Circuit Judge, concurring in part and dissenting in part:
I would affirm in full the district court’s dismissal of plaintiff’s common law claims against United States Steel based upon the offensive physical and verbal acts of Bryan, its employee.
If those acts were done — as some of them concededly were done — they were utterly reprehensible, no matter what their particular coloration or context and even if, as alleged by defendant, some may not have been directly resisted or contemporaneously objected to by plaintiff. Certainly the common law of South Carolina makes *214such acts tortious and remediable as against the employee.
But the question here is not whether those acts are tortious at the hands of the actor, but whether under the common law of South Carolina, on the evidence before the court, and under the legal/factual theories advanced in the trial court and here, the employer, United States Steel, should be held liable for those acts. I agree with the district court that on the common law theories advanced and the evidence laid before the court on the summary judgment motion, United States Steel was not liable as a matter of law.
Only two theories of liability were advanced below, and are urged on this appeal: vicarious liability under respondeat superi- or, and liability by virtue of “ratification” or affirmance by United States Steel of Bryan’s acts.
I agree with the district court and with the lead opinion on this appeal, that as a matter of South Carolina law the employer here is not liable vicariously on a responde-at superior theory. Specifically, I agree that Rabon v. Guardsmark, Inc., 571 F.2d 1277 (4th Cir.1978), is direct precedent for the proposition that under South Carolina law, an employee’s acts, such as those here, that are wholly prompted by personal sexual desires, even though committed on the employer’s premises during work hours, cannot constitute acts done “in furtherance of the employer’s business,” within fair contemplation of the controlling common law principle. That principle cannot properly be extended to include such acts simply on the basis that the employee’s acts were made possible or more easily accomplished by virtue of the employment setting and relationships. To do so would effectively convert the principle into no more than an on-the-job test, which would wholly undercut it as a limiting principle.
Similarly, the principle cannot properly be extended to include such acts simply on the basis that in retrospect they can be seen to have “furthered the master’s business” in some purely fortuitous and attenuated way, as by speeding a subordinate along by a physical assault, including an offensive sexual touching, that was never intended for that purpose.
The only other legal/factual theory actually advanced by the plaintiff below, either in her pleadings or in resisting the defendant’s summary judgment motion, was that of “ratification” or “affirmance” by United States Steel of Bryan’s conduct. This was claimed to have occurred by the defendant’s failure to take punitive action against Bryan, including possibly discharging him. Again, I agree with the district court in its well-reasoned explanation of why that common law agency principle simply was not invoked by the evidence before the court. As the court pointed out, ratification can only occur with respect to a servant’s acts which did not then bind the master but which were done or professedly done on the master’s account, citing, quite properly, Restatement (Second) of Agency, § 82. Here, where the acts done were not done or professedly done on United States Steel’s account, they simply are not subject to ratification under the basic principle.
At this point I differ with the lead opinion, which, without dealing directly with the ratification theory, finds a basis for establishing liability on still a third theory which, though related to ratification, is distinctly not that theory. Relying on Restatement (Second) of Torts, § 317, the lead opinion (possibly joined by Judge Butzner in his separate opinion) holds that there was evidence from which a jury might have found United States Steel directly liable in tort for its negligent failure or refusal to control Bryan after it learned, through Stoutz and directly from plaintiff, of Bryan’s conduct.
As indicated, this theory is distinct from that of ratification, and in critical ways. Its proof turns upon quite different circumstances than those giving rise to ratification, and its establishment would lead to a quite different basis of liability, hence remedy. Most critically, it would relate only to employee acts continued after the employer learned of the employee’s conduct, while ratification and its remedy *215would relate to prior acts. In this very case, that could make a very great difference, both in respect of liability and remedy.
Not only was this “negligent failure to control” theory not pleaded or otherwise advanced below, it was not urged upon us on this appeal. Here, as in the district court, the plaintiff only claimed liability on respondeat superior and ratification theories.
We should not of course automatically refuse to consider theories of claim or defense not properly before us because of procedural defaults or unintended “elections.” Though procedural default rules advance important ongoing values of the adversarial system, they are not ultimate values. Where a more expansive view of the theories effectively presented by a party can be indulged in fairness to the other party, and where justice in the case requires it, courts should not be unwilling to consider them simply because of a purely technical failure to have raised them by name. Civil Rule 15(b)’s litigation-by-consent principle, for example, embodies that very notion.
But here I think we cannot, in fairness to United States Steel, sua sponte, and at this late stage, invoke this negligent failure to control theory, never suggested by the plaintiff, as a basis for possible liability. As indicated, proof of this theory, and correspondingly of the defenses to it, would require quite different evidence in detail than could have been anticipated by defendant from the theories actually advanced. The district judge did not consider it, for the obvious reason that the plaintiff did not invoke it. Under the circumstances, I think we should not do so, still without party prompting.*
The impulse to try to find a fair means for holding a financially responsible employer liable for this sort of egregious misconduct of its presumably less financially responsible employee is understandable. But because of the plaintiff’s own tactical choice to forego the more expansive substantive bases for employer liability under Title VII, the only remaining means in this case are those provided by the common law. That law imposes wise, not unreasonable, limits on the range of an employer’s liability for its employee’s conduct. I believe that, faithfully applied here, they preclude holding United States Steel liable under either of the common law theories of liability advanced by plaintiff in this litigation.
I regret, therefore, that I cannot join fully in the lead opinion’s announced result.
At most, I think we should not do more than point out that upon the remand required by the court’s decision, the plaintiff might of course move to amend her pleadings to introduce this quite different theory of liability. This would at least allow the district court to exercise the discretion which should first be exercised by that court and not by ours, to consider whether, in terms of practical fairness, this new theory should be allowed belatedly into the case. And if allowed by that court, this would at least give defendant an opportunity not yet had to marshal and present any evidence it may have available to avoid liability under that theory.