520 So.2d 715 (1988)
Jessie HOWARD, Individually and as Administrator of the Estate of His Minor Child, Jessie J. Howard, and Constance Mae Howard, Individually and as Administratrix and Legal Custodial Parent of Her Minor Children, Tina Louise Holloway and Sherry Lynn Holloway
v.
ALLSTATE INSURANCE COMPANY, Chris Barcelona, and His Spouse, Peggy Barcelona, Individually and as Administrators of the Estate of Their Minor Child, Christi Barcelona.
Nos. 87-C-2117, 87-C-2145.
Supreme Court of Louisiana.
February 29, 1988.
Glenn Diaz, Chalmette, for applicant.
Christopher Lawler, Arthur O'Keefe, Donovan & Lawler, Metairie, for respondent.
MARCUS, Justice.
This lawsuit involves an incident in which Tina Louise Holloway, an eleven-year-old girl, was attacked and bitten by a dog belonging to Chris and Peggy Barcelona while she was playing with the Barcelonas' daughter, Christi, in their yard. The jury found in favor of the plaintiffs, assessing total damages of $28,000 but finding Tina 50% at fault, thus reducing her award to $14,000. On plaintiffs' motion for judgment notwithstanding the verdict or for a new trial, the trial judge increased the damages to $32,000 and eliminated the jury's finding that Tina was at fault and the proportionate reduction of her recovery. The court of appeal reversed the judgment notwithstanding the verdict and reinstated the jury's finding of comparative fault on the part of Tina (50%) and reduced the award from $32,000 to $25,300, with one judge dissenting.[1] Upon applications of plaintiffs and defendants, we granted certiorari to review the correctness of the judgment of the court of appeal.[2]
The record reflects that Chris and Peggy Barcelona and their daughter Christi lived in a mobile home in rural St. Bernard Parish. They owned a German shepherd dog, Nick, which they kept in their yard as a watchdog. The Barcelona's yard was enclosed by a wooden fence. The enclosed yard consisted of two areas: a small first yard adjoining the porch of the home and a larger second yard where Nick stayed during the day. The two areas were separated by a latched gate. Thus, Nick was kept in an area which is beyond two gates: the first allowing entry to the first yard from the parking area and the second allowing entry from the first yard to the second yard. A "Beware of Dog" sign was posted on the fence beside the gate which opened into the first yard. At this time, there was no warning sign posted on the fence beside the gate which opened into the second yard. The second yard contained a pen into which Mr. or Mrs. Barcelona customarily placed Nick when they had company.
On May 19, 1983, Tina Holloway, Christi Barcelona, and Tina's younger sister and brother were playing in a vacant lot adjacent to the Barcelona yard. Tina, eleven years old at the time, was the oldest child in the group. Her family had moved into a house across the road from the Barcelona residence approximately a month before this incident. However, May 19 was the first day that Christi and Tina had played together. On that day, after the children became tired of playing in the lot, Christi Barcelona, who was then seven years old, invited the other children to play on her swing set which was in the second yard where Nick was kept. At this point, there is contradiction among the three children who testified. Christi said she told the other children to wait while she went inside the house to get one of her parents to put Nick in his pen. Tina and her sister Sherrie deny being told to wait. They testified that Christi told them that it was all right to go into the yard because Nick would not bite. Whatever the instruction given Tina, she preceded the younger children into the yard where she was immediately attacked by Nick, whom she had not seen before he attacked her. There was some conflict in the evidence as to whether Tina was in the first yard when she was attacked and the dog jumped the fence or whether she opened the second gate and went into the second yard. Most of the witnesses, however, placed her well inside the second yard.
Tina sustained several serious lacerations on the side of her face. The healing process for Tina was slow and painful, and disfiguring scars developed. Dr. Onyx P. Garner, Jr., the plastic surgeon who treated Tina, testified that he recommends surgery and other procedures which will partially remove the scars. Apparently, the dog bite incident exacerbated Tina's pre-existing anxiety and shyness. Dr. Garner supported his evaluation of Tina's psychological condition with the report of a consulting psychiatrist whom he had asked to examine her. The report stated that she appeared two to three years younger than her age and that the incident had a depressed and anxious affect on her.
The Barcelonas' liability in this case arises primarily from La.Civ.Code art. 2321 which provides in pertinent part that the "owner of an animal is answerable for the damage he has caused; ..." This article was interpreted by this court in Holland v. Buckley, 305 So.2d 113, 119 (La.1974):
When a domesticated animal harms another, the master of the animal is presumed to be at fault. The fault so provided is in the nature of strict [footnote omitted] liability, as an exception to or in addition to any ground of recovery on the basis of negligence, Article 2316. The owner may exculpate himself from such presumed fault only by showing that the harm was caused by the fault of the victim, by the fault of a third person for whom he is not responsible, or by a fortuitous event.
Thus, the injured person need not prove the negligence of the owner. Liability arises solely from the legal relationship between the owner and the animal. Rozell v. Louisiana Animal Breeders Cooperative, Inc., 434 So.2d 404 (La.1983).
In this case, the jury, the trial judge and the court of appeal found the defendants liable under art. 2321 but applied comparative fault which the legislature established by rewriting La.Civ.Code art. 2323, effective August 1, 1980, to provide:
When contributory negligence is applicable to a claim for damages, its effect shall be as follows: If a person suffers injury, death or loss as the result partly of his own negligence and partly as a result of the fault of another person or persons, the claim for damages shall not thereby be defeated, but the amount of damages recoverable shall be reduced in proportion to the degree or percentage of negligence attributable to the person suffering the injury, death or loss.
As stated before, the jury found that Tina was guilty of substandard conduct or negligence which contributed to her own injury and assigned 50% of fault to her. The trial judge then granted a judgment notwithstanding the verdict and stated that he was "unable to understand how twelve reasonable persons could conclude that the 11 year old plaintiff, with a report of mental ability of lesser years ... in any way contributed to her own injury." The court of appeal, after holding that the concept of comparative fault would apply to cases arising under La.Civ.Code art. 2321, held that the trial judge erred in entering the judgment notwithstanding the verdict and reinstated the jury's verdict finding Tina 50% at fault.
We first review the court of appeal's holding that comparative fault should apply to this strict liability case and result in a reduction of plaintiffs' recovery.
Before the legislature rewrote La.Civ. Code art. 2323 establishing comparative fault, the only defenses to cases arising under art. 2321 were (1) fault of the victim; (2) fault of a third person; (3) irresistible force. Rozell v. Louisiana Animal Breeders Cooperative, Inc., 496 So.2d 275 (La. 1986). In Rozell, a case arising under art. 2321 prior to the enactment of the comparative fault statute, this court held the defense of victim fault was not available because the victim of a charging bull had not assumed the risk of his injuries. This court held that for victim fault to be a legal defense, it must be "at least a substantial cause." Thus, in cases like Rozell, arising before the institution of comparative fault, courts must consider whether the plaintiff's conduct amounted to a substantial cause.
Tina was attacked and bitten by the Barcelonas' dog in 1983, well after the effective date of the legislature's establishment of comparative fault under art. 2323. Thus, if we find that the concept of comparative fault should apply to cases arising under art. 2321, comparative fault will apply to Tina's case.
Article 2323 does not apply to the determination of liability; it applies only to the apportionment of damages. The article states that it is to be utilized only in cases in which "contributory negligence is applicable to a claim for damages," but the article does not designate the cases in which courts can or cannot allow a defense of contributory negligence. Nor does the article define fault of another person, or limit the concept to negligent actions. Hence, courts have discretion in determining when damages are to be apportioned under the statute. This discretion allows the application of comparative fault to a claim previously insusceptible to the defense of contributory negligence, such as strict products liability or cases arising from art. 2317. See Bell v. Jet Wheel Blast, Division of Ervin Industries, 462 So.2d 166 (La.1985); Landry v. State, 495 So.2d 1284 (La.1986).
This court has already begun to apply comparative fault in strict liability cases. In Bell v. Jet Wheel Blast, we decided that comparative fault may be applied in some strict products liability situations:
On the other hand, pure comparative fault principles would seem to coincide with and further the goals of products liability doctrine in some cases. Where the threat of a reduction in recovery will provide consumers with an incentive to use a product carefully, without exacting an inordinate sacrifice of other interests, comparative principles should be applied for the sake of accident prevention.
462 So.2d at 171. In the recent case of Landry v. State, this court applied the doctrine of comparative fault to a case arising under art. 2317 and compared the fault of the strictly liable Orleans Levee Board with that of the plaintiff who fell and was injured because he failed to see a hole next to the Lake Pontchartrain sea wall.
In view of the foregoing statutory and case law, we hold that comparative fault applies in cases such as this where a domesticated animal inflicts injuries for which its owner is held liable under art. 2321. It seems only fair that the damages recovered by negligent victims in dog bite cases should be reduced by their percentage of fault.
However, there is a problem with applying comparative fault to a strict liability case. Strict liability is based on a theory of responsibility which requires no finding of negligence. The "fault" of the defendant does not involve blameworthiness or culpability. The theory of strict liability does not lend itself to a comparison of culpability. If a defendant liable under a strict liability rule can mitigate damages by showing the plaintiff's negligence, then the plaintiff will be forced to show the defendant's culpability in order to lessen by comparison his own. This inappropriately reintroduces the element of negligence into a strict liability action. There is also the conceptual difficulty of comparing the two types of legal fault. We agree with the court of appeal that the most satisfactory result to this problem seems to be the principle of comparative causation. Under this principle, the factfinder compares the causal effect of the plaintiff's conduct with that of the defendant's nonnegligent fault. This comparison is supported by this court's citation in Watson v. State Farm Fire and Casualty Insurance Co., 469 So. 2d 967 (La.1985), to the Uniform Comparative Fault Act, Section 2(b), which provides:
In determining the percentages of fault, the trier of fact shall consider both the nature of the conduct of each party at fault and the extent of the causal relation between the conduct and the damages claimed.
Thus, the extent to which each party contributed to the damages should be the measure by which the loss is apportioned.
Having found that the concept of comparative fault will apply to this case, it must now be determined whether the court of appeal was correct in reinstating the jury's finding that Tina was fifty percent at fault for the dog's attack. As we held in Watson, 469 So.2d at 972, a finding of fact by a trial court should be upheld "unless it is clearly wrong.... Proper review requires that the court of appeal determine from the record that the trial court finding is not clearly wrong or manifestly erroneous."
At the time of the dog attack, Tina was eleven years old with a report of a mental ability of two or three years younger than her age. In Dufrene v. Dixie Auto Ins. Co., 373 So.2d 162, 165, n. 4, (La.1979), a case decided before the legislature established comparative fault, the court stated that "contributory negligence of a child is not measured by the standard of self-care expected of an adult but rather only by the self-care expected of a child of his age, intelligence and experience under the particular circumstances presented to him." Thus, in applying the concept of comparative fault to Tina, the self-care expected of her will be that of a child with a mental ability of eight or nine years old.
According to the record, Tina had been invited by Christi, the dog owner, to play on the swingset in the second yard where Nick was kept. Tina testified that she had not seen Nick before the attack. There was a "Beware of Dog" sign on the first gate that Tina admitted seeing. However, there was no sign on the second gate at this time. Whether or not Christi instructed Tina to wait before going through the second gate while Mrs. Barcelona put Nick in his pen, we fail to see how under these circumstances an eleven-year old girl, with a report of a mental ability of two to three years younger than her age, could be considered 50% at fault for the subsequent attack in the second yard. Hence, the jury was clearly wrong in deciding that Tina was 50% at fault. Thus, the court of appeal erred in reinstating this jury finding.
Having reached this conclusion, we are empowered by La.Code Civ.P. art. 2164 to "render any judgment which is just, legal and proper upon the record on appeal." Watson, 469 So.2d at 973. Because Tina saw the "Beware of Dog" sign on the first gate and may have been told by Christi to wait while the dog was put in its pen, we find her slightly at fault and assign her 10% of the fault for this incident. To the Barcelonas, we therefore assign 90% of the fault.
Next, we consider the question of quantum. The jury awarded damages of $28,000 which the trial judge increased to $32,000. The court of appeal then lowered the damages to $25,300. Upon review of the record, we find no abuse of discretion in the jury's award. Therefore, both the trial judge and the court of appeal erred in changing the jury award.
In sum, we find the court of appeal erred in finding Tina 50% at fault and in lowering the jury's assessment of damages to $25,300. We must reverse that portion of the judgment. We find that damages should be $28,000 and that Tina should be assessed only 10% at fault. In all other respects, the judgment of the court of appeal will be affirmed.
DECREE
The judgment of the court of appeal is reversed insofar as it found Tina Louise Holloway 50% at fault and reduced the damages to $25,300. Tina Louise Holloway is assessed 10% at fault and damages are fixed at $28,000. Accordingly, plaintiffs are entitled to recover $25,200 in damages. Otherwise, the court of appeal judgment is affirmed.
DENNIS, J., concurs in part and dissents in part for reasons assigned.
DENNIS, Justice, concurring in part and dissenting in part.
I respectfully dissent from the majority's decision that the plaintiff's recovery should be reduced. I agree that the meliorating principle of Baumgartner, which was properly invoked under the regime of contributory negligence, should not carry over automatically to the comparative negligence regime. See Turner v. NOPSI, 476 So.2d 800 (La.1985). However, when the plaintiff is in a class of persons who cannot protect itself from the risk in question, and who inflicts no substantial risks upon others by his contributory fault, the defendant who either knows or has reasonable ground to foresee the plaintiff's inability to protect himself comes under a duty to protect the plaintiff from his own fault. D. Dobbs, Accountability and Comparative Fault, 47 La.L.Rev. 939, at 962 (1987); A. Johnson, Comparative Negligence and the Duty/Risk Analysis, 40 La.L.Rev. 319, 333 (1980); Turner v. NOPSI, supra, and concurring opinion; Bell v. Jet Wheel Blast Indus., 462 So.2d 166, 171 (1985); Mercer v. Fruehauf Corp., 492 So.2d 538, 542-543 (La.App. 3d Cir.1986); Argus v. Scheppegrell, 472 So.2d 573 (La.1985); Travelers Ins. Co. v. Southwestern Transport, 488 So.2d 978 (La.1986). See also Bexiga v. Havir Mfg. Co., 60 N.J. 402, 290 A.2d 281 (1972). In the present case, Tina was of mentally diminished capacity and limited ability to care for herself, her conduct inflicted no risk upon others, and the defendant had reasonable ground to foresee that such a child, while visiting his children, would be exposed to the risk of injury by his dangerous dog. Under these circumstances, the plaintiff should be entitled to a full, rather than reduced recovery. Further, in my opinion the threat of reduction of recovery will not provide young children such as Tina with an incentive to exercise greater care around their friends' domestic animals, but it will drastically reduce the dangerous animal owner's incentive to protect children from harm. See Bell v. Jet Wheel Blast, supra at 171; Mercer v. Fruehauf Corp., supra at 542-543; See also Argus v. Scheppegrell, 472 So.2d 573 (La. 1985); Travelers Ins. Co. v. Southwestern Transport, 488 So.2d 978 (La.1986). See cases cited at Prosser & Keeton on Torts, § 67, n. 93 (5th ed.).
Although I would not reduce the plaintiff's recovery in this case, I believe that more than causation must be compared in order to intelligently apportion responsibility in those cases in which plaintiff's negligence may properly be considered. Causation in fact is an absolute concept, which by its nature is incapable of being divided into comparative degreesit either exists or it does not. Prosser, § 67 p. 474. Therefore, the trier of fact must consider and quantify not only causation but also policy factors pertaining to the risk created by each party in rationally comparing the fault or negligence of each. I recognize that there has been much debate over the exact form this process should follow, R. Pearson, Apportionment of Loss, 40 La.L.Rev. 341; Prosser, Comparative Negligence 1953, 51 Mich. L.Rev. 465, 481 (negligence); Twerski, From Defect to Cause to Comparative Fault, 1977, 60 Marq.L.Rev. 297, 326 (causation). See generally Borgo, Causal Paradigms in Tort Law, 1979, 8 J. Legal Stud. 419, 448-52; Calabresi, Concerning Cause and the Law of Torts: An Essay for Harry Kalven, Jr., 1975, 43 U.Chi.L.Rev. 69, 71-73; Epstein, Defenses and Subsequent Pleas in a System of Strict Liability, 1974, 3 J. Legal Stud. 165, 179-80, and that there has been too little experience and analysis to date to assert with confidence which approach is best. Prosser & Keeton, supra, § 67 p. 474 (5th ed.). However, the best method would appear to be one which in some way involved both a consideration of causation and the magnitude of unreasonableness of the risk created by each party that contributed to the accident. See Barham, The Viability of Compartive Negligence as a Defense to Strict Liability in Louisiana, 44 L.La.Rev. 1171, 1190 (1984).
I respectfully concur in the rest of the majority opinion, and I heartily agree that, if the plaintiff's recovery is to be diminished, a reduction of only 10% is much more just than one of 50%.
NOTES
[1] 510 So.2d 685 (La.App. 4th Cir.1987).
[2] Both applications were granted. 514 So.2d 1166 (La.1987); 514 So.2d 1167 (La.1987). Subsequently, they were consolidated for oral argument before this court.