McCORMICK, Justice.
The question here is whether the common law of Iowa should be changed to substitute the doctrine of comparative negligence for the existing doctrine of contributory negligence. In the present case the trial court [744]*744submitted a contributory negligence defense over an objection by plaintiff in which she advocated the change. We now abandon the doctrine that contributory negligence is a complete bar to recovery. In its place we adopt the doctrine of comparative negligence under which an injured party’s recovery is diminished in proportion to that party’s contributory negligence, and recovery is not barred unless the injured party’s negligence is the sole proximate cause of the damages. We reverse and remand this case.
This is a medical malpractice action. The facts are relevant only in fixing the context for consideration of the legal question. Plaintiff Mary E. Goetzman alleged in her petition that defendant Homer E. Wichern, M.D., was negligent in the diagnosis and treatment of her breast cancer over a period from 1974 to 1978. She alleged his negligence made it necessary for her to have a mastectomy and to undergo radiation therapy and chemotherapy, proximately causing substantial damages for which she asked judgment. Defendant admitted plaintiff was his patient during the period involved but' denied the other material allegations of the petition. Shortly before trial he amended his answer to allege generally that plaintiff was barred from recovery because she was guilty of negligence which was a proximate cause of her injuries and damages. At the beginning of trial, defendant amended his answer to specify that plaintiff was negligent in failing to follow his advice and cooperate with his diagnostic and treatment recommendations.
The case was submitted on plaintiff’s claim and defendant’s defense, and the jury returned a verdict for defendant upon which judgment was entered. This appeal followed. The only question presented is whether Iowa should now abandon the defense of contributory negligence as an absolute bar to recovery in favor of a system of comparative negligence. Plaintiff contends the change should be made as a development of the common law and on constitutional grounds.
No dispute exists that the evidence at trial was sufficient for jury submission of the allegations of the petition and the contributory negligence defense. Controversy does exist, however, concerning whether plaintiff preserved error on her present contentions, whether she was prejudiced by the court’s instruction on the contributory negligence defense, whether the doctrine of contributory negligence is statutory in Iowa, whether any change in the law should be made by the legislature, and whether the contributory negligence doctrine is constitutional. We address all of these issues except the constitutional one. In addition, because we adopt the principle of comparative negligence, we decide the form of the system and the cases to which it is applicable.
I. Preservation of error. In its instruction stating the issues, the court characterized the contributory negligence defense by saying:
The defendant doctor asserts additionally the affirmative defense that the plaintiff herself was negligent and that such negligence was a proximate cause of any injury or damage sustained by her and thus under law she may not recover.
The plaintiff denies this claim. The defendant doctor has the burden of proof.
In its instruction on the defense, the court said:
If you find the plaintiff has proved her case against the defendant doctor under the prior instruction you should consider the defendant’s affirmative defense.
The law of Iowa provides that in such a situation if the plaintiff was also negligent and if such negligence of the plaintiff was also a proximate cause of any injury or damage, then the plaintiff may not recover.
The defendant doctor relies on this law in this case. The defendant has the burden of proof.
To establish it the defendant must prove each and all of the following propositions:
A. That the plaintiff was herself negligent. The defendant claims that the [745]*745plaintiff was negligent in failing to undergo biopsy and possible further surgery at the time recommended by the defendant doctor.
B. That such negligence was a proximate cause of any injury and damage to the plaintiff.
If the defendant doctor has proved both the above propositions then your verdict should be for him and against the plaintiff as to all, or that part of any damage suffered by the plaintiff, which in whole or in part was proximately caused by the plaintiff’s own negligence.
If the defendant doctor has failed to prove both the above propositions then this affirmative defense has not been established.
Plaintiff’s counsel made a timely objection to the relevant paragraph of the statement of issues by alleging “the defense of contributory negligence is not a proper defense in that the proper defense would be comparative negligence .. . . ” He proceeded to add a detailed objection based on the alleged unconstitutionality of the contributory negligence defense, alleging that equal protection and due process are denied by making comparative negligence applicable in common carrier and employer-employee cases but not in other cases such as the one on trial. He objected to the instruction on contributory negligence on the same grounds “as to the constitutional questions and as to the misstatement of the law.”
In determining the sufficiency of an objection to preserve error, “the test is whether the exception taken alerted the trial court to the error which is urged on appeal.” Dutcher v. Lewis, 221 N.W.2d 755, 759 (Iowa 1974). The purpose is “to afford the trial judge an opportunity to catch exactly what is in counsel’s mind and thereby determine whether the objection possesses merit to an extent the instruction should be recast.” State v. Baskin, 220 N.W.2d 882, 886 (Iowa 1974). The issues presented by plaintiff on appeal are the same as those presented in her trial court objection. She contends the doctrine of comparative negligence should be substituted for the absolute defense of contributory negligence, and she separately repeats her constitutional challenge. Any doubt that plaintiff was advocating a change in the law in the trial court was removed by the joinder of her common law ground with her constitutional ground.
Moreover, the objection was not made in a vacuum. The question whether Iowa should adopt the doctrine of comparative negligence was addressed in two cases decided by this court in the two years before the present case was tried. See Fuller v. Buhrow, 292 N.W.2d 672 (Iowa 1980); Stewart v. Madison, 278 N.W.2d 284 (Iowa 1979). Because the question was not settled by those decisions, it is reasonable to believe the bench and bar were aware the question was still viable when the present case was tried. In these circumstances it is unlikely a trial court would not know what a lawyer meant in advocating substitution of comparative negligence for contributory negligence as a statement of the “proper” law.
Defendant argues the objection was insufficient because it did not point out a specific deficiency in the language of the court’s instructions and because plaintiff did not offer a substitute instruction setting forth plaintiff’s theory of comparative negligence. Plaintiff’s attack, however, was not on particular language in the court’s instructions but on the principle of contributory negligence incorporated in them. The attack was thus on all of the language outlining the contributory negligence defense in the two instructions involved. When an instruction concerns a feature of the case on which the court has a duty to instruct without a request, it is sufficient to allege the ground of error without offering a substitute instruction. It is only “when the issue is not integral to the case or when an instruction on an integral issue is correct as proposed but is not as explicit as a party desires” that the objecting party must offer an instruction. State v. Templeton, 258 N.W.2d 380, 382 (Iowa 1977). The contributory negligence defense was clearly integral to the present case.
[746]*746Furthermore, plaintiff’s failure to offer a substitute instruction should not have been a factor in the court’s ruling on the objection. The question raised by the objection was whether contributory negligence should be abandoned in favor of some form of comparative negligence. It did not raise the next question which is what form the new system should take. The objection was obviously made with little realistic hope that the trial court would refuse to follow this court’s precedents dictating an adverse response on the first question. See Kersten Co. v. Department of Social Services, 207 N.W.2d 117, 121-22 (Iowa 1973). Because the trial court gave a negative answer to the first question, it had no occasion to be concerned about plaintiff’s view on the second question. Plaintiff’s failure to suggest a specific answer to the second question thus did not affect the sufficiency of her objection to raise the first one.
We hold that plaintiff’s objection was sufficient to preserve the issue for review.
II. Existence of prejudice. Defendant contends plaintiff was not prejudiced by the court’s ruling on her objections to the instructions because the instruction on the contributory negligence defense was actually a comparative negligence instruction. Thus, according to defendant, plaintiff actually received what she was asking for. In urging this contention, defendant relies on the sentence of the instruction informing the jury that if defendant proved plaintiff was negligent and that her negligence was a proximate cause of her injuries its verdict was “to be for him and against the plaintiff as to all, or that part of any damages suffered by the plaintiff, which in whole or in part was caused by the plaintiff's own negligence.”
The instruction differs from the uniform instruction on contributory negligence, which defendant had requested of the court, only in that sentence. See Iowa State Bar Association, Uniform Jury Instructions (Civil) No. 2.2B. In objecting to this language variance, defendant alleged it improperly mingled the concepts of contributory negligence and proximate cause and had the effect of converting the instruction to a comparative negligence instruction. In overruling the objection, the trial court said:
The part he wishes to have stricken serves the purpose of delineating the possible range of finding by the jury — specifically that situation [where it is] claimed by the plaintiff the doctor was [negligent] from 1976 onward and not negligent merely from the two times in 1978 when the doctor testified he told the plaintiff she ought to have the biopsy, and she refused — according to the testimony.
It is possible they will accept the doctor’s testimony on that, in which case any injury or damage to the plaintiff after those two times in 1978 would not be attributable to the doctor and that there wouldn’t be any proximate cause as to claims for damages.
The claim of negligence before 1978 and back to 1976 [is] supported by the testimony of the doctor from Chicago [who said defendant] should have assessed or diagnosed the plaintiff’s true situation and recommended a biopsy at that time.
So that in sum the jury could accept the doctor’s testimony that in 1978 he told her she should have a biopsy and she refused and [he] could be free from that point on, but they could find that from mid-1976 to those dates in 1978 that there was a case made out against the doctor of negligence and proximate cause.
So that part of the next to last paragraph of number seven is intended to, and I believe it does, take into account that possible range of finding by the jury.
In additional discussion the court repeated that the purpose of the language was merely to permit the jury to consider the contributory negligence defense separately as to events before and after defendant’s alleged 1978 advice to plaintiff concerning a biopsy.
Thus the trial court did not intend the challenged language to authorize a finding for plaintiff based on comparative negli-[747]*747genee, and, with the court’s position having been made clear in the colloquy with defense counsel, we have no basis to believe comparative negligence was argued to the jury. Indeed, plaintiff having objected to the instruction because it was not a comparative negligence instruction and the court having confirmed it was not, we believe defendant’s assertion that it was clear to plaintiff and the trial court that the instruction was a comparative negligence instruction is untenable.
Defendant now also argues that the instruction did not prejudice plaintiff because it was ambiguous. He acknowledges that except for the disputed statement the instruction accurately stated the existing law making contributory negligence a bar to recovery. In its statement of issues, the court cháracterized contributory negligence as if it were a complete defense. In the second paragraph of the contributory negligence instruction the court told the jury: “The law of Iowa provides that [even if plaintiff proved her case against the doctor] if the plaintiff was also negligent and if such negligence of the plaintiff was also a proximate cause of any injury or damages, then the plaintiff may not recover.” Assuming defendant is correct that the instruction contained language inconsistent with this statement, the most that can be said is that the jury was given a correct statement and an incorrect statement of the existing contributory negligence law.
When a court gives a jury inconsistent instructions and the jury returns only a general verdict, it is impossible to tell which instruction the jury followed. See Childers v. McGee, 306 N.W.2d 778, 780 (Iowa 1981). In the present case, assuming the court’s instruction stated the law of contributory negligence as a complete defense in one place and the law of comparative negligence in another, it is impossible to say the jury did not apply the law of contributory negligence as a complete defense. The record thus does not support a finding that plaintiff was not prejudiced by the court’s ruling on her objection to submission of contributory negligence as a bar to recovery.
III. Effect of statutes. Defendant alleges that the common law doctrine of contributory negligence was codified by the legislature in enacting section 619.17 in 1965. See 1965 Iowa Acts ch. 430, § 1. In relevant part, section 619.17 provides:
In all actions brought in the courts of this state to recover damages of a defendant in which contributory negligence of the plaintiff, actual or imputed, was heretofore a complete defense or bar to recovery, the plaintiff shall not hereafter, have the burden of pleading and proving his freedom from contributory negligence, and if the defendant relies upon negligence of the plaintiff as a complete defense or bar to plaintiff’s recovery, the defendant shall have the burden of pleading and proving negligence of the plaintiff, if any, and that it was a proximate cause,of the injury or damage.
We believe this provision modified the common law doctrine of contributory negligence but did not codify it.
In Rusch v. City of Davenport, 6 Iowa 443 (1858), this court adopted the common law contributory negligence rule. The court said that in order to recover for injuries sustained in an accident caused by a defendant’s negligence the plaintiff had the burden to prove “that the accident happened without any want of reasonable care on his part.” Id. at 453. Subsequently it was held that an injured party was barred from recovery if guilty of negligence that “contributed in any way, or in any degree directly” to his injury. See, e.g., Hogan v. Nesbit, 216 Iowa 75, 80, 246 N.W. 270, 272 (1933).
In 1909 the doctrine' was abrogated by statute for railroad employees in actions against their employers, and the doctrine of comparative negligence was substituted. See 1909 Iowa Acts ch. 124, § 1. The present version of this statute is section 327D.188. The doctrine was also abrogated for employees covered by workers’ compensation in 1913. See 1913 Iowa Acts ch. 147, § 3. In 1915 the legislature enacted a statute applicable to employee-employer and [748]*748passenger-common carrier actions shifting the burden of pleading and proving contributory negligence to the defendant and making the rule one of comparative negligence. 1915 Acts ch. 277, § 1. The present form of this provision is in Iowa R.Civ.P. 97.
The 1965 statute was simply a further amelioration of the common law doctrine. It made a procedural change in shifting the burden to plead and prove contributory negligence to the defendant, and it made a substantive change in requiring contributory negligence to be established as a proximate cause to bar recovery. Schultz v. Gosselink, 260 Iowa 115, 148 N.W.2d 434 (1967). It did not define the doctrine or limit it in any other respect. It is plain from the amendment that it presupposed the existence and continuation of the common law doctrine except as modified.
This court’s subsequent decisions have continued the development of the common law doctrine. For example, the court held in Rosenau v. City of Estherville, 199 N.W.2d 125 (Iowa 1972), that assumption of risk could not be used as a separate defense with contributory negligence. It abrogated the no eyewitness rule in Froman v. Perrin, 213 N.W.2d 684 (Iowa 1973). In Handeland v. Brown, 216 N.W.2d 574 (Iowa 1974), the court held the contributory negligence defense is inapplicable to a parental claim for medical expense and loss of services, companionship and society under Iowa R.Civ.P. 8. The court held in Fuller v. Buhrow, the defense is unavailable to bar of a loss of consortium claim. Presumptions of incapacity of children for negligence or contributory negligence were rejected in Peterson v. Taylor, 316 N.W.2d 869 (Iowa 1982).
It would be strange if the legislature, by stepping in to modify the harshness of the common law defense in two respects, should be held to have preempted the field, freezing the development of the common law at that point, and disabling the courts from subsequently changing the doctrine in any other respect. We hold that section 619.17 modifies the contributory negligence defense in only the two respects noted in Schultz v. Gosselink. It does not otherwise affect the judicial authority and responsibility to develop the common law contributory negligence doctrine. A similar conclusion was reached in analogous circumstances in Scott v. Rizzo, 96 N.M. 682, 687, 634 P.2d 1234, 1239 (1981) (“[Ljegislative enactments designed to make the judge-made rule work or ameliorate its harshness cannot be taken as legislative integration of the rule into statutory law.”).
IV. Deference to the legislature. Defendant raises three arguments in support of a contention that the court should leave it to the General Assembly to decide whether the defense of contributory negligence should be abolished and, if so, what system should replace it. One is that the court previously indicated a willingness to defer the issues to the legislature and should continue to wait for legislative response. Second is that the change will raise collateral legal issues that the court can answer only on a case-by-case basis whereas the legislature could answer them in advance by enacting a comprehensive statute. Third is that the legislature is better equipped than the courts to make the choice among the alternative comparative negligence systems.
These arguments are substantial but not new, and they have been rejected elsewhere by courts that have adopted comparative negligence by judicial decision. They have also been rejected by commentators who are virtually unanimous in criticizing the contributory negligence defense and advocating judicial adoption of comparative negligence.
The first state to make the change by judicial decision was Florida in Hoffman v. Jones, 280 So.2d 431 (Fla.1973). The court noted that the contributory negligence doctrine was of judicial origin in Florida. It discussed the doctrine’s amelioration through subsequent judicial decisions establishing last clear chance and degrees of negligence. It recognized the duty of the judiciary to modify the common law as circumstances warrant and identified analogous situations in which the court had done so. It quoted with approval a statement [749]*749from a prior case that: “Legislative action could, of course, be taken, but we abdicate our own function, in a field peculiarly non-statutory, when we refuse to reconsider an old and unsatisfactory court-made rule.” Id. at 436.
The Florida court also said:
The contemporary conditions must be met with contemporary standards which are realistic and better calculated to obtain justice among all of the parties involved, based upon the circumstances applying between them at the time in question. The rule of contributory negligence as a complete bar to recovery was imported into the law by judges. Whatever may have been the historical justification for it, today it is almost universally regarded as unjust and inequitable to vest an entire accidental loss on one of the parties whose negligent conduct combined with the negligence of the other party to produce the loss. If fault is to remain the test of liability, then the doctrine of comparative negligence which involves apportionment of loss among those whose fault contributed to the occurrence is more consistent with liability based on a fault premise.
We are, therefore, of the opinion that we do have the power and authority to reexamine the position we have taken in regard to contributory negligence and to alter the rules we have adopted previously in light of current “social and economic customs” and modern “conceptions of right and justice.”
Id.
The court discussed the harshness of the “all or nothing” contributory negligence defense. It traced judicial and legislative efforts to ameliorate it. The court also responded to those who contended juries tend to ignore the defense in an effort to achieve justice. It pointed out the contradiction and subversion of principle involved in defending a system on the basis that juries can be trusted to disregard it. Id. at 437.
In deciding to make the change, the court noted that the legislature’s earlier failure to do so might be explained by the legislature’s belief that the problem is judicial. In answering arguments concerning a number of collateral issues raised by adopting comparative negligence the court held that those issues could be resolved in the context of concrete cases in the usual way. It observed that prior case law dealing with comparative negligence 'under an earlier railroad statute would provide guidance. Id. at 439.
The second state to adopt comparative negligence by judicial decision was California in Li v. Yellow Cab Co. of California, 13 Cal.3d 804, 532 P.2d 1226, 119 Cal.Rptr. 858 (1975). The California court was confronted with a statute that had codified the contributory negligence doctrine as it existed in 1872. In rejecting a separation of powers argument, the court examined the context and history of the enactment, concluding that the legislature’s purpose was merely to “formulate existing common law principles and definitions for purposes of orderly and concise presentation and with a distinct view toward continuing judicial evolution.” Id. at 814, 532 P.2d at 1233, 119 Cal.Rptr. at 865. The court rejected the argument that the change should be left to the legislature because of practical considerations involving unanswered collateral issues. It said: “The existence of the foregoing areas of difficulty and uncertainty ... has not diminished our conviction that the time for revision of the means for dealing with contributory fault in this state is long past due and that it lies within the province of this court to initiate the needed change by our decision in this case.” Id. at 826, 532 P.2d at 1241, 119 Cal.Rptr. at 873.
A similar conclusion was reached by the Alaska court in adopting comparative negligence for that state in Kaatz v. State, 540 P.2d 1037 (Alaska 1975). The court said: “Judicial administration of the rule has not presented insuperable difficulties in those jurisdictions which have long employed it. Experience has not borne out the argument that comparative negligence is difficult for courts and juries to apply.” Id. at 1048. The court also observed that, although in many instances the doctrine has been adopt[750]*750ed by legislative action, “increasingly it is perceived that a rule which is judicial in origin can be, and appropriately should be, altered by the institution which was its creator.” Id. at 1049.
The fifth state to adopt comparative negligence by judicial decision was Michigan in Placek v. City of Sterling Heights, 405 Mich. 638, 275 N.W.2d 511 (1979). The court recognized the authority of both the court and legislature to change the common law and reviewed instances in which it had previously recognized its prerogative. Then it recited, discussed and rejected three arguments against judicial adoption of comparative negligence.
, These arguments included the legislature’s alleged superior investigative capacity, ability to handle collateral problems, and power to give the bench and bar time to adjust to the change by passing a statute that will be effective at a future date. In responding to these arguments the court adopted an analysis of them from Fleming, Foreward: Comparative Negligence at Last—By Judicial Choice, 64 Calif.L.Rev. 239, 279-81 (1976). As to the first argument, Fleming asserted that the judiciary has had unique experience and preoccupation with the problem, access to the same quantitative data as the legislature, and an ability to pass independent and sensitive judgment on what he described as “preeminently lawyer’s law.” See 64 Calif.L.Rev. at 280. Concerning the second argument, Fleming pointed out that the legislature’s supposed superior ability to anticipate and resolve collateral issues in advance is not reflected in the many comparative negligence statutes that have been enacted. He noted that almost all of them are “in the briefest conceivable form,” leaving the additional questions for resolution by judicial decision. Id. at 281. He also observed that courts, if they choose, are equipped to adopt a comprehensive rule. Id. Addressing the third argument, Fleming referred to the authority of courts to determine the retroactive effect of common law changes. Id. The Michigan court also noted that a prior split decision by the court on the issue put the bench and bar on notice of possible future change. See 405 Mich. at 658, 275 N.W.2d at 518.
In the sixth state to make the change to comparative negligence by judicial decision, West Virginia, the court rejected the idea that a separation of powers problem existed and recognized its historic authority to alter or amend the common law. See Bradley v. Appalachian Power Co., 256 S.E.2d 879, 881 (W.Va.1979). Having recognized its authority, the court proceeded without hesitation to exercise it.
New Mexico, in Scott v. Rizzo, 96 N.M. 682, 634 P.2d 1234 (1981), and Illinois, in Alvis v. Ribar, 85 Ill.2d 1, 52 Ill.Dec. 23, 421 N.E.2d 886 (1981), became the seventh and eighth states to adopt comparative negligence through judicial decision.
The New Mexico court did so by endorsing an intermediate court opinion on the issue. In addressing the familiar separation of powers argument, the court referred to its duty to decide the New Mexico common law, pointed out the inconclusive significance of legislative inaction on the issue, and rejected the argument that the doctrine had been codified by statutes changing it to ameliorate its harshness. 96 N.M. at 687, 634 P.2d at 1239. The court also characterized the asserted future difficulties as “greater than the reality” and left adaptations of the rule to subsequent case-by-case adjudication. Id. at 687-88, 634 P.2d at 1239-40.
The Illinois court also rejected an argument for deference to the legislature. It noted the judicial authority to make such a change in the common law. See 85 Ill.2d at 21, 52 Ill.Dec. at 32, 421 N.E.2d at 895. In disagreeing that the legislature was better equipped to address the issue comprehensively the court said: “An examination of the States from which comparative negligence statutes have emerged reveals that such statutes are very general and brief and do not address collateral issues. Rather, the legislators apparently deemed it wise to leave the solution of collateral issues to the courts.” Id. at 22, 52 Ill.Dec. at 32, 421 N.E.2d at 895. The court also refused to [751]*751read determinative significance into legislative inaction, observing that it might reasonably be attributable to legislative deference to the judiciary. It also rejected an argument that the contributory negligence rule had been codified by statutes making it applicable under various liability statutes: “We believe that in enacting such statutes the legislature did not focus on the merits of the contributory negligence rule, but, rather, conformed the statutes to the then-existing law as announced by the court.” Id. at 23, 52 Ill.Dec. at 33, 421 N.E.2d at 896. The court concluded:
We believe that the proper relationship between the legislature and the court is one of cooperation and assistance in examining and changing the common law to conform with the ever-changing demands of the community. There are, however, times when there exists a mutual state of inaction in which the court awaits action by the legislature and the legislature awaits guidance from the court. Such a stalemate is a manifest injustice to the public. When such a stalemate exists and the legislature has, for whatever reason, failed to act to remedy a gap in the common law that results in injustice, it is the imperative duty of the court to repair that injustice and reform the law to be responsive to the demands of society.
Id. at 23-34, 52 Ill.Dec. at 33, 421 N.E.2d at 896.
The United States Supreme Court adopted comparative negligence for admiralty cases in United States v. Reliable Transfer Co., 421 U.S. 397, 95 S.Ct. 1708, 44 L.Ed.2d 251 (1975). In doing so the court found no merit in an argument that the task should have been entrusted to Congress rather than the Court. Id. at 409-11, 95 S.Ct. at 1714-15, 44 L.Ed.2d at 261-62.
Arguments urging judicial inaction have generally been criticized by commentators. See, e.g., Comments on Maki v. Frelk—Comparative v. Contributory Negligence: Should the Court or Legislature Decide?, 21 Vand.L.Rev. 889 (1968); Fleming, Foreward: Comparative Negligence at Last—By Judicial Choice, 64 Calif.L.Rev. 239, 273-82 (1976); Keeton, Creative Continuity in the Law of Torts, 75 Harv.L.Rev. 463, 506-09 (1962); Peck, The Role of the Courts and Legislatures in the Reform of Tort Law, 48 Minn.L.Rev. 265, 304-07 (1963).
A remark by Professor (now Judge) Robert E. Keeton in Comments has particular application to defendant’s arguments in the present case. He asserted that courts “are better situated than legislatures for developing a doctrine of comparative negligence gradually, resolving questions of detail as they are presented in the context of concrete cases, rather than attempting — always unsuccessfully — to anticipate and answer, at once every question that may arise in the future.” 21 Vand.L.Rev. at 915. Professor (former Judge) Robert A. Leflar contended that the history of comparative negligence statutes did not support the view that legislative change will establish a comprehensive comparative negligence system. Id. at 927-28. He concluded that, “When the socio-legal need for a change in judge-made law is evident, when the courts which made the law in the first place can do as good a job and possibly a better one than the legislature is likely to do in updating it, and when the legislature gives no sign of acting on the matter, it seems that judicial action is fully justified.” Id. at 929.
Although not expressly declared to be part of the law of Iowa by constitution or statute, the common law has always been recognized as in force in this state. See Iowa Civil Liberties Union v. Critelli, 244 N.W.2d 564, 568 (Iowa 1976). The authority of the court to alter or modify it has never been questioned: “To say that every principle of that law, however inapplicable to our wants and institutions, is to continue in force, until changed by some legislative rule, we believe has never been claimed, neither indeed could it be, with any degree of reason.” Wagner v. Bissell, 3 Iowa 396, 402 (1856). The common law may, of course, be changed by constitution or statute but also may be changed by judicial decision “where . . . varied by custom, not founded in reason, or not consonant to the genius and manners of the people .... ” [752]*752Id. at 403. It is “our duty to monitor and interpret the common law, and to abandon antiquated doctrines and concepts.” Bearbower v. Merry, 266 N.W.2d 128, 129 (Iowa 1979). We have said the genius of the common law “is its flexibility and capacity for growth and adaptation.” Handeland v. Brown, 216 N.W.2d 574, 577 (Iowa 1974).
Recent examples abound of this court’s responsiveness to its duty to develop and declare the common law, even when it means changing long-established doctrines. See, e.g., In re Marriage of Schneckloth, 320 N.W.2d 535 (Iowa 1982) (abolishing rule excluding spousal testimony of non access in a paternity case); Walker v. Clark Equipment Co., 320 N.W.2d 561 (Iowa 1982) (extending bystander claims for emotional distress to strict liability and warranty actions); Long v. McAllister, 319 N.W.2d 256 (Iowa 1982) (changing damages rules on loss of use in motor vehicle accident cases); National Steel Service Center, Inc. v. Gibbons, 319 N.W.2d 269 (Iowa 1982) (applying strict liability for ultrahazardous activity to common carriers); Weitl v. Moes, 311 N.W.2d 259 (Iowa 1981) (recognizing minor child’s independent cause of action for loss of consortium of injured parent); Fundermann v. Michelson, 304 N.W.2d 790 (Iowa 1981) (abolishing alienation of affections tort); Turner v. Turner, 304 N.W.2d 786 (Iowa 1981) (partial abrogation of parent-child immunity); Barnhill v. Davis, 300 N.W.2d 104 (Iowa 1981) (recognizing bystander tort for emotional distress).
Among the significant common law changes in earlier years were abrogation of governmental immunity in contract cases in Kersten Co. v. Department of Social Services, 207 N.W.2d 117 (Iowa 1973), adoption of a warranty of habitability in landlord-tenant cases in Mease v. Fox, 200 N.W.2d 791 (Iowa 1972), adoption of the doctrine of strict liability in tort in Hawkeye-Security Insurance Co. v. Ford Motor Co., 174 N.W.2d 672 (Iowa 1970), abrogation of the privity requirement in warranty cases in Hahn v. Ford Motor Co., 256 Iowa 27, 126 N.W.2d 350 (1964), establishment of a right of equitable contribution between negligent tortfeasors in Best v. Yerkes, 247 Iowa 800, 77 N.W.2d 23 (1956), and abrogation of charitable immunity in Haynes v. Presbyterian Hospital Association, 241 Iowa 1269, 45 N.W.2d 151 (1950).
We believe the arguments for deference to the legislature in the present case are substantially outweighed by the considerations reflected in the decisions of other courts that have addressed the issue, the analyses of the commentators, and the concept of the judicial role exemplified in past decisions of this court.
V. The merits of the proposed change. The main reason for changing the doctrine of contributory negligence as a complete bar to recovery to one of comparative negligence is fairness:
The predominant argument for its abandonment rests, of course, upon the undeniable inequity and injustice in casting an entire accidental loss upon a plaintiff whose negligence combined with another’s negligence in causing the loss suffered, no matter how trifling plaintiff’s negligence might be. Liability based on fault is the cornerstone of tort law, and a system such as contributory negligence which permits one of the contributing wrongdoers to avoid all liability simply does not serve any principle of fault liability.
Scott v. Rizzo, 96 N.M. 682, 689, 634 P.2d 1234, 1241 (1981).
In its pure form, the doctrine of comparative negligence assigns responsibility for damages in proportion to a party’s fault in proximately causing them. The rule thus operates to reduce rather than bar recovery. See Hoffman v. Jones, 280 So.2d 431, 438 (Fla.1973).
The main alternative comparative negligence systems apportion damages on the basis of fault either up to the point at which a plaintiff’s negligence equals defendant’s negligence or exceeds it. When that point is reached, plaintiff is barred from recovery. These systems thus lower, but do not eliminate, the bar of contributory negligence. Li v. Yellow Cab Co. of California, 13 Cal.3d 804, 828, 532 P.2d 1226, 1243, 119 Cal.Rptr. 858, 875 (1975).
[753]*753We have experienced more than 100 years of judicial and legislative efforts to ameliorate the harshness of the contributory negligence doctrine in Iowa. The doctrine of last clear chance, the no eyewitness rule, the rescue doctrine, and several rules against imputed negligence exemplify the judicial effort. See Willson, Contributory, Concurrent, or Comparative Negligence Under the New Iowa Statute?, 15 Drake L.Rev. 97, 99 (1966). The comparative negligence provision for two categories of cases in Iowa R.Civ.P. 97 represents a joint judicial and legislative effort. The legislature eliminated the doctrine altogether in workers’ compensation cases, provided for comparative negligence in railroad employee-employer cases in section 327D.188, and modified the doctrine in two respects in enacting section 619.17. As demonstrated in other jurisdictions, abrogation of the “all- or-nothing” rule of contributory negligence is a logical extension of this process. It is analogous in principle to abrogation of the rule against contribution between negligent joint tortfeasors in Best v. Yerkes.
Moreover, stare decisis does not preclude the change. That principle does not require blind imitation of the past or adherence to a rule merely because “it was laid down in the time of Henry IV.” See Kersten Co. v. Department of Social Services, 207 N.W.2d 117, 121 (Iowa 1973). We must reform common law doctrines that are unsound and unsuited to present conditions.
Most of the major civil and common law nations in the world, including England, now have comparative negligence systems. See Turk, Comparative Negligence on the March, 28 Chicago-Kent L.Rev. 189 (1950). Since the Supreme Court’s decision in United States v. Reliable Transfer Co., 421 U.S. 397, 95 S.Ct. 1708, 44 L.Ed.2d 251 (1975), all major maritime nations apply the principle in admiralty cases. Thirty-eight states in this country now have comparative negligence by statute or judicial decision. See Ohio Rev.Code Ann. § 2315.19 (1981) and Alvis v. Ribar, 85 Ill.2d 1, 52 Ill.Dec. 23, 421 N.E.2d 886 (1981). The subject has also been exhaustively analyzed and discussed in the cases and by the commentators. See Fuller v. Buhrow, 292 N.W.2d at 673-74. As a result, comparative negligence has a long history, wide application, and strong support. It is demonstrably superior in theory and practice to the absolute contributory negligence defense.
We are convinced that comparative negligence is a fairer system. It diminishes but does not defeat the right to. recover damages caused by another party’s fault. We are also convinced that the pure form of comparative negligence should be adopted in Iowa. It gives full rather than partial effect to the principle of comparative fault by reducing a person’s recovery based on another party’s fault by the percentage of the person’s own fault in the occurrence.
Of the eight other states that have adopted comparative negligence by judicial decision, only West Virginia adopted a modified version. See Bradley v. Appalachian Power Co., 256 S.E.2d 879 (W.Va.1979). In doing so, the West Virginia court asserted that the practical result of a pure comparative negligence system “is that it favors the party who has incurred the most damages regardless of his amount of fault or negligence.” 256 S.E.2d at 883.
Examples can be conceived of a seriously negligent plaintiff who is seriously injured suing a slightly negligent defendant who is slightly injured, and who counterclaims. The result may be that the seriously negligent plaintiff will achieve a substantial net recovery from the slightly negligent defendant. In responding to a contention that this result is unfair, the New Mexico court said:
How can it be argued that such a result would be unfair, when each party would be held responsible to the other person for harm caused to that other person by his proportionate fault? It surely is a fairer allocation of liability than the “modified” forms which require plaintiff to have been less negligent than or not more than equally as negligent as defendant. Those formulae punish either the plaintiff or counter-plaintiff who is but [754]*754slightly more negligent with bearing his own loss and about one-half of the losses of the other party as well.
Scott v. Rizzo, 96 N.M. at 689-90, 634 P.2d at 1241-42. We agree with this response. Under a system of comparative negligence, the keystone to fairness is proportionate responsibility for fault, not the relative severity of injuries. Each party’s recovery of damages is reduced in proportion to that party’s responsibility for them. As a result, no one is unjustly enriched.
Only the pure form of comparative negligence proportions a reduction of recovery to a person’s fault in all cases and prevents a party at fault from escaping liability in any case. As demonstrated in the reasoning of the courts that have adopted the pure form, it is the fairest, most logical and simplest to administer of all available systems. It is the system endorsed by most of the commentators. See, e.g., Comments on Maki v. Frelk—Comparative v. Contributory Negligence: Should the Court or the Legislature Decide?, 21 Vand.L.Rev. 889 (1968); Haugh, Comparative Negligence: A Reform Long Overdue, 49 Or.L.Rev. 38 (1969); James, Contributory Negligence, 62 Yale L.J. 691 (1953); Prosser, Comparative Negligence, 41 Calif.L.Rev. 1 (1953). It is also the form recommended by the National Commissioners on Uniform State Laws in the Uniform Comparative Fault Act (1977). The pure form has already been adopted for the categories of cases subject to rule 97 and section 327D.188. Moreover, it is consistent with the burden of proof and causation requirements of section 619.17. It is the system we adopt for Iowa.
We hold that in all cases in which contributory negligence has previously been a complete defense, it is supplanted by the doctrine of comparative negligence. In such cases contributory negligence will not bar recovery but shall reduce it in the proportion that the contributory negligence bears to the total negligence that proximately caused the damages.
Like most other courts that have adopted the comparative negligence doctrine, we do not decide in advance collateral issues which eventually may be raised. Those issues can best be addressed and resolved in the context of concrete cases. This is the traditional approach of courts to doctrinal change, and nothing in the experience of other jurisdictions indicates that a different approach is warranted when the comparative negligence doctrine is adopted. Our holding does not merely choose a rule. It delineates the rule, the category of cases to which it applies, and the only doctrine it necessarily displaces. We believe that is sufficient.
VI. Retroactivity. The other courts that have adopted comparative negligence have all made the doctrine effective to pending cases to some extent. Considerations affecting that determination are discussed in them. See, e.g., Placek v. City of Sterling Heights, 405 Mich. at 662-68, 275 N.W.2d at 520-22; see also Keeton, Creative Continuity in the Law of Torts, 75 Harv.L.Rev. 463, 506-09 (1962). We conclude that the doctrine shall apply to (1) the present case, (2) all cases tried or retried after the date of filing of this opinion, and (3) all pending cases, including appeals, in which the issue has been preserved.
REVERSED AND REMANDED.
All Justices concur except CARTER, Le-GRAND, HARRIS and McGIVERIN, JJ., who dissent.