Clyde Arnold Weldon was found guilty by the court of conspiracy to commit robbery.1 He had given a confession to the crime which was introduced in evidence and formed the basis of his conviction. The ground of his unsuccessful motion for dismissal below, and on appeal, is .that there was not sufficient evidence, independent of his confession to establish the corpus delicti •of the crime and justify his conviction.

The pertinent facts are: Shortly after midnight on May 15, 1956, Jack F. Miller, a policeman of Cedar City, Utah, acting on information that a robbery of the Safeway Store was planned, went with one Robinson to a hotel room occupied by defendant, Weldon and Robert Harke. These men were found in the room and each had in his possession a loaded revolver, Weldon’s being in the pocket of his jacket, which was hanging over the back of a chair. Questioning these men revealed that they were strangers in town; they did not offer any explanation of any proper business or employment, nor of any lawful purpose for having the loaded pistols. Upon being confronted with the information the police had received concerning the contemplated robbery, Weldon admitted • that such was their intent and signed a written confession outlining the details of the planned robbery.

The rule is quite universal that an extrajudicial confession, by itself, is not sufficient to sustain a conviction of a crime, but there must be evidence, independent of the confession to establish the corpus delicti.2

This rule had its origin in the English courts. It began in cases of homicide and was first pronounced by Sir Mathew Hale when he wrote, “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.” 3 The purpose of the rule was to safeguard against convicting the innocent on the strength of false confessions. It appears that there were several actual cases where persons innocent of the crime were convicted of murder and executed and the supposed victims later appeared alive.4

*374Though the rule was extended to apply to other crimes, in practical application the courts have attempted to hold it within the bounds of reason consistent with its original purpose of guarding against convicting the innocent and so that it is not applied to create a device for protecting defendants who reek with guilt. For example, the circumstances of possession of property suspected of having been stolen, may have been of such character as to justify conviction of theft with a minimum of separate proof of the corpus delicti where suspicious circumstances and the- reasonable inferences derivable therefrom would justify reasonable minds in believing the crime had been committed.

Said Justice Maulé in the case of Regina v. Burton,5 “If a man were to go into the London docks quite sober, and shortly afterwards were to be found very drunk, staggering out of one of the cellars, in which above a million gallons of wine are stowed, I think that this would be reasonable evidence that the man had stolen some of the wine in the cellar, though no proof were given that any particular vat had been broached and that any wine had actually been missed.” In that case a prisoner accused of larceny was seen coming out of a warehouse in the London docks near where pepper was kept, and where he had no business to be. A constable who suspected him from the bulky state of his pockets stopped him and said, “I think there is something wrong about youupon which the accused said, “I hope you will not be hard upon meand then threw a quantity of pepper out of his pocket on the ground. The constable stated that he could not say whether any pepper had been stolen, nor that any pepper had been missed; but that which was found upon the prisoner was of like description to the pepper in the warehouse. It was held by all of the judges that the prisoner was properly convicted of larceny.

In another early case6 a prisoner was found about 1200 yards from a “fowl house” with dead fowls in his possession, of which he could give no account. On the floor of the coop were some feathers corresponding with the feathers of a fowl found on the prisoner, from the neck of which feathers had been removed. The prosecutor, not knowing the number of fowls kept, could not swear that he had lost any but it was held that there was evidence to support a conviction for larceny.

In this country, the corpus delicti rule is applied to all crimes although there is considerable variation in the manner of its application. Notwithstanding its universality, eminent authorities have gravely *375-doubted its validity. As one writer aptly put it, “Legal doctrines, while appropriate in one setting, may become a deterrent to justice when overextended.”7 Wigmore makes this comment: 8

“The policy of any rule of the sort is questionable. * * * Common intelligence and caution, in the jurors’ minds, will sufficiently appreciate it, without a laying on of the road in the shape of a rule of law. Moreover, the danger which it is supposed to guard against is greatly exaggerated in common thought. That danger lies wholly in a false confession of guilt. Such confessions, however, so far as handed down to us in the annals of our courts, have been exceedingly rare * * * this rule, and all such rules, are today constantly resorted to by unscrupulous counsel as mere verbal formulas with which to entrap the trial judge into an error of words in his charge to the jury. These capabilities of abuse make it often a positive obstruction to the course of justice.”

Another well-known authority in the field of evidence, Charles T. McCormick, had the following to say about the rule: 9

“The English courts and text writers have warned that, in homicide cases, it is dangerous to convict upon a confession alone without some additional evidence such as the finding of the body, * * *. This rather flexible practice has, in most American courts hardened into a rigid rule based upon the premise that confessions generally are an unreliable class of evidence. This premise as suggested above, seems to be quite unfounded, and the cases applying the rule give for the most part the impression of measuring the proof according to a mechanical yardstick, rather than by a standard aimed at convicting the innocent.

* !* * * *

“It is submitted that hard-and-fast rules requiring corroboration are as likely to obstruct the punishment of the guilty as they are to safeguard the innocent.”

The rule has also been subjected to criticism on the ground that persons who falsely claim to have committed crimes are much more likely to confess to some identifiable crime the corpus delicti of which can easily be proved.10 There is even more doubt of its value as to minor crimes because false confessions are more likely to be of murder or the more sensational ones.

*376The arguments presented by those who criticize the rule are not without some merit, but in deference to the time honored and important precept of our law that it is better that ten guilty go free, than that one innocent be punished, we feel that there is wisdom in requiring proof of the corpus delicti independent of a confession as a precaution against convicting the innocent in the few' instances when persons mentally disturbed or impelled by strange motivations, might confess to crimes they did not commit. However, we also realize that due to the natural compunction people have against confessing to or implicating themselves in crimes, such false confessions are extremely rare. Therefore, the rule should be applied with caution and not permitted to be used as a technical obstruction to the administration of justice.

The above considerations have caused courts to adopt a variety of tests as to the quantum of proof necessary to satisfy the requirement. Although they vary, it seems quite generally agreed that the evidence of the corpus delicti need not be “beyond reasonable doubt,” “conclusive” or “sufficient to warrant a conviction,” independent of other evidence.11 From a perusal of such authorities it seems to us that the generally accepted view, to which we give our approval, is that the evidence independent of the confession need not establish the corpus delicti by separate, full or positive proof, and that the whole evidence, including the confession, may be considered together in determining whether the corpus delicti has been satisfactorily established.12 As to the sufficiency of this outside evidence, several jurisdictions require that the independent evidence of the fact, that the crime was committed need only be “slight,”13 or “prima facie,”14 while others say that it must be “substantial” 15 or that it must “establish the corpus delicti to a probability.” 16

The authorities we have examined, considered in the light of the problems involved in making this determination in a criminal proceeding, lead the writer to believe that the rule best suited to the administration of justice is that there must be *377substantial separate evidence of the corpus delicti, such that reasonable minds could believe that the crime is a real one which was in fact committed, and not one which is fanciful or imaginary.17 This seems a sufficient safeguard against convicting the innocent on a spurious confession, inasmuch as the burden ultimately rests upon the state to prove the whole offense beyond a reasonable doubt.

In our own jurisdiction there has been some variation in the statements of this court concerning the degree of proof requisite to separately establish the corpus delicti.18 Notwithstanding the writer’s preference for the rule just discussed above, I recognize that in the most recent pronouncement of this court on the subject, we gave approval to the expression of the rule that there “must be independent, clear and convincing evidence of the corpus delicti, * * * ” 19 which rule indeed has the approval of some other courts, and to which I attorn as the law of this jurisdiction.

It is appreciated that a plausible .agrument can be made that the facts here shown, independent of the confession, constitute sufficient independent evidence of the corpus delicti of the crime of conspiracy to commit a planned robbery. Two strangers in town, in a hotel room with loaded pistols, a somewhat rare circumstance nowadays, in the light of ordinary experience and common sense, seems to require some explanation. Under the prerogative of the trial court to draw every reasonable inference that may fairly arise from the evidence, it seems to the writer a close question whether the evidence was sufficient to stand up against the motion for dismissal. However, my concurring colleagues are of the opinion that there is not sufficient evidence to meet the requirements as to independent proof of the corpus delicti, whatever rule as to the quantum of proof is applied, and I therefore resolve my doubts by deferring to their judgment on that point.

The judgment of conviction is reversed.

(concurring and dissenting).

*378I concur in the result reached by the majority opinion, but am compelled to disagree both with the reasoning and the rule which its author prefers in this type of case, even though it is conceded that there is not sufficient evidence to make the rule operative. I am of the opinion that this court should repudiate the rule preferred by the author of the majority opinion.

In the case of State v. Ferry,1 our most recent declaration on this point, we, in October, 1954, observed:

“An accused cannot be convicted on his confession alone. We believe mid hold that in addition there must be independent, clear and convincing evidence of the corpus delicti, although we and the authorities generally do not require it to be convincing beyond a reasonable doubt.”

Under the rule contended for in the majority opinion, every guide post and measuring stick appropriate for the trial judge to use are cast aside and there is substituted in their stead belief by reasonable minds. If the judge is satisfied that the crime charged is a real one and has been committed, the defendant has had his day in court and enjoyed to the top degree his constitutional guarantees. It needs only to quote the rule contended for — which seems to have found very limited support — in order to see the fallacy of the same. As-stated in the majority opinion, it is as-follows:

“ * * * The rule best suited to the administration of justice is that there must be substantial separate evidence of the corpus delicti, such that reasonable minds could believe that the crime is a real one which was in fact committed, and not one which is fanciful or imaginary.”

This rule rejects the idea of requiring-separate proof by clear and convincing evidence notwithstanding we said and held in the Ferry case, supra, that there must be additional evidence which is clear and con- • vincing.

The opinion is nothing less than an attempt to weaken an announced and accepted rule of this court in a situation where the same cannot have any bearing, since it is ■ conceded that there is not sufficient evidence to meet the requirements of the rule.

I am of the opinion that the concluding - statement in the majority opinion in effect concedes that the rule as stated cannot be accepted. In this case the evidence was • sufficient to support belief by reasonable minds that the crime charged is a real one - which has been committed and not one-which is fanciful or imaginary.

*379It seems an unjustified position to say that the evidence warrants a belief by reasonable minds that the crime charged is a real one which has been committed and then íurn^ — after reasonable minds have been fully satisfied as to the sufficiency of the •evidence for that purpose — and say that the reasonable minds should not have been convinced.

(concurring in the result).

I concur in the result but differ as to ■the pertinency of some of the citations in the main opinion, and do not subscribe to the dicta concerning rules in other jurisdictions.

The rule in State v. Ferry was unanimously adopted by this court two years ago and represents the rule to which the main opinion, as well as this writer, adheres, and my concurrence in the result stems from such reaffirmation.

I believe the fact situations in Regina v. Burton and Regina v. Mockford, cited in ■ the main opinion, are complete misanal-ogies. Also, that our decision in State v. Ferry makes unnecessary any reference to tests or rules enunciated by Connecticut, California, Washington, Maine or any other states, since we have not accepted, but 'ihave rejected them, now re-affirming the .rule in the Ferry case.