly or after a long period of ill health. If the question turns only on his sudden death, both parties will have still freer scope for discussion.

At times the character of the accused may be adduced to prove the fact, and to make it likely that it was or was not a case of poisoning because the accused is or is not a likely person to have committed such an act. When, on the other hand, the enquiry concerns both the accused and the act, the natural order for the accuser to pursue is to commence by proving that the act has been committed and then to go on to show that it was committed by the accused. If, however, proofs of the authorship of the crime are more in number than the proofs of the commission, this order may be reversed.

On the other hand, the accused will always begin by denying the act, since if this can be successfully proved, there is no need to say anything more, while if it is not proved, there remain other means of defence. Similarly, when the dispute turns solely on the act and, the act being proved, there can be no doubt as to the author, arguments may be drawn in like manner both from the person and the facts, although with reference to the question of fact alone.

Take the following controversial theme as an example, for it is best to employ scholastic themes as illustrations since they are more familiar to the student. A man who had been disinherited by his father took to the study of medicine. His father fell sick and, his life being despaired of by the other doctors, the son was called in, and said he would cure him if he would take a draught prescribed by himself. The father after drinking part of the draught said that he had been poisoned: the son drank the remainder of the draught. The father died and the son is accused of parricide.

There is no doubt who administered the draught, and, if it was poison, there is no question as to the author: but the problem as to whether the draught was poison can only be decided by arguments drawn from the character of the accused. There remains a third type of conjectural case where the fact is admitted, and the only question is as to the author. It is unnecessary for me to quote examples, since such cases are of frequent occurrence. For example, it may be clear that a man has been killed or that sacrilege has been committed, but the person accused of the crime may deny his guilt. It is from such circumstances that cases of mutual accusation arise, where it is admitted that the crime has been committed, but each party charges the other with being the author.

With regard to this class of case Celsus points out that they cannot actually occur in the courts, a fact which I imagine is familiar to all: for the jury is empanelled to try one accused person only, and even though the defence and the prosecution may accuse each other of tile crime, the first case must be tried before the second.

Apollodorus again stated that mutual accusation involved two separate disputes, and this is of course in conformity with the practice of the courts, which insists on two separate trials. On the other hand, mutual accusation is possible in cases tried before the senate or the emperor, and even in the courts the fact of mutual accusation will involve no difference in the pleadings, since the same methods will be required whether the verdict is given on both charges simultaneously or only on one.

In such cases the defence must always come first for three reasons. In the first place, we naturally prefer to secure our own safety than to injure our opponent, while secondly, our accusation will carry greater weight if we have first proved our own innocence, and thirdly, we shall thus secure a double line of defence. For the man who says, I did not kill him, is then free to go on to say, You killed him, whereas it is superfluous for the man who says, You killed him, to go on to say, I did not kill him.

Further, such cases consist of comparison, which may be effected in different ways. For we may either compare our case in its entirety with that of our adversary, or we may compare individual arguments. The choice between these two methods can only be determined by the requirements of the case. For example, in the pro Vareno, Cicero, in dealing with the first charge, compares the individual arguments: for it would have been rash in connexion with the second charge to compare the position of a stranger with that of a mother. The pro Vareno being lost, it is impossible to say to what this refers, and for the same reason Halm's conjecture must be regarded as quite uncertain. It is therefore best, if possible, to refute argument by individual argument: if, however, our individual arguments are weak, we shall try to secure success by comparison of case with case as a whole.

But whether the case is one of mutual accusation, or the accused throws the guilt upon his opponent without making any formal accusation (as Roscius Roscius Amerinus, accused of parricide and defended by Cicero. did without indicting his accusers), or the responsibility for the deed be placed on the victims themselves, whom we allege to have perished by their own hand, the arguments for both sides of the case will be compared in exactly the same way as in cases of mutual accusation.