IN THE UNITED STATES COURT OF APPEALS

                            FOR THE FIFTH CIRCUIT


                                    No. 94-41292




ANTHONY GUILLORY,
MARY GUILLORY,
                                Plaintiffs-Appellees,

                                        versus

DOMTAR INDUSTRIES INCORPORATED,
                         Defendant-Third Party Defendant Intervenor-Appellee,

                                        versus

JOHN DEERE INDUSTRIAL EQUIPMENT COMPANY,
                          Defendant-Third Party Defendant Plaintiff-Appellant,


                                          and


JOHN DEERE COMPANY,
                                Defendant,

DEERE & COMPANY,
                                Defendant-Third Party Plaintiff-Appellant,

                                        versus

HARLO PRODUCTS CORPORATION,
                        Defendant-Third Party Defendant Intervenor,

TRANSAMERICA INSURANCE COMPANY,
                        Third Party Defendant Intervenor-Appellee.


_________________________________________________________________________________
ANTHONY GUILLORY,
MARY GUILLORY,
                                          Plaintiffs-Appellees,

                                                  versus

DOMTAR INDUSTRIES INCORPORATED,
                         Defendant-Appellee,


JOHN DEERE COMPANY,
                                          Defendant,


JOHN DEERE INDUSTRIAL EQUIPMENT COMPANY,
                          Defendant-Appellant.



                             Appeals from the United States District Court
                                for the Western District of Louisiana


                             September 27, 1996
               *
Before LAY , HIGGINBOTHAM, and STEWART, Circuit Judges.

CARL E. STEWART, Circuit Judge:

          In this Louisiana products liability action, a jury awarded the plaintiffs over $6 million for

injuries sustained because a fork fell off a forklift and struck Anthony Guillory in the head. The jury

found that the forklift was improperly installed and maintained by Anthony Guillory’s employer,

Domtar Industries. The jury also found that the manufacturers, Deere & Company and John Deere

Industrial Equipment Company (“Deere”), provided inadequate warnings regarding the forklift. The

district court granted summary judgment in favor of Domtar because the Louisiana workers’



   *
       Circuit Judge of the Eighth Circuit, sitting by designation.

                                                     2
compensation scheme limited the plaintiffs’ recovery against Guillory’s employer. Thus, though the

jury apportioned eighty percent fault to Domtar and twenty percent fault to Deere, Louisiana’s laws

of solidary obligation required Deere to pay 100% of the judgment. Deere appeals challenging

several rulings of the district court as well as the findings of the jury. After thoroughly reviewing

the record and finding no error with either the rulings or the findings, we affirm the jury’s award for

the plaintiffs.




                                                  3
                                            BACKGROUND

          Anthony Guillory, an employee of Domtar Industries, worked as a welder in a salt mine

located on Cote Blanche Island in Iberia Parish, Louisiana. On September 8, 1990, Guillory was

injured while assisting two co-employees, Irvin Boutte and Stafford Caesar, in the replacement of a

section of conveyor belt framework.1 Boutte was operating a John Deere 380 forklift to move the

section of conveyor system. The men attached the section of the conveyor system to the left fork with

a chain. When the conveyor system got caught on a suspension cable, Guillory went to the right side

of the forklift, freed the conveyor, and signaled Boutte to continue the lift.     After he freed the

conveyor, Guillory stood about three feet from the conveyor and seven feet from the front tire of the

forklift; he was not standing underneath either the conveyor or the fork. Moments later, the right

fork fell from the forklift and struck Guillory in the head.

          Domtar purchased the John Deere 380 forklift from a John Deere dealership in 1980. Harlo

Products, Inc., one of the defendants, manufactured the mast system and component parts that Deere

incorporated into the 380 forklift. The 380 forklift originally came equipped with a set of forks with

capacities of 4,000 and 5,000 pounds. Harlo also manufactures a set of forks with a 6,000 pound

capacity, which is similar in appearance and dimension to the 4,000/5,000 pound fork. The John

Deere 380 forklift parts catalog improperly listed the 6,000 pound fork as an appropriate fork for the

380 forklift. Before Guillory’s accident, when the original forks became bent, Domtar replaced them

with the 6,000 pound forks listed in the catalog. Deere never advised its users that the listing of the

6,000 pound fork in the 380 parts catalog was erroneous.




   1
       The conveyor is used to haul salt to various sections in the salt mine.

                                                    4
        The forks were designed t o be attached to the forklift with a dual retention system, which

consists of a one-inch backing plate using either two spiral pins or two spring loaded pins. Though

Harlo provided detailed instructions to Deere regarding the operation and maintenance of the

retaining system, neither Deere nor Harlo provided instructions to users regarding the importance of

maintenance of the system or recommended that users use lubricants on the backing plate. Domtar

conceded, however, that it did not read the Deere manuals or provide training in the operation of the

forklift.

        Consequently, the Domtar employees sometimes only partially used the dual retention system.

Occasionally, they substituted the dual retention system with nuts and bolts, while other times they

used nothing to secure the forks on the forklift.2 It was common knowledge in the mine that forks

had fallen off the 380 forklifts several times prior to Guillory’s accident. However, the forks had not

fallen off during a lifting operation, only when the forklift was traveling over a bumpy terrain.

Further, most miners believed that the bolts used on the forklift’s rack would prevent the forks from

falling off the forklift. The fork that injured Guillory was conducting a lifting operat ion and was

secured by bolts at the time of the accident. The employees misunderstood the actual cause of falling

forks, and apparently most had not reported the incidents to the Domtar management because they

did not consider the falling forks a hazard -- until Guillory’s accident.

        Guillory’s injuries rendered him permanently quadriplegic. Guillory experiences painful

spasms in almost all parts of his body that must be controlled with medication. He will require



    2
     The plaintiffs presented evidence establishing that the fork retaining system would not work
properly with either a small amount of wear or when the mismatched 6,000 pound forks were
installed. The evidence also showed that the conditions in the salt mine caused the retaining system
to wear very rapidly.

                                                   5
twenty-four hour attendant care and continuing medical treatment for the rest of his life. Guillory

sued Domtar (his employer), Deere and John Deere (the forklift manufacturer), and Harlo (the

manufacturer of component parts for the forklift). The district court granted Domtar’s motion for

summary judgment on the ground that workers’ compensation was Guillory’s exclusive remedy

because he was injured while in the course and scope of his employment.3

       Before trial, a settlement conference was arranged. Deere offered to settle the case for

$100,000. The district court concluded that Deere’s unrealistic offer and subsequent comment that

there was never a chance at settling represented a failure to act in good faith. The magistrate

sanctioned Deere $8,500, the amount of expenses incurred by the parties in preparing for the

settlement conference.

       Also, before trial, Deere’s trial expert became unavailable for trial due to medical reasons, and

Deere selected Dr. Walter Reed as a substitute. The plaintiffs filed a motion in limine seeking to

exclude or limit Dr. Reed’s testimony. The district court denied the motion. On the sixth day of trial,

the district court, on its own motion, reconsidered the plaintiffs’ motion regarding Dr. Reed’s

testimony. The court allowed Dr. Reed to testify but excluded the videotape and model containing

a demonstration conducted by Dr. Reed.

       The jury found Domtar eighty percent at fault and Deere twenty percent at fault. Deere, being

solidarily obligated with Domtar, became responsible for the entire judgment under Louisiana law.

Deere timely appealed the adverse rulings and verdict.




  3
   The district court denied Deere’s request to appeal the summary judgment pursuant to 28 U.S.C.
§ 1292(b).

                                                  6
                                           DISCUSSION

I.     SUMMARY JUDGMENT FOR DOMTAR INDUSTRIES.

       We review the granting of a motion for summary judgment de novo. Christophersen v.

Allied-Signal, 

, 1109 (5th Cir. 1991) (en banc), cert. denied, 

(1992).

Summary judgment shall be granted if the record, taken as a whole, “together with the affidavits, if

any, show that there is no genuine issue as to any material fact and that the moving party is entitled

to a judgment as a matter of law.” FED. R. CIV. P. 56. When evaluating the summary judgment

evidence, we resolve factual controversies in favor of the nonmoving party, but only when there is

an actual controversy; that is, when both part ies have submitted evidence of contradictory facts.

Little v. Liquid Air Corp., 

, 1075 (5th Cir. 1994). Review of the summary judgment

requires the same analysis employed by the district court. Turnage v. General Elec. Co., 

, 212 (5th Cir. 1992).

       A.      Issue of Intent.

       Deere argues that summary judgment is improper when a determination of intent is necessary

 to evaluate the existence of a fact issue. Deere asserts that there is disputed evidence regarding

whether Domtar knew of an unsafe working condition and failed to take action to protect its

employees. We disagree.

       Though summary judgment is rarely proper when an issue of intent is involved, the presence

of an intent issue does not automatically preclude summary judgment; the case must be evaluated like

any other to determine whether a genuine issue o f material fact exists. See Krim v. BancTexas

Group, Inc., 

, 1449 (5th Cir. 1993) (“when the nonmoving party has not produced an

affidavit or a request for a continuance outlining how further discovery would enable him to make


                                                  7
the question of intent an issue of material fact, summary judgment is proper even if intent is an

essential element of the nonmoving party’s case”). In International Shortstop, Inc. v. Rally’s, Inc.,

, 1265-66 (5th Cir. 1991), cert. denied, 

(1992), the court detailed why

summary judgment usually is inappropriate for issues involving state of mind;4 however, the court

quickly clarified that “[t]his is not to say that the court can never enter summary judgment when

intent or state of mind is at issue, . . . to be sure, [summary judgment] may be appropriate if the

moving party rests merely on conclusory allegations, improbable inferences, and unsupported

speculation.” Accordingly, the allegation of intent does not spontaneously shield a case from

summary judgment. See S.W. Erectors, Inc. v. Infax, Inc., 

, 495 (5th Cir. 1996) (“It is

well established in this circuit that an allegation of fraud does not create an impenetrable shield

through which the sword of summary judgment cannot pierce. A fraud case can be wounded and

killed like any other case; it receives no special privileges or protections in the battle for summary

dismissal.”). We must, therefore, determine whether the intent allegation in this case raises a genuine

issue of material fact.

        Our review of a granted motion for summary judgment is limited to the evidence available to

the district court at the time it ruled on the motion. See 7547 Corp. v. Parker & Parsley Dev’t

Partners, LP , 

, 220 (5th Cir. 1994); Murphy v. Georgia-Pacific Corp., 

, 866


  4
    “When state of mind is an essential element of the nonmoving party’s claim, it is less fashionable
to grant summary judgment because a party’s state of mind is inherently a question of fact which turns
on credibility. Credibility determinations, of course, are within the province of the fact-finder. . . .
Only through live cross-examination can the fact-finder observe the demeanor of a witness and assess
his credibility. A cold transcript of a deposition is generally no substitute because it cannot unmask
the veracity of a testifying witness clad in a costume of deception; it cannot unveil that a seemingly
well-groomed witness is coming apart at the seams: ‘that he fidgets when answering critical
questions, his eyes shift from the floor to the ceiling, and he manifests all other indicia traditionally
attributed to perjurers.’” International 

(collecting cases).

                                                   8
(5th Cir. 1980); and Feminist Women’s Health Center, Inc. v. Mohammad, 

, 546 n.15

(5th Cir. 1978), cert. denied, 

(1979). We cannot consider facts and evidence uncovered

during trial when assessing the propriety of the summary judgment. We, therefore, review only the

deposition excerpts and the notes and reports of the Mine Safety Health Act (“MSHA”) investigator,

which are attached as exhibits to the memorandum in opposition as well as the deposition excerpts

submitted during the summary judgment hearing.

         Here, the evidence present at the time of the motion does not demonstrate a genuine issue

capable of saving this case from summary judgment. Though Deere adopted the arguments raised

in defendant Harlo’s memorandum in opposition, the summary judgment evidence presented by Harlo

does not support the claim that Domtar intentionally injured Guillory. Simply put, the evidence does

not establish that Domtar consciously desired to injure Guillory or knew that Guillory’s injury was

substantially certain to follow from its actions. See Bazley v. Tortorich, 

, 482 (La.

1981).

         The substantially certain test is satisfied when an employer consciously subjects an employee

to a hazardous or defective work environment where injury to the employee is nearly inevitable --

that is, injury is “almost certain” or “virtually sure” to occur or is incapable of being avoided. See

Kent v. Jomac Products, Inc., 

, 100 (La. App. 1st Cir. 1989). The evidence, viewed

in the light most favorable to Deere, shows that some miners and supervisors were aware that several

forks had detached from forklifts when retaining bolts were not used. The evidence indicates that

some miners knew that the forks on the No. 6 forklift, which injured Guillory, had fallen off while

they were operating the forklift. Miners testified that the forks on the No. 6 came off when there was

no load on the forks, and indicated that the right fork was the fork that usually fell off this machine.


                                                   9
However, by itself, knowledge of the falling forks cannot establish “intent” on the part of Domtar.

Deere’s summary judgment evidence does not demonstrate that it was common knowledge among

miners or known by management that the forks would fall when secured by nuts and bolts present on

the upper corners of the forklift’s fork rack.5 The miners generally believed that the nuts and bolts

would secure the forks, and the uncontroverted evidence shows that at the time of the accident the

nuts and bolts were present in the upper corners of the fork rack on the No. 6 forklift.

       We find that any issue regarding Domtar’s intent is foreclosed by the absence of summary

judgment evidence establishing that it was commonly known that forks would fall despite the

presence of nuts and bolts. Because the falling forks were viewed as a nuisance rather than a hazard,

because Domtar’s supervisors believed that the retaining bolts prevented the falling forks, and

because the No. 6 had retaining bolts present at the time of the accident, Domtar did not consciously

place Guillory in a wo rk environment where injury was an unavoidable and a nearly inevitable

consequence.6

       Courts narrowly interpret the intentional act loophole to the workers’ compensation system.

Bridges v. Carl E. Woodward, Inc., 

(La. App. 4th Cir. 1995), writ denied, 

674 (La. 1996). Even if the falling forks in this case created a “high probability” of injury, this



   5
    Two miners testified in their depositions that a fork detached while they operated forklifts even
though the fork had been secured by a bolt. Also, one miner testified that the right fork on the No.
6 forklift detached during an operation despite the presence of a bolt in the upper right corner of the
rack. The summary judgment evidence does not demonstrate, however, that Domtar’s management
was aware that the forks could bypass the nuts and bolts and detach from the forklift.
   6
    We find this case distinguishable from the cases cited by Deere. The prior incidences of falling
forks did not adequately put Domtar on notice that the forks would fall despite the presence of nuts
and bolts. By contrast, the facts of the cases cited by Deere involved notice from which the employer
knew that the propensity for a nearly inevitable injury existed.

                                                  10
would not establish “intent” sufficient to bypass the workers’ compensation system. See Armstead

v. Schwegmann Giant Super Markets, Inc., 

, 1142 (La. App. 4th Cir. 1993), writ

denied, 

(La. 1993). Similarly, even if Deere could establish that Domtar’s acts or

omission constituted “gross negligence,” “intent” still would be lacking. See King v. Schuylkill

Metals Corp., 

, 302 (La. App. 1st Cir.), writ denied, 

(La. 1991).

To qualify for the very narrow intent exception, Louisiana jurisprudence requires a strong link

between the employer’s conduct and the employee’s injury. Williams v. Gervais F. Favrot Co., 

, 541 (La. App. 4th Cir.), writ denied, 

(La. 1991). Deere failed to provide

this “link” in its summary judgment evidence. We refuse to take the leap that would be necessary to

reach the conclusion Deere asks of us. At best, Deere’s evidence establishes negligence. An issue

of intent cannot be construed from the summary judgment evidence presented.

       Further, Deere has not refuted the availability of summary judgment in the context of a

workers’ compensation dispute. The district court held that summary judgment was the proper

procedural method to consider an allegation that an employee’s injury resulted from an intentional

act of his employer. We agree. See Mayer v. Valentine Sugars, Inc., 

, 618 (La

1984); and 

. Because no issue of intent exists, Guillory is relegated to the

workers’ compensation statutes to recover damages against Domtar.




       B.      Summary Judgment Shortly Before Trial.

       Deere contends that because summary judgment was granted forty-five days before the

original trial date the judgment affected their trial preparation and strategy. Deere argues that it was


                                                  11
prejudiced by the ruling because the court prevented it from presenting evidence of Domtar’s

intentional act, and the jury never received an instruction regarding Domtar’s intent. Moreover,

prejudice resulted because the plaintiffs discontinued the assertions presented in their pleadings and

throughout discovery regarding Domtar’s intentional acts; instead, the plaintiffs presented evidence

of stellar conditions in the mine.

          Quite to the contrary, the appellees respond that the district court’s ruling almost three and

one-half months before trial came only after Deere failed to present any evidence of an intentional act

by Domtar.

          The timing of the summary judgment does not warrant reversal in this case. Rule 56(b)

allows a defendant to move for summary judgment at any time. Accordingly, the court may grant

summary judgment any time before trial. See Bynum v. FMC Corp., 

, 577-78 & n.32

(5th Cir. 1985) (appellate court affirmed judgment where the district court initially denied summary

judgment but upon reconsideration granted summary judgment the day of trial). However, the license

to grant summary judgment close to the date of trial comes with some restraint. See Prudhomme v.

Tenneco Oil Co., 

, 392-96 (5th Cir.), cert. denied, 

(1992). In

Prudhomme, when discussing court-induced prejudice in the context of an amended complaint, this

court demonstrated when elevent h hour rulings can cause prejudice and function as an abuse of

discretion.7 Extending Prudhomme by analogy, if the circumstances are such that a belated granting


      7
       The Prudhomme court found that the district court abused its discretion by sua sponte
reconsidering and then reversing on the morning of trial its prior ruling which had denied the
plaintiffs’ motion to amend its complaint to allege an additional claim against a defendant. The
plaintiffs moved to amend its complaint to add a claim of strict liability against Booker after the
expiration of discovery, the pre-trial conference, the submission of the Joint Pre-trial stipulations, and
the setting of the trial 

. The court granted the plaintiff’s motion to dismiss, filed
to avoid losing the trial date. The strict liability claim was not raised or mentioned until the pretrial

                                                   12
of summary judgment in favor of one defendant will prejudice the trial preparation of another

defendant, the district court should continue the trial in order to allow an interlocutory appeal under

28 U.S.C. § 1292(b) or to allow time for the prejudiced defendant to adjust its trial strategy,

witnesses, and evidence.

       Though recognizing that a belated summary judgment could cause prejudice where only one

of several defendants obtains summary judgment, we cannot say that prejudice or abuse of discretion

has occurred in the present case. The district court ruled on May 6, 1994; trial began on August 22,

1994. Deere had over three months to adjust its trial strategy or to move for a continuance when the

district court refused to certify an interlocutory appeal. Deere’s appellate brief demonstrates the lack

of prejudice in the timing. They noted that the plaintiffs took a 360° turn and completely altered their

position at trial from criticism to almost praise of the mine conditions. The plaintiffs had the same

amount of time as Deere to adjust their case, and they did it. Three months provided Deere sufficient

time to modify trial tactics in a way that would accommodate Domtar’s absence from trial. Further,

even without the evidence of intent, the jury found Domtar eighty percent negligent. Thus, the timing

of the summary judgment did not create prejudice warranting reversal of the judgment. Under the

law and the facts presented, we agree that Domtar was entitled to summary judgment.

II.    EVIDENTIARY RULINGS.



conference held the morning of trial when the district court announced that it would consider the
strict liability claim against Booker. The district court noted Booker’s objection but began trial the
same morning. 

The appellate court in Prudhomme agreed that the district court’s
dismissal of the supplemental and amended complaint misled Booker and induced prejudicial inaction.
 

Booker had prepared to defend the action against negligence only. It would have
produced eyewitnesses, retained experts, and located physical evidence if it had known strict liability
would be encountered at trial. 

Accordingly, the court vacated the judgment to the extent
it held Booker strictly liable. 

                                                  13
        We review a district court’s exclusion or admission of evidence for an abuse of discretion.

See Mac Sales, Inc. v. E.I. du Pont de Nemours & Co., 

, 753 (5th Cir. 1994) (excluded

expert); United States v. Stephens, 

, 239 (5th Cir. 1985) (admitted exhibits); and United

States v. Follin, 

, 375 (5th Cir. 1992) (admitted photograph). The district court is given

wide discretion regarding evidentiary rulings and, consequently, we review evidentiary decisions for

manifest error. United States v. Massey, 

, 140 (5th Cir. 1995); London v. Mac Corp.,

, 318 (5th Cir. 1995). In other words, the district court does not abuse its discretion if

the error is merely harmless. Grizzle v. Travelers Health Network, Inc., 

, 271 (5th Cir.

1994). We "will not disturb an evidentiary ruling, albeit an erroneous one, unless it affects a

substantial right of the complaining party." Polythane Sys. Inc. v. Marina Ventures Int’l, Ltd., 

, 1208 (5th Cir. 1993), cert. denied, 

(1994); see also Edmonds v. Illinois

Cent. Gulf R. Co., 

, 1287 (5th Cir. 1990).

        A.      Limits on Dr. Reed’s Exhibits and Testimony.

        Deere argues that the district court abused its discretion when, after trial had commenced, it

excluded the videotape prepared by Deere’s expert, significantly limited his testimony, and limited

the number of photographs to be introduced. Deere also asserts that the exclusion and limitations

adversely affected its trial strategy.

        When arguing their motion in limine, the plaintiffs in this case argued that Dr. Walter Reed,

a mechanical engineer and accident reconstruction expert, was not an equal substitute for Deere’s

originally listed trial expert, Ronald Brass, an expert in the field of agricultural engineering. The

district court’s minute entry rejected, without explanation, the plaintiffs’ arguments to restrict Dr.

Reed’s testimony to the same scope as Brass’s testimony. The minute entry also noted the court’s


                                                 14
denial of the plaintiffs’ oral motion to exclude the models and exhibits prepared by Dr. Reed. Though

the plaintiffs’ challenged the reliability of the models, the court ruled that “after considering the

arguments of counsel, I conclude that the issues raised are properly addressed on cross examination.”

The court then responded to plaintiffs’ arguments of prejudice by offering to continue once more the

trial date, which the plaintiffs rejected.

          After hearing some of Dr. Reed’s testimony during trial, the district court limited his

testimony to a discussion of only one of ten models and excluded Dr. Reed’s videotape. The court

found that Dr. Reed’s conclusions were not sufficiently grounded in scientific methodology or the

facts as presented in the testimony of other witnesses.8 Similarly, after recognizing that under

Daubert9 vigorous cross-examination is the appropriate means to attack shaky but admissible

evidence, the district court prohibited Deere from introducing Dr. Reed’s videotape. The court

reasoned that Dr. Reed used a forklift model that was no t sufficiently similar to the forklift which


  8
  Regarding the testimony, the district court expressed its dissatisfaction with the foundation upon
which Dr. Reed based his conclusions:

          All right, Counsel, I have an obligation under the Daubert case to exercise control
          over the scope of expert testimony, and Dr. Reed’s testimony for the last many
          minutes has been a series of, basically, speculations concerning what probably
          happened, what possibly could have happened based upon not even testimony in
          certain circumstances but his version of what might have happened based upon some
          foundation which is not clear to me. I have a problem with letting this continue in
          this manner. If there is some testimony that you all have or can elicit from Dr. Reed
          based upon scientific methods that are not speculative, I will allow him to continue,
          but if this is the way all the testimony is going to proceed, the court cannot allow it
          to continue. It’s highly prejudicial. It’s not sufficiently based upon scientific methods
          that we discussed earlier. . . . [Dr. Reed’s] testimony regarding the fact that probably
          the reason that the plaintiff and Mr. Caesar were on the left side of the machine was
          to help support a seven hundred and fifty pound section of conveyer, I don’t recall
          any testimony to that effect. . . . That’s just not in the record, as far as I can recall.
   9
       Daubert v. Merrell Dow Pharmaceuticals, Inc., 

(1993).

                                                     15
caused the accident. Because the model involved highly technical information and because Dr.

Reed’s model differed in several respects from the forklift and the accident scene, the court found that

the video would confuse the jury.10

          The Daubert Court explained the district court’s role in administering Federal Rule of

Evidence 703. Citing Daubert, this court summarized the district court’s responsibilities as follows:

(1) to ensure that an expert’s testimony rests upon a reliable foundation, (2) to ensure that all



   10
        The court detailed why it deemed the model confusing and highly prejudicial:

          Because we are dealing with a videotape, the court has to take into consideration that
          this evidence could be very powerful to the jury. . . . My concerns with the evidence
          itself are a number of things. First of all, Dr. Reed, I am confident that he made as
          accurate a measurement as he could of the forklift in question and that his model is
          based upon a scientific method that is appropriate for introduction to the jury vis-a-vis
          expert testimony. However, Dr. Reed does state that his demonstration, first of all,
          is not in [sic] accident reconstruction, although I think it could be viewed by the jury
          as such even with a limiting instruction. Second of all, Dr. Reed testified that the
          demonstration shows only one possible mechanism of how the forks fell off. Certainly
          could not exclude the whole universe of possibilities in this case. We do have an
          unusual case in that there are no eyewitnesses that saw the fork fall off. However,
          there is no doubt the fork did fall off. Dr. Reed used as a basis for his demonstration
          a slope of four degrees. All the testimony in this case indicates that the slope may
          have been twelve degrees at the time. The model has a painted surface; and although
          Dr. Reed testified that the coefficient of friction would be lowered with painted
          surfaces, no measurements were ever made with a coefficient of surfaces involved.
          So there is no way to really accurately compare them. The video demonstration did
          not involve any upward or downward movement of the forks as witnesses have
          testified in this case, or at least one witness who was in a position to see. In addition,
          the -- one of the critical measurements involved in this case is the bottom end of the
          bottom backing plate where the lower hook on the fork would catch. In this case,
          that is a critical measurement; and Dr. Reed testified that that was not measured here.
          He also, after moving the forks on the model in question, testified that the positioning
          of the forks on the actual forklift could be done in a certain way that could not be
          done on the model, which indicates another difference, a critical difference. So my
          ruling, I anticipate ruling and I think the way I’m going to go is to exclude the
          videotape presentation because I do find that the probative value is outweighed by the
          prejudicial value in [sic] the potential to mislead or confuse the jury in this case.

                                                     16
scientific testimony or evidence is reliable and relevant, and (3) to exclude scientific evidence “if its

probative value is substantially outweighed by the danger of unfair prejudice.” Marcel v. Placid Oil

Co., 

, 567 (5th Cir. 1994). In Mac Sales, the expert flatly stated that he had conducted

no research on foreign 

. He later reiterated that his research was limited to

domestic markets. The court concluded that the district court had not abused its discretion in holding

that the expert “lacked a proper foundation from which to testify on foreign markets.” 

        We find that the district court properly excluded Dr. Reed’s expert testimony, which was not

based upon the facts in the record but on altered facts and speculation designed to bolster Deere’s

position.

        Certainly nothing in Rule 703 requires a court to admit an opinion based on facts that
        are indisputably wrong. Even if Rule 703 will not require the exclusion of such an
        unfounded opinion, general principles of relevance will. In other words, an opinion
        based totally on incorrect facts will not speak to the case at hand and hence will be
        irrelevant. In any event such an opinion will not advance the express goal of
        “assisting the trier of fact” under rule 702.

See 

. Though the district judge serves as a gatekeeper for expert

evidence, it is an important role designed to extract evidence tainted by farce or fiction. Expert

evidence based on a fictitious set of facts is just as unreliable as evidence based upon no research at

all. Both analyses result in pure speculation. We find the testimony properly excluded on this

ground.

        Additionally, in the present case, the differences bet ween the model and the actual forklift

made the model unreliable because of the technical nature of the evidence. Dr. Reed constructed a

forklift containing a different design than the No. 6 forklift. It included post-accident modifications

added to the forklifts. Dr. Reed videotaped the model performing several tests and concluded that



                                                   17
it was impossible for the forks to fall with a retaining bolt present. Dr. Reed admitted, however, that

though the forks could not rotate off his model this could be accomplished on the actual forklift.

Further, demonstrations with the real forks and real forklift, without the post-accident modifications,

revealed that Dr. Reed’s conclusions were unfounded and misleading.

           Normally, the truth regarding differences in models and demonstrations surfaces with

vigorous cross-examination; however, where technical information is involved, it is easier for the jury

to get lost in the labyrinth of concepts. We are convinced that cross-examination of Dr. Reed could

not salvage the truth. Equipment and procedures in a salt mine are foreign to the average juror. The

jury, frantically grasping at complex forklift and mining concepts, could easily miss subtle distinctions

revealed on cross-examination and then drown in the untrue and the unproven. This is especially true

because the unreliable evidence here would have been presented in a format resembling a recreation

of the event that caused the accident. See Fusco v. General Motors Corp., 

, 264 (1st

Cir. 1993). Therefore, we find that the district court properly applied the law and exercised its

discretion in limiting Dr. Reed’s testimony and excluding the videotape.11

         Further, we find that Deere was not unduly prejudiced by the fact that the limitations and

exclusions came during trial. We note at the outset that the district court in the present case provided

notice on the record that it would revisit the admissibility issue. When Deere confronted the court

about reconsidering the admissibility of the videotape, the court responded as follows:



    11
      We note that part of the district court’s concern was the prejudicial nature of Dr. Reed’s
testimony. Even if the district court had found Dr. Reed’s testimony admissible under rule 703 and
Daubert, the court still would have been well within its discretion to exclude the testimony under rule
403. See 

(“Assuming the witness’s testimony is relevant and the
witness is qualified, and assuming his testimony has the appropriate factual basis and his methodology
is well founded, the only remaining inquiry is the balancing authorized in rule 403[.]").

                                                   18
       [W]e have had a number of discussions throughout the trial, some of them are on the
       record some are not, that the court was going to revisit this question, view the
       videotape, and make a final ruling. So to that extent your statement is not exact ly
       accurate. . . . [T]he Court advised counsel that it was going to view the videotape last
       week and make a ruling at that time. The issue is there is no issue of surprise.

       In spite of the reservation given by the court in this case regarding the initial ruling, the district

court had a duty to reconsider its erroneous denial of the plaintiffs’ motion in limine. See Xerox

Corp. v. Genmoora Corp., 

, 356 (5th Cir. 1989). In Xerox, the court held that a district

court abuses its discretion if it fails to reconsider a grant of summary judgment when later-developed

evidence demonstrates that the judgment was improper. Xerox suggests that the duty to reconsider

a ruling is triggered when the court receives posit ive proof that its prior ruling was erroneous.

Regarding the motion for summary judgment at issue in Xerox, the court noted:

       Because these documents reflect the existence of genuine issues of material fact, the
       trial judge knew positively at that time that his earlier grant of summary judgment
       could no longer be justified. He could not turn his back on such an overwhelming
       showing. He was obliged to reconsider and rescind the interlocutory summary
       judgment order. Failure to do so constituted an abuse of discretion.

court in the present case could only realize that Dr. Reed’s testimony and

videotape would confuse the jury after hearing some of his testimony and learning specifics about Dr.

Reed’s model. See 

. Thus, the court properly discharged its duty to

reconsider its prior ruling upon realizing that it was made in error. Deere, therefore, cannot claim

prejudice from the timing of the court’s reconsideration.

       B.      Excluded photographs.

       Though between 500 and 700 photographs were taken, the district court allowed each party

to introduce only ten photographs, even though most of the parties’ experts relied upon the pictures.


                                                    19
The district court’s limitation came after the commencement of trial. Deere argues that it was

prejudiced by the exclusion. None of the jurors had ever entered a salt mine and were not

knowledgeable about the mine, forklifts, and other equipment relevant to the litigation. Deere claims

that pictorial demonstratives were necessary to articulate the various concepts to the jury.

        It is well established that the district court is entitled the manage its court room and docket.

The limitation placed upon the number of pictures is justifiable, unless a party demonstrates that the

absence of a particular depiction has caused undue prejudice. That showing has not been made here.

We cannot say that the district court abused its discretion by requiring all parties to limit the number

of photographs introduced at trial.




III.    JURY FINDINGS.

        Deere contends that the jury was clearly erroneous in finding that Deere’s negligence was the

cause in fact of Guillory’s injuries. Deere had no duty to warn of obvious dangers and no duty to

warn where the warnings would be futile. Similarly, Deere asserts that the jury was clearly erroneous

in failing to find Guillory contributorily negligent.

        First, Deere claims that because Guillory and his co-workers qualify as sophisticated users of

the forklift, it had no duty to warn them about the retention system. Further, even if they cannot be

deemed sophisticated users, the danger was obvious or should have been obvious to the ordinary user

and obviated the need for a warning. Finally, Deere maintains that its failure to warn in the manual

was not the cause in fact of Guillory’s injury because the warning would not have been heeded. It

is undisputed that the mine employees never read the manuals.


                                                   20
         We disagree. The jury was amply armed with evidence establishing Deere’s failure to warn

its consumers of the improper listing of the 6000 pound fork as an appropriate replacement part for

the 380 forklift in the Deere catalogue. The jury’s finding of liability for failure to warn is properly

supported by the record. We, therefore, need not address the other failure to warn points raised by

Deere.

         Second, Deere argues that there is no evidence that the fork hit Guillory. No one actually saw

the fork hit Guillory, and Guillory did not know what hit him. Further, Deere’s expert in

reconstruct ion and in the field of biomechanics testified that the 210 pound fork could not travel

laterally eight feet and then hit Guillory. He testified that only the conveyor could have struck

Guillory and caused him to fall in the direction in which he was discovered.

         More than enough evidence in the record allowed the jury to infer that the fork hit Guillory.

Immediately after the accident Guillory said the fork hit him. The testimony from Guillory and Boutte

indicated that Guillory was not standing under the conveyor load or the forks. He was standing six

inches from the fork and an arm’s length from the conveyor before the accident. The testimony

clearly provided that the fork, not the conveyor lay inches from Guillory immediately after the

accident. True, no one saw the fork hit Guillory; however, enough circumstantial evidence surfaced

at trial to allow the jury to reject all other explanations and to infer that the fork was the cause in fact

of the accident.

         Finally, Deere contends that the jury erred in finding Guillory free from fault. Guillory was

aware that forks had fallen off the forklift, and he never reported it. Guillory positioned himself near

the conveyor load. Also, Guillory failed to use tag lines and failed to use other safer equipment to




                                                    21
complete the task. Finally, Guillory was aware of the purpose of the retaining system but failed to

use it.

          We cannot agree with Deere’s assertions. Ample evidence supports the jury’s conclusion that

Guillory was not comparatively negligent. The jury received conflicting testimony regarding whether

Guillory was comparatively negligent. For example, the MSHA representative’s report indicates that

during an interview with Guillory after the accident Guillory said he saw forks fall from forklifts and

that he was aware of forks falling eight to ten times a week. Guillory testified to the contrary that

he had no knowledge of the falling forks. Additionally, no miners testified that they related their

stories regarding the falling forks. The jury was free to believe Guillory was more credible than the

MSHA representative.

          Further, other evidence supports the jury’s finding that Guillory did not contribute to his

injury. Guillory had an accident-free record with Domtar. In the over twenty years that he worked

in the salt mine, he had not even smashed a finger. Guillory positioned himself six inches from the

load and an arm’s length from the conveyor; he was not standing underneath either object at the time

of the accident. Additionally, witnesses testified that the other equipment would not have been better

suited or safer to move the section of the conveyor. Guillory testified that Domtar did not have a

cherry picker suitable for the conveyor job. The stinger,12 which Guillory built, was not appropriate

because the roads in t he area were too narrow. Moreover, the jury could have discounted

contributory negligence regarding Guillory’s failure to use a retaining system. The plaintiffs’ expert,

Robert Lipp, testified that he found bolts being used to secure forks in the John Deere dealership

instead of the retaining system. Further, testimony from miners such as Boutte indicated that forks


   12
        Guillory explained that a stinger is an extension hooked on the forks in order to reach farther.

                                                    22
had not fallen when a bolt was present on the backing plate. The jury was free to find that Guillory

was not contributorily negligent.

        The jury chose to believe Guillory and other witnesses whose testimony indicated that

Guillory was hit by the fork, that Deere breached its duty to warn, and that Guillory did not

contribute to his injury. These findings were based upon credibility calls which are completely within

the province of the jury. We cannot disturb the jury’s credibility determinations. See United States

v. Garcia, 

, 398 (5th Cir. 1996) (noting that the appellate court must accept the credibility

choices supporting the verdict); and United States v. Straach, 

, 237 (5th Cir. 1993)

(noting that the appellate court cannot weigh and assess the credibility of the witnesses). We cannot

say that any of the jury’s determinations were clearly erroneous.

IV.     RULE 16 SANCTIONS.

        Deere argues that the district court abused its discretion by imposing sanctions pursuant to

rule 16. Deere alleges that it informed the court that settlement possibilities were doubtful. Further,

Deere increased its initial settlement offer by 100% in an attempt to achieve a settlement. Moreover,

the law does not support the imposition of sanctions simply because Deere did not offer what the

court considered “serious money.”

        Federal Rule of Civil Procedure 16(f) authorizes the imposition of sanctions “if a party or

party’s attorney fails to participate [in a pretrial settlement conference] in good faith.” Fed. R. Civ.

Pro. 16(f); see also Newton v. A.C. & S., Inc., 

, 1126 (3d Cir. 1990). We cannot

allow parties to waste the court’s dispute resolution assets by pretending to support settlement while

never intending to settle the case. The district court provided detailed reasons for the imposition of

sanctions in its order. The court emphasized the resources the court and parties wasted orchestrating


                                                  23
the futile settlement conference and identified why it concluded that Deere approached the conference

in bad faith. The court made the following conclusions regarding Deere’s lack of good faith:

         As the settlement conference progressed, it became apparent that Deere did not intend
         to make any substantial contribution to a settlement fund, and that Deere had never
         intended to do so. In fact, Mr. Kraft informed the court that Deere had always taken
         the position that this was a case which they would rather try. In addition, Deere had
         never considered settlement a realistic possibility because the present value of
         plaintiff’s workmen’s compensation benefits was so large. In order for the plaintiff
         to settle, and therefore give up these benefits, the settlement fund would have to be
         in the millions of dollars. In order to achieve such a fund, Deere, Harlo and the
         intervenor would each have to make a substantial contribution to the fund. This,
         Deere had never been prepared to do.

In a later memorandum ruling, the district court noted that any time before the settlement conference

Deere could have informed the court of its position and t he court would have canceled the

conference. Prior to the conference, the district court requested written statements from the parties

regarding their positions and settlement prospects. Deere gave no indication that it believed

settlement to be a useless endeavor.

         The district court could infer a lack o f good faith from actions and statements made by

Deere’s representatives. We find that the record does not demonstrate that the district court

sanctioned Deere for failing to make a serious offer,13 but rather it imposed the sanctions because


    13
      Refusal to make a bonafide offer at a scheduled settlement conference is an impairment to
reaching settlement that can be likened to attending a conference unprepared. See In re Novak, 

, 1404-06 (11th Cir. 1991) (discussing wastefulness of resources in the context of
participants coming to settlement conferences unprepared). However, imposition of sanctions for
making bad faith settlement offers is within the spirit of rule 16 only where the court receives
information demonstrating that the party attended the conference intending not to reach settlement.
In the absence of objective evidence of bad faith, sanctions cannot be justified simply because a party
does not offer what the court considers a “bonafide offer” or “serious money.” See Dawson v.
United States, 

, 897 (5th Cir. 1995) (reversing sanctions imposed on good faith grounds
because the defendants failed to make a monetary settlement, and noting that “there is no meaningful
difference between coercion of an offer and coercion of a settlement”); see also G. Heileman Brewing
Co. v. Joseph Oat Corp., 

, 653 (7th Cir. 1989) (en banc) (suggesting that sanctions

                                                  24
Deere concealed its true position that it never intended to settle the case. Deere’s refusal to make

a settlement offer with a realistic potential of being accepted by the plaintiff confirmed the Deere

representative’s statements regarding settlement. Under these circumstances, the imposition of

sanctions was well within the court’s discretion.14

                                          CONCLUSION

        For the foregoing reasons, we AFFIRM summary judgment granted in favor of Domtar

Industries. Likewise, we AFFIRM the district court’s and magistrate’s evidentiary rulings and

$8,500.00 award for sanctions against Deere. Finally, we AFFIRM the jury’s findings and its award

of $6,982,315.00 in favor of the plaintiffs. Costs are rendered against Deere.




cannot be based on the refusal to make a monetary offer); Hess v. New Jersey Transit Rail
Operations, Inc., 

(2d Cir. 1988) (suggesting that sanctions would be inappropriate
where the defendant failed to make what the court considered a “bonafide offer”); and Kothe v.
Smith, 

(2d Cir. 1985) (fine reversed where the defendant failed to offer the amount
recommended by the court).
   14
    Deere takes issue with statements that it requested the settlement conference. Whether Deere
requested the conference or merely attended, it had a duty to attend and to participate in good faith.

                                                 25