tity of handwriting, are known to have much difficulty in divesting their minds from the predisposition to accept the view as to such identity which is unconsciously received by them from the party who first puts the papers in their hands.

Of the inadequacy of memory and language exactly to represent a particular scene as it really took place, we have constant illustrations in the cross-examinations and recallings of witnesses during every protractedly contested trial. There is one probability in a thousand that a witness may be perjured; there is one in fifty or twenty or ten that he may be so prejudiced as unconsciously to misstate; there is a far higher probability that his statement may not be exactly true. All these probabilities the jury have to weigh; and the conclusions they reach must be inferences from circumstances. Even in the case of the abstract witness, without antecedents or circumjacents, whom this hypothesis presents to us, the jury would infer, if such a witness were possible, a want of credibility from the very circumstance that the witness comes forward in this anomalous isolation. But no such witness exists or can exist. Every witness has some circumstances about him from which inferences as to his veracity and capacity may be made. Hence every case depending nominally upon what is called direct testimony, depends really upon that which is circumstantial. Hence if we are to hold that in circumstantial evidence there can be no conviction if the facts "are incapable of explanation upon any other reasonable hypothesis than that of guilt," we must hold this to be the case with all evidence. If we do not hold this as to evidence in general, we must not hold it as to that kind of evidence called peculiarly circumstantial.

§ 743. The conclusion to which the reasoning we have been following leads us is this: All evidence is more or less circumstantial; all statements of witnesses, all conclusions of juries, are the results of inferences. There is therefore no ground for the distinction between "circumstantial and "direct" evidence. All evidence admitted by the court is to be considered by the jury in making up their verdict; and their duty is to acquit if on such evidence there is reasonable doubt of the defendant's guilt; if otherwise to convict.j

(j) The following cases may be consulted as to the points made in the

text: People v. Strong, 30 Cal. 151; People v. Dick, 32 Cal. 213; People v.

It is important to observe that wherever circumstantial evidence is introduced to connect the defendant with the criminal act, the circumstantial facts themselves must be connected with the defendant, or they will be incompetent as evidence.j1

Where the only question is as to the identity of the prisoner with the guilty party, the jury may be justified in returning a verdict of guilty, although no witness will swear positively to the identity.2

In weighing evidence, whether direct or circumstantial, the following rules must be kept in mind:

1. The Onus of proving Everything essential to the Establishment of the Charge lies on the Prosecutor, though the Non-production of Explanatory Evidence, clearly in the Power of Defendant, must weigh against him. m

§ 744. In every criminal case, as has been seen, the defendant's guilt must be made out by evidence sufficiently conclusive to exclude any reasonable supposition of his innocence.p

The neglect of a defendant to produce evidence of good characacter does not, it should be observed, afford ground for a presumption of law against him, and it should not be so left to the jury by the court. q

The non-production, by a defendant, of evidence to prove his whereabouts, at the time when the offence was committed, and when there is strong circumstantial evidence implicating him as guilty, is not, in point of law, a conclusive presumption of his guilt. It is a presumption of fact to be left with the jury.r

Cronin, 34 Cal. 191; Orr v. State, 34 Ga. 342; Martin v. State, 38 Ga. 293; State v. Johnson, 19 Iowa, 230; State v. Collins, 20 Iowa, 85; Hall v. State, 40 Ala. 698; Mose v. State, 36 Ala. 211; Chisholm v. State, 45 Ala. 66; Phipps v. State, 3 Cold. (Tenn.) 344; Com. v. Drum, 58 Penn. 9; Com. v. Annis, 15 Gray, 197; Conner v. State, 34 Texas, 659; Com. v. James, 37 Conn. 355; Bowler v. State, 41 Missis. 570; Com. v. Hanlon, 3 Brewster, 461; State v. Ford, 21 Wis. 610; State v. Daley, 41 Vt. 564; Schuster v. State, 29 Ind. 394.

j1 People v. Kennedy, 32 N. Y. 141. j2 Com. v. Cunningham, 104 Mass. 545 (1870).

m R. v. Burdett, 4 Barn. & Ald. 140; Starkie on Evid. 436; Bowler v. State, 41 Miss. 570; Dranquet v. Prudhomme, 3 Louisiana R. 83, 86; Jones v. Kennedy, 11 Pick. 125, 132; Wills on Circum. Ev. 183.

p See ante, § 707; State v. Newman, 7 Ala. 69; Tomkins v. State, 32 Ala. 573.

9 See ante, § 637.

r Toler v. State, 16 Ohio St. R. 583. See ante, § 709, 716; and see White

On the other hand, the burden of proving an alibi is on the defendant. But this is only on the supposition that the prosecution has already proved the corpus delicti, and given proof of the defendant's participation.

In proving an alibi its not necessary to establish coincidence in any point of artificial time. 81

2. There must be Clear and Unequivocal Proof of the corpus delicti.

§ 745. The fact of the commission of the offence must necessarily be the foundation of every criminal suit, and until that fact is proved, most dangerous would it be to convict. t "I would

never," says Lord Hale, “convict any person for stealing the goods of a person unknown, merely because he would not give an account how he came by them, unless there were due pro of made that a felony had been committed. I would never convict any person of murder or manslaughter, unless the fact were proved to be done, or at least, the body found dead." u Equally emphatic was the language of another great judge: "To take presumptions, in order to swell an equivocal and ambiguous fact into a criminal fact, would, I take it, be an entire misapplication of the doctrine of presumptions." v And the Roman law is the same: "Diligenter cavendum est judici, ne supplicium præcipitet, antequam de crimine constiterit." w "De corpore interfecti ne

cesse est ut constet." x The death in such a case should be distinctly proved, either by direct evidence of the fact, by inspection of the body, y or by circumstantial evidence strong enough to leave no ground for reasonable doubt. z The proof must be clear and distinct. Thus in a case of horse-stealing, a mere declaration in evidence that the horse had been stolen is not sufficient to prove theft. The facts must appear, so that the judge and jury may see whether such facts in point of law amounted to a felonious taking and carrying away of the property in question. a

377; Smith v. Com. 21 Grat. 809; State v. Keeler, 28 Iowa, 553; People v. Bennett, 49 N. Y. 137.

u 2 Hale P. C. 290; and see Tyner v. State, 5 Humphreys, 383; and see for an interesting case on the corpus delicti in larceny, R. v. Burton, 24 Eng. Law & Eq. 551; 6 Cox C. C.

293.

v Lord Stowell, in Evans v. Evans, Tittman Handb. iii. p. 495; Kitka Be1 Hagg. C. R. 105.

weisler, p. 13. Of the old jurists, see

w Matth. de Crim. in Dig. lib. 48, Blanci de indiciis, Venet, 1545; Bruni tit. 16, ch. 1.

z Matth. Probat. ch. 1, n. 4, p. 9. y 1 Stark. Evid. 575, 3d ed. c. 5. z People v. Rulloff, 3 Parker C. R. (N. Y.) 401.

a Tyner v. State, 5 Humphreys, 383; see Mitchum v. State, 11 Geor. 615. The reference to facts, which, having in themselves no bearing upon the guilt or innocence of the party, are

Guido de Suzaria de indiciis et tortura, Lugd. 1546; Crusuis de tortura et indiciis, Francof. 1704; Menochius de Præsumt., Colon. 1686; Tabor de indiciis delict, Giess. 1767; Cocieji de fallæ, Crim. indic. in ejus exeri. cur. p. 1. uro. 75; Reinhardt de eo quod circa reum ex Præsumt. Convinc et Cond. Just est Erford, 1732; Woltaer femiol, Crim. quæd Capita Ital. 1790; Puett

§ 746. In most cases the proof of the crime is separable from that of the criminal. Thus, the finding of a dead body, or a house in ashes, indicates the probable crime, but does not necessarily afford any clew to the perpetrator, and here it is necessary to draw a distinction relative to the effect of presumptive evidence. The corpus delicti in such cases is made up of two things; first, certain facts forming its basis; and secondly, the existence of criminal agency as the cause of them. With respect to the