Felicia Blake was permanently crippled in an automobile collision on November 4,1992, when the driver of a taxicab lost control of the cab, crossed over into the oncoming lane of traffic, and crashed head-on into the 1988 Chevrolet Spectrum automobile driven by Blake. Blake sued the driver, the taxicab company, and General Motors Corporation, the maker of the Spectrum, claiming that crippling leg injuries she suffered in the accident were caused by a defective seat belt in the Spectrum which failed to restrain her. The other defendants were not involved in the trial, in which a jury returned a verdict in favor of Blake in the amount of $3,800,000.
1. In its first enumeration of error, General Motors (GM) claims the court erred by denying a continuance sought again by GM on the day of trial, on the basis that it was surprised and prejudiced by *427Blake’s identification of an expert witness the week before, which witness Blake used allegedly to present a new theory of the case.
A motion for continuance of a trial is properly addressed to the “sound legal discretion” of a trial judge, who is in control of the management of the case in court. OCGA § 9-10-167; Work Clothes Outlet v. M & S Purchasing, 188 Ga. App. 179, 181 (2) (372 SE2d 509) (1988). The appellate court, which is far removed from the unfolding development in the life of a case in court and does not participate in its ongoing journey, is therefore bound to respect the exercise of the trial court’s discretion and reverse it only if it is “manifestly abused.” Simmons v. Simmons, 265 Ga. 183, 184 (453 SE2d 696) (1995).
This Court, like the Supreme Court as discussed in Mills v. State, 188 Ga. 616, 623-625 (4 SE2d 453) (1939), is a court for the correction of errors of law made by the trial courts.1 Oasis Goodtime Emporium I v. Cambridge Capital Group, 234 Ga. App. 641, 642 (1) (507 SE2d 823) (1998). As was said of the Supreme Court in Mills,
[w]e can not undertake to correct errors of fact. We are prohibited from so doing, and can only decide law questions. But no such limitation is placed on the judges of the superior [or, in this case, state] court. . . . Ours is the function merely to see that the law is followed. Theirs is the solemn responsibility to see to it that justice is administered according to the evidence as well as the law. ... A judge of the superior [or state] court has authority which [the Supreme Court and likewise this Court] [do] not possess.
188 Ga. at 623-624. The great power entrusted to the trial judge is a judicial discretion, justified by skill and by functional commitment to the singular purpose to serve “the great object for which courts are established — to administer justice according to law.” Id. at 624.
The judicial conscience is applied when a trial judge is called upon to exercise discretion. To say that as a matter of law that discretion was abused is a matter of no small moment. As related to denial of a trial continuance, not only does a legal conclusion of abuse derail the case from its journey to finality and require costly and lengthy and wearing retrial, but with hindsight it second-guesses a trial court in managing the case.
The denial of a continuance in this case did not exceed the broad bounds of discretion. We cannot say that it was outside the law’s contemplation to require General Motors Corporation to proceed.
The complaint was filed in October 1994, and the first motion for *428continuance sought a delay beyond Monday, April 28, 1997. The motion had been filed on Thursday, April 17. It gave as reason that GM did not yet know the identity of plaintiff’s experts on the subject of seat belt manufacture, design, and performance and thus could not depose them to discover their opinions on those subjects as well as vehicle speed, force of impact, occupant kinematics and injury causation. The motion was argued in chambers, without being reported, on Tuesday, April 22 (defendant’s recollection) or Thursday, April 24 (the court’s recollection). At that time it was denied.
Defendant renewed the motion in a five-sentence argument on Monday, trial day, based on a vague reference to deposition testimony given on Friday. Plaintiff responded, defendant briefly replied, and the court, unpersuaded by the recitation, again denied the motion. Defendant had filed a number of other motions in the interim between filing the motion for continuance and the commencement of trial, and a pretrial order had been entered. The consolidated pretrial order entered April 23 states: “Since the case will take longer than one week to try, General Motors requests that it be specially set or held in place until it becomes the first case on a trial calendar.”
The renewal of the motion was based on the fact that on the previous Friday, GM “deposed for the first time [plaintiffs] liability expert, and that’s the expert who is testifying as to what they believe was the defect in this automobile, and who is an expert who had been identified a week or so earlier.” Defendant stated that at the deposition it “learned for the first time” the theory of defect which plaintiff would travel on, which it had not known previously and that there were other experts still to be deposed. GM did not explain how the theory differed or how it would be prejudiced. It simply stated that severe prejudice to defendant would result unless there was time for additional discovery.
Plaintiff responded by pointing out that there was no change in the allegation of negligence that had been advanced all along, i.e., that the seat belt did not hold her or her passenger upon impact. She did not know exactly why it did not hold and thought at first it was attributable to a design or manufacturing defect in the buckle. But that had been fairly conclusively eliminated by the experts during arbitration as the locus of the failure. Plaintiff pointed out that a number of experts on both sides of the case had examined the apparatus and that although there was a change in the theory pinpointing exactly what failed, the substance of the allegation remained the same. She assured the court that although one expert on each side still had to be deposed, that could be done without again delaying the trial. Plaintiff expressed the fear that, because of what had been said at the chambers hearing, defendant would further elongate and *429expand the whole proceeding by pursuing a motion for summary judgment and was using the motion for continuance to achieve that purpose. Defendant replied briefly that its try for summary judgment was based on a lack of expert evidence of defect, and that additional discovery was necessary to prevent prejudice to GM.
The court took all of this into account, including the fact the case had been set for trial once before but had come off the calendar with the consent of all parties, and it denied the renewal motion.
The court cannot be faulted. First, GM’s argument for continuance did not elucidate the “surprise” of the new theory allegedly discovered the preceding Friday. The key to its request to stop the process was shrouded in the single sentence that it had learned that “the theory of defect had changed from the pleadings and responses to discovery . . . previously . . . served and filed.” Without giving the court more detail of the magnitude or nature of this change, GM could have had little expectation that the court would consent. The superficial argument, unsupported by brief or written motion, did not compel a decision to comply with the request. The trial court can hardly be expected to fathom the depth of what defendant contended it was faced with, absent more than the sparse explanation offered.2 As for the discovery, plaintiff showed that what remained to be done could be accomplished without interrupting the trial schedule. Defendant did not disagree.
Second, the evidence accepted by the jury clearly shows that the seat belt failed to perform its function as anticipated, which led to the shattered leg, confinement to a wheelchair, and permanent disability of plaintiff. True, she was uncertain during trial preparation and at times inconsistent as to how the seat belt failed, but she steadfastly maintained that it did not restrain her as it was designed to do, given that she had the belt on and fastened at impact. Throughout the course of the trial discovery and preparation she kept trying to pin down exactly what part of the mechanism went wrong. She thought for a long time that the buckle did not hold but rather unlatched. Then she realized that was not the problem. She kept searching and, before trial, determined that the problem was unstopped spooling which resulted in an overloaded cinch and consequent failure to hold her back. That theory was actively investigated by both sides on the Friday before the Monday trial was to begin.
Firestone Tire &c. Co. v. King, 145 Ga. App. 840 (244 SE2d 905) (1978), involves similar facts. The tire at issue clearly failed, resulting in a blowout and injuries to plaintiff. The plaintiff’s experts *430“could not specify the exact nature of the defect which, in their opinion, caused the tire to fail. Instead they merely speculated as to possibilities, such as contamination of materials or undervulcanization.” Id. at 842 (1). This Court ruled in favor of plaintiff, holding:
We do not agree that it was necessary for the [plaintiff] to specify the nature of the defect in order to meet her burden of proof. It has often been held that the existence of a manufacturing defect in a products liability case may be inferred from circumstantial evidence. [Cits.]
Id. King indicates that precise specificity is not essential for recovery so long as plaintiff can show that the seat belt did not operate as intended and this was the proximate cause of her injuries. That being so, the absence of precise specificity when unknown gives no cause for trial delay. In the present case, GM did know the specifics several days before trial but, more importantly, could have tested the entire seat belt apparatus for months, if not more than a year, before.
That leads to the third reason. GM had every opportunity to completely examine the belt, its retractor and its cinch latch plate as well as its buckle, particularly after it eliminated the buckle as the problem by the time of the arbitration hearing in August 1996. Its expert for the arbitration, a former GM testing engineer, examined Blake’s seat belt system and found no deformation of the buckle or defect so that it would inertially unlatch. He did find marks on the guide loop assembly and the seat belt webbing and deformation of the guide loop assembly bushing at the attaching bolt hold. He also found that the body anchorages for the belt system bolts for the driver’s seat belt system were deformed. He attributed all of this to the collision impact.
Fully aware that the issue was why did the seat belt not restrain plaintiff, if it did not do so,3 GM apparently tactically decided not to conduct a fuller and more complete investigation of the apparatus. Defendant should not be permitted to avoid learning the facts by lack of thoroughness and then claim surprise. Its experts were as equipped as plaintiff’s, if not more so, to discover the cause. It had the capacity and resources to do all kinds of examinations and tests on the seat belt mechanisms. It apparently did not explore the possibility that the source was the retractor.
Cases concerning surprise are not on point because in this *431instance the surprise was of GM’s own making. See, e.g., Jones v. Atkins, 120 Ga. App. 487, 491 (1) (171 SE2d 367) (1969). The rules of procedure are concerned with prevention of surprise, to be sure, but the surprise must be genuine and justified. Kamensky v. Stacey, 134 Ga. App. 530, 532 (1) (215 SE2d 294) (1975) states:
[Wjhere the complaining party cannot legitimately claim surprise, either because he knew of the existence of the [facts to be testified to] or had equal means of knowing, it is not error to fail to invoke the sanctions of postponement, mistrial, barring the witness, etc. [Cits.]
Danforth v. Danforth, 156 Ga. App. 236 (274 SE2d 628) (1980), a case dealing with a reasonable time for discovery under OCGA § 9-11-40 (a), is inapposite. “[D]efendant had had only 29 working days of discovery and 18 of those had already passed when he was informed that he would have only 11 more, and only 5 more after plaintiff was to answer the defendant’s first interrogatories.” Id. at 238. What is on point is that the Court went on to say: “If the defendant . . . had already had a reasonable and adequate time for discovery . . ., then the defendant might [not be entitled to a continuance].” Id.
The judgment, based on a jury verdict for plaintiff after a trial which consumed the better part of nine days, should not be reversed nor should a new trial be required.
2. We have carefully reviewed the other enumerations of error and find them each to be without merit.
Judgment affirmed.
McMurray, P. J, Pope, P. J., Blackburn, Ruffin and Eldridge, JJ, concur. Andrews, J., dissents.The Georgia Constitution does not expressly so state but does describe the nature and function of this Court as being “a court of review.” Ga. Const, of 1983, Art. VI, Sec. V, Par. III.
The expansion of this on appeal, given defendant’s hindsight and reflection, is not relevant to the basis upon which the trial court was called to rule.
There was ample evidence early on that the plaintiff and her three-year-old goddaughter, who was killed, were wearing seat belts. Even defendant’s expert so concluded. Yet defendant claimed surprise and says it “was prepared to defend against a case that assumed plaintiff was not buckled, not a case that assumed she was.”