dissenting in part, concurring in part.
I respectfully dissent, in part, from the majority opinion affirming the trial court judgment in this case for the reasons which follow.
Initially, I do not agree that the trial court properly found no fault on the part of either the driver of the vehicle or plaintiff, the passenger. With regard to the driver of the vehicle, the trial court found that he “inadvertently dropped off the edge of the roadway....” Defendant, the State of Louisiana, contends on appeal that “inadvertent” means “[n]ot attentive or observant ... [or] ... [d]ue to oversight ... ”; I agree. Allowing one’s vehicle to stray off the traveled portion of the highway because of inattentiveness merits a percentage of fault. Brooks v. City of Baton Rouge/Parish of East Baton Rouge, 558 So.2d 1177 (La.App. 1st Cir.), writ denied, 566 So.2d 982 (La.1990). I would assign 16% fault to Mr. Graham. Further, the passenger and plaintiff herein, Ms. Jenkins was not wearing her seatbelt at the time of the accident.1 The testimony established that Ms. Jenkins was thrown from the vehicle and it was clear her injuries resulted from being ejected from the vehicle. Mr. Graham was not thrown from the vehicle and testified to sustaining only minor injuries in the accident. The trial court refused to assign fault to plaintiff stating that the “[s]tate did not offer any evidence to minimize the plaintiff’s claim for damages.” I believe such reasoning is manifestly erroneous; there was simply no testimony of record which would lead one to conclude plaintiff would have been ejected from the vehicle regardless of whether she wore the safety belt. Therefore, I would therefore assign 2% fault, as limited by the statute in effect at the time of the accident, to Ms. Jenkins. See Demarest v. Progressive American Insurance Company, 552 So.2d 1329 (La.App. 5th Cir.1989).
With regard to additional allegations of fault on the part of the plaintiff in contributing to the accident, two versions of the circumstances immediately leading to the accident were given by Mr. Graham. Louisiana State Trooper Allen L. Coburn testified2 as follows:
I asked him what happened and he said that they had ran off the road in a curve. She had fell over on his lap ... and kind *1203of pinned the steering wheel where he couldn’t negotiate. And it caused him to run off the road, at which time when he did run off the road he had flipped over.
[[Image here]]
... When I arrived at the hospital, I talked to him in the Emergency Room. The nurses were in and out. And he recognized me at that point and called me by my first name. And we talked about the accident, and again he told me that — the same thing — that he had — was coming around the curve and she had fell over against him and — in his lap basically — and pinned the steering wheel and he couldn’t turn it and they ran off the road.
[[Image here]]
... [T]hat prior to him being with her that he had been home, or some location, and that he had received a phone call telling him that she was at a barroom at Springfield called Our Place, I believe is the name if I can remember correctly, and said whoever was on the phone had told him to come get her because she was extremely intoxicated.... And he had gone down there and — and gotten her and was coming back home with her.
[[Image here]]
... To the best of my knowledge, he mentioned at the scene that she had fell over on him, but he didn’t — at that point we didn’t discuss much more as to why, where they were coming from, or anything to that effect. That came later at the hospital when he was in the — when I was talking to him in depth about the accident.
Danny Graham’s version of the accident, in his depositions on October 18, 1988, and November 27, 1989, was admittedly given after he had settled with plaintiff and had been released of any liability. Mr. Graham testified, on October 18, 1988, as follows:
Well, I went around the curve, and — and, you know, I thought the road kept going straight, you know. It looked like that, you know, when you went around the curve, you know, that — that, you know, the road was straight, you know. And so I was, you know, started going straight there, you know, an [sic] it was — it was more of the curve, you know. It was like the curve, you know, went on around. And so I kind of, you know, went like that (indicating), you know. I thought it was straight, and then when I figured out that it wasn’t straight, well, then I went back that a way (indicating), and that’s when my wheel dropped off, you know. And then I tried to get back on the road, and every time I’d try to go back on the road, you know, well, it kept shifting me and kept kind of throwing me down in the ditch, you know. And so then, you know, I figured well, I couldn’t get — it felt like a hole, or something, you know. And so then I just — it just kept — every time I’d try to go back on it, you know, I could hear the bottom of my, you know, my truck dragging, it sounded like, you know. And so I tried it a couple of times, you know, and I couldn’t get back on the road. So then I just — it just kept throwing me back down in the ditch, you know. So then I just — finally just had to follow it, you know and go down in the ditch, you know. And then I didn’t hit my brakes because if I did, I knew that I would either slide into all them pine trees, you know, or something worse would happen, you know. So I just kind of let it go until I could get, you know, either get back on the road or stop, you know. And then that’s when I hit that gravel entrance, you know, coming in there, and then it just flipped, you know....
[[Image here]]
... All I know is when I — like I was going down in the ditch, you know, and I started kind of going back up, and that’s when I hit it, you know. And then I guess if you’d hit on the right side, you’d start flipping, you know—
[[Image here]]
... I’d hit my brakes, you know, to try to slow down, but, you know, it was dew on the grass, you know, and you know when you just even barely hit your brakes, or anything, you know, you could feel it just trying to slide, you know. So it was best just to kind of ride it out and stop it, you *1204know, the best you could, you know. It just happened so quick, you know.
When questioned whether he remembered telling the investigating officer that plaintiff was very intoxicated and that as they rounded the curve, she fell over in his lap and grabbed the steering wheel causing him to drive off the roadway, Mr. Graham responded:
I remember telling him something like that, but not exactly like that, you know. In other words, I was, you know, it was going by kind of like what happened, you know, from, you know, me trying to, you know, from the — me, you know, not knowing the curve was there, you know, that the curve continued, and all that, you know, and trying, you know, thinking the road went one way, and it went, you know, the other way and jerking, you know, her around I imagine, you know.
[[Image here]]
... [S]he didn’t cause, you know, me to leave the roadway.
[[Image here]]
... [S]he didn’t grab no steering wheel, you know.
In Mr. Graham’s second deposition, given on November 27, 1989, he testified as follows:
I was just going down the road and it kind of looked like the road went straight, you know, and then it was a big curve there, you know. And it looked like you went out of the curve, or something, you know. It’s kind of hard to say, you know. It just looked like it was going straight, and I run off the road.... It didn’t have a shoulder on it and I couldn’t get back on the road....
[[Image here]]
... I tried to get back on the road, and it felt like my truck was dragging the bottom or something, you know. And every time I tried to get back on the road it kept throwing me back off the road, you know. You could feel it just, you know jerking — almost jerked the wheel out of my hand. I tried a couple of times, then it just — finally I went down in the ditch.... I wouldn’t hit the brakes, you know, because I knew if I hit the brakes, you know, that it would be worse, you know. I don’t remember hitting the brakes — put it that way, you know.
[[Image here]]
... That’s — just trying to get back on the road and then I just followed the ditch, you know, kind of straight until I could try to stop it and everything. And then I hit some kind of embankment or something over there, right when I was trying to get control of it.
When questioned again as to whether plaintiff fell over into his lap or grabbed the steering wheel, Mr. Graham responded:
You know, it’s kind of hard to say, you know, because, you know, it happened so fast and, you know, and she was getting throwed around so much, you know, because them cars — them trucks is so light, you know. You can just be riding down the road, you know, and they’re light, you know, and a passenger just jumps up and down automatically anyway, you know.
Mr. Graham denied telling the officer that plaintiff fell over and grabbed the steering wheel prior to the accident and stated, “She was in pretty well control of herself, you know I’d say, through the whole deal.”
Two versions of the circumstances leading up to the accident were presented to the trial court. Mr. Graham first told a version which would tend to inculpate plaintiff, on the date of the accident when the events were fresh in his memory. Some seven months later, after Mr. Graham had been released from the lawsuit, he began to tell a different version of plaintiff’s actions which was exculpatory. This convoluted and belated second version of the accident is remarkably in tune with plaintiff’s theory of the case.3
Courts should be realistic in their evaluation of the evidence. However, under what some legal authorities are calling a new standard of appellate review, the supreme *1205court in Stobart v. Louisiana Department of Transportation and Development, 617 So.2d 880 (La.1993), has said that the appellate power to review facts is severely limited and any facts found by a trial court are beyond our pale of review, however incredible. Even though I would be inclined to reverse the trial court on this issue, and find that Mr. Graham’s second version of the events was not worthy of belief, I am compelled to join in the majority affirmance on this issue in accordance with Stobart.
I disagree with that portion of the majority opinion affirming the finding of design defects in the highway. The trial court found the following design defects in the roadway at the scene of the accident:
(1) The curve was compound with varying radii which gave the appearance to an oncoming motorist that the curve was about to straighten out when in fact it would not;
(2) The slope of the curve in certain portions was actually higher in the center than on the outside aiding centrifugal forces in moving the car to the outside of the roadway;
(3) The curve had varying super elevation measurements making it harder for a motorist to control his vehicle;
(4) The curve had a substandard lane width;
(5) The curve was not marked with edge striping des[pite] the fact that it was narrower than portions of the same road close by that were striped; and
(6) The road had no warnings of a dangerous curve or a low shoulder.
These findings by the trial court were based on the testimony of plaintiffs expert, Mr. James R. Clary, Sr., a civil engineer, who gave his deposition on February 28, 1989. Mr. Clary testified with regard to the compound curve with varying radii: “[I]t appears that the radius changes — the sharpness of the curve changes within the curve. And I — I’m going to do some more geometry on that curve to find out just what it is.” (Emphasis added.) No further testimony was elicited from Mr. Clary after the date of his deposition. Mr. Clary went on to state in his deposition the results of his examination of the accident site:
It does not appear to be a continuous curve. The radius from what we’ve picked up appears to change. I want to check the radius of the curve, and from the cross-sections I took, apply the contract amount of super elevation required to see that the contractor complied with the state regulations on that matter.
[[Image here]]
... I read the signs and I didn’t apply the station of the signs in that to the map that I drew as of yet. I read the testimony about the shoulder and also about lack of the edge drop-off signs, which would be the hazard markers that would be adjacent to the paving; I didn’t see any reference to them. But I have not applied the stationing of the signs in this diary to my drawing yet.
[[Image here]]
... It’s not a uniform curve; it appears to be of several different radiuses, and the sharpness of the curve increases, decreases and it is just not your normal curve. And I’m going to do some geometry on that curve; I haven’t done it as of yet.
[[Image here]]
... The curve, as we picked it up — we picked it up with theodolite and with electronic distance measuring equipment — and the curve is not a uniform curve. It varies in sharpness, which means it varies in radius, which means it’s a compound curve of some type.
[[Image here]]
... To a person using the highway, when you have a curve, you expect it to have a beginning and you go through a curve that’s uniform and you expect it to have an end. This one, you come into it and you have one degree of sharpness, and it changes to another, and then where you would expect it to end it does not end; it — there’s just something — there’s just something strange about it and I — I want to investigate it further.
[[Image here]]
... The striping as of January 26, 1989, indicates center striping, no passing type *1206stripping. It does not have edge striping, which I think is a mistake. This road is only 20 feet in width. It’s a, according to the police report, a posted 55 mile an hour speed limit. And from the daily traffic information, I think the road should be wider. And in lack of width, the edge striping gives the motorist an extra safety feature to determine exactly where the roadway ends — the roadway width ends.
[[Image here]]
... So if for any reason somebody would go off the paved surface and onto the drop-off that I’ve read existed at that time, then they’d have a problem. But I’m sure that a lot of people did negotiate it safely.
[[Image here]]
... [A cross-section of the roadway and the scene of the accident] shows that one edge of the road, which would be the left-hand side, south edge, is depressed almost a foot lower than the one on the right-hand side. The crown of the road is somewhere between them but it’s not a uniform slope. It doesn’t conform to either a normal cross-section nor does it conform to super elevation.
[[Image here]]
... [T]he center line of the road is lower then [sic] the outer edge and that’s exactly opposite from the way a normal crown would be unless you’re in a super elevated section.
[[Image here]]
... In the super elevation requirements on sheet 1 of 3, which is page 4 of the plans, number 2 in the super elevated section, they show required leveling to achieve super elevation. So the — in the super elevated sections, you are to level it up — and I believe that’s a pay item. And sheet 6 of plans, which is 3 of 3 of the super elevation, they give you super elevation values for rural overlays for various speeds. This is a 55 mile an hour posted speed road for normal — not during construction times. And for whatever degree of curve that is, they give you an absolute maximum and minimum that you can have for super — super elevated sections. And the — it is the contractor’s responsibility to level that as required to achieve super elevation according to plans.
[[Image here]]
... Now the super elevation rates — I’m going to check to see whether or not they comply with the design and the absolute minimum required by the state as fulfilling the contract. I’m going to go back on each one of these (indicating) and figure out what the crown is, et cet-era_ I haven’t done that yet, ... But with a set of plans it would be just as easy for, you know, if you had somebody that you want to check it, they could do it also.
[[Image here]]
... Super elevation in a curve is to — to drain water off the road and to give a driver going around a curve a comfort level whereby centrifugal force will not force him off the road. It will allow him to travel through that curve at certain speeds — whatever the design speed is for the curve with centrifugal force equaling — centrifugal force where he will have no tendency to slide. [Emphasis added.]
Although this expert did state that the lanes should have been wider and that there should have been edge striping on the roadway, he did not state that the absence of these things caused the accident in question; nor did he state that the absence of these items violated any industry standard. Also, Mr. Clary did state that the surface of the highway was not level; however, he stated that it was the contractor’s responsibility to level the highway in accordance with the plans and specifications provided by the state. Further, this expert did not state definitely that any of the other design defects found by the trial court actually existed. Mr. Clary found only that the curve was unusual and merited further investigation for possible design defects.4
*1207Further, although the general vicinity of the accident was established and referred to by witnesses as encompassing a “curve”, the precise location of the accident was never established. The most accurate description of the location was provided by the investigating officer, Trooper Coburn, who testified that the accident occurred on Highway 42, “18 tenths” of a mile east of Highway 63; however, Trooper Coburn admitted that his measurements were not completely accurate.
Because of the failure of the plaintiff to establish her case by a preponderance of the evidence as to the design defects, I would find the trial court committed manifest error in holding design defects existed at the accident site; there was simply no direct evidence to support such findings.
The only remaining defects in the highway (i.e., low shoulder, lack of signs, and unlevel highway surface) were all testified to be within the responsibility of the contractor to perform per his contract with the state. Thus, La. Paving actually committed the tortious offenses which caused the accident in this case, but because the state has a non-delegable responsibility to maintain public highways in a condition that is reasonably safe, the state is also liable for the negligence of its contractor.5 Robinson v. Louisiana Department of Transportation and Development, 454 So.2d 257 (La.App. 1st Cir.), writ denied, 458 So.2d 122 (La.1984). For.this reason, I would assign 82% fault to the state. However, La. Paving was released from tort liability by plaintiff. Therefore we must examine the effect this release has on the liability of the state.
In Rowell v. Carter Mobile Homes, Inc., 482 So.2d 640 (La.App. 1st Cir.1984), affirmed on other grounds, 500 So.2d 748 (La.1987), this court held that the release of the primarily liable obligor extinguishes the obligation, and precludes recovery from the derivatively liable obligor. The Rowell court reasons that these two obligors are solidarily liable, and as such, the liability of the tortfeasor remaining in the lawsuit should be reduced by the fault of the released tortfeasor. Since the percentage of fault for these two obligors would be the same, reduction by one would result in zero liability for the other.6 482 So.2d at 648.
*1208The Third Circuit would reach the same result under a different rationale. In Cas-son v. Hartford Fire Insurance Co., 548 So.2d 66 (La.App. 3d Cir.1989), that court concluded that the vicariously or technically at-fault defendant is entitled to contractual indemnification from the principal tort-feasor who has been released by the plaintiff, and distinguished subrogation and indemnity as follow:
Contribution and indemnity are different legal tenets. One guilty of fault is not due indemnity.... Indemnity, unlike contribution, is not dependent upon sub-rogation to the right of the creditor, but finds its basis in the concept of unjust enrichment, i.e., the party primarily at fault is unjustly enriched when one held liable vicariously or by reason of technical fault discharges the indebtedness.
But the court went on to find that since the plaintiff had agreed to indemnify and hold harmless the principal tortfeasor against any third party demands or claims in return for its settlement, indemnification from the principal tortfeasor to the vicariously at-fault tortfeasor could be followed by indemnification from the plaintiff to the principal tortfeasor. Unfortunately, in the instant case, the actual release between La. Paving and plaintiff was not made a part of the record, though some evidence that a release did in fact occur is in the record.
Nevertheless, we are compelled to follow the holding of this court as expressed in Rowell:
[W]e hold that in a principal-mandatary relationship the release of the primarily liable person or entity accomplishes the release of the vicariously liable person or entity.
Although the supreme court in affirming Rowell on another basis, stated that:
[TJhere appeared to be possible error in the court of appeal’s application of our decision in Sampay ... and in its conclusion as to the effect of a servant’s settlement upon the solidary liability of his master. These questions cannot be addressed in this case, however, because a proper application of the correct legal precepts render them abstract or purely academic.
Thus, the expression of the First Circuit in Rowell has not been overruled and remains the law of this circuit.7 Therefore, I would hold that La. Paving is the principal tort-feasor in this case, and that, the state as a derivatively at-fault tortfeasor is relieved of liability as a result of plaintiff’s release of La. Paving. Accordingly, I would dismiss plaintiff’s claims against the state.
In all other respects, I agree with the views expressed in the majority opinion, and in particular, the majority holding applying the statutory limit on general damage recovery to $500,000.00 in personal injury suits against the state. La.R.S. 13:5106(B)(1).
Accordingly, I dissent, in part, and concur in part.
*1209APPENDIX "A"
(From the record as "Plaintiffs Exibit 3")
[[Image here]]
. La.R.S. 32:295.1(E.)(4) as it appeared at the time of the accident allows the assignment of a percentage of fault for failure to wear a safety belt when the use of the safety belt would have reduced the injured party’s damages.
. Under C.E. 803, this statement is probably an excited utterance which can be given substantive weight.
. It is also interesting to note that in a case of this magnitude all testimony was given by deposition; no live witnesses were called at the trial of the matter.
. Plaintiff’s expert, Mr. Clary, testified that a compound curve is one which is not uniform but changes in degree of sharpness and does not end as expected; however, he did not state cate*1207gorically that the curve where the accident occurred was, in fact, a compound curve. A copy of Plaintiffs Exhibit 3 is appended hereto as Appendix "A”, and purports to be an aerial photograph of the curve in question. Review of this photograph would not lead a layperson to conclude that the curve changes in degree of sharpness, or appears to end and yet does not. Hence, this exhibit alone is not of sufficient probative value to establish that such a design defect existed without expert testimony to substantiate the claim.
. I believe the majority opinion also errs in its assignment of fault, 50% to La. Paving and 50% to the state. Since the state is not relieved of its duty to maintain public roads and highways by contracting with another entity to perform the maintenance and/or repairs, whatever fault La. Paving has in failing to perform the necessary repairs is also imputable to the state. See Ishee v. Louisiana Department of Transportation and Development, 413 So.2d 1362 (La.App. 1st Cir. 1982). Where there is no negligence of the state independent of that of the contractor, the fault of the state is the same as that of the contractor. See Rowell v. Carter Mobile Homes, Inc., 482 So.2d at 648. Thus, under these circumstances and in accordance with the majority findings, the state would be 100% liable for plaintiffs injuries, not 50%. (The 50% fault of La. Paving is imputable to the state, which, when added to the 50% independent fault of the state (attributable to the state according to the majority opinion), equals 100% fault.)
. The Rowell court states:
Implicit in Sampay [v. Morton Salt Co., 395 So.2d 326 (La.1981) ] is the conclusion primary and derivative obligors are each liable for virile shares. Prior to our adoption of a comparative fault system, there was no other basis for fixing an employer’s liability when his negligent employee was released by the plaintiff and the employer itself was not guilty of any independent fault.... However, in our view, the subsequent adoption of a comparative fault system in this state negates this position. La.Civ.Code art. 2103, as amended by Act 431 of 1979, effective August 1, 1980, and in effect at the time of plaintiff’s injury, provided a solidary obligation resulting from an offense or a quasi-offense shall be divided in proportion to each debtor’s fault. Where the derivative obligor is not guilty of any independent fault, no additional percentage of fault can be assessed to it under La.Civ. Code art. 2103 or its successor articles, other than the primary obligor’s percentage of fault. Thus in such cases, the obligation of the derivative obligor is identical to that of the primary obligor.
. The conclusion reached in the First Circuit opinion of Rowell has been cited with approval in David W. Robertson, Comparative Fault in the 1990s: Solidarity, Contribution, Indemnity.