Theophelis v. Lansing General Hospital

430 Mich. 473 (1988) 424 N.W.2d 478

THEOPHELIS
v.
LANSING GENERAL HOSPITAL

Docket No. 78166, (Calendar No. 2).

Supreme Court of Michigan.

Argued May 4, 1987. Decided June 6, 1988.

Farhat, Story & Kraus, P.C. (by Richard C. Kraus), and Church, Wyble, Kritselis & Robinson, P.C. (by William N. Kritselis), for the plaintiffs.

Kitch, Saurbier, Drutchas, Wagner & Kenney, P.C. (by Ronald E. Wagner and Susan Healy Zitterman), for defendant Lansing General Hospital.

GRIFFIN, J.

In this medical malpractice action we are required to determine whether a 1974 amendment^[1] of the Michigan contribution among tortfeasors act, MCL 600.2925; MSA 27A.2925, abrogated the common-law rule that settlement with, and release of, an agent operates to discharge the principal from vicarious liability for the agent's acts. We conclude that the amendment had no such effect.

We are asked also to decide whether releases executed by the plaintiffs in this case should be reformed as covenants not to sue as a means of avoiding the common-law rule. Our answer is in the negative.

I

Gene Christopher Schneider, a seven-year-old boy, was admitted to Lansing General Hospital for a tonsillectomy and bilateral tympanotomy. On June 16, 1978, following the tympanotomy and before the tonsillectomy, the child suffered a cardiac arrest. After half an hour of resuscitative effort, the child resumed a spontaneous heartbeat, and was taken to the intensive care unit where he was placed on a respirator and cardiac monitor. Several physicians on the hospital staff examined the child and directed treatment in the intensive care unit. On the following day, a second cardiac arrest occurred. The child's condition continued to deteriorate until his death six days later, on June 22, 1978.

Plaintiffs, the personal representative and parents of the decedent, filed this wrongful death action in the Ingham Circuit Court. Among defendants named were Jana Palmer, a certified nurse anesthetist who administered anesthesia for the surgery, and Jack Gilmore, D.O., the supervising anesthesiologist. Both were employed by Capital Anesthesiologists, P.C., which had a contract to perform anesthesia in the hospital. Additional defendants included the Lansing General Hospital; Gerald Gilroy, D.O., who performed the surgery; David Sciamanna, D.O., a resident pediatrician; and several other physicians.

Capital Anesthesiologists and physicians other than those named above were dismissed for reasons unrelated to this appeal. Prior to trial plaintiffs entered into a settlement and release agreement with Nurse Palmer in exchange for a payment of $85,000. Shortly after the trial began, plaintiffs also entered into a settlement and release agreement with Dr. Gilmore. The latter was a structured settlement providing for a cash payment of $72,000 and for installment payments over a period of twenty years or more, depending upon the life of the decedent's father, with a guaranteed total of at least $417,000. In both instances, a release was executed in consideration of the settlement payments and a provision was included in the release specifically reserving plaintiffs' claims against the remaining defendants.^[2]

Plaintiffs' complaint included allegations of independent negligence^[3] on the part of the hospital. The complaint also charged the hospital with vicarious liability for negligence on the part of Palmer and Gilmore who, although not employees of the hospital, were alleged to be ostensible agents of the hospital under Grewe v Mt Clemens General Hosp, 404 Mich. 240; 273 NW2d 429 (1978). During the course of the trial the hospital moved to strike all allegations of negligence on the part of Palmer and Gilmore upon the theory that the settlements operated to release the hospital from vicarious liability for the torts of Palmer and Gilmore. Although the trial judge recognized the common-law rule that release of an agent discharges the principal, he nevertheless denied the motions to strike and admitted evidence of negligence on the part of Palmer and Gilmore on the ground that it was necessary to the jury's understanding of independent negligence claims against the remaining defendants, Lansing General, Dr. Sciamanna, and Dr. Gilroy.^[4]

After a two-week trial, the jury returned a verdict against the hospital of $1 million, and found no cause of action against Drs. Gilroy and Sciamanna. The trial court reduced the amount of the verdict to $742,261, allowing for a setoff of amounts received from the settlements with Palmer and Gilmore.

Initially, the judgment below was affirmed by the Court of Appeals; however, on rehearing it was reversed. While holding that the release of Palmer and Gilmore discharged the hospital from liability on a theory of respondeat superior, the Court recognized that such a release provided no protection from liability for independent acts of negligence. However, after noting that the trial court had granted the hospital a directed verdict on all but one of plaintiffs' claims of independent negligence against the hospital, the panel's majority concluded on rehearing that as to the single remaining claim^[5] there was "no competent evidence or expert testimony" to support it. Finding the evidence insufficient to submit a question of the hospital's independent negligence to the jury, the Court of Appeals set aside the verdict and remanded the case for a new trial.^[6]

We granted leave to appeal, limited to two issues: (1) whether the settlements reached with Palmer and Gilmore released the hospital from vicarious liability for their actions under MCL 600.2925d; MSA 27A.2925(4); and (2) whether the releases, if not so protected, should be reformed as covenants not to sue. 426 Mich. 864 (1986).

II

Since it was determined below that the hospital was not guilty of any independent act of negligence ÔÇö an issue that is not before us on appeal ÔÇö recovery against the hospital is precluded except on a theory of vicarious liability.

Although plaintiffs alleged that Palmer and Gilmore were agents of the hospital, they were actually employees of Capital Anesthesiologists, P.C. However, since Capital had an exclusive contract to perform anesthesia within the hospital, and the choice of an anesthetist or anesthesiologist was made without consultation with the decedent's parents, it is contended that Palmer and Gilmore were ostensible agents, or agents by estoppel, under Grewe, supra. We proceed with our analysis on that basis.

At common law a valid release of an agent for tortious conduct operates to bar recovery against the principal on a theory of vicarious liability, even though the release specifically reserves claims against the principal. 53 Am Jur 2d, Master and Servant, ž 408, pp 416-418; 126 A.L.R. 1199; 76 CJS, Release, ž 50, p 689. See Bacon v United States, 321 F2d 880 (CA 8, 1963); Max v Spaeth, 349 S.W.2d 1 (Mo, 1961). Michigan courts have adhered to this common-law rule. Geib v Slater, 320 Mich. 316; 31 NW2d 65 (1948), overruled on other grounds Moore v Palmer, 350 Mich. 363; 86 NW2d 585 (1957); Lincoln v Gupta, 142 Mich App 615; 370 NW2d 312 (1985); Willis v Total Health Care, 125 Mich. App. 612; 337 NW2d 20 (1983); Drinkard v William J Pulte, Inc, 48 Mich. App. 67; 210 NW2d 137 (1973).

Plaintiffs argue that this common-law rule is inapplicable in the instant case by reason of the Michigan contribution act, MCL 600.2925; MSA 27A.2925. In particular they point to ž 2925d which, as amended in 1974, reads in pertinent part:

When a release or a covenant not to sue or not to enforce judgment is given in good faith to 1 of 2 or more persons liable in tort for the same injury or the same wrongful death:

(a) It does not discharge any of the other tortfeasors from liability for the injury or wrongful death unless its terms so provide. [Emphasis supplied.]

Plaintiffs argue that the word "tortfeasors" in the statute includes persons whose liability is based solely upon the theory of respondeat superior, as in the case of principal and agent. We disagree.

The Michigan contribution act does not include a definition of the terms "1 of 2 or more persons liable in tort," or "other tort-feasors," as used in the above-quoted ž 2925d. The Uniform Contribution Among Tortfeasors Act, 12 ULA 63, ž 4 (1955 rev), upon which the Michigan act is based, likewise fails to define the term "tortfeasor."^[7] Hence, the present question has arisen, and a split has developed among the jurisdictions^[8] as to whether a vicariously liable principal is a "tortfeasor" for purposes of ž 2925d.

At common law and prior to enactment in 1941 of the Michigan contribution statute,^[9] 1941 PA 303, contribution was not recoverable among or between tortfeasors. O'Dowd v General Motors Corp, 419 Mich. 597, 603; 358 NW2d 553 (1984). Also, at common law the release of one joint tortfeasor automatically released other joint tortfeasors, although the release of a concurrent tortfeasor did not release other concurrent tortfeasors. See Witucke v Presque Isle Bank, 68 Mich. App. 599; 243 NW2d 907 (1976); McBride v Scott, 132 Mich. 176; 93 N.W. 243 (1903). In MacDonald v Henry Hornblower & Weeks, 268 Mich. 626, 628-629; 256 N.W. 572 (1934), this Court explained:

The law of joint tortfeasors has been the subject of many refinements of reasoning and much conflict of opinion.... The underlying principle is that the tort is single and the cause of action indivisible and exists against all feasors or none....

[W]hen the cause of action is destroyed as to one tortfeasor, it falls as to the others, even though it is attempted to preserve the liability of the others. So, where one tortfeasor is released from liability on payment of part of the damage, the others are discharged although the contract expressly reserves right of action against them. [Citations omitted.]

However, common-law rules which once governed contribution rights among joint and concurrent tortfeasors should not be confused with the deeply rooted common-law doctrine that release of an agent discharges the principal from vicarious liability. The rationale for the latter rule is entirely different and is grounded on the very nature of the principal's derivative liability.

Vicarious liability is indirect responsibility imposed by operation of law. In modern jurisprudence, vicarious liability is justified as "a rule of policy, a deliberate allocation of a risk." Prosser & Keeton, Torts (5th ed), ž 69, p 500. As explained in Dessauer v Memorial General Hosp, 96 NM 92, 108; 628 P2d 337 (1981),

"Vicarious liability is based on a relationship between the parties, irrespective of participation, either by act or omission, of the one vicariously liable, under which it has been determined as a matter of policy that one person should be liable for the act of the other. Its true basis is largely one of public or social policy under which it has been determined that, irrespective of fault, a party should be held to respond for the acts of another." [Citing Nadeau v Melin, 260 Minn 369, 375-376; 110 NW2d 29 (1961).]

More bluntly stated, "``[i]n hard fact, the reason for the employers' liability is the damages are taken from a deep pocket.'" Prosser & Keeton, supra, p 500, quoting Baty, Vicarious Liability (1916), p 154. The principal, having committed no tortious act, is not a "tortfeasor" as that term is commonly defined. According to Black's Law Dictionary (5th ed), a tortfeasor is "a wrong-doer; one who commits or is guilty of a tort."

Although enactment in 1941 of Michigan's contribution statute narrowed the scope of concomitant doctrines applicable to contribution and release, this Court made clear in Geib, supra, that the statute did not abrogate the common-law rule that release of an agent discharges the principal.

The defendant in Geib was guilty of no wrongdoing. However, he was vicariously liable as the owner of an automobile which was negligently operated by another.^[10] Prior to trial, plaintiff settled with and released the negligent driver. However, by its terms the release did not discharge the owner-defendant.

The defendant in Geib moved for summary judgment on the ground that release of the negligent driver operated to release the owner. Opposing the motion, plaintiff pointed to ž 2 of the 1941 Michigan contribution statute which in essence provided that a claimant's release of one joint tortfeasor would not impair the claimant's right of action against other joint tortfeasors who were not released.^[11]

Holding the statute inapplicable because the defendant was not a joint tortfeasor, the Geib Court, in a unanimous opinion,^[12] reasoned that the "[d]efendant is guilty of no tortious act; he did not participate in the commission of the tort; and his liability arises only by operation of law." The Court compared the vicarious liability of the automobile owner "with that of a surety for the honesty of an employee, whose obligation differs from that of his surety." 320 Mich. 321.

A brief review of the development of Michigan's contribution statute will demonstrate that the Legislature focused on other concerns and evinced no intention to abrogate the common-law rule that release of an agent discharges the principal from vicarious liability. After it was enacted, the 1941 statute was criticized for several reasons. First, the statute allowed contribution only if a "joint judgment" was obtained, thereby allowing the plaintiff to favor one joint tortfeasor over others. Second, it made contribution available to the insurer of one who was jointly or severally liable, thereby differentiating unfairly between insured and uninsured persons. Third, contribution under the statute did not apply to concurrent tortfeasors but only to "joint tortfeasors." See discussion in Moyses v Spartan Asphalt Paving Co, 383 Mich. 314, 327-328; 174 NW2d 797 (1970); Morgan v McDermott, 382 Mich. 333, 371-372; 169 NW2d 897 (1969).

In 1961 the Legislature enacted a virtual redraft of the 1941 act as part of the Revised Judicature Act. 1961 PA 236. The few changes made dealt principally with the "joint judgment" criticism, Morgan, supra, pp 374-375, and the accommodation of court rule revisions applicable to third-party practice. Moyses, supra, p 329. However, as Justice BLACK, writing for the Moyses Court, saw it,

Section 2925 [of 1961 PA 236] is a near-verbatim redraft of PA 1941, No 303 (CL 1948, ž 691.561 et seq.). It turned out in the interval that the act of 1941 was considerably short of its supposedly purposeful aim; that of providing the substantive right of contribution between or among jointly liable or severally liable but not wilful tortfeasors. Section 2925 is correspondingly deficient, it having provided the desired right only to the extent of contribution between or among litigant "joint tortfeasors," and that only upon the nod of the plaintiff or of the possibly whimsical circuit judge. [383 Mich. 327.]

In Moyses this Court considered whether the defendant in a negligence action was entitled to contribution from another tortfeasor whose liability to the plaintiff was several, but not joint. Frustrated by the fact that the 1961 statute limited contribution rights to "joint tortfeasors," the Moyses Court announced:

[W]e have decided to overrule what is left of Michigan's common-law bar of contribution between or among "wrongdoers," wilful or intentional wrongdoers excepted. [383 Mich. 334.]

In addition to its use of the word "wrongdoers," the Court provided other guidance concerning the scope of its overruling action:

When one speaks specifically of joint tortfeasors, he does not refer (a) to wrongdoers the liabilities of whom arise out of variant legal positions, the concurrently applied but legally different derelictions of whom make them severally responsible to the plaintiff in damages, or (b) to the acts or omissions of several who act independently rather than in concert, or (c) to those who may ÔÇö under present rules of court ÔÇö be joined as defendants, by the plaintiff (see GCR 1963, 206), and held responsible to him for damages sustained on account of their causally cooperating but non-joint acts or omissions, say by the negligence of one, the violation of a statute like the dramshop act by another, and the breach by still another of an express or legally implied warranty. [Id., pp 331-332. Emphasis in original.]

Thereafter, within a few years of the Moyses decision, the Legislature enacted the present version of the contribution statute. 1974 PA 318. As this Court explained in O'Dowd, supra, 604,

The former limitation on contribution among tortfeasors to the class of tortfeasors where a judgment was obtained against two or more persons jointly was thought to preclude contribution where (i) the acts of the tortfeasors were separate, independent, or concurrent rather than joint or in concert, or (ii) where the tortfeasors were liable in tort on different legal theories. The revised act by explicitly providing for contribution among tortfeasors "severally" liable in tort extended contribution to these situations.

There can be no doubt that the Legislature intended by its 1974 revision to expand the right of contribution. However, there is no basis in the legislative history for the contention that it also intended to abrogate the common-law doctrine that release of an agent discharges the principal. Rather, the Legislature's more evident purpose was to satisfy the criticisms of Moyses by adopting its holding, which extended contribution as a matter of common law to wrongdoers, whether "jointly" or "severally" liable. By its use throughout the 1974 statute of "tortfeasor," rather than "joint tortfeasor," the Legislature made evident its purpose to incorporate, not abrogate, the then-existing common law as fashioned in Moyses.

Despite the fact that there were three opportunities to do so, the Legislature has never expressed an intent to make the change in law contended for by plaintiffs in this case. Obviously, the Legislature could easily have accomplished such a purpose by defining the word "tortfeasor" as used in the act to include one who, though guilty of no wrong, is vicariously liable for the tort of another. Instead, the Legislature chose in its 1961 and 1974 versions to eliminate altogether a definition of "joint tortfeasor" which had been included in the 1941 act.

Of primary importance to this discussion is the fundamental principle that

statutes in derogation of the common law must be strictly construed, ... and will not be extended by implication to abrogate established rules of common law.... The statute ... must be construed sensibly and in harmony with the legislative purpose. [Rusinek v Schultz, Snyder & Steele Lumber Co, 411 Mich. 502, 508; 309 NW2d 163 (1981), reh den 412 Mich. 1101 (1981). See also Hack Investment Co v Concrete Wall Co, 356 Mich. 416; 97 NW2d 106 (1959).]

In interpreting a statute containing language virtually identical to ž 2925d(a), the Illinois Court of Appeals in Bristow v Griffitts Const Co, 140 Ill App 3d 191; 488 NE2d 332 (1986), reached a conclusion similar to ours. Noting that "[t]he common law is not to be deemed abrogated by statute unless it clearly appears that was the legislative intent," id. at 193, the court said:

The resolution of this case depends upon the meaning of the word "tortfeasors" as used in section 2(c). The plaintiffs maintain the word is synonymous with the phrase "one or more persons liable in tort arising out of the same injury." A tortfeasor has also been defined as a "wrong-doer; one who commits or is guilty of a tort." (Black's Law Dictionary 1335 [5th ed, 1979].) Under the doctrine of vicarious liability, an employer is held liable to a third party even when the employer is free from all fault. The doctrine is justified as a deliberate allocation of risk. The employer, having engaged in a business and seeking to profit from it, is better able to foresee harm caused by the negligence of his employees and to insure against it than the innocent third party. (Prosser & Keeton, Torts sec. 69, at 499-501 [5th ed, 1984].) Thus, the employer is held liable as a matter of policy, but he is not a wrongdoer. The liability of the master and servant for the acts of the servant is deemed that of one tortfeasor and is a consolidated or unified one. (Towns v Yellow Cab Co [1978], 73 Ill 2d 113, 124; 382 NE2d 1217, 1221.) The master, therefore, would not be "any of the other tortfeasors." [Id., pp 193-194.]

We are convinced that our interpretation of ž 2925d harmonizes with other provisions of the contribution act.^[13] For example, ž 2925b provides:

Except as otherwise provided by law, in determining the pro rata shares of tortfeasors in the entire liability as between themselves only and without affecting the rights of the injured party to a joint and several judgment:

(a) Their relative degrees of fault shall be considered.

(b) If equity requires, the collective liability of some as a group shall constitute a single share.

(c) Principles of equity applicable to contribution generally shall apply. [Emphasis supplied.]

The Commissioners' Comment to ž 2 of the Uniform Contribution Among Tortfeasors Act, 12 ULA 87, upon which this part of the Michigan statute is based, indicates that the vicarious liability of a principal/employer is subsumed by that of a tortfeasor agent/employee:

This section in positive terms resolves several difficult questions of policy.

* * *

Second, it invokes the rule of equity which requires class liability, including the common liability arising from vicarious relationships, to be treated as a single share. For instance the liability of a master and servant for the wrong of the servant should in fairness be treated as a single share. Other examples are those situations involving co-owners of property, members of an unincorporated association, those engaged in a joint enterprise and the like....

Taking a similar view with respect to its own contribution statute, another court in Horejsi v Anderson, 353 NW2d 316, 318 (ND, 1984) has stated:

The "percentage of negligence" attributable to the conduct of the servant constitutes the entire "single share" of liability attributable jointly to the master and servant. Thus, when the plaintiff releases the servant he gives up his right to recover, from either the master or servant, damages caused by the servant's negligence. Because this percentage of negligence represents the "single share" of liability covered by the common liability of the master and servant, the master is necessarily released from vicarious liability for the released servant's misconduct. [Emphasis supplied.]

See also Comment, 43 Boston U L R 417, 423 (1963).

Put in another context, if A, B, and C are sued because each is guilty of negligence which resulted in injury to a plaintiff, their pro-rata shares of the common liability are to be determined under the contribution statute without regard to whether A's principal, not a wrongdoer, is also joined as a fourth defendant. As between A, an agent, and his principal, there is only one tortfeasor, and they represent only one share of the common liability.

Section 2925d of the contribution act which is invoked by the plaintiff in this case makes clear that a release or a covenant not to sue given to A would not discharge the "other tortfeasors" (B and C) from liability. However, the statute says no more, leaving in place the deep-rooted common-law principle that the release of A would discharge his principal. Any other result would be illogical and unjust because release of the agent removes the only basis for imputing liability to the principal.

Accordingly, we conclude that the release by plaintiffs of Nurse Palmer and Dr. Gilmore had the legal effect of discharging the hospital from vicarious liability for their acts or omissions.

III

Finally, we consider the argument that the releases executed in favor of Nurse Palmer and Dr. Gilmore should be reformed as covenants not to sue.^[14] We recognize, as plaintiffs have pointed out, that at common law a covenant not to sue given to an agent did not discharge his principal. Boucher v Thomsen, 328 Mich. 312, 321-322; 43 NW2d 866 (1950). However, in this case we decline to reform the releases for several reasons.

Courts are required to proceed with utmost caution in exercising jurisdiction to reform written instruments. Ivinson v Hutton, 98 U.S. 79; 25 L. Ed. 66 (1878), see also 66 Am Jur 2d, Reformation of Instruments, ž 3, p 528. In this case the plaintiffs' request for reformation is made for the first time on appeal. See Napier v Jacobs, 429 Mich. 222, 233; 414 NW2d 862 (1987); Durant v State Bd of Ed, 424 Mich. 364, 396; 381 NW2d 662 (1985).

We note that Nurse Palmer and Dr. Gilmore, who were directly involved in the settlement process and the releases, are no longer parties in the present action. See Citizens' Mutual Automobile Ins Co v Downing, 339 Mich. 434, 439-440; 64 NW2d 660 (1954). 66 Am Jur 2d, Reformation of Instruments, ž 100, pp 631-632.

As this Court has stated, "the burden of proof is upon one seeking reformation of a written instrument." River Rouge Bank v Fisher, 372 Mich. 558, 562; 127 NW2d 426 (1964). Absent a mutual mistake or a unilateral mistake coupled with fraud, shown by clear and convincing evidence, this Court has declined to reform written instruments. Windham v Morris, 370 Mich. 188, 192; 121 NW2d 479 (1963).

Finally, a unilateral misunderstanding of the legal effect of an instrument is not a sufficient ground for reformation. Wiersma v Nordella, 260 Mich. 574; 245 N.W. 520 (1932); Malone v SCM Corp, 63 Mich. App. 11; 233 NW2d 872 (1975). In the absence of a showing that the releases were not what was intended by the parties, the documents will not be rewritten. See Denton v Utley, 350 Mich. 332; 86 NW2d 537 (1957).

IV

Accordingly, we would affirm the decision of the Court of Appeals except in one respect. Instead of remanding for a new trial, we would remand to the circuit court for entry of judgment in favor of the hospital.

RILEY, C.J., and BRICKLEY, J., concurred with GRIFFIN, J.

BOYLE, J.

I agree with the conclusion that the Court of Appeals was in error in remanding this case for a new trial. Assuming that the document was a release, I would also agree with Justice GRIFFIN'S conclusion that MCL 600.2925d; MSA 27A.2925(4) does not abrogate the common-law rule that the release of the agent of a vicariously liable principle operates to discharge the principal. I have written separately because despite close examination of the trial court record, I am unable to reach a firm conclusion that the trial court held that the document in question was a release. I am, therefore, not persuaded that this case presents an appropriate record for resolution of the substantive issue. In my judgment, this matter should be remanded to the trial court for a determination as to whether the document was intended to release the principal or merely was intended as a covenant not to sue certain agents.

The trial court denied the defendant hospital's motion to strike and stated that the statute did not apply. The trial court, at the same time, acknowledged the common-law rule that release of an agent bars suit against the principal. However, although it should have logically followed that the respondeat superior claim would be barred, the jury was instructed, without objection, that Lansing General could be found vicariously liable for the acts of Gilmore and Palmer. The matter was submitted to the jury on a general verdict, and the hospital did not move for a directed verdict or judgment notwithstanding the verdict on the vicarious liability theory.

These facts raise a question as to how the parties themselves viewed the trial court's characterization of the document. It is possible that the parties understood that the court had held that the document operated to release the hospital and that the defense felt that it was unnecessary to object to the vicarious liability instruction in view of the fact that the motion to strike had been made and that the defendant moved for directed verdict on the independent claims of negligence. It is possible that the failure to submit the special verdict questions was sound trial strategy, an oversight in a protracted and technical trial, or is explained by the trial judge's conclusion that if counsel had not requested the verdict form, the trial judge should not intervene. It also could be speculated, however, that these facts are explained by misgivings on the part of the trial court as to whether the document was a release.^[1]

If, indeed, the document was a release rather than a covenant not to sue, there might be a reason for admitting the evidence of the negligence of Palmer and Gilmore to facilitate the jury's understanding of the plaintiffs' theory of independent negligence. However, if this was the only reason for admission, it seems illogical that the court instructed on vicarious liability and presented the jury with alternative grounds for a verdict in favor of the plaintiffs.

Portions of the trial court's colloquy with counsel at the time of the ruling on the motion to strike underscore the difficulty of resolving this threshold issue. Thus, the trial court observed:

The case law seems to indicate that's the situation [the hospital is released], but it seems to me great pains were taken in this particular instance to attempt to avoid the obvious effect of the law, so it's difficult to fault the Plaintiffs when they took care to avoid the pitfalls, by their actions, in preparing this release and the petition in court here, so I think that my first suggestion is the one that I prefer to adopt, proceed to trial and then let the Court determine whether or not, from the facts, there is [sic], in fact, any independent acts of negligence on the part of the other Defendants that will put them in the category of a joint tort feasor, as opposed to a master-servant, and if there are, that we submit that to the jury on a separate and special verdict that they find specifically regarding these Defendants as they apply to the acts, or lack of acts, on the part of Jana Palmer. Okay.

The parties' arguments have assumed that this document was a release or that the trial court held it to be a release. Whether the document is considered a contract or a judgment, its construction is, of course, a matter of law, Bradway v Miller, 200 Mich. 648; 167 N.W. 15 (1918).

The upshot of all of this is that both the opinion for affirmance and the opinion for reversal in this case have focused on the effect of a release when it is by no means clear whether there is a release in this case.

Turning to the document itself, it is clear that the word "release" was used, but it is not clear in what sense it was used. For instance, the document indicates that the plaintiffs gave Gilmore and Palmer a "release indemnity agreement" holding these individual defendants harmless from any potential cause of action. A future action for indemnity could only be premised upon a theory of vicarious liability, since joint or several liability would give rise instead to a right to contribution. Also, as the trial court noted at the time of motion, the document seems to take away in the second sentence whatever it granted by way of "release" in the first sentence.

The discussions of the parties also undermine a firm conclusion that the document was intended to release the hospital. Thus, at the settlement hearing, plaintiffs' counsel questioned his client in regard to the document as follows:

Q. [Mr. Kritselis] And, further, you are relying upon the fact that we are not dismissing anybody else other than Dr. Gilmore only, Capital Anesthesiologists only, Dr. Mahoney only, and that all claims against Lansing General Hospital and Dr. Gilroy and Dr. Sciamanna under the various theories as set forth in the pleadings, that those claims will continue. Do you understand that?

A. [Mr. Schneider] Yes.

This statement lends credibility to Mr. Kritselis' own, unsworn statement at a postverdict motion that he intended the consent judgment to preserve his vicarious liability theories against the hospital. These statements are also consistent with the fact that the hospital did not contribute to the settlement amounts underlying the "releases."

[W]hether an instrument is a release or a covenant not to sue, the designation of the instrument as "release" or "covenant not to sue" is immaterial. To determine the nature of the document it is necessary to look to its terms, the words used, the amount paid, the substance of the agreement, and the intention of the parties. [73 ALR2d 403, 420. See also 4 Restatement Torts, 2d, ž 885, comment c.]

In my view, the correct resolution of this case requires a definite determination as to whether the document was, in fact, a release. To do this properly, the matter should be remanded to the trial court, since it is quite likely that an ambiguity will be found and that parol evidence will be required.

I would reverse the determination of the Court of Appeals to grant a new trial,^[2] vacate the judgment of the trial court, and remand for proceedings consistent with this opinion.

LEVIN, J.

At common law, where two or more tortfeasors contributed to a plaintiff's injury, a settlement reflected in a document containing language of release of the settling tortfeasor necessarily discharged all other tortfeasors.^[1] And ordinarily, a tortfeasor could not obtain contribution from another tortfeasor.^[2]

The common-law rule has been changed in most jurisdictions by statute or by subsequent common-law decision. It is now generally provided that a settlement with one tortfeasor, although reflected in a document containing language of release, does not discharge other tortfeasors. And a tortfeasor who pays more than his share of the liability may now generally enforce contribution by other tortfeasors.^[3]

The question presented is whether the common-law rule, that a settlement with a tortfeasor reflected in a document containing language of release necessarily discharges other tortfeasors, continues to govern where the person released is an employee or agent of another person who is vicariously liable under the common-law doctrine of respondeat superior for the torts of the employee or agent, with the result that a release of the employee or agent discharges the employer or principal although the release reserves claims against other persons who are subject to liability for the alleged injury or wrongful death.

In Geib v Slater, 320 Mich. 316; 31 NW2d 65 (1948), this Court considered a section of 1941 PA 303,^[4] modeled on a provision of the 1939 Uniform Contribution Among Tortfeasors Act,^[5] providing that an injured person may settle with a "joint tortfeasor" and yet maintain an action against another joint tortfeasor. The Court held that this remedial provision was not applicable in a case where the settlement was with a tortfeasor for whose torts the defendant tortfeasor was vicariously liable because their liabilities were several and not joint, and thus they were not "joint tortfeasors."

The lead opinion would hold that the defendant hospital ÔÇö found to be vicariously liable for the torts of a physician and nurse with whom plaintiff Theophelis entered into a settlement, reflected in a document containing language of release ÔÇö is not a tortfeasor within the meaning of 1974 PA 318,^[6] substituting language of the 1955 Uniform Contribution Among Tortfeasors Act for the 1941 act.

We would hold that a settlement with any person subject to tort liability, including a person vicariously liable for the torts of another, reflected in a document containing either language of release or language covenanting not to sue, does not ÔÇö unless otherwise agreed in the document ÔÇö discharge any other person subject to tort liability without regard to whether the plaintiff's theory of liability against the other person is that he is vicariously liable.

I

The operative effect of the common-law rule concerning the effect to be given a settlement with a tortfeasor had been ameliorated by judicial decision before the enactment of the 1941 legislation incorporating provisions of the 1939 Uniform Act and the 1974 legislation incorporating provisions of the 1955 Uniform Act.

A

In 1935, this Court in Cook v City Transport Corp, 272 Mich. 91, 92; 261 N.W. 257 (1935), observed that by the great weight of authority the common-law rule could be avoided by substituting a covenant not to sue for a release. The Court acknowledged that "[t]he distinction between a release and an agreement not to sue is finespun and seems overtechnical. However, the whole law of joint tortfeasors is of like character." "[T]he overwhelming weight of authority draws the distinction and denies the effect of an agreement not to sue one as a release of other joint tortfeasors."

This Court concluded in Cook that it would "follow the weight of authority," not because it approved "the distinction, except as a mental exercise," id., p 93, but for policy reasons:

[B]ecause such effect of an agreement not to sue offers a way for a party to buy his peace and allows an opportunity to compromise a doubtful claim without requiring an injured party to forego the right of full compensation against known wrongdoers.

No longer need a settlement with one tortfeasor necessarily discharge all other tortfeasors. The operative effect of the common-law rule could be avoided by substituting one piece of paper (a covenant not to sue) for another (a release).^[7]

B

One would think that what was left of the common-law rule would by now be history. The Restatement now provides that a release does not, as a matter of common law, discharge others liable for the same harm unless it is agreed that it will so discharge them.^[8] This Court, having by common-law decision in Cook changed the operative effect of the common-law rule, could, by common-law decision, as has the United States Supreme Court,^[9] complete the unfinished work by adopting the revision of the common-law rule as set forth in the Restatement of Torts, 2d.

This Court's construction of the 1941 act in Geib is seen by the lead opinion as standing in the way. There the Court held that a settlement with a negligent driver of an automobile reflected in a release necessarily discharged the owner of the automobile because they were not "joint tortfeasors" within the meaning of the 1941 legislation providing that a person having a claim against "2 or more joint tort-feasors" may settle with and discharge one or more of them without impairing the right of such person to maintain an action against the remaining joint tortfeasors.

Although this Court had frequently used the term "joint tortfeasor" to describe persons severally liable in tort to the injured plaintiff,^[10] the Court in Geib construed the 1941 legislation as in effect applicable only to tortfeasors acting in concert and not to those severally liable to the plaintiff.

C

In Moyses v Spartan Asphalt Paving Co, 383 Mich. 314, 334; 174 NW2d 797 (1970), this Court considered Geib's construction of the term "joint tortfeasor" in the context of an action for contribution. In pertinent part, the question was whether an action for contribution could be maintained by a severally liable tortfeasor against another severally liable tortfeasor under the section of the 1941 act^[11] modeled on the 1939 Uniform Act^[12] providing for contribution among "joint tortfeasors."

The Court in Moyses held, as a matter of common law, that the Geib construction would no longer bar an action for contribution. In Moyses, the owner of an automobile sought contribution from the manufacturer of an allegedly defective tire. The Court reviewed the history of the 1941 and subsequent legislative efforts to rewrite common-law doctrine (by providing a substantive right of contribution among joint tortfeasors and by permitting settlement and discharge of one or more of them without impairing rights against the others) and Geib's construction of those legislative efforts, and decided "to overrule what is left of Michigan's common-law bar of contribution between or among ``wrongdoers,' wilful or intentional wrongdoers excepted."

The Court thus by common-law decision superseded, as to actions for contribution, Geib's limiting construction of the term "joint tortfeasor." Henceforth, an action for contribution could be maintained by one severally liable tortfeasor against another.

It is contended, nevertheless, in the lead opinion that, except as provided in the 1974 legislation, common-law doctrine and the Geib analysis still control where the question is whether the release of one severally liable tortfeasor releases other severally liable tortfeasors.

II

The first Restatement of Torts, promulgated in 1932, stated that a release discharged all others liable for the same harm, unless ÔÇö as in the instant case ÔÇö "the parties to the release agree that the release shall not discharge the others."^[13] It was further stated that a "covenant not to sue one tortfeasor for a harm does not discharge any other liable for the harm."^[14]

By 1981, it was observed in the commentary to the Restatement Torts, 2d, that the "arbitrary distinction" between a release and a covenant not to sue had "frequently resulted in the unintended and unpaid-for discharge of one of the tortfeasors. This earlier rule is not consistent with the modern American point of view."^[15]

The rule stated in the Restatement Torts, 2d, is that "[a] valid release of one tortfeasor from liability for a harm, given by the injured person, does not discharge others liable for the same harm, unless it is agreed that it will discharge them."^[16] In the instant case, the release provided that it did not release "any other persons or entities who may be liable in tort for the alleged wrongful death."

The United States Supreme Court has held that a release that clearly intends ÔÇö as did the release in the instant case ÔÇö to save the releasor's rights against another person subject to tort liability does not automatically surrender these rights. The Court said that the "ancient common-law rule" was "grounded upon a formalistic doctrine that a release extinguished the cause of action."^[17] The Court declared:

To adopt the ancient common-law rule would frustrate such partial settlements, and thereby promote litigation, while adoption of the First Restatement rule would create a trap for unwary plaintiffs' attorneys. The straightforward rule is that a party releases only those other parties whom he intends to release. Our conclusion that this is the appropriate rule for giving effect to releases under the antitrust laws is further buttressed by the Restatement's abandonment in a tentative draft of the rule requiring express reservation of rights in order to save them, and its adoption of the rule to which we adhere.[^[18]]

I see no reason to perpetuate "the ancient common-law rule,"^[19] where the piece of paper reflecting the settlement is in the form of a release rather than in the form of a covenant not to sue, as "a trap for unwary plaintiffs' attorneys"^[20] and their unfortunate clients.

III

The 1974 act, substituting the 1955 Uniform Act for the 1941 act, enacted RJA, ž 2925d concerning the effect of a release of one tortfeasor on the liability of another:

When a release or a covenant not to sue or not to enforce judgment is given in good faith to 1 of 2 or more persons liable in tort for the same injury or the same wrongful death:

(a) It does not discharge any of the other tortfeasors from liability for the injury or wrongful death unless its terms so provide.[^[21]] [Emphasis supplied.]

The lead opinion would read RJA, ž 2925d as superseding Geib generally ÔÇö for severally as well as jointly liable tortfeasors ÔÇö but apply a different rule where the non-settling defendant is not himself a "wrongdoer" and is subject to vicarious liability for the wrongdoing of another.

A

The basis of the proposed distinction is not policy. The Court considered in Boucher v Thomsen, 328 Mich. 312, 316; 43 NW2d 866 (1950), and declined to accept, policy arguments advanced in the instant case for excepting vicariously liable defendants from the rule that another tortfeasor is not relieved of liability if the injured person settles, by covenant not to sue, with another tortfeasor. In policy terms, there is no difference between a covenant not to sue and a release.

In Boucher, as in Geib, a person injured in an automobile accident had settled with the driver of the automobile and his employer with whom the automobile had been left for servicing and sought to maintain an action against the owner of the automobile.^[22] The only factual difference between Geib and Boucher was that the settlement was reflected in a release in Geib and a covenant not to sue in Boucher.

This Court observed in Boucher that the owner of the automobile was not himself guilty of negligence and that his liability arose "solely by virtue of the statute."^[23] Rejecting policy arguments similar to those advanced in the instant case that the owner would, if recovery should be obtained against him, have "the right to insist on reimbursement [indemnity] from the other defendants, and that in effect his status is that of a surety,"^[24] and that "the entire burden of defending the action should not be cast on him,"^[25] this Court declined to carve out an exception for the vicariously liable owner of the automobile, relieving him of his statutory liability on the basis that the injured person had settled with the driver of the automobile and his employer.

The policy arguments that have been advanced in support of the proposed distinction, excepting a vicariously liable defendant from the operative effect of RJA, ž 2925d, were thus urged upon and rejected by this Court in Boucher just two and a half years after Geib was decided.

B

The basis of the proposed distinction is the rule of construction that statutes in derogation of the common law must be strictly construed and not extended by implication.

The proposed distinction finds no support in the common law as stated in the Restatement Torts, 2d.^[26]

In both Geib and Boucher the vicarious liability arose by virtue of the statute subjecting the owner of an automobile to liability for the negligent driving of a person who drives the automobile with the owner's permission.^[27]

In the instant case, the vicarious liability arises if at all pursuant to the common-law doctrine of respondeat superior subjecting an employer/principal to liability for the acts of his employee/agent.

Geib equated the vicarious liability that arises by statute with the vicarious liability that arises by virtue of the common-law doctrine of respondeat superior. The Court said of the defendant owner of the automobile:

[H]is statutory liability is based upon the doctrine of respondeat superior. It may be compared with that of a surety for the honesty of an employee, whose obligation differs from that of his surety.[^[28]]

The lead opinion nevertheless draws a distinction between vicarious liability arising by statute, as in Geib and Boucher, and vicarious liability arising at common law. The lead opinion speaks of "the deeply rooted common-law doctrine that release of an agent discharges the principal from vicarious liability."^[29] The opinion states that "this Court made clear in Geib, supra, that the [1941] statute did not abrogate the common-law rule that release of an agent discharges the principal."^[30] The opinion further states that there is no basis in the legislative history of the 1974 act incorporating the 1955 Uniform Act for the contention that the Legislature "intended to abrogate the common-law doctrine that release of an agent discharges the principal."^[31] The opinion states that "[o]f primary importance to this discussion is the fundamental principle that ``statutes in derogation of the common law must be strictly construed, ... and will not be extended by implication to abrogate established rules of common law....'"^[32]

There is to be sure such a rule of construction, but it is just that, a rule of construction. The ultimate question is what did the Legislature intend.

Neither the language of RJA, ž 2925d nor the analysis in Geib^[33] permits drawing a distinction between tortfeasors vicariously liable as a result of common-law doctrine and those vicariously liable by virtue of statute. RJA, ž 2925d should not be read as superseding Geib where the injured person settles with the actual wrongdoer in the factual situation presented in Geib ÔÇö an owner of an automobile statutorily liable for the driving of another who was held to have been discharged from liability by a release of the driver ÔÇö but as inapplicable in the situation, adverted to in Geib as analogous and relied on in the lead opinion as a statement by this Court of the common-law rule, of a master liable under the doctrine of respondeat superior for the torts of his servant.

There is no reason to suppose that the Legislature sought to provide one rule where the vicarious liability arises by statute and another when it arises at common law: If one settles by a document containing language of release with a person for whose tort another tortfeasor is vicariously liable by statute, the other tortfeasor is discharged from liability, but if the vicarious liability arises at common law the other tortfeasor is discharged. Lawyers and judges ÔÇö not legislators ÔÇö conjure and draw such distinctions.

Construing RJA, ž 2925d as applicable in both cases does not "extend by implication" (emphasis supplied) RJA, ž 2925d. "Strictly construed," (emphasis supplied) RJA, ž 2925d applies in both cases. RJA, ž 2925d provides that either "a release or a covenant not to sue" two or more "persons liable in tort" "does not discharge any of the other tort-feasors from liability for the injury or wrongful death unless its terms so provide." (Emphasis supplied.) Read as a whole, "tortfeasor," as used in RJA, ž 2925d, is shorthand for and means "persons liable in tort." Construing "any of the other tortfeasors" (emphasis supplied) as meaning all "persons liable in tort" without distinction is the strict construction. So construing "tort-feasor" does not extend that term by implication. Carving out a narrow exception for a situation adverted to by analogy in Geib is to narrow, not extend, the "strict" construction "by implication."

C

The lead opinion argues that "[t]he principal, having committed no tortious act, is not a ``tortfeasor' as that term is commonly defined. According to Black's Law Dictionary (5th ed), a tortfeasor is ``a wrong-doer; one who commits or is guilty of a tort.'"^[34]

The Legislature can be its own lexicographer.^[35] The term "tortfeasor" is used throughout the 1974 act, incorporating the 1955 Uniform Act. If it does not include within the meaning of "tortfeasor" a person subject vicariously to tort liability for the acts of another then (i) a person vicariously liable may not enforce contribution under the provisions of RJA, ž 2925a (enacted by the 1974 act)^[36] (or possibly even Moyses, as the lead opinion would restrict it^[37]) against a severally liable "wrongdoer,"^[38] and (ii) RJA, ž 2925b (also enacted by the 1974 act),^[39] providing for the determination of the pro-rata shares of contribution to be made by "tortfeasors," does not apply to a person vicariously liable where a judgment is about to enter against two or more "wrongdoers" and a person vicariously liable for the wrong of one of them.

The commentary to the 1955 Uniform Act accompanying the language incorporated as RJA, ž 2925b states, however, that "the liability of a master and servant for the wrong of the servant should in fairness be treated as a single share."^[40] RJA, ž 2925b provided:

In determining the pro rata shares of tort-feasors in the entire liability:

(a) Their relative degrees of fault shall not be considered.

(b) If equity requires, the collective liability of some as a group shall constitute a single share.

(c) Principles of equity applicable to contribution generally shall apply. [1974 PA 318. Emphasis supplied.]

The word "some" in the phrase "some as a group" clearly refers to "tortfeasors," and means "some tortfeasors as a group." As set forth in the commentary, such a "group" includes a master and servant. Accordingly, "the collective liability of some [tortfeasors, e.g., master and servant,] as a group shall constitute a single share." It thus appears that the drafters of the Uniform Act intended that those vicariously liable would be subject to the language enacted as RJA, ž 2925b and hence that they be regarded as tortfeasors within the meaning of that language.

The intention of the drafters of the 1955 Uniform Act that a master be regarded as a "tortfeasor" is further manifested by the language of the 1955 Uniform Act adopted by the 1974 Legislature as RJA, ž 2925a(7) and the accompanying commentary. RJA, ž 2925a(7) provides:

This section does not impair any right of indemnity under existing law. Where 1 tort-feasor is entitled to indemnity from another, the right of the indemnity obligee is for indemnity and not contribution, and the indemnity obligor is not entitled to contribution from the obligee for any portion of his indemnity obligation.[^[41]]

The commentary accompanying the foregoing provision states:

Where a master is vicariously liable for the tort of his servant, the servant has no possible claim to contribution from the master; and the master does not need contribution from the servant and will not seek it, since he is entitled to full indemnity. The master, of course, may recover contribution from any third tortfeasor against whom he has no right of indemnity. [Emphasis supplied.][^[42]]

Reading the commentary together with RJA, ž 2925a(7), it is clear that the drafters regarded a master, subject to vicarious liability under the common-law doctrine of respondeat superior for the tort of his servant, as a "tortfeasor" within the meaning of the phrase "[w]here 1 tort-feasor is entitled to indemnity from another." The master is the tortfeasor "entitled to indemnity," and the servant is the "another" tortfeasor obliged to indemnify the master.

The reference in the commentary to the master's right to "recover contribution from any third tortfeasor" indicates that the drafters of the 1955 Uniform Act regarded a master as a tortfeasor within the meaning of RJA, ž 2925a(2) providing that "[t]he right of contribution exists only in favor of a tort-feasor who has paid more than his pro-rata share of the common liability...."^[43] (Emphasis supplied.)

In adopting the language of the 1955 Uniform Act, the 1974 Legislature evidenced an intention that an employer (master) as well as an employee (servant) was a "tortfeasor" for the purposes of RJA, ž 2925a, concerning the enforcement of contribution against a "wrongdoer,"^[44] RJA, ž 2925b, concerning the determination of pro-rata shares of contribution, and RJA, ž 2925d, concerning the effect of a release of one person subject to tort liability on the liability of other persons subject thereto. To repeat, "tortfeasor" is shorthand for and means "persons liable in tort."

IV

In the penultimate paragraph of its analysis the lead opinion departs somewhat from the derogation of the common-law analysis and may be read as, in effect, carving out an exception to RJA, ž 2925d, concerning the operative effect of a "release or a covenant not to sue," for a release, and possibly even a covenant not to sue, where the settlement is with a tortfeasor for whom another is vicariously liable, without regard to whether the basis of the vicarious liability is a statute or the common law.

A

The lead opinion argues on the basis of RJA, ž 2925b, concerning pro-rata shares, that "the vicarious liability of a principal/employer is subsumed by that of a tortfeasor agent/employee,"^[45] and that "[a]s between A, an agent, and his principal, there is only one tortfeasor, and they represent only one share of the common liability."^[46] (Emphasis added.) The lead opinion continues that RJA, ž 2925d, concerning the effect of a release on the liability of other tortfeasors, "makes clear that a release or a covenant not to sue given to A would not discharge the ``other tortfeasors' (B and C) from liability." The lead opinion continues that the "statute says no more, leaving in place the deep-rooted common-law principle that the release of A would discharge his principal. Any other result would be illogical and unjust because release of the agent removes the only basis for imputing liability to the principal."^[47]

The arguments that there is "only one share of the common liability" and that "[a]ny other result would be illogical and unjust because release of the agent removes the only basis for imputing liability to the principal" apply with equal force whether the basis of the vicarious liability is a statute or the common law, or whether the document reflecting the settlement is a release or a covenant not to sue.

Those arguments were, I repeat, considered and rejected by this Court in Boucher in the context of a covenant not to sue sought to be interposed as a defense by a vicariously liable defendant. RJA, ž 2925d applies by its terms to "a release or a covenant not to sue." The lead opinion, in declaring on policy grounds that RJA, ž 2925d does not overrule the common-law principle that a release of an agent discharges the principal,^[48] casts a shadow on the common-law principle set forth in Boucher, decided two and one-half years after Geib, that a covenant not to sue does not discharge even a vicariously liable defendant.

B

The arguments that there is "only one share of the common liability" and that it is "illogical and unjust" to hold liable the vicariously liable person after the release of the actual wrongdoer because the release removes the "only basis for imputing liability" to the person vicariously liable, are refinements, in the vicarious liability context, of the arguments that were made in rationalizing the development of the common-law doctrine that the release of one tortfeasor necessarily released all other tortfeasors.

The theories of the common-law doctrine as enunciated in the case of persons acting in concert ÔÇö true joint tortfeasors ÔÇö were that a release of one "necessarily released the other, since there was in the eyes of the law but one cause of action against the two, liable for the same acts, which was surrendered."^[49] While there was no reason to apply the same analysis to independent wrongdoers, it was extended generally to all releases of all tortfeasors for, among other reasons, "the cause of action is by its nature indivisible in the eyes of the law,"^[50] and "the essential unity of the injury, and the fact that the injured party is entitled to but one compensation therefor, make it impossible for the injured person to settle with one tortfeasor without discharging the other."^[51] The common-law rule relied on in the lead opinion was based on the same theories, that "the damages recoverable for the tort are entire and not severable" or that "the injured person is entitled to receive but one compensation for his injury."^[52]

These theories and the result, the unintentional release of a person otherwise subject to tort liability, were "justly condemned."^[53] Today, there can be no double recovery because any amount recovered in settlement from one tortfeasor must be deducted from a judgment rendered against another.^[54] Piecemeal collection from different defendants is clearly permitted after judgment.^[55] The "one cause of action," "indivisibility" of the cause of action, and "common liability" arguments can be avoided by employing a covenant not to sue to reflect the settlement.

C

While the single, indivisible cause of action analysis set forth in the lead opinion has, it might be argued, more substance in the context of vicarious liability than it had in the more general context in which the common-law doctrine concerning the effect of a release developed, such analysis and argument was rejected by the drafters of the 1955 Uniform Act on policy grounds: injured persons and those subject to tort liability should be encouraged to enter into partial settlements and a person subject to tort liability should not, by declining to join in a settlement of all claims, be able to preclude partial settlements with tortfeasors willing to buy their peace.

By incorporating in 1974 the 1955 Uniform Act, the Legislature enunciated as the public policy of this state the policy grounds underlying the 1955 Uniform Act. This Court should not resurrect an outdated and repudiated analysis in support of a result inconsistent with that public policy.

Unless the analysis of the lead opinion is extended to a covenant not to sue, it is but a trap for unwary lawyers and their unfortunate clients. To extend that analysis to covenants not to sue would be to overrule Boucher.^[56]

D

The analysis of the lead opinion in effect attributes to the 1974 Legislature an intent to enact "name and retain" legislation for the benefit of hospitals and other defendants vicariously liable at common law (and possibly as well for the benefit of defendants vicariously liable by statute) although such legislation was enacted for the first time years later in the context of dramshop liability^[57] and has not otherwise been enacted.

If it is thought that, on policy grounds, an injured person should not be able to enter into a partial settlement with a wrongdoer for whose acts another is vicariously liable at common law, then Boucher should, as a matter of common-law decision, be overruled. The injured person should, by further common-law decision, then no longer have the option of maintaining an action only against the person vicariously liable; the injured person should be required to maintain an action against the actual wrongdoer as well as the vicariously liable person.

The injured person would then be barred from partially settling with the actual wrongdoer by covenant not to sue or by release, and the actual wrongdoer would be required to be named and retained throughout the litigation. The vicariously liable person would then be empowered to stymie a settlement that the actual wrongdoer was willing to enter into. Physicians and nurses, like those in the instant case, would as a practical matter be precluded from attempting to buy their peace respecting independent as well as vicarious claims of negligence without the consent of the hospital.

To be sure, a hospital might, by commencing or indicating that it will commence an action for indemnification as provided for in RJA, ž 2925a(7), be able to frustrate the efforts of a physician or nurse to settle without the consent of the hospital. It is, however, somewhat unusual for a hospital to seek indemnification from a physician or nurse.

V

Two cases from other jurisdictions have construed the 1955 Uniform Act to reach the same result as does the lead opinion but did so on a different analysis.^[58]

The Supreme Court of Tennessee in Craven v Lawson, 534 S.W.2d 653, 656 (Tenn, 1976), concluded that the act does not apply in the master/principal, servant/agent relationships. The court rested that conclusion on the language of the 1955 Uniform Act, adopted in Michigan as RJA, ž 2925a(7), providing in effect that the master/principal is entitled to indemnification from the servant/agent. The court said:

Where no right of contribution exists, the act does not purport to intrude. Thus, the section providing that a covenant not to sue discharges the covenantee from contribution and does not discharge any other tort feasor, has no application to the master-servant, principal-agent relationship, where liability is solely derivative.

An Illinois intermediate appellate court adopted the same view in Bristow v Griffitts Const Co, 140 Ill App 3d 191, 198; 488 NE2d 332 (1986).^[59] The court declared that the language of the 1955 Uniform Act, enacted in Michigan as RJA, ž 2925d,

was designed to encourage settlements. Because we find an action for indemnity remains viable in cases involving vicarious liability, the employee in this case would gain nothing in return for his $20,000 and relinquishing his right to defend unless the covenant not to sue also extinguished the employer's vicarious liability. We, therefore, find a party whose liability is solely derivative is not "any of the other tortfeasors" within the meaning of [the language enacted in Michigan as RJA, ž 2925d].

The fallacy in the reasoning of the Tennessee and Illinois courts is that the drafters of the 1955 Uniform Act carefully excepted from the operative effect of that act the master/principal and servant/agent relationships only insofar as, but for such exception, it might be thought that an action for contribution might be maintained by one against the other.^[60]

The question whether a release or a covenant not to sue a person for whose tort another is vicariously liable discharged the person so vicariously liable had arisen and been decided by a number of courts,^[61] including this Court in Geib, in the construction of the 1939 act or language modeled thereon before the 1955 act was drafted. There is no basis for supposing that the drafters of the 1955 act, who so carefully excepted^[62] the master/principal, servant/agent relationships from the operative effect of the provisions enacted in Michigan as RJA, ž 2925a concerning contribution, overlooked making, although they intended, a like exception regarding the master/principal, servant/agent relationships in drafting the provisions enacted in Michigan as RJA, ž 2925d concerning the effect of a release or covenant not to sue on the tort liability of another.

The argument advanced by the Illinois intermediate appellate court that the employee would "gain nothing in return for his $20,000" unless the employer's vicarious liability is also extinguished ignores that the employee may not have expected that his employer's claim for indemnity would be extinguished. In all events, there is no claim here that the nurse and physician were misled or expected that their settlements with Theophelis extinguished the hospital's claim for indemnity. It is again relevant that it is unusual for a hospital to seek indemnity from a nurse or physician; the nurse and physician in the instant case may have gained something by their settlements with Theophelis.

VI

The majority of jurisdictions that have considered this issue have determined that the release of a servant/agent does not release the vicariously liable master/principal.^[63]

In Alaska Airlines, Inc v Sweat, 568 P2d 916 (Alas, 1977), the Supreme Court of Alaska construed the 1955 Uniform Act and concluded that a settlement with an agent did not release a vicariously liable principal. Plaintiffs were injured in the crash of an airplane operated by Chitina Air Service, an independent contractor of Alaska Airlines. They entered into a covenant not to sue^[64] with Chitina and a stipulation dismissing the claim with prejudice. The injured person brought an action against Alaska Airlines, which claimed that it was not subject to vicarious liability for Chitina's act because the "release of Chitina automatically released Alaska."^[65]

The Alaska court considered the Alaska adoption of the 1955 Uniform Act, identical to RJA, ž 2925d, that the act refers to "two or more persons liable in tort" and "other tortfeasors" without defining "tortfeasors," and concluded that the act was "intended to include those vicariously liable. It may be that Alaska Airlines is not technically a ``tort-feasor,' but it is ``one of two or more liable in tort for the same injury.'"^[66]Alaska Airlines, supra, p 930.

The Nevada Supreme Court, in Van Cleave v Gamboni Const Co, 101 Nev 524, 530; 706 P2d 845 (1985), held that Nevada, in adopting the 1955 Uniform Act, abrogated the common-law rule that release of an employee automatically discharges the vicariously liable employer. Approving Alaska Airlines and criticizing the opposing view of Craven, the court said:

We recognize that the expressed public policy established by the Uniform Act is "to encourage rather than discourage settlements." 12 ULA 65. One commentator has recognized that the "release provisions of the 1955 Revised Act were intended to promote settlements, so a construction of the statute should be directed to achieving that objective." Comment, Torts ÔÇö Vicarious Liability ÔÇö Covenant Not to Sue Servant or Agent as Affecting Liability of Master of [sic] Principal, 44 Tenn L Rev [188] 197 (1976).

... If we determined that the Uniform Act does not apply to [plaintiffs] claim against the employer, we would be discouraging prompt resolutions of actions, not encouraging such settlements, in contravention of the expressed public policy of the Uniform Act. "An injured party probably would be reluctant to settle with the servant or agent, and thereby extinguish his cause of action against the master or principal, unless he could settle with the servant or agent for an amount sufficient to compensate him for his entire loss." Id. at 198.

A United States district court in Harris v Aluminum Co of America, 550 F Supp 1024, 1030 (WD Va, 1982) (construing Virginia law), also rejected Craven and adopted the view of Alaska Airlines that "the act's plain language appears to apply to a party who is vicariously liable, as the act's coverage extends to ``one of two or more persons liable in tort for the same injury...." The Virginia Supreme Court confirmed Harris' view of Virginia law in Thurston Metals & Supply Co, Inc v Taylor, 230 Va 475, 483-485; 339 SE2d 538 (1986). The court ruled that the plaintiff's settlement with the defendant's insured did not discharge the vicariously liable insurer-principal. The court found that the insurer corporation was a "tortfeasor" within the meaning of the Uniform Act and followed Harris and Alaska Airlines in so holding.^[67]

The Supreme Court of Utah held in Krukiewicz v Draper, 725 P2d 1349, 1351-1352 (Utah, 1986), that a release given one "joint tortfeasor" does not release another joint tortfeasor who is vicariously liable for his servant's negligence. Utah had enacted a statute patterned after the 1939 Uniform Act which included the Uniform Act's definition of "joint tortfeasor" as "one of two or more persons, jointly or severally liable in tort for the same injury to person or property...." The court relied on several cases^[68] that construed the 1939 Uniform Act's definition of joint tortfeasor to include the master-servant relationship, and continued:

In 1955 the Uniform Act was revised. The 1955 version of the Uniform Act does not contain a definition of "joint tort-feasor." However, the 1955 Act does state that the release provisions apply "where two or more persons become jointly or severally liable in tort for the same injury to person or property...." All jurisdictions but one which have construed this language, which is essentially identical to the definition of joint tortfeasor contained in the 1939 Act, have also held that the master-servant relationship is subject to the 1955 Act's language making it applicable to those jointly and severally liable in tort. See, e.g., Harris v Aluminum Co of America, 550 F Supp 1024, 1030 (WD Va, 1982); Van Cleave v Gamboni Const Co, 706 P2d 845, 847 (Nev, 1985); Alaska Airlines, Inc v Sweat, 568 P2d 916, 929-30 (Alas, 1977). Only one court has held that the language of the 1955 Act which deals with the release of joint tort-feasors does not apply to liability based on a master-servant relationship, Craven v Lawson, 534 S.W.2d 653, 656 (Tenn, 1976), and that opinion has been criticized. Harris v Aluminum Co of America, 550 F Supp at 1029-30; Van Cleave v Gamboni, 706 P2d at 848; Comment, The Covenant Not to Sue: Virginia's Effort to Bury the Common Law Rule Regarding the Release of Joint Tortfeasors, 14 U Rich L Rev 809, 833 (1980); Comment, Torts ÔÇö Vicarious Liability ÔÇö Covenant Not to Sue Servant or Agent as Affecting Liability of Master or Principal, 44 Tenn L Rev 188, 199 (1976). [Krukiewicz, supra, p 1352. Emphasis added.][^[69]]

Additionally, the majority of courts that have considered the effect of a release of a principal on the liability of an agent have construed the Uniform Acts to include the vicariously liable principal as a "tortfeasor." Several have stated that the effect of a release does not depend on whether it is given to the agent or to the vicariously liable principal.^[70]

Only Bristow and Craven, criticized in Krukiewicz, supra, support the result of the lead opinion.^[71] All the other cases cited in the lead opinion either do not construe a uniform act or construe statutes not based on a uniform act.^[72]

VII

We would hold that the release of an employee/agent does not release a vicariously liable employer/principal, and that the vicariously liable employer/principal is a "tortfeasor," one of two or more "persons liable in tort" for the same injury within the meaning of RJA, ž 2925d. We would reverse the Court of Appeals and reinstate the jury's verdict. There are not, however, four votes to do so. Nor are there four signatures on the lead opinion. In order to resolve this appeal, we join in the remand provided for in Justice BOYLE'S opinion.

CAVANAGH and ARCHER, JJ., concurred with LEVIN, J.

ARCHER, J. (dissenting).

We granted leave to consider whether a release, containing a provision expressly reserving plaintiffs' claim against the employer hospital-principal, and executed in consideration of settlement payments received from independent contractors-agents, nevertheless released the hospital from vicarious liability under MCL 600.2925d; MSA 27A.2925(4),^[1] and, if the release of the independent contractors-agents is not protected by MCL 600.2925d; MSA 27A.2925(4), the release should be reformed as a covenant not to sue, thereby effectuating the intent of the settling parties which would not discharge the hospital.

I would hold that the plain language of Michigan's contribution statute, MCL 600.2925d; MSA 27A.2925(4), as amended in 1974, applies to all tortfeasors liable for the same injury, including those vicariously liable. Therefore, I need not reach the issue whether to reform the release as a covenant not to sue. I would further hold that the evidence introduced at trial, viewed in its entirety, was sufficient for the jury to determine that the hospital was liable for independent as well as vicarious acts of negligence. Finally, I would hold that the hospital's failure to request special verdicts as urged by the trial court impermissibly invited error, and thereby waived the hospital's right to appeal the issue.

I would reverse the decision of the Court of Appeals and reinstate the jury verdict against defendant Lansing General Hospital.

FACTS

On June 16, 1978, Gene Christopher Schneider, a seven-year-old boy, was admitted to Lansing General Hospital, on the recommendation of his family physician, Robert Wirt, D.O., for a tonsillectomy and bilateral tympanotomy. After the tympanotomy, and before commencement of the tonsillectomy, the child developed a drastic increase in heart rate and his legs and fingers became cyanotic. Jana Palmer, the certified nurse anesthetist who had administered anesthesia for the surgery, called in Jack Gilmore, D.O., the supervising anesthesiologist. Dr. Gilmore had not been present in the operating room during the surgery. The child subsequently suffered a cardiac arrest. After half an hour of resuscitative efforts, the child resumed a spontaneous heartbeat, and was taken to the intensive care unit where he was placed on a respirator and cardiac monitor. Various physicians on staff at the hospital examined the child and directed treatment in the intensive care unit. The following day, a second cardiac arrest occurred. While maintained on a respirator in the intensive care unit, the child's condition continued to deteriorate until his death six days later, on June 22, 1978.

Plaintiffs, the personal representative and the parents of the child, filed this wrongful death action in the Ingham Circuit Court on June 13, 1980. The defendants named were Jana Palmer, the certified nurse anesthetist, and Dr. Gilmore, the anesthesiologist, both of whom were employed by Capital Anesthesiologists, P.C. Capital had the exclusive contract to perform anesthesia within the hospital. The other defendants included Lansing General Hospital, Gerald Gilroy, D.O., the ear-nose-throat specialist who performed the tympanotomy, David Sciamanna, D.O., the assisting resident pediatrician, and various medical personnel.

Prior to trial, a number of physicians, as well as Capital Anesthesiologists, were dismissed for reasons unrelated to this appeal. Also prior to trial, a motion for an order authorizing settlement in the amount of $85,000 against nurse Palmer, only, was granted. After trial began, plaintiffs entered into a similar settlement in the amount of $172,739 with Dr. Gilmore, the anesthesiologist. In both instances, a release was executed in consideration of the settlement payments and a provision was included in the release specifically reserving plaintiffs' claims against the remaining defendants.^[2] There were no objections raised by the defendants as to the settlements, to the language of the release, or to the court's order authorizing the release.

Subsequently, during trial, the hospital moved to strike all allegations concerning the actions of Palmer and Gilmore as ostensible agents of the hospital (under Grewe v Mt Clemens General Hosp, 404 Mich. 240; 273 NW2d 429 [1978]). The motions were based upon the theory that the settlements released the hospital from vicarious liability for the actions of Palmer and Gilmore despite the express reservation of those claims in the settlement order. While the trial court ruled that the settlements released the hospital from vicarious liability for the actions of Palmer and Gilmore, it permitted the evidence concerning the conduct of the released defendants to go to the jury, reasoning that the evidence was necessary to a jury's understanding of the independent allegations of negligence against the remaining defendants, Lansing General, Dr. Sciamanna, and Dr. Gilroy. The trial court also recommended that the hospital request the return of special verdicts to determine whether the jury verdict was based upon the hospital's vicarious or independent liability. The defendants rejected the court's request.

During trial, at the close of plaintiffs' proofs, and upon defendants' motion, the trial court struck certain allegations of independent hospital negligence from plaintiffs' complaint, but granted plaintiffs' motion to amend their complaint to conform to the proofs. The amended complaint continued to include allegations that the hospital failed to require use of a precordial stethoscope in pediatric cases.

After a two-week trial, the jury returned a verdict against defendant Lansing General Hospital in the amount of $1 million. It found no cause of action as to Drs. Gilroy and Sciamanna. The verdict was reduced by the settlement amount of $257,739, and, on June 2, 1983, judgment was entered against Lansing General Hospital for $742,261.

On March 4, 1985, the Court of Appeals affirmed the jury verdict, concluding that the acts of the settling tortfeasors were properly introduced where "critical to the ability of plaintiffs to prove their independent claims against the hospital." Theophelis v Lansing General Hosp, 141 Mich. App. 199, 205; 366 NW2d 249 (1985). On rehearing, a divided Court of Appeals set aside the jury verdict and remanded the case for a new trial. Having found that the trial court granted a directed verdict on all but one of plaintiffs' independent claims against the hospital, the Court of Appeals decided there was now "no competent evidence or expert testimony" to support the issue of independent liability on the part of the hospital to send to the jury. Theophelis v Lansing General Hosp (On Rehearing), 148 Mich. App. 564, 567; 384 NW2d 823 (1986). Judge MICHAEL J. KELLY dissented, continuing to adhere to the analysis set forth in the original Court of Appeals opinion. Id., 568.

I

The Court of Appeals found that under common law, "the release of nurse Palmer and Dr. Gilmore released the hospital from any liability based on a theory of respondeat superior." 141 Mich. App. 204; 148 Mich. App. 566. I disagree. An appreciation for the historical development of Michigan's contribution statute, MCL 600.2925d; MSA 27A.2925(4), is necessary for an understanding of our disagreement with the Court of Appeals.

The common-law rule in Michigan is that a release of one or more joint tortfeasors releases the others, regardless of an express reservation of rights against another tortfeasor. McBride v Scott, 132 Mich. 176; 93 N.W. 243 (1903); Lindsay v Acme Cement Plaster Co, 220 Mich. 367; 190 N.W. 275 (1922); Moffit v Endtz, 232 Mich. 2; 204 N.W. 764 (1925); MacDonald v Henry Hornblower & Weeks, 268 Mich. 626; 256 N.W. 572 (1934). This rule is based upon the theory that there can be but one compensation for a joint wrong, and the action is satisfied once the injured party has been paid by any joint tortfeasors.^[3]

The harshness of and dissatisfaction with the common-law rule has resulted in a modern trend towards various devices designed to avoid its effect.^[4] These devices included the use of covenants not to sue,^[5] uniform legislation abrogating the rule, such as the Uniform Contribution Among Tortfeasors Act,^[6] and state statutes,^[7] among other things.^[8] Generally, these devices obviate the need for the common-law rule by setting off the amount of any settlement.

In 1941, the Michigan Legislature chose to ameliorate the rigidity of the common-law rule by enacting a variation of the Uniform Contribution Among Tortfeasors Act. The statute, 1941 PA 303, ž 2, 1948 CL 691.562; MSA 27.1683(2), provided:

It shall be lawful for all persons having a claim or cause of action against 2 or more joint tortfeasors to compound, settle with, and discharge, at any time prior to rendition of a judgment in said action, any and everyone or more of said joint tortfeasors for such sum as such person may deem fit, without impairing the right of such person or persons to demand and collect the balance of said claim or cause of action from the remaining joint tort-feasors, against whom such person, or persons, has such claim or cause of action, and not so released. [Emphasis added.]

Unlike the 1941 Uniform Contribution Among Tortfeasors Act, the original Michigan statute did not include the definition of "joint tortfeasor." The use of the term "joint tortfeasor" without definition resulted in problems of interpretation. See Geib v Slater, 320 Mich. 316; 31 NW2d 65 (1948), overruled on other grounds Moore v Palmer, 350 Mich 363; 86 NW2d 585 (1957); Moyses v Spartan Asphalt Paving Co, 383 Mich. 314; 174 NW2d 797 (1970).

In Geib, supra, 320, the issue was whether the defendant owner of an automobile driven by a negligent employee was a "joint tort-feasor" within the statute. This Court held that the statute did not apply to the liability of the automobile owner because the owner was not a "joint tort-feasor."

While the 1955 revision of the uniform act eliminated the term "joint tortfeasor" and substituted the term "tortfeasor," the language of the 1961 Michigan revision authorizing release of a joint tortfeasor without discharging any others remained identical to the original 1941 act.^[9] Later, in an effort to resolve the definitional problem, this Court in Moyses, supra, 329, defined "joint tortfeasors," stating,

[W]here two (or more) persons owe to another the same duty and by their common neglect of that duty such other is injured, the two (or more) have committed a joint tort and therefore are joint tortfeasors.

The Legislature expressed its dissatisfaction with the various judicial interpretations that evolved following its failure to define "joint tortfeasors" in the 1941 statute and its 1961 amendment by amending MCL 600.2925d; MSA 27A.2925(4) to totally eliminate use of the term "joint tortfeasor." The 1974 amendment provides in pertinent part:

When a release or a covenant not to sue or not to enforce judgment is given in good faith to 1 of 2 or more persons liable in tort for the same injury or the same wrongful death:

(a) It does not discharge any of the other tortfeasors from liability for the injury or wrongful death unless its terms so provide. [Emphasis added.]

The 1974 change in language to include "persons liable in tort" and to eliminate the term "joint tortfeasors" indicates the Legislature's intention to apply the statute to all tortfeasors liable in tort for the same injury regardless of whether their liability was joint. To read the present statute, as amended in 1974, otherwise would render nugatory a related provision of the statute, ž 2925d(b). Section 2925d(b) provides that a release or covenant not to sue "reduces the claim against the other tort-feasors to the extent of the amount stipulated by the release or the covenant...." See n 1.^[10] Every part of a statute should be given meaning and no part should be rendered nugatory. Baker v General Motors Corp, 409 Mich. 639, 665; 297 NW2d 387 (1980); Stowers v Wolodzko, 386 Mich. 119, 133; 191 NW2d 355 (1971).

The 1974 amended statute, unlike its predecessor, does not limit its application to joint tortfeasors. Rather, the express language of the current statute applies to "1 of 2 or more persons liable in tort for the same injury or the same wrongful death." Liability "in tort for the same injury" would include claims founded in vicarious liability. The defendant hospital is "1 of 2 or more ... liable in tort for the same ... wrongful death." Thus, its vicarious liability falls within the statute. Other jurisdictions with statutes having language identical or similar to the 1974 amendment agree.^[11] Accordingly, I would find the principle established in Geib, supra, prior to the 1974 amendment to be in conflict with the plain language of ž 2925d, and therefore, no longer valid.

My interpretation is further supported by the recent decisions interpreting related provisions of ž 2925a. See O'Dowd v General Motors Corp, 419 Mich. 597; 358 NW2d 553 (1984), and Piper Aircraft Corp v Dumon, 421 Mich. 445; 364 NW2d 647 (1984). In both O'Dowd and Piper, we acknowledged that the statute's focus is on the relationship between the defendants and the injured party rather than the relationship between the defendants themselves.

More importantly, a statute must be construed in light of legislative intent. O'Donnell v State Farm Mutual Automobile Ins Co, 404 Mich. 524; 273 NW2d 829 (1979). Clearly, the Legislature's intention in enacting MCL 600.2925d; MSA 27A.2925(4) was to "(1) encourag[e] settlements and (2) assur[e] that a plaintiff is fully compensated for injuries sustained." Mayhew v Berrien Co Road Comm, 414 Mich. 399, 411-412; 326 NW2d 366 (1982). This policy enables a plaintiff to settle with an agent without losing his cause of action against a party liable under a theory of respondeat superior, while allowing the remaining defendant complete access to the courts for a full adjudication of his liabilities and rights to indemnification. See Knutson v Morton Foods, Inc, 603 S.W.2d 805 (Tex, 1980).

To allow the common-law rule to survive would deter the parties from settling because they would risk losing other claims necessary to obtain full compensation. Any inequities argued by the defendant are obviated by the setoff requirements imposed by related provisions of the statute. See MCL 600.2925a et seq.; MSA 27A.2925(1) et seq. I reject the hospital's argument that enforcement of the release in these cases would encourage a circuity of action in view of its right to indemnification. MCL 600.2925a(7); MSA 27A.2925(1)(7). I believe the remedial nature of ž 2925d ÔÇö to encourage settlements and assure full compensation to the plaintiff for injuries ÔÇö takes precedence over possible indemnity problems. The defendant to be indemnified suffers no prejudice because the settlement actually results in the reduction of its possible liability by the amount paid in settlement. See Knutson, supra. Loss of indemnification is especially not a problem in cases where the agent or employee settles for the full amount of the policy limits under his liability insurance. See Munson v United States, 380 F2d 976 (CA 6, 1967).^[12] Furthermore, in cases where, as here, there is a question of active negligence as well as vicarious negligence, the defendant may have no right to indemnity.^[13]

I also disagree with the hospital's argument that this view would create an atmosphere ripe for collusion.^[14] If there is any indication that a settlement involved collusion, a trial court can rely on the "good faith" requirement of ž 2925d to find that the settlement was not made in good faith. To further avoid any impropriety, challenges to the good faith of the settlements should be made at the time of settlement or prior to trial.

Accordingly, under MCL 600.2925d; MSA 27A.2925(4), the release of defendants Palmer and Gilmore did not release the defendant hospital from vicarious liability for the actions of its agents or employees. I would reverse the decision of the Court of Appeals and reinstate the judgment against defendant Lansing General Hospital.

II

Because I conclude that MCL 600.2925d; MSA 27A.2925(4) is applicable in cases of vicarious liability, I do not find it necessary to pursue the issue whether the releases, if not protected by the statute, should be reformed as covenants not to sue which discharge the hospital to effectuate the intent of the settling parties.

III

Notwithstanding the applicability of MCL 600.2925d; MSA 27A.2925(4) to defendant hospital's vicarious liability, the record also reflects that it would have been possible for the jury to find the hospital liable for independent acts of negligence.

While the trial court struck certain allegations of independent hospital negligence from plaintiffs' complaint upon the hospital's motion to direct a verdict, the trial court simultaneously allowed the plaintiffs to amend their complaint to conform to the proofs. The hospital concedes that the issue whether defendant hospital had been independently negligent in failing to require the use of the precordial stethoscope as standard operating procedure remained a viable issue for the jury. When testifying at trial, plaintiffs' expert, Dr. Zsigmond, recognized, albeit in a different context, that failure to use the precordial stethoscope in this case was a "very severe major deviation from the good standard of practice." The precordial stethoscope issue and Dr. Zsigmond's testimony were submitted, without objection, to the jury. If there is sufficient evidence tending to support the jury's verdict, a court cannot set it aside even if it might be in doubt as to the ultimate facts. Baldwin v Nall, 323 Mich. 25, 29; 34 NW2d 539 (1948). Hence, I agree with Judge KELLY'S dissent on rehearing that when viewed in its entirety, the evidence introduced at trial regarding use of the precordial stethoscope could support the jury's verdict.

IV

Finally, the hospital contends that a new trial would be required even if this Court held that the hospital could be vicariously liable for Palmer's and Gilmore's actions, as it cannot be determined from the jury's verdict, upon which basis it found liability. Baker v Alt, 374 Mich. 492, 497; 132 NW2d 614 (1965). However, the record reflects that the trial judge repeatedly recommended and encouraged the hospital to request the return of special verdicts at the close of proofs to determine upon what basis liability was found. The hospital rejected the request.

Had the hospital made the request, any confusion as to the basis upon which the jury found the hospital liable would have disappeared. "A party cannot sit back and after the jury decides against him claim error after he had been given the fullest opportunity to correct any alleged error at the trial." Baldwin v Nall, supra, 31; Watson v Dax, 334 Mich. 320, 330; 54 NW2d 674 (1952). Accordingly, I would find that the hospital's failure to request special verdicts at the urging of the trial court invited impermissible error, thereby waiving its right to complain on this issue.

CONCLUSION

I would reverse the decision of the Court of Appeals and reinstate the judgment against Lansing General Hospital.

NOTES

[1] 1961 PA 236, žž 2925a-d, added by 1974 PA 318, ž 1.

[2] The releases provided in pertinent part:

It is intended that this Release shall be operative as to Jana Palmer only and shall not be construed to release any other persons or entity who may be liable in tort for the same wrongful death. MCL 600.2925d; MSA 27A.2925(4).

It is intended and understood that this release shall be operative as to Jack Gilmore, D.O. only, and shall not be construed to release any other persons or entities who may be liable in tort for the alleged wrongful death of Gene Christopher Schneider, deceased, and any and all other damages alleged to result from the alleged wrongful death. MCL 600.2925d; MSA 27A.2925(4).

[3] Our use of the term, "independent negligence," is intended only for the purpose of identifying those claims by plaintiffs which were directed to hospital policies and procedures as opposed to those claims which rested directly upon the negligent acts of Nurse Palmer and Dr. Gilmore and are referred to as "vicarious liability" claims.

Obviously, if claims directed to hospital policies and procedures involved acts of directors, officers or agents of the hospital during the scope of corporate activity, any liability of the hospital for such acts would also be "vicarious." See Henn & Alexander, Corporations (3d ed), ž 234, pp 624-625. Plaintiffs did not name the particular officers or agents of the hospital in their complaint. See Jones v Martz & Meek Const Co, 362 Mich. 451, 455; 107 NW2d 802 (1961).

[4] The trial court recommended to the hospital that it request a special verdict which would require the jury to indicate the extent, if any, to which its verdict would be based upon vicarious liability or liability for independent acts.

[5] The single remaining allegation against the hospital of independent negligence, as determined by the Court of Appeals on reconsideration of the record, was a claim that the hospital negligently failed to "ascertain that in all pediatric cases a precordial stethoscope shall be used."

[6] KELLY, J., dissented, maintaining that there was sufficient evidence to support a claim of independent negligence against the hospital, particularly in light of the fact that the trial court had granted a motion by plaintiffs to amend their complaint to conform to the proofs.

[7] The 1939 version of the UCATA defined the term "joint tortfeasors" as two or more persons jointly or severally liable in tort for the same injury to person or property, whether or not judgment has been recovered against all or some of them. However, the 1955 revised act eliminated the definition and the term itself. See Commissioners' Comment, 12 ULA 64.

[8] Compare Alaska Airlines, Inc v Sweat, 568 P2d 916 (Alas, 1977), and Aherron v St John's Mercy Medical Center, 713 S.W.2d 498 (Mo, 1986), with Bacon v United States, 321 F2d 880 (CA 8, 1963), Seaboard Air Line R Co v Coastal Distributing Co, 273 F Supp 340 (D SC, 1967), Bristow v Griffitts Const Co, 140 Ill App 3d 191; 488 NE2d 332 (1986), Dickey v Estate of Meier, 188 Neb 420; 197 NW2d 385 (1972), Jacobson v Parrill, 186 Kan 467; 351 P2d 194 (1960), Barsh v Mullins, 338 P2d 845 (Okla, 1959), and Mid-Continent Pipeline Co v Crauthers, 267 P2d 568 (Okla, 1954).

[9] The 1941 contribution statute was amended by 1961 PA 236, and by 1974 PA 318.

[10] 1929 CL 4648.

[11] Section 2 of 1941 PA 303 provided as follows:

It shall be lawful for all persons having a claim or cause of action against 2 or more joint tort-feasors to compound, settle with, and discharge, at any time prior to rendition of a judgment in said action, any and everyone or more of said joint tort-feasors for such sum as such person may deem fit, without impairing the right of such person or persons to demand and collect the balance of said claim or cause of action from the remaining joint tort-feasors, against whom such person, or persons, has such claim or cause of action, and not so released.

[12] The decision in Geib, supra, has been criticized. See the opinion of Judge LEVIN in Duncan v Beres, 15 Mich. App. 318, 327-329; 166 NW2d 678 (1968).

[13] Some courts have concluded that to hold a principal liable despite his agent's release would frustrate a second goal of the contribution act, i.e., the early and final settlement of claims. Because the act preserves the right of indemnity, see ž 2925a(7), it is argued that such a result actually spawns litigation and leads to circuity of action. See 24 ALR4th 547, 552, 567. The possibility of indemnity actions has been at least recognized in Craven v Lawson, 534 S.W.2d 653, 656 (Tenn, 1976); Ritter v Technicolor Corp, 27 Cal App 3d 152, 155; 103 Cal Rptr 686 (1972); Swanigan v State Farm Ins Co, 99 Wis 2d 179, 201-203; 299 NW2d 234 (1980); Van Cleave v Gamboni Const Co, 101 Nev 524, 528-529; 706 P2d 845 (1985).

[14] The use and significance of a covenant not to sue has been explained as follows:

A covenant not to sue is to be distinguished from a release in that it is not a present abandonment or relinquishment of the right or claim but is merely an agreement not to sue on an existing claim. It does not extinguish the cause of action. As between the parties to the agreement, the final result is the same in both cases. The difference is primarily in the effect as to third parties, and is based mainly on the fact that in the case of a release there is an immediate release or discharge, whereas in the other case there is merely an agreement not to prosecute a suit. [66 Am Jur 2d, Release, ž 2, p 679. See also 76 CJS, Release, žž 1-3, pp 629-630. 1974 PA 318, ž 2925d refers to both releases and covenants not to sue.]

[1] These observations are not intended as a reflection on the trial court or counsel for the parties. The perspective of an appellate court focused on a single issue is, by definition, more precise than the rich and shifting range of issues dealt with in a lengthy trial.

[2] Somewhat inexplicably, the Court of Appeals remanded for a new trial, despite holding that the issue of vicarious liability was improperly submitted to the jury and despite its holding that there was insufficient evidence to support the plaintiffs' alternative theory of independent negligence.

[1] Prosser & Keeton, Torts (5th ed), ž 49, p 332. See Geib v Slater, 320 Mich. 316; 31 NW2d 65 (1948).

[2] Prosser & Keeton, Torts (5th ed), ž 50, p 336. See Moyses v Spartan Asphalt Paving Co, 383 Mich. 314, 334; 174 NW2d 797 (1970).

[3] Prosser & Keeton, Torts (5th ed), ž 49, p 332, ž 50, p 336; 12 ULA 63; 4 Restatement Torts, 2d, ž 885, p 333.

[4] 1948 CL 691.561, carried forward in pertinent part in 1970 CL 600.2925.

[5] 12 ULA 57.

[6] MCL 600.2925a-600.2925d; MSA 27A.2925(1)-27A.2925(4). Section 2925b has been amended by 1982 PA 147 and 1986 PA 178.

[7] Similarly see Slinkard v Nat'l Machine & Tool Co, 274 Mich. 662; 265 N.W. 494 (1936); Garstka v Republic Steel Corp, 294 Mich. 387; 293 N.W. 691 (1940); Kallas v Lincoln Mutual Casualty Co, 309 Mich. 626; 16 NW2d 99 (1944); Larabell v Schuknecht, 308 Mich. 419; 14 NW2d 50 (1944); Geib v Slater, supra.

[8] 4 Restatement Torts, 2d, ž 885(1), p 333; see part II, post, p 504, n 16.

[9] See n 17.

[10] See, e.g., Cook v City Transport Corp, supra, p 95, where this Court held in an action by a passenger on a bus, apparently against the driver of another automobile, that a covenant not to sue the owner or operator of the bus did not release the other driver and described, as set forth in the text preceding n 4, the owner and operator of the bus and the other driver as "joint tortfeasors," and where the dissenting opinion said "[t]he reason for this rule is that all joint tortfeasors are jointly and severally liable." (Emphasis supplied.)

See also cases collected in Duncan v Beres, 15 Mich. App. 318, 326, n 15; 166 NW2d 678 (1968).

Courts in other jurisdictions accepted the common jargon (see e.g., n 16) of lawyers and judges who used the term "joint tortfeasor" to describe those severally as well as jointly liable, and held that remedial legislation based on the 1939 Uniform Act was applicable to severally as well as jointly liable tortfeasors. See Smith v Raparot, 101 RI 565; 225 A2d 666 (1967); Holve v Draper, 95 Idaho 193, 195; 505 P2d 1265 (1973).

[11] See n 4.

[12] 12 ULA 57.

[13] 4 Restatement Torts, 2d, ž 885, p 460.

[14] Id.

[15] 4 Restatement Torts, 2d, comment b, p 334.

[16] 4 Restatement Torts, 2d, ž 885(1), p 333.

The rule stated in the first Restatement of Torts regarding covenants not to sue was restated in Restatement Torts, 2d, ž 885(2).

The comment states:

c. Rationale. At present, persons whose independent tortious conduct contributes to a tort are in most cases (see ž 879) regarded as joint tortfeasors, without distinction between them and those who cooperate in producing a tortious result. Likewise today releases are effective if given for consideration although not under seal and are therefore regarded as contracts rather than grants. Furthermore, the tendency is to give effect to the intent of the parties to a transaction rather than to regard as controlling the formalities with which the transaction is executed. The law has now developed so that it is possible to carry out the intent of the parties as expressed in the document by which the release is given. Thus, if it is the understanding of the parties that the payment by the tortfeasor will be full compensation for the injury suffered and therefore full satisfaction of the claim, there is no longer an enforceable claim against any tortfeasor.

d. Limits and effect of rule. A covenant not to sue clearly indicates by its form that it is not intended to discharge the liability of other tortfeasors jointly liable. On the other hand, a document in the usual form of a release given to one of them is normally construed as intended to discharge all claims for the tort and operates to discharge others also liable for the same harm. If, however, there is language in the release that manifests that the releasor intended to preserve his rights against the others, effect is given to this manifestation.

It is not essential that the reservation of rights against other tortfeasors be expressed in the document itself. Such a requirement might easily prove a boobytrap for the unwary, since releases are commonly signed by individuals without legal advice. The agreement as to the effect of the release may be proved by external evidence; and the objection of the parol evidence rule is met by the fact that the second tortfeasor who raises the question is not a party to the instrument.

A release obtained as the result of a basic mistake or of fraud or duress may be rescinded and after the rescission the claim against all the tortfeasors revives. On the circumstances under which there may be rescission of an agreement, see Restatement, 2d, Contracts, žž 301-319 (Tent. Draft).

[17] Zenith Radio Corp v Hazeltine Research, Inc, 401 U.S. 321, 343; 91 S. Ct. 795; 28 L. Ed. 2d 77 (1971); Aro Mfg Co, Inc v Convertible Top Replacement Co, Inc, 377 U.S. 476, 581; 84 S. Ct. 1526; 12 L. Ed. 2d 457 (1964).

[18] Zenith, n 17 supra, p 347.

[19] Id., p 343.

[20] Id., p 347.

[21] RJA, ž 2925d continues:

(b) It reduces the claim against the other tort-feasors to the extent of any amount stipulated by the release or the covenant or to the extent of the amount of the consideration paid for it, whichever amount is the greater.

(c) It discharges the tort-feasor to whom it is given from all liability for contribution to any other tort-feasor. [MCL 600.2925d; MSA 27A.2925(4).]

Neither the 1939 nor the 1955 Uniform Acts were adopted in this state without substantive additions and deletions. There is, however, no substantive difference between RJA, ž 2925d and the corresponding language of the 1955 Uniform Act.

[22] There was an express reservation of the right to sue the owner of the automobile, Boucher, supra, p 315.

[23] Boucher, supra, p 316.

[24] Id., p 317.

In Geib, supra, p 321, the Court had advanced much the same argument which two and one-half years later it rejected in Boucher. In Geib, in finding that the release of the driver discharged the owner, the Court said that the owner's liability "may be compared with that of a surety for the honesty of an employee, whose obligation differs from that of his surety."

[25] Boucher, supra, p 317.

[26] See n 16.

[27] MCL 257.401; MSA 9.2101.

[28] Geib, supra, p 321.

[29] Ante, p 483.

[30] Id., p 484.

[31] Id., p 487.

[32] Id., p 488.

[33] Geib did not hold that the owner of the automobile was discharged from liability because the 1941 act was not intended to apply to a release of a tortfeasor for whose wrong another person was vicariously liable. It held that the 1941 act was not applicable because it applied only to "joint tortfeasors," persons jointly liable, and that the act did not apply to a release of a severally liable tortfeasor. In so holding the Court did not distinguish between tortfeasors vicariously liable as a result of common-law doctrine and those vicariously liable by virtue of statute.

In Geib, the Court found that there was not the requisite oneness in the bases of the owner's and driver's liabilities for a finding that the owner and driver were joint tortfeasors. The lead opinion declares that the owner and driver have a "common liability" and constitute "only one tortfeasor."

[34] Ante, p 483.

[35] 2A Sands, Sutherland Statutory Construction (4th ed), ž 45.01, p 1; ž 45.14, p 70 ("Certainly a legislature is not compelled by any superior force to obey dictionary definitions or the rules of grammar"); ž 46.01, p 74; ž 46.02, p 81; ž 46.05, p 92; ž 46.07, p 110; ž 47.07, p 133; cf. 3 Corbin, Contracts, ž 572B, p 198 (Supp, 1971).

[36] right of contribution exists only in favor of a tortfeasor who has paid more than his pro rata share of the common liability.... [RJA, ž 2925a(2). Emphasis supplied.]

See ns 42 and 43 and accompanying text.

[37] Moyses overruled what was left of the common-law bar of "contribution between or among ``wrongdoers.'" Id., p 334. (Emphasis supplied.) The lead opinion (ante, pp 487-488) stresses that a vicariously liable person is not a "wrongdoer." It can accordingly be argued on the basis of that analysis that the common-law doctrine barring contribution between severally liable tortfeasors was not overruled in Moyses where the action for contribution is brought by a person subjected to vicarious liability. And, thus, that the vicariously liable tortfeasor continues to be barred by the Geib analysis from maintaining a common-law action for contribution.

[38] Such as a staff physician not employed by the hospital.

[39] MCL 600.2925b; MSA 27A.2925(2).

[40] 12 ULA 87. 1974 PA 318 was amended by 1982 PA 148, adding the words "as between themselves only and without affecting the rights of the injured party to a joint and several judgment" to the introductory phrase, and deleting the word "not" in subdivision (a). The act was additionally amended by 1986 PA 178, adding the words "Except as otherwise provided by law," to the beginning of the introductory phrase.

[41] MCL 600.2925a(7); MSA 27A.2925(1)(7).

[42] 12 ULA 66.

[43] See n 36 and accompanying text.

[44] See ns 36 and 38.

An employer (master) is entitled to seek indemnification from an employee (servant). Indemnity is excepted from the operative effect of RJA, ž 2925a:

(7) This section does not impair any right of indemnity under existing law. Where 1 tort-feasor is entitled to indemnity from another, the right of the indemnity obligee is for indemnity and not contribution, and the indemnity obligor is not entitled to contribution from the obligee for any portion of his indemnity obligation. [MCL 600.2925a; MSA 27A.2925(1).]

[45] Ante, p 490.

[46] Id., p 491.

[47] Id.

[48] Id., p 487.

[49] Prosser & Keeton, Torts (5th ed), ž 49, p 332.

[50] Id., p 333.

[51] 66 Am Jur 2d, Release, ž 37, p 716.

[52] 53 Am Jur 2d, Master and Servant, ž 408, p 416.

[53] Prosser & Keeton, Torts (5th ed), ž 49, p 333.

[54] O'Dowd v General Motors Corp, 419 Mich. 597, 609, n 24; 358 NW2d 553 (1984). See RJA, ž 2925d(b), n 21 supra.

[55] Prosser & Keeton, Torts (5th ed), ž 49, p 333.

[56] The lead opinion assumes that the Legislature was aware of Geib and by failing to expressly overrule it in effect signaled approval. Ante, pp 487-488. The same analysis would be applicable to Boucher: the Legislature was aware of Boucher and by failing to expressly overrule it in effect signaled approval.

[57] MCL 436.22; MSA 18.993.

[58] See ante, p 489, n 13.

[59] In both Craven and Bristow, the document evidencing the settlement was a covenant not to sue. The language of the 1955 Uniform Act enacted as RJA, ž 2925d speaks of both a release and a covenant not to sue without distinction. See n 21 and accompanying text.

[60] See ns 36 ff., and accompanying text.

[61] See, e.g., Smith v Raparot, supra; Holve, supra, p 195.

[62] By RJA, ž 2925a(7).

[63] Indeed, of the authorities cited in the lead opinion, only Craven v Lawson and Bristow v Griffitts Const Co, (following Craven), supra, discuss the Uniform Act and reach the result of the lead opinion. Craven's holding and reasoning have been criticized. See, e.g., Van Cleave v Gamboni Const Co, 101 Nev 524; 706 P2d 845 (1985).

[64] The 1955 Uniform Act does not distinguish between a release and a covenant not to sue. By its terms, it applies to both. See n 21 and accompanying text.

[65] Alaska Airlines, supra, p 923.

[66] The court relied on Smith v Raparot, n 10 supra, where the Rhode Island Supreme Court reached the same conclusion on the basis of the definition of "joint tort-feasor" in the 1939 Uniform Act. Similarly see Holve, n 10 supra.

[67] In both Harris and Thurston, the document evidencing the settlement was a covenant not to sue. The language of the 1955 Uniform Act enacted as RJA, ž 2925d speaks of both a release and a covenant not to sue without distinction. See n 21 and accompanying text.

[68] Blackshear v Clark, 391 A2d 747, 748 (Del, 1978); Smith v Raparot, n 10 supra; Holve, n 10 supra; Clark v Brooks, 377 A2d 365 (Del Super, 1977).

[69] Also adopting the Alaska Airlines-Harris line of cases and rejecting Craven is Mesler v Bragg Management Co, 39 Cal 3d 290; 216 Cal Rptr 443; 702 P2d 601 (1985). The California Supreme Court held that plaintiff's settlement with a subsidiary crane operator company did not release the parent corporation.

The California version of the Uniform Act has two important differences: (1) the California act does not speak of "joint tortfeasors" but instead of "tortfeasors claimed to be liable for the same tort," and (2) the act applies to "one or more persons ... liable solely for the tort of one of them or of another, as in the case of the liability of a master for the tort of his servant." Id., p 302 (emphasis added).

The Texas Supreme Court also adopted the view expressed in Alaska Airlines in Knutson v Morton Foods, Inc, 603 S.W.2d 805 (Tex, 1980) (covenant not to sue). The court held that a release must identify the parties to be released. The court extended this requirement to cases in which the nonreleased defendant was liable solely on the theory of respondeat superior. The plaintiffs in that case who had released the agents could thus sue the nonreleased principal who was vicariously liable.

[70] Three state supreme courts have dealt with the effect of a plaintiff's release of the principal/master on the liability of the agent/servant. The courts held that their state's versions of the 1939 and 1955 Uniform Acts applied. In Aherron v St John's Mercy Medical Center, 713 S.W.2d 498, 500, 501 (Mo, 1986), the Missouri Supreme Court held that the plaintiff's release of the vicariously liable hospital did not release the servant/agent doctor under a Missouri statute "substantially the same as" the 1955 Uniform Act and said: "[a]n employee and a vicariously liable employer are ``persons liable in tort for the same injury or wrongful death'"; "a vicariously liable employer as well as its employee" are "other tortfeasors" under the act. (Emphasis added.)

The Missouri court relied on Blackshear v Clark, 391 A2d 747 (Del, 1978), where the Delaware Supreme Court had held that a plaintiff's release of the hospital in a medical malpractice action did not release the physician-employee under a Delaware statute modeled on the 1939 Uniform Act. Referring to the language "2 ... persons ... [are] severally liable in tort for the same injury," the court said, in language that has been repeated in other opinions:

The basis of liability is not relevant, nor is the relationship among those liable for the tort. In short, it makes no difference whether the Center's liability is based upon the doctrine of respondeat superior or any other legal concept. The point is that both it and the Doctor are (at least) "severally" liable for the same injury to plaintiff. Therefore, the Uniform Contribution Among Tort-Feasors Act applies. [Id., p 748.]

The New Hampshire Supreme Court, in Waters v Hedberg, 126 NH 546; 496 A2d 333 (1985), followed Blackshear and Alaska Airlines. Further, the same day, the New Hampshire court declared in Chamberlain v Town of Kingston, 126 NH 553; 496 A2d 337 (1985):

This ... raises the question whether an injured plaintiff, who has settled a claim with an employee tortfeasor and has covenanted not to sue the employee, may maintain an action against the employer where no independent fault of the employer is alleged. The sole basis of the employer's alleged liability is the doctrine of respondeat superior.

Our decision of this date in Waters v Hedberg, 126 NH 546; 496 A2d 333 (1985) presented the obverse of this case. The release in Waters was given to the employer, and the suit was then permitted to be maintained against the employee. The considerations are no different in this case, and the suit likewise is not barred.

[71] In Bristow, an Illinois intermediate appellate court considered an Illinois statute identical to RJA, ž 2925d. The court noted the competing views represented by Alaska Airlines and Craven on the question whether the Uniform Act included situations involving vicarious liability. The Bristow court noted that in Craven "the court reasoned that the Uniform Act did not purport to intrude where a right to indemnity, not contribution, existed." Bristow, p 194. The Bristow court then observed that the act "envisions the sharing of liability between two culpable defendants," emphasizing "the culpability of the parties rather than the precise legal means by which the plaintiff is ultimately able to make each defendant compensate him for his loss." Id., p 195. The opinion, however, appears to turn not on the question of "culpability" or "wrongdoing" but on the reasoning of Craven, regarding the distinction between the right to indemnity and the right of contribution. See the language quoted from Bristow in part v.

The lead opinion also cites Ritter v Technicolor Corp, 27 Cal App 3d 152; 103 Cal Rptr 686 (1972), Swanigan v State Farm Ins Co, 99 Wis 2d 179; 299 NW2d 234 (1980), and Van Cleave, supra.

In Ritter, the court declared: "By California statute, release of an agent before trial does not discharge his principal from tort liability, even though the sole basis alleged for recovery from the principal is his vicarious liability for the acts of his agent." Id., p 153.

In Swanigan, the court relied not on the Uniform Act but instead on a provision of a Wisconsin statute governing the liability of sponsors of minors on driver's license applications. The court specifically rejected defendant's characterization of the sponsor-minor relationship as that of master-servant and noted that even if it accepted the characterization, it "need not accept [defendant's] conclusion that a release of the agent which reserves rights against the principal operates as a matter of law to release the principal." Id., p 187, n 2.

While Van Cleave v Gamboni Const Co, supra, also noted that the Uniform Act does not impair any right of indemnity, it followed the line of cases opposed to Craven and characterized the master-servant relationship as one of joint tortfeasors.

[72] Dickey v Estate of Meier, 188 Neb 420; 197 NW2d 385 (1972), did not construe a statute, relying instead on the Restatement, Bacon v United States, 321 F2d 880 (CA 8, 1963), discussed below, and Max v Spaeth, 349 S.W.2d 1 (Mo, 1961). The continuing vitality of Max is questionable after the Missouri Supreme Court's decision in Aherron, supra, pp 500-501 ("Max... includes dicta that ``release of the master releases the servant.' We need not determine whether that dicta correctly states the law in Missouri in cases involving vicarious liability. The pertinent rule is provided by [the Missouri version of the Uniform Act]").

No statute was discussed by Jacobson v Parrill, 186 Kan 467; 351 P2d 194 (1960), Barsh v Mullins, 338 P2d 845 (Okla, 1959), Mid-Continent Pipeline Co v Crauthers, 267 P2d 568 (Okla, 1954), or Seaboard Air Line R Co v Coastal Distributing Co, 273 F Supp 340 (D SC, 1967).

Bacon v United States, supra, p 883, is distinguishable not only because the Missouri statute was not based on a uniform act but also because the statute spoke of "two or more joint tort-feasors or wrongdoers." (Emphasis supplied.) Unlike the 1955 Uniform Act and RJA, ž 2925d, which speak of two or more "persons liable in tort," the language of the Missouri statute may have appeared to offer more support for the view that a release discharged the vicariously liable principal who was not "culpable" in that he personally "did no wrong." In all events, following the decision of the Missouri Supreme Court in Aherron, Bacon like Max may no longer represent the law of Missouri.

Horejsi v Anderson, 353 NW2d 316 (ND, 1984), is distinguishable because the court relied on language peculiar to North Dakota's version of the Uniform Act.

[1] a release or a covenant not to sue or not to enforce judgment is given in good faith to 1 of 2 or more persons liable in tort for the same injury or the same wrongful death:

(a) It does not discharge any of the other tort-feasors from liability for the injury or wrongful death unless its terms so provide.

(b) It reduces the claim against the other tort-feasors to the extent of any amount stipulated by the release or the covenant or to the extent of the amount of the consideration paid for it, whichever amount is the greater.

(c) It discharges the tort-feasor to whom it is given from all liability for contribution to any other tort-feasor.

[2] The releases provide in pertinent part:

It is intended that this Release shall be operative as to Jana Palmer only and shall not be construed to release any other persons or entity who may be liable in tort for the same wrongful death. MCL 600.2925d; MSA 27A.2925(4).

* * *

It is intended and understood that this release shall be operative as to Jack Gilmore, D.O. only, and shall not be construed to release any other persons or entities who may be liable in tort for the alleged wrongful death of Gene Christopher Schneider, deceased, and any and all other damages alleged to result from the alleged wrongful death. MCL 600.2925d; MSA 27A.2925(4).

[3] See 66 Am Jur 2d, Release, ž 1.

[4] See anno: Release of, or covenant not to sue, one primarily liable for tort, but expressly reserving rights against one secondarily liable, as bar to recovery against latter, 24 ALR4th 547, 551; anno: Release of one joint tortfeasor as discharging liability of others: Modern trends, 73 ALR2d 403, 407-408.

[5] A covenant not to sue, as opposed to a release, is a collateral undertaking which indicates the parties' specific intention not to discharge the remaining joint tortfeasors, without expressly reserving rights against the joint tortfeasors. See Robinson v Godfrey, 2 Mich. 408 (1852); Morgan v Butterfield, 3 Mich. 615 (1855); Cook v City Transport Corp, 272 Mich. 91; 261 N.W. 257 (1935); Boucher v Thomsen, 328 Mich. 312; 43 NW2d 866 (1950); annotations, n 4 supra; anno: Release of (or covenant not to sue) master or principal as affecting liability of servant or agent for tort, or vice versa, 92 ALR2d 533, 538.

[6] In 1939, the National Conference of Commissioners on Uniform State Laws approved the Uniform Contribution Among Tortfeasors Act, 12 ULA 63, which provided in ž 4 that a release given to one of several persons liable in tort for the same injury does not discharge any other tortfeasor unless its terms so provide. A 1955 revision of the act extended ž 4 to covenants not to sue. It states, in pertinent part:

When a release or covenant not to sue or not to enforce judgment is given in good faith to one of two or more persons liable in tort for the same injury or the same wrongful death:

(a) It does not discharge any of the other tortfeasors from liability for the injury or wrongful death unless its terms so provide; but it reduces the claim against the others to the extent of any amount stipulated by the release of the covenant, or in the amount of the consideration paid for it, whichever is the greater...."

The uniform act has been adopted in at least sixteen jurisdictions. See anno: Uniform contribution among tortfeasors act, 34 ALR2d 1107, 24 ALR4th 547, 551, 565-567, and 73 ALR2d 403, 434-435.

The Uniform Joint Obligations Act, (which preceded the Uniform Contribution Among Tortfeasors Act) and which has been adopted by a minority of states, provides that the release of one joint obligor, which includes a person liable in tort, shall not release all unless it contains a reservation of right to the contrary. See anno: 73 ALR2d 432, n 4 supra. See, also, anno: Federal Tort Claims Act: Judgment against or settlement with negligent employee as releasing United States, or vice versa, 42 ALR2d 960.

[7] See n 4.

[8] Other devices abrogating the common-law rule include the Restatement Torts, 2d, ž 885(1) and (2), which provides that releases and covenants not to sue do not discharge any other tortfeasor liable for the same harm. See 24 ALR4th 547, 551, and Restatement Judgments, 2d, ž 51, which provides that a settlement with an agent generally does not discharge the liability of the principal.

[9] See 1961 PA 236, MCL 600.2925; MSA 27A.2925. See also Duncan v Beres, 15 Mich. App. 318, 328, n 18; 166 NW2d 678 (1968) (opinion of LEVIN, J.).

[10] See, also, Mayhew v Berrien Co Road Comm, 414 Mich. 399, 410; 326 NW2d 366 (1982).

[11] See Alaska Airlines, Inc v Sweat, 568 P2d 916 (Alas, 1977); Ritter v Technicolor Corp, 27 Cal App 3d 152; 103 Cal Rptr 686 (1972); Clark v Brooks, 377 A2d 365 (Del Super, 1977); Blackshear v Clark, 391 A2d 747 (Del, 1978); Van Cleave v Gamboni Const Co, 101 Nev 524; 706 P2d 845 (1985); Aherron v St John's Mercy Medical Center, 713 S.W.2d 498 (Mo, 1986); Waters v Hedberg, 126 NH 546; 496 A2d 333 (1985).

[12] Cf. Ritter v Technicolor Corp, n 11 supra.

[13] While I would not rule on that issue today, I note that recent Court of Appeals decisions have held that a party seeking commonlaw indemnity must be free from any active negligence. Pontious v E W Bliss Co, 102 Mich. App. 718, 721; 302 NW2d 293 (1981); Hill v Sullivan Equipment Co, 86 Mich. App. 693, 696-697; 273 NW2d 527 (1978); Palomba v East Detroit, 112 Mich. App. 209, 215-216; 315 NW2d 898 (1982); Johnson v Bundy, 129 Mich. App. 393, 399; 342 NW2d 567 (1983); Reed v St Clair Rubber Co, 118 Mich. App. 1, 10; 324 NW2d 512 (1982); Skinner v D-M-E Corp, 124 Mich. App. 580; 335 NW2d 90 (1983); McIntyre's Mini Computer Sales Group, Inc v Creative Synergy Corp, 644 F Supp 580 (ED Mich, 1986).

[14] Lansing General Hospital admits in its brief that no collusion occurred in the settlements involved herein.