This case arises from the tragic death of Jordan Garris. In June, 1999, Jordan shot himself while playing with his father’s handgun. Jordan’s mother, petitioner here, seeks to hold the manufacturer of the handgun, respondent Sturm, Ruger & Co. (Sturm Ruger), liable for Jordan’s death. The Circuit Court for Baltimore City, by granting respondent’s motion for summary judgment, found no liability. A divided Court of Special Appeals affirmed. Halliday v. Sturm, Ruger & Co., 138 Md.App. 136, 770 A.2d 1072 (2001). We shall do likewise.
BACKGROUND
The handgun in question is a Ruger P89 semi-automatic pistol. To fire the gun, one must place a loaded magazine into it, pull the slide at the top of the gun as far to the rear as possible and then release it, ensure that a safety lever is in the “fire” position, and then pull the trigger. Even when loaded, the gun will not fire unless the trigger is pulled with the safety lever in the “fire” position.
Jordan’s father, Clifton Garris, purchased the gun in March, 1999, from On Target, Inc., a retail firearms store. With the purchase of the gun came an instruction manual, the offer of a free safety course, which Garris declined, a pamphlet entitled “Youth Handgun Safety Act Notice” published by the Federal Bureau of Alcohol, Tobacco and Firearms, a lock box in which to store the gun and the magazine, and a padlock for the box. *189There was a dispute as to whether On Target recommended that Garris purchase a separate trigger lock for the gun — in an affidavit filed in support of the motion for summary judgment, the salesman from On Target stated that such a device was recommended, but in a responding affidavit, Garris said that it was not.
The instruction manual provided multiple warnings and instructions regarding the storage and use of the gun. On the cover of the manual, and embossed on the barrel of the gun itself, was an admonition to read the manual before using the gun. Among other warnings and instructions in the manual is a highlighted box entitled “WARNING — STORAGE” in which, in red letters, is the statement “Firearms should always be stored securely and unloaded, away from children and careless adults” and the statement, in capital letters, “STORE SECURELY AND UNLOADED.” In the part on “THE BASIC RULES OF SAFE FIREARMS HANDLING,” which itself is in red capital letters, is a section headed, in red capital letters, “FIREARMS SHOULD BE UNLOADED WHEN NOT IN USE,” and in that section is the warning:
“Firearms and ammunition should be securely locked in racks or cabinets when not in use. Ammunition should be safely stored separate from firearms. Store your firearms out of sight of visitors and children. It is the gun owner’s responsibility to be certain that children and persons unfamiliar with firearms cannot gain access to firearms, ammunition, or components.”
Garris signed an acknowledgment that the On Target salesperson explained the instruction manual, the safety lever, and the action of the gun. The Youth Handgun Safety Act Notice warned Garris, in highlighted letters, that the misuse of handguns was a leading contributor to juvenile violence and fatalities and that “[sjafely storing and securing firearms away from children will help prevent the unlawful possession of handguns by juveniles, stop accidents, and save lives.”
Garris disregarded virtually every one of these warnings and opportunities. He did not store either the gun or the *190magazine in the lock box but rather placed the gun under his mattress and kept the loaded magazine on a bookshelf in the same room, so that it was visible and accessible to Jordan. Jordan found the handgun under his father’s mattress. He also found the loaded magazine. From watching television, the child knew how to load the magazine into the gun, and he did so. While playing with the gun, he apparently pulled the slide and thereby placed a bullet into the chamber. Either the safety lever was in the “fire” position already or Jordan moved it there. He then pulled the trigger, shot himself in the head, and died two days later. He was three years old.
Petitioner filed suit in the Circuit Court for Baltimore City against Sturm Ruger and On Target, alleging both a design defect in the gun (Counts I and II) and inadequate warnings (Counts III and IV). We are no longer concerned here with Counts III and IV, which have been abandoned, or with the action against On Target, which petitioner dismissed. Our focus, therefore, will be only on Counts I and II, which sought liability against Sturm Ruger based on the allegation that the gun was defective and unreasonably dangerous when placed in the stream of commerce.
Petitioner alleged that the gun was defective and unreasonably dangerous because its design “failed to incorporate reasonable devices to prevent its use by young children,” in particular “one or more of the following: a grip safety, a heavy trigger-pull, a child-resistant manual safety, a built-in lock, a trigger lock, and/or personalized gun technology that would have substantially reduced the likelihood that a child could fire the gun.... ” Citing data released by the Centers for Disease Control and Prevention to the effect that 1,641 children under ten were accidentally killed by handguns between 1979 and 1996, petitioner averred in her complaint that “[i]t was foreseeable that the gun would be found and handled by a young child, and that it would be fired by a young child, with resulting foreseeable grievous or fatal injury to the child and/or others.” Petitioner contended that the handgun industry was aware of the problem of young children finding and injuring themselves with handguns and, in the 1880’s, had *191developed a childproof grip safety, but that Sturm Ruger manufactured the gun without that, or any other, childproof device.
Sturm Ruger responded to the complaint with a motion to dismiss or, in the alternative, for summary judgment, the latter based on the assertions that (1) as a matter of law, the gun was not in a defective condition or unreasonably dangerous, and (2) it was used in a manner that was contrary to the clearly worded instructions and warnings that accompanied the product when sold, and was therefore mis used. Attached to a memorandum that accompanied the motion were copies of the instruction manual, the Youth Handgun Safety Act Notice, an affidavit from the On Target salesman attesting, among other things, to his recommendation that Garris purchase a trigger lock, and a picture of the lock box. Sturm Ruger argued that the gun did not malfunction but rather performed exactly as it was designed to function, and that the accident occurred because of Garris’s failure to heed clear warnings.
In response, petitioner argued that it is “inconsistent with the proper function of a gun to design it so that it can be fired by young children.” The lack of child-resistant features, she claimed, made the gun foreseeably dangerous to small children, and, “[bjecause firing by small children is not one of the proper functions of a gun, a gun fired by a small child has not performed properly.” She urged that Garris’s failure to heed the warnings and keep the gun securely locked was not a defense to liability because that also was foreseeable. In that regard, she attached an affidavit from Garris that, when he bought the gun, he “glanced through the owner’s manual” but did not recall reading “any warnings regarding storage of the gun.” Also attached was an affidavit from Stephen Teret, a professor at the Johns Hopkins School of Hygiene and Public Health, who recounted statistics dealing with accidental gun-related deaths of children and evidence that a substantial number of gun owners, including those with children in the home, store their guns in an unsafe manner. Based on this data, Teret offered the opinion in his affidavit that reliance by Sturm Ruger on the instruction manual and the provision of a *192lock box was inadequate to protect children from the risk of unintended gun death.
The essence of petitioner’s case was that, when dealing with design defects in a strict liability claim, the court should apply a “risk-utility” analysis in lieu of a “consumer expectation” test and hold that the gun in question failed that preferred test because (1) the risk of excluding child safety features outweighs the utility of that exclusion, and (2) alternative safer designs could have been adopted economically. She argued to the Circuit Court that “[t]he central thing that Sturm Ruger did wrong in designing this gun ... is to sell a gun a three year old could shoot.” The court rejected that argument, holding instead that, under Maryland law, the risk-utility test applied only when the product malfunctioned and that the gun in question did not malfunction. Although at times indicating that the warnings given in the instruction manual and Federal notice were not really adequate, the court concluded that Garris clearly knew that the gun was dangerous.
In affirming the summary judgment, the panel majority for the Court of Special Appeals first cast doubt on the admissibility of much of the material included in the Teret affidavit, noting that, although, in support of Teret’s ultimate conclusion that an accidental shooting by a child was foreseeable, the affidavit mentioned eleven studies and one editorial, none of those documents were attached to the affidavit, and thus the court had no basis upon which to assess the reliability of Teret’s conclusion. Halliday, supra, 138 Md.App. at 154, 770 A.2d 1072. The panel pointed out that the debate over whether the warnings actually given to Garris were adequate was irrelevant in light of petitioner’s argument that no warnings would be adequate — that the only thing that could make the gun not defective was the inclusion of a child-resistant device of some kind.
The core of the intermediate appellate court’s ultimate conclusion was the holding of this Court in Kelley v. R.G. Industries, Inc., 304 Md. 124, 497 A.2d 1143 (1985), that the risk-utility test did not apply to a product, including a gun, *193that did not malfunction and its own determination that the gun in this case operated exactly as it was supposed to operate and therefore did not malfunction. Alternatively, the court stated that, even if it were to find that the failure to include a child safety device constituted a design defect, there would be no liability because of misuse. As to that, the court held that Garris’s failure to heed the clear warnings supplied with the gun relating to its safe storage constituted a misuse that was not reasonably foreseeable by Sturm Ruger. Two judges on the nine-judge en banc panel dissented. They concluded that the consumer expectation test that we applied in Kelley was no longer Maryland law, that the alleged design defect should be considered under a risk-utility analysis, and that, under that analysis, there was a triable issue.
DISCUSSION
The principal issue presented here is whether, in examining whether a product in general, or a handgun in particular, is defective for purposes of a strict liability action, this Court should continue to apply the “consumer expectation” test, as urged by Sturm Ruger, or should adopt instead a version of the “risk-utility” analysis, as requested by petitioner. It would be helpful, therefore, at the outset, to define these two standards.
The consumer expectation test emanates from § 402A of the Restatement (Second) of Torts which, under certain circumstances, makes the seller of a product that is in a “defective condition unreasonably dangerous” to the consumer liable for the physical harm caused to the consumer by that product. The test defines what is meant by the terms “defective condition” and “unreasonably dangerous.” Comment g to § 402A defines “defective condition” as a “condition not contemplated by the ultimate consumer, which will be unreasonably dangerous to him.” Comment i, in speaking to the term “unreasonably dangerous,” states that the article must be dangerous “to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its char*194acteristics.” Thus, Prosser and Keeton explain that, under the consumer expectation or contemplation test set forth in § 402A, a product is defectively dangerous “if it is dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchased it with the ordinary knowledge common to the community as to the product’s characteristics.” W. Page Keeton et al„ Prosser and Keeton on the Law of Torts, § 99, at 698 (5th ed.1984).
The “risk-utility” test, which has been applied principally to alleged defects in the design of a product, regards a product as defective and unreasonably dangerous, for strict liability purposes, if the danger presented by the product outweighs its utility. Where this test is applied, the issue usually becomes whether a safer alternative design was feasible, for, if so, that would likely alter the balance by reducing the extent of the danger. Indeed, § 2 of the Restatement (Third) of Torts: Product Liability, which adopts this test for design defect cases, goes directly to that issue:
“A product ... is defective in design when the foreseeable risks of harm posed by the product could have been reduced or avoided by the adoption of a reasonable alternative design by the seller ... and the omission of the alternative design renders the product not reasonably safe.”
This Court first adopted the concept of strict liability, as articulated in Restatement (Second) of Torts § 402A, in Phipps v. General Motors Corp., 278 Md. 337, 363 A.2d 955 (1976). Paraphrasing the language in that section, we said that, to recover in an action for strict liability,
“it must be established that (1) the product was in a defective condition at the time that it left the possession or control of the seller, (2) that it was unreasonably dangerous to the user or consumer, (3) that the defect was a cause of the injuries, and (4) that the product was expected to and did reach the consumer without substantial change in its condition.”
Id. at 344, 363 A.2d at 955.
Expounding upon those conditions and the nature of the action, we added that, for a seller to be liable under § 402A, *195the product must be both in a “defective condition” and “unreasonably dangerous” at the time it was placed on the market and that both of those conditions were “explained in the official comments in terms of consumer expectations.” In that regard, we called attention, in particular, to Comments g and i to § 402A. Id. at 344, 363 A.2d at 955.
Phipps involved an alleged design defect in an automobile— latent defects in the accelerator mechanism, the carburetor and its components, and the motor mounts — that caused the accelerator to stick and the car to be driven off the road. We observed that, when dealing with an alleged error in the manufacturing process, there was less difficulty in applying the consumer expectation test, but that when “the alleged defect is the result of the design process so that the product causing injury was in a condition intended by the manufacturer, the test has proved more difficult to apply.” Id. at 344-45, 363 A.2d at 959. That difficulty, we added, had caused some commentators to suggest that the theory of strict liability articulated in § 402A was not really applicable to design defects and that a design defect case still required “a weighing of the utility of risk inherent in the design against the magnitude of the risk,” which was a test generally affiliated with negligence actions. Id. at 345, 363 A.2d at 959. We observed, however, that there were certain kinds of conditions that, whether caused by design or manufacture, would never be said to involve a reasonable risk, giving as examples the kinds of situations then before us-failing brakes (Elmore v. American Motors Corporation, 70 Cal.2d 578, 75 Cal.Rptr.652, 451 P.2d 84 (1969)), separating drive shafts in new cars (Henningsen v. Bloomfield Motors, Inc., 32 N.J.358, 161 A.2d 69 (1960)), and sticking accelerators. Those kinds of conditions, we said, whether resulting from a defect in design or manufacture, were defective and unreasonably dangerous “without the necessity of weighing and balancing the various factors involved.” Id. at 346, 363 A.2d at 959.
Although we applied the consumer expectation test set forth in § 402A to the situation before us in Phipps, that language certainly left open whether there were other conditions in *196which, in a design defect case, a risk-utility analysis might be appropriate, and, indeed, we also stated in Phipps that “in some circumstances the question of whether a particular design is defective may depend upon a balancing of the utility of the design and other factors against the magnitude of that risk.” Id. at 348, 363 A.2d at 961. With that opening, the Court of Special Appeals and the U.S. Court of Appeals for the Fourth Circuit alluded to or applied such an analysis in a number of design defect cases, usually involving the absence of some form of safety device that would, allegedly, have made a potentially dangerous product less so. See, for example, Banks v. Iron Hustler Corp., 59 Md.App. 408, 475 A.2d 1243 (1984) (failure to provide shield on conveyor belt to guard against contact with “nip” points where belt came into contact with rollers); Troja v. Black & Decker Mfg. Co., 62 Md.App. 101, 488 A.2d 516 (1985) (absence of safeguard to prevent radial arm saw from operating when guide fence removed); Valk Manufacturing v. Rangaswamy, 74 Md.App. 304, 537 A.2d 622 (1988) (protruding hitch in snowplow); Singleton v. International Harvester Co., 685 F.2d 112, 115 (4th Cir.1981) (holding that, under Phipps, design defects were to be treated as negligence rather than under strict liability principles).
We revisited the issue of which test to apply in Kelley, supra, 304 Md. 124, 497 A.2d 1143, in connection with handguns. The plaintiff there was injured when he was shot during a robbery attempt. He sued the gun manufacturer; claiming, among other things, that the gun was “abnormally dangerous.” Responding to questions certified to us by the U.S. District Court, we opined on whether (1) the manufacturer of a handgun, in general, was liable under any strict liability theory to a person injured as a result of the criminal use of its product, and (2) the manufacturer of a particular category of small, cheap handgun, sometimes referred to as “Saturday Night Specials,” which were regularly used in criminal activity, was strictly liable to a person injured by such a handgun during the course of a crime. In answering the first question, we cited Phipps for the proposition that “[i]n determining whether a product is defective, in its design or manufacture, *197Maryland cases have generally applied the ‘consumer expectation’ test,” id. at 135, 497 A.2d at 1148, and we concluded that “[a] handgun manufacturer or marketer could not be held liable under this theory.” Id. at 136, 497 A.2d at 1148. We explained:
“[A] handgun is not defective merely because it is capable of being used during criminal activity to inflict harm. A consumer would expect a handgun to be dangerous, by its very nature, and to have the capacity to fire a bullet with deadly force. Kelley confuses a product’s normal junction^ which may well be dangerous, with a defect in a product’s design or construction.”
Id. (emphasis in original).
We noted, then, that, following the decision of the California Supreme Court in Barker v. Lull Engineering Co., 20 Cal.3d 413, 143 Cal.Rptr. 225, 573 P.2d 443 (1978), the risk-utility test of which we spoke in Phipps had been adopted, at least as an alternative test to the consumer expectation test, by a number of courts, but we concluded that that test was inapplicable to the situation before us, as it is “only applied when something goes wrong with a product.” Kelley, supra, 304 Md. at 138, 497 A.2d at 1149. We added that, in the cases in which the risk-utility test had been applied, the product had malfunctioned in some way, but that,
“in the case of a handgun which injured a person in whose direction it was fired, the product worked precisely as intended. Therefore, the risk/utility test cannot be extended to impose liability on the maker or marketer of a handgun which has not malfunctioned.”
Id.
On that premise, we concluded that, “regardless of the standard used to determine whether a product is ‘defective’ under § 402A, a handgun which functions as intended and as expected is not ‘defective’ within the meaning of that section,” noting that “[t]his has been the consistent conclusion in other jurisdictions which have confronted the issue.” Id. That conclusion, in turn, led to the ultimate holding that, under existing *198strict liability principles, as applied in Maryland, a handgun manufacturer or marketer was generally not liable for gunshot injuries resulting from a criminal’s use of the product.
Having so concluded, we took note that the common law was dynamic and could be changed, suggesting the prospect of adopting new theories that might create liability, but we pointed out that “we have consistently recognized that common law principles should not be changed contrary to the public policy of the State set forth by the General Assembly.” Id. at 141, 497 A.2d at 1151. We looked, then, to the handgun laws then in force, which allowed persons to own and possess handguns and, under certain circumstances, to carry them. From that, we concluded that “to impose strict liability upon the manufacturers or marketers of handguns for gunshot injuries resulting from the misuse of handguns by others, would be contrary to Maryland public policy as set forth by the Legislature.” Id. at 144, 497 A.2d at 1153.
Our response to the second question posed was different. We determined that there was a “limited category of handguns which clearly is not sanctioned as a matter of public policy” and that to impose strict liability upon the manufacturers and marketers of those handguns, which we denoted as “Saturday Night Specials,” would not be contrary to the public policy set by the General Assembly. Id. Those kinds of guns, characterized by short barrels, low weight, easy concealability, cheap quality, inaccuracy, and unreliability, rendered them particularly attractive for criminal use but virtually useless for any legitimate purpose. Id. at 144-46, 497 A.2d at 1153-54. After surveying both Federal and State legislation, we determined that those types of guns really were in a separate category, that their use for criminal purposes was entirely foreseeable by their manufacturers and marketers, and that holding such manufacturers and marketers strictly liable for injuries to innocent persons from the criminal misuse of those guns would be consistent with public policy. Whether a particular gun fell within that limited category, we said, was an issue of fact for a trial court to determine.
*199In Simpson v. Standard Container Co., 72 Md.App. 199, 527 A.2d 1337 (1987), the Court of Special Appeals properly followed our pronouncements in Kelley. The case involved a gasoline can that, despite clear warnings to the contrary, the buyer stored, full of gasoline, in his basement, where a four-year-old child found and opened it, causing the gasoline to spill, ignite, and severely burn the child. The action against the manufacturer alleged that the can was defective because it was designed without a childproof cap, and the argument was presented that liability should attach under a risk-utility analysis because (1) the danger of a child spilling gasoline and igniting a fire was foreseeable, (2) a container equipped with a childproof cap was available in the industry at nominal cost, and (3) the product was defective because the risk outweighed the utility of the container without the cap. Affirming the dismissal of the action, the intermediate appellate court correctly noted that “[t]o determine whether a product is defective in its design, Maryland cases have generally used the ‘consumer expectation’ test.” Id. at 203, 527 A.2d at 1340. It also concluded that, even if a risk-utility analysis was employed, the result would be the same “because misuse of the product and failure to read or follow the product’s warnings and instructions are defenses to strict product liability.” Id. at 204-05, 527 A.2d at 1340.
In a string of other cases, however, the Court of Special Appeals continued to apply the risk-utility test in design defect cases involving the lack of a safety device, sometimes, unfortunately, by misconstruing, side-stepping, or ignoring what we said in Kelley. See C & K Lord v. Carter, 74 Md.App. 68, 86, 536 A.2d 699, 707 (1988) (language in Kelley that risk-utility test does not apply in absence of malfunction is itself inapplicable where alleged design defect is failure to include safety device); Ziegler v. Kawasaki Heavy Industries, 74 Md.App. 613, 622-23, 539 A.2d 701, 706 (1988), cert. denied, 313 Md. 32, 542 A.2d 858 (1988) (same); Klein v. Sears Roebuck, 92 Md.App. 477, 485, 608 A.2d 1276, 1280 (1992) (in design defect case, § 402A requires “weighing of the utility risk inherent in the design against the magnitude of the risk,” *200citing Phipps but not Kelley); and Nissan v. Nave, 129 Md.App. 90, 118-19, 740 A.2d 102, 117 (1999) (same). We, however, have made no such pronouncements and have in no way modified what we said in Kelley. The holdings in Kelley, that the risk-utility test does not apply to a design defect unless the product malfunctions in some way and that a handgun does not malfunction when it shoots a bullet into a person in whose direction it is fired, remain the law of Maryland.
Petitioner raises four questions in her brief but really presents five propositions. First, she urges that we depart from Kelley, abandon the consumer expectation test, and adopt the risk-utility test in strict liability actions based on design defects, as the Court of Special Appeals has done, as a number of other States have done, and as the Restatement (Third) of Torts: Product Liability has done. Second, she asks either that, in applying that test, we do not require that a product malfunction as a prerequisite or that we regard the use of the gun by a three-year-old as a malfunction. Third, she requests that we not carve out an exception to the risk-utility test for handguns. Fourth, she contends that, because Garris’s conduct in leaving the gun and the magazine accessible to Jordan was foreseeable, it did not constitute a misuse of the product, and, finally, she argues that the warnings contained in the instruction manual do not suffice to shield Sturm Ruger from liability.
There has been a great deal of ferment regarding these issues, both in Maryland and elsewhere. Some of the debate is grounded in theory — whether, on the one hand, a consumer expectation test is either relevant or workable in a design defect situation, especially when the product is inherently dangerous, or, on the other, whether a departure from the consumer expectation test necessarily reintroduces negligence concepts, by focusing on the manufacturer’s conduct rather than the product itself, and thus becomes inconsistent with the notion and function of strict liability. That debate has a practical significance. The concept of strict liability, especially as formulated in § 402A of Restatement (Second), *201was regarded as an important pro-consumer advance; relieving persons injured by products from the requirement of proving negligence on the part of manufacturers or others in the distribution chain and focusing, instead, on the product itself, made it easier to obtain a recovery for a defectively designed or manufactured product. Substitution of a risk-utility analysis, however, especially as formulated in the Restatement (Third), has attracted considerable criticism and has been viewed by many as a retrogression, as returning to negligence concepts and placing a very difficult burden on plaintiffs. See John F. Vargo, The Emperor’s New Clothes, The American Law Institute Adorns a “New Cloth” for Section I02A Products Liability Design Defects — A Survey of the States Reveals a Different Weave, 26 U. Mem. L.Rev. 493 (1996); Patrick Lavelle, Comment, Crashing Into Proof of a Reasonable Alternative Design: The Fallacy of the Restatement (Third) of Torts: Products Liability, 38 Duq. L.Rev. 1059 (2000); Frank J. Vandall, Article, Constructing a Roof Before the Foundation is Prepared: The Restatement (Third) of Torts: Products Liability Section 2(b): Design Defect, 30 Mich. J.L. Ref. 261 (1997); Angela C. Rushton, Comment, Design Defects Under the Restatement (Third) of Torts: A Reassessment of Strict Liability and the Goals of A Functional Approach, 45 Emory L.J. 389 (1996). Much of this debate predated Kelley and was considered by us in that case; some of it, especially that focusing on the drafting and adoption of the Restatement (Third), has occurred since Kelley was decided.
In an extensive comment to § 2 of the Restatement (Third), the two Reporters for that work (Professors James A. Henderson, Jr. and Aaron D. Twerski) surveyed the legal landscape and concluded that the formulation in Restatement (Third), § 2, represents the majority thinking in the United States, and they are not alone in that view. See 1 M. Stuart Madden, Products Liability § 8.3, at 299 (2d ed.1988), cited in Hernandez v. Tokai Corp., 2 S.W.3d 251, 257 (Tex.1999). That view is not universally shared, however, and, to the extent it is shared, it has been criticized as representing an unwanted ascendancy of corporate interests under the guise of *202tort reform. In his comprehensive article, John Vargo, supra, 26 U. Mem. L.Rev. 493, notes the wide range of tests, and permutations of tests, used by the various States. Vargo concluded that, as of 1996, ten States continued to apply the ordinary consumer expectation test for strict liability design defects,, id. at 538-40, 952 (Index 3), and that six States applied a modified consumer expectation test, which, according to Vargo, engrafts on the consumer expectation test certain risk-utility balancing factors and substitutes the “reasonable consumer” for the “ordinary consumer.” Id. at 540-41, 952 (Index 4). Nine States, according to Vargo, have adopted a version of Dean Wade’s test, first articulated in 1973, which imputes knowledge of the product’s dangerousness to the manufacturer and then balances the product’s risks against its utility, the question being whether, with the imputed knowledge of the product’s dangers, the manufacturer was negligent in marketing the product. Id. at 541-44, 952 (Index 5).
In Barker v. Lull Engineering Co., 20 Cal.3d 413, 143 Cal.Rptr. 225, 573 P.2d 443 (1978), California adopted the consumer expectation and risk-utility tests as parallel alternatives:
“[A] product may be found defective in design, so as to subject a manufacturer to strict liability for resulting injuries, under either of two alternative tests. First, a product may be found defective in design if the plaintiff establishes that the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner. Second, a product may alternatively be found defective in design if the plaintiff demonstrates that the product’s design proximately caused his injury and the defendant fails to establish, in light of the relevant factors, that, on balance, the benefits of the challenged design outweigh the risk of danger inherent in such design.”
Id. at 455-56.
Several States have followed that approach. Others have, modified the second part so as not to shift the burden to the defendant. See Vargo, supra, at 544-45, 953 (Index 6 and 7).
*203Mr. Vargo observes that only seven States apply a pure risk-utility analysis in design defect cases, id. at 545-46, 953 (Index 8), and that four States have adopted strict liability rules that do not fit neatly into any of these various tests. Id. at 546-47, 953 (Index 9). Thirty-two States, he contends, apply a risk-utility analysis under any test for design defects, and he includes Maryland in that category. Id. at 550, 955 (Index 13). In Vargo’s view, only three States (Alabama, Maine, and possibly Michigan) require evidence of a reasonable alternative design in all design defect cases. Id. at 536.
To some extent, the debate over where the country is on this issue depends on how one counts and categorizes— whether intermediate appellate court decisions are regarded as definitive, how to deal with decisions based on statutes, how to deal with situations where one test is used for one type of case but not another, and whether the author is correctly reading the cases. There is also the overlay of §§ 3 and 4 of Restatement (Third), which may excuse the plaintiff from having to establish a reasonable alternative design when a defect may be inferred through a form of res ipsa loquitur analysis (§ 3) or where the product fails to comply with a product safety statute or regulation (§ 4).
The courts still seem to be split with respect to gun cases. Some follow the approach of Kelley, apply the consumer expectation test, and hold that a manufacturer may not be held liable for design defect on a risk-utility analysis unless the gun malfunctions. See Armijo v. Ex Cam, Inc., 656 F.Supp. 771 (D.N.M.1987); Caveny v. Raven Arms, 665 F.Supp. 530 (S.D.Ohio 1987); Perkins v. F.I.E. Corp., 762 F.2d 1250 (5th Cir.1985); Addison v. Cody Wayne Williams, 546 So.2d 220 (La.App.1989); McCarthy v. Olin Corp., 119 F.3d 148 (2d Cir.1997); Koepke v. Crosman Arms Co., 65 Ohio App.3d 1, 582 N.E.2d 1000 (1989). In some States, Texas and California among them, that approach is governed by statute. See Keene v. Sturm, Ruger & Co., 121 F.Supp.2d 1063 (E.D.Tex.2000); Hernandez v. Tokai Corp., 2 S.W.3d 251 (Tex. 1999); Merrill v. Navegar, Inc., 26 Cal.4th 465, 110 Cal.Rptr.2d 370, 28 P.3d 116 (Cal.2001). Others have, as *204petitioner urges, adopted a risk-utility analysis without regard to malfunction and held gun manufacturers liable, even when the gun operates precisely as intended, for failure to attach an available safety feature that might have precluded the gun from firing. See LeMaster v. Glock, Inc., 610 So.2d 1336 (Fla.App.1992); Hurst v. Glock, Inc., 295 N.J.Super. 165, 684 A.2d 970 (App.Div.1996). We have discerned no significant shift or coalescence of views in this regard since our decision in Kelley.
The one arena in which Kelley, itself, and the question of gun safety in general, has produced the most significant and relevant debate has been the Maryland General Assembly. Immediately on the heels of Kelley, bills representing nearly opposite viéwpoints were introduced into the Legislature— one, SB 151 (1986), would have directly overturned the second part of Kelley and expressly precluded liability on the part of manufacturers and merchants of “Saturday Night Specials” for injuries caused by another’s use of such a weapon, and another, SB 98 (1986), would have made it a misdemeanor to sell a “Saturday Night Special.” Both failed.
In 1988, the Legislature adopted a different approach. By 1988 Md. Laws, ch. 533, it (1) created a Handgun Roster Board within the Department of Public Safety and Correctional Services and charged that Board with creating a Handgun Roster — a listing of the kinds of handguns that could lawfully be sold in Maryland, (2) established certain standards for the Board to consider in determining whether to include a particular handgun on the Roster, including ease of concealment, accuracy of the weapon, quality of materials and manufacture, reliability as to safety, and utility for legitimate sporting activities, self-protection, and law enforcement, and (3) prohibited the manufacture and sale in Maryland of any handgun not listed on the Roster. The law also added two provisions dealing with civil liability. It provided that a person “may not be held strictly liable for damages of any kind resulting from injuries to another person sustained as a result of the criminal use of a firearm by a third person,” absent evidence of a conspiracy between the two, but added that “[tjhis section may *205not be construed to otherwise negate, limit, or modify the doctrine of negligence or strict liability relating to abnormally dangerous products or activities and defective products.”
The clear thrust of this law was to overturn the second part of Kelley and preclude strict liability actions based on the criminal use of handguns but to attempt to control the distribution and sale of handguns that were particularly dangerous or that had no legitimate utility through the device of the Handgun Roster. It would appear that the Legislature adopted some form of risk-utility analysis in establishing the standards for the Roster Board to consider in deciding whether to include particular kinds of handguns on the Roster.
In 1992, the Legislature dealt specifically with the problem of guns falling into the hands of children. 1992 Md. Laws, ch. 439 made it a criminal offense for a person to store or leave a loaded firearm in any location where the individual knew or should have known that an unsupervised minor would gain access to the firearm. As introduced, the bill would have defined the offense as leaving a loaded firearm where “it may reasonably be expected” that an unsupervised minor “may” gain access to the firearm, but that was deleted, along with an exception for a firearm “secured with a trigger lock or other similar device which prevents the firearm from firing ammunition” and a requirement that firearm dealers, with the sale of a firearm, must offer to sell or give to a purchaser such a trigger lock or device. The Legislature added to the bill provisions (1) that a violation may not be considered evidence of either primary or contributory negligence, limit the liability of a party or insurer, or diminish recovery for damages arising out of the ownership, maintenance, or operation of a firearm, and (2) precluding parties and others from making reference to a violation in the trial of a civil action for property damage, personal injury, or death.
In 1994, a bill (HB 1167) was introduced that would have expressly created strict liability in tort for all direct and consequential damages arising from bodily injury or death resulting from the discharge of a firearm wilfully sold or *206transferred in violation of State law. Interestingly, the bill would have precluded such an action for damages if the injury or death was self-inflicted or if the victim assumed the risk of the injury or negligently contributed to that injury. The bill did not pass. In 1996, and again in 1997, 1998, and 1999, bills were introduced that, after a certain date, would have precluded the sale of any handgun that did not meet “State performance standards for child resistance.” See HB 318 (1996), HB 292 (1997), HB 266 (1998), HB 267 (1999). Each of those bills required that the performance standards for child resistance, “(1) shall be designed so that a child under 6 years of age would have significant difficulty in firing a handgun; and (2) shall require a grip-safety mechanism, a combination lock mechanism, or other device to be attached to a handgun” and further directed the Secretary of State Police to adopt implementing regulations. Those bills also did not pass.
In 1999, the Governor, by Executive Order, directed that all law enforcement officers be provided with a locking device that would render their issued handguns inoperable while stored in the home. The Executive Order also created a Task Force on Childproof Guns and charged it with drafting legislation “to implement measures that prevent the unintentional and criminal misuse of handguns by children and other unauthorized users.” Executive Order 01.01.1999.18. The product of that Task Force were bills (HB 279 and SB 211) submitted to the 2000 Session of the General Assembly, one of which was enacted under the name “Responsible Gun Safety Act of 2000.”
As introduced, the bills, among other things, would have precluded a dealer, from and after January 1, 2002, from selling or transferring any handgun manufactured after December 31, 2001, unless the gun had “an integrated mechanical safety device or other incorporated design technology that is designed to prevent children and other unauthorized users from discharging the handgun.” The term “integrated mechanical safety device” was defined to mean “a disabling or locking device that ... is built into a handgun; and ... is designed to prevent the handgun from being discharged un*207less the device has been removed or deactivated.” The bills also would have created a Commission on Personalized Handgun Technology, which was to consider and report to the Legislature by July 1, 2002 whether “personalized handguns” are commercially available. A “personalized handgun” was defined as a handgun “manufactured with incorporated design technology allowing it to be fired only by a person who is the authorized user of the handgun and that prevents any of the safety characteristics from being readily activated.” If the Commission were to report that such handguns were commercially available and the Governor concurred in that judgment, the Governor was to so inform the Legislative Policy Committee, in which event a dealer could not sell or transfer a handgun after May 31 of the following year unless it was a “personalized handgun.”
The two bills were the subject of extensive discussion in the Legislature; literature and testimony were offered from all sides of the controversy, including Dr. Teret. The House Bill was killed; the Senate Bill was heavily amended and enacted as 2000 Md. Laws, ch. 2. In the enacted version, the references to “personalized handguns” were removed, along with the requirement that handguns have an “incorporated design technology that is designed to prevent children and other unauthorized users from discharging the handgun.” As enacted, the bill precluded dealers from selling or transferring handguns manufactured before December 31, 2002 unless they were sold with “an external safety lock” and, beginning January 1, 2003, from selling or transferring a handgun that does not contain an “integrated mechanical safety device.” At no point did the bill attempt to deal with handguns already in the hands of the public but merely placed restrictions of one kind or another on the future sale of handguns by dealers.
About a dozen bills are pending in the current 2002 session of the General Assembly dealing with handgun safety and with the right to sue for damages caused by handguns. Some would strengthen the penalties attached to the 1992 law dealing with the storage of loaded guns where children might find them. One (SB 381) would reserve to the State alone the *208right to sue gun manufacturers and dealers for damages or other relief.
CONCLUSION
It is clear that, under the consumer expectation test that we applied in Kelley, no cause of action had been stated in this case. There was no malfunction of the gun; regrettably, it worked exactly as it was designed and intended to work and as any ordinary consumer would have expected it to work. The gun is' a lawful weapon and was lawfully sold. What caused this tragedy was the carelessness of Jordan’s father in leaving the weapon and the magazine in places where the child was able to find them, in contravention not only of common sense but of multiple warnings given to him at the time of purchase.
We are asked to modify Kelley in various ways that would permit an action to proceed against the manufacturer of the weapon. We are asked to modify the common law to impose liability on gun manufacturers who have failed to incorporate into their products one or another kind of device that would make the weapon childproof, quite apart from the inclusion of other safety devices, clear warnings regarding the storage of the weapon, and the offer of a lock box in which to store it. Although, as we noted, some courts have done that, there is no consensus in that regard. We were asked in Kelley to extend and create new theories of liability, which we declined to do, noting that “we have consistently recognized that common law principles should not be changed contrary to the public policy of the State set forth by the General Assembly of Maryland.” Kelley, supra, 304 Md. at 141, 497 A.2d at 1151. See also State v. Sowell, 353 Md. 713, 728 A.2d 712 (1999) (declining to abolish distinction between principals and accessories); Gaver v. Harrant, 316 Md. 17, 557 A.2d 210 (1989) (declining to recognize action permitting child to recover damages for loss of parental society when parent disabled by negligence of third person); Felder v. Butler, 292 Md. 174, 438 A.2d 494 (1981) (declining to adopt Dram Shop liability as matter of common law); Harrison v. Mont. Co. Bd. of Educ., 295 Md. *209442, 456 A.2d 894 (1983) (declining to abrogate common' law doctrine of contributory negligence); Murphy v. Baltimore Gas & Elec., 290 Md. 186, 428 A.2d 459 (1981) (declining to alter common law principles governing duty of care owed trespasser by property owner); Austin v. City of Baltimore, 286 Md. 51, 405 A.2d 255 (1979) (declining to abrogate common law doctrine of governmental immunity in tort actions); Howard v. Bishop Byrne Council Home, 249 Md. 233, 238 A.2d 863 (1968) (declining to alter common law rule of charitable immunity); Creaser v. Owens, 267 Md. 238, 297 A.2d 235 (1972) (declining to alter judicially created “boulevard rule”); White v. King, 244 Md. 348, 223 A.2d 763 (1966) (declining to change common law rule of lex loci delicti in tort actions).
That caution is especially appropriate here. Given the controversy that continues to surround the risk-utility standard articulated for design defect cases in § 2 of the Restatement (Third), we are reluctant at this point to cast aside our existing jurisprudence in favor of such an approach on any broad, general basis. Nor is there a need to do so in this case, which deals with more specific issues that have been presented on several occasions to the General Assembly and have been considered and debated in that arena. So far, the Legislature has chosen not to place these burdens on gun manufacturers but has attempted to deal with the problem in other ways. We shall respect that policy choice.
JUDGMENT OF COURT OF SPECIAL APPEALS AFFIRMED, WITH COSTS.
BATTAGLIA, J., dissents and files opinion.