Under the decision of this court in Long’s case, 38 Georgia, 491, the charge of the judge was error. We do not care to go over the argument. The Code of 1873, by dropping the parts of the section judged to be obsolete under the constitution of 1868, does not present the section as it was passed. To give the section the meaning contended for would be to say that if the case be one of circumstantial evidence, the fate of the prisoner is wholly in the discretion of the judge, whatever the jury may do; since if the former case means, “when the conviction is upon circumstantial evidence,” then it is wholly in the discretion of the judge. We cannot, believe such was the intention of the legislature. There is the same high reason why the jury should have the right of recommendation in such cases as that the judge should exercise his discretion. The whole clause, in our judgment, had in view *124cases of circumstantial evidence. If the jury in such cases saw fit to recommend, the judge was bound; if they did not, he might still, in his discretion, lessen the penalty. The language of the section is not very exact, in any view of it: “In the former case it is discretionary with the judge; in the latter, it is not.” “It” what? In either case the sentence is with the judge. The former case evidently meant, is the case where there is no recommendation, and the latter case where there is. To hold otherwise would be to say that, as we have said, if the conviction is founded solely on circumstantial evidence, the judge may or may not commute the sentence whatever they, the jury, may recommend; whereas, in all other cases, the recommendation of the jury is binding. This, we think, is absurd. As we said in Long’s case, the key to the section is the old law and the object of the Code. So vital' a change in the law as is contended for, is not to be supposed in a law passed under the circumstances of the Code, unless the change is patent and undoubted. Nearly four years have elapsed since the decision in the Long case, and the acquiescence of the legislature in the construction we then put upon the section indicates strongly that the construction is right, and accords with the legislative will.

Assuming the charge to be wrong, we think the prisoner entitled to a new trial. The conviction was clearly had under the belief that the recommendation would save his life. Archer’s case, in 35 Georgia, 5, is exactly in point; and Long’s case, in 38 Georgia, is not against this view. In Long’s case, the jury refused to recommend, though the court told them they might do it. The charge did Long no possible harm. It was more favorable to him than the law warranted, and the jury, confident of his guilt, and the absence of any mitigating circumstance, thought he was not entitled to mercy. Who shall say the jury would have found this verdict had the judge not misled them?

It is said that the question of his guilt of the crime of murder is wholly independent of the opinion of the jury that he ought not to suffer death. Our law makes a jury man, who *125has conscientious scruples on the subject of capital punishment, a good cause of challenge in cases where the^ penalty may be death. Why? Because the law wisely recognizes the fact that such opinions do affect the verdict. And this is the law of human nature. One of the old writers quaintly though wisely remarks: “He whose stroke is death should be very wary in striking.” And'all experience shows that men of the highest virtue and' conscientiousness, act upon the advice. Is it not the demand, not only of humanity, but of sound sense, to deliberate more seriously, to weigh more cautiously, our acts, in proportion as their results are more important. The law is full of just such principles. Ordinary care and extraordinary care are familiar terms. One is the care we take about trifles; the other the care we exercise in deciding important matters. And even in matters of mere business, men are justified by law in dealing less solemnly with unimportant than with important things.

This jury deliberated on the evidence under the impression given them, intentionally, by the judge in his charge, that if they found him'guilty, it did not follow that the prisoner would be sentenced to death. Their verdict shows that they found him guilty with that understanding. It is not necessary to read the affidavits to learn this; it is expressed in terms in the verdict, and was the law for their action, as laid down by the court. Who shall say that the verdict would have been guilty had there been no such instructions? Who shall say there would not have been more hesitation, more caution, in weighing the testimony ? We do not go into the evidence. From some of it, if worthy of credit, it is not clear that the offense is not manslaughter merely; and -whatever a clearheaded, confident man might think of his capacity to come to an opinion of the truth of a fact, entirely independently of the consequences of his conclusion, we all know that this is not true of the average of men, and I doubt if it be true of any man.

In our judgment, the effect of this charge was to put the prisoner at a disadvantage — to lessen in the minds of the jury *126that care, deliberation and caution, which it is not only their right, but their duty, to exercise in cases so important a conclusion ; and as the record itself shows that the jury thought lie ought not to suffer death, we think it is plain that he was damaged by the charge.

Judgment reversed.