HATCHER, JUDGE:

Plaintiff was injured by an automobile owned by defendant. This is an action to recover damages for that injury. A verdict was returned in favor of defendant which was set aside by the trial court. The defendant seeks here to sustain that verdict.

The place of the injury was near Fairmont on a public road which is surfaced with concrete a width of eighteen feet; the time was between six and seven P. M., January 31, 1928. The plaintiff was walking along or near the right edge of the concrete and was struck from behind. The automobile was proceeding in the same direction as the plaintiff and was driven by an employee of defendant. The excuse given by the driver for not avoiding the accident is that he was partially blinded by the headlights of an approaching car. The driver was employed to sell used cars, at the price and on the terms fixed by defendant, who is in the garage business at Fairmont; the driver was authorized by defendant to go anywhere within reason on his own initiative in search of a purchaser, subject to the order that he should return the car to the garage by eight P. M. He was privileged to "pick up" his friends on his trips. At the time of the accident, the driver was taking a friend (with three other persons) to Fairmont to consult a doctor. The friend lived at Everson, which is about seven miles from Fairmont. None of the passengers were prospects for a sale. Under these circumstances the defendant contends that the relation of master and servant between him and the driver did not exist at the time of the accident and that he is not liable to the plaintiff, as a master.

With great diligence counsel for each litigant has cited many cases as bearing on this question. The citations lead us to no solution, however, as they are in hopeless conflict. This condition prevails not only in the several jurisdictions in the United States but sometimes in the same jurisdiction. Hundreds of opinions have been written on this subject. Nice and subtle distinctions ("almost too fine to be acceptable." Pollock, Torts, (13th Ed.), p. 90), have been drawn, as to just when a servant steps without the ambit of his duty and the precise moment when he re-enters it. The mental attitude of the servant as well as his acts, the distance, the time and the direction travelled from what has been termed "the path of authorized conduct" have each been accorded material significance in solving the riddle. Most of the decisions are merely the result of stare decisis. In the leading cases, it is apparent that their diversity is due to different judicial conceptions of the master's liability. This difference is plausibly explained by Young B. Smith in a thoughtful article in the Col. Law Review, 23 Vol. 463: "If a court doubts the wisdom of a rule, a narrow and restricted application is to be expected. On the other hand, if the court approves the policy underlying a rule, a broad and liberal application generally follows. Then, again if the court entertains no convictions about a rule, a more or less mechanical application ensues, ofter leading to inconsistencies and confusion." In order to avoid, in this case, a purely mechanical application of the rule of vicarious liability, we have made some research into its history and its rationale.

Again we are met with confusion. The rule generally bears the impress of a Latin formula, respondeat superior, (let the principal answer). The Roman law, however, never extended vicarious responsibility beyond the (a) liability of masters for acts of their slaves; (b) absolute liability of shipowners, innkeepers and stablekeepers for property intrusted to their charge; (c) absolute liability of occupants of buildings for injuries caused by things thrown or poured down from the building; (d) liability of a contractor for the negligence of his servants in performing his contract. 6 Labatt's, Master Servant, Ch. XCV. Justice Holmes cynically calls the modern rule a bare fiction — "a form of words" — implanted upon the Patria Potestas (Paternal authority) recognized by the Roman law. See article on Agency, 4th Harvard Law Review, 345. He finds partial support in Shearman Redfield, The Law of Negligence, (6th Ed.), sec. 142, and 2 Kent. Com. (12th Ed.), 260, note 1. Prof. Wigmore takes a different view, attributing the rule to a gradual development of English thought commencing about the year 1300 and not based on or influenced by the Roman law. He traces it through three phases: (1) when the act of the servant was directly commanded by the master (about 1300); (2) when the act was impliedly so commended (about 1700); (3) when the act was within the scope of authority or course of employment (about 1800). See, article on Responsibility for Tortious Acts, 7 Harvard Law Review, 381-2. Pollock and Maitland in their History of the English Law, 2 Vol. 528, etc., find no definite trace of our present doctrine prior to 1688. They agree with Wigmore thus far, that the doctrine had its inception "in phrases which teach that a master is liable for acts that he has 'impliedly' as well as for those which he has 'expressly' commanded" (p. 529). Labatt, supra, sec. 2234, reaches about the same conclusion as Pollock and Maitland. It is generally conceded among legal historians that the modern law of respondeat superior is founded on expressions of Holt, C. J. In Boston v. Sanford, (1689) 2 Salk. 440, he is reported as saying: "Whoever employs another is answerable for him." In Turburville v. Stampe (1698), 1 Ld. Raym. 264, the Chief Justice was more exact: "If my servant doth anything prejudicial to another, it shall bind me, where it may be presumed that he acts by my authority, being about my business." These pronouncements have been called "unwarranted dicta". Be that as it may, the latter statement of Holt was accepted by the English judges and furnishes a fair conception of the master's liability as it exists even to this day.

Many different foundations for the rule have been suggested. A very common one among judges and writers is the maxim, qui facit per alium facit per se. ("He who acts through another acts by himself.") 1 Blackstone, Com. 429, 430; 18 R. C. L., subject, Master and Servant, sec. 247; Wood, Master and Servant, sec. 277. This maxim, however, as pointed out by Pollock, "states the effect of the rule, not any reason for it." One reason offered by this author is that whoever "exposes others to risk (per alium) should abide the consequences if the risk ripens into actual harm." Essays in Jurisprudence, p. 122. And then again, "I am answerable for the wrongs of my servant or agent, not because he is authorized by me or personally represents me, but because he is about my affairs, and I am bound to see that my affairs are conducted with due regard to the safety of others." Pollock, Torts, p. 81. The earlier decisions attributed the rule to such principles as: "Whenever one of two innocent persons must suffer by the acts of a third, he who has enabled said third persons to occasion the loss must sustain it," ( Lickbarrow v. Mason, 1787, 2 T. R. 70); and "he who expects to derive advantage from an act which is done by another for him, must answer for any injury which a third person may sustain from it," (Hall v. Smith, 1824, 2 Bing. 145). The first principle above was adopted by Bishop in sec. 608, Non-Contract Law. The second principle seemed sufficient to the present savant of our own Court, JUDGE LIVELY, in Wills v. Gas Co., 104 W. Va. 12, 17. "The rule is founded on the principles of justice between man and man," declared Caldwell, J., in R. R. Co. v. Stevens, 20 Ohio 415, 432, (Lawrence). Many other judges have taken this view. The law writers generally, however, have based the rule entirely on consideration of public policy. "The rule making the master liable does not depend upon foundations of natural justice, but is defended upon consideration of expediency." Floyd R. Mechem on Employers' Liabilities, 44 Am. Law Review, 228; Pothier on Obligations, sec. 121. Labatt says in section 2248, supra: "The considerations which in this point of view (public policy) have been adverted to as justifying the adoption of the rule are these: Its efficacy as a means of 'insuring vigilance in selecting and superintending servants;' 'the expediency of throwing the risk upon those who can best guard against it;' the circumstance that adequate compensation would very seldom be obtainable if suits against servants were the only remedies available to persons aggrieved by their tortious acts; and the general consideration that 'in no other way could there be any safety to third persons in their dealings, either directly with the principal, or indirectly with hint through the instrumentality of agents.' " Public policy was the ground inferentially adopted by another great judge of this Court, JUDGE BRANNON, when he quoted from Story (Agency, sec. 452), that the rule of respondent superior "is founded on public policy and convenience." Gregory's Adm'r. v. R'd. Co., 37 W. Va. 606, 614.

Our research has not induced us to take issue with JUDGE LIVELY and the host of other judges who would support the rule on principles of justice. We find no reason to combat the conclusion of JUDGE, BRANNON Professor Mechem and other literati, that the rule is based on public policy. Because these two theories have each appealed to jurists and pundits of the highest standing, it would seem that the rule need rely on neither theory alone, but may claim both in its foundation. This conclusion is supported by the opinion of Shaw, C. J., in Farwell v. R. R., (Mass.) Mete. 49, 38 Am. Dec. 339, 340 (a leading case), who attributed the rule both to "the great principle of social duty," etc., and to "general considerations of policy and security." When a rule combines in its support both principles of natural justice and public policy, we are of opinion that it should be liberally applied in favor of those who invoke it.

We have found no more comprehensive advice on the application of the rule than that given by Mechem in his great treatise on Agency in section 1879, (2nd Ed.), which is as follows:

"Since no act can be completely isolated from its surroundings, since every act must have its penumbra of incident and attribute, it is essential that some term shall be found which shall include, not merely the act itself, but this train of attendant circumstances. For the lack of a better term it is said that in order to charge the master with the servant's negligence, the servant must be acting 'in the course of his undertaking' or 'within the course of his employment.' This term 'course of his employment,' like the corresponding term 'the scope of the authority' in cases of agency, and 'the scope of business' in cases of partnership, is one not capable of precise definition although many attempts have been made to define it. It is largely a question of fact and its determination may vary in each case in view of the particular circumstances. The utmost that can ordinarily be said is that a servant is acting within the course of his employment when he is engaged in doing, for his master, either the act consciously and specifically directed or any act which can fairly and reasonably be deemed to be an ordinary and natural incident or attribute of that act or a natural, direct and logical result of it. If in doing such an act, the servant acts negligently, that is negligence within the course of the employment." Accord: 39 C. J., subject, Master and Servant, sec. 1472.

It is apparent that at the time of the accident the driver was not engaged directly in defendant's business, in a narrow and restricted sense; but, as we have seen from Mechem, the rule of respondeat superior is not limited by such a restriction. See also Cooley on Torts, (3rd Ed.), sec. 1027 (632). It may be conceded that the driver did not even intend to further the master's business by the course he was pursuing at that time; but his mental attitude, alone, is not controlling. Riley v. Oil Company, 213 N.Y. 301; 39 C. J., supra, sec. 1475. The real test is "in the relation which the act done bears to the employment." Mechem, supra, sec. 1880. Much of the driver's information as to prospective purchasers necessarily came through his friends and acquaintances. One purpose of his "roving commission" was evidently to allow him to visit at will among his friends and acquaintances, in search of this information. It is reasonable that the defendant had this in mind in giving the driver permission to pick up friends. A friend picked up became an eager informant as well as a partisan of the driver, and the interest of the defendant was thus promoted. Authority to pick up friends impliedly included permission for the driver to put aside momentarily the direct search for purchasers in order to serve the friends. It is a common trait of servants to usurp authority, in the absence of the master. Knowledge of this trait is chargeable to the defendant, as an employer of labor. Mere abuse of authority by the servant, does not sever the relation of master and servant. Labatt, supra, sec. 2277; 39 C. J., supra, sec. 1476. It is not necessary that the defendant should have anticipated the exact circumstances surrounding this accident. It is sufficient if it was probable that the driver would at times exceed his express authority and carry a friend (a greater or less distance) solely as an accommodation to the friend. Mechem, supra, sec. 1884. We are of opinion that such an act of accommodation was a natural consequence of the broad powers conferred on the driver, and that the defendant should have anticipated that at some time some place the driver would do just that thing. If so, the act which occasioned the plaintiff's injury was within the scope of the driver's employment, and the defendant is responsible for that act. As there was no conflict in the evidence on this subject, it was a question of law for the trial court, and should not have been submitted to the jury as was done by several instructions given on behalf of defendant. This was error and warranted the court in setting aside the verdict. 39 C. J., supra, sec. 1593.

We observe no special prejudice to plaintiff because of the modification of his instructions Nos. 1 and 2 by the court. We suggest that counsel examine the case of Keller v. N. W. Ry. Co., recently decided by this Court (advance sheets 156 S.E. 50), if the question of the last clear chance arises upon another trial of this case.

Much of the briefs is devoted to questions of negligence and contributory negligence. These questions are for jury instead of judicial determination.

The judgment of the trial court is affirmed.

Affirmed.