Because I believe that consistent with well-established principles of products liability jurisprudence the plaintiffs fulfilled their burden of demonstrating that a safer alternative was not only feasible but, in fact, was manufactured, I must respectfully dissent. Evidence adduced at trial sufficiently established that the safer alternative ultra light was the same as a regular cigarette in all respects save for its nonaddictive levels of nicotine and cancer-causing tar. Thus, I find no legal merit in the defendants’ assertion, which the majority supports, that the plaintiffs had an additional burden to show consumer acceptability of the nonaddictive product. In my view, this is nothing more than a cynical effort by the defendants to maintain the commercial advantages of continuing to sell unreasonably dangerous addictive products to addicts.
This appeal arises from a jury verdict of liability on the grounds of negligent design defect. It presents a singular issue: whether the defendant cigarette manufacturers negligently designed and marketed cigarettes containing addictive levels of nicotine, and whether the design increased the plaintiff Norma Rose’s exposure to cigarette tar and known carcinogens.
In my view, the plaintiffs fulfilled their burden of showing defective design by establishing that there was a feasible safer alternative and that the defendant manufacturers should have ceased manufacturing and marketing cigarettes with .4 milligram of nicotine and more per cigarette (hereinafter referred to as regular cigarettes).
In summary, evidence adduced at trial established that the defendants knew that .4 milligram of nicotine per cigarette was a threshold level for creating and sustaining addiction; that defendants were able to manipulate the levels of nicotine in cigarettes as far back as the 1950s and 1960s; and that defendants were able to, and in fact, did produce cigarettes with lower levels of nicotine and lesser amounts of tar (hereinafter referred to as ultra lights).
*91Further, the testimony of expert witnesses established that in virtually every respect ultra lights were the same product as regular cigarettes. The essential difference lay in the different amounts of nicotine and tar, rendering ultra lights with the lesser and nonaddictive levels of nicotine the safer alternative because they would not create or sustain addiction. Thus, in my opinion, the evidence established that ultra lights were a feasible safer alternative within the generally accepted meaning of feasibility.
It is the defendants’ contention, however, that such a traditional analysis is insufficient. The defendants argue, and the majority supports their view, that an additional element is required to establish feasibility in this case. Evidence notwithstanding, they argue that utility, and therefore feasibility, may only be demonstrated by evidence of consumer acceptability. Moreover, the defendants submit that they acted reasonably in marketing both regular cigarettes and ultra lights, and that they were not obligated to cease manufacturing and marketing cigarettes with addictive levels of nicotine because consumers rejected ultra lights.
I do not find this argument persuasive. First, consumer acceptability cannot be a factor in determining feasibility when the consumers are nicotine addicts—a class of consumer created by the defendants through their admitted manipulation of nicotine levels. It is hardly illuminating that sales and marketing data would show that nicotine addicts prefer cigarettes with sufficient levels of nicotine to sustain their addiction with minimum effort. Second, much of the evidence that the defendants wanted to proffer to establish consumer rejection of ultra lights amounted to testimony of consumer complaints about taste and the difficulties with “draw.” None of the proffered evidence would have established that given the choice between ultra lights or no cigarette at all, smokers would have rejected the ultra lights and quit smoking, and thus that ultra lights were unacceptable to consumers.
Further, the uncontroverted evidence was that plaintiff Norma Rose was determined to quit smoking but that she engaged in 15 failed attempts because she was severely addicted. Thus, a jury could rationally conclude that it was the addictive level of nicotine in regular cigarettes that thwarted her attempts to quit, and so it was the design defect that led to more than 50 years’ continued exposure to cancer-causing tar which was a substantial factor in causing her lung cancer.
*92It is undisputed that Norma Rose, now 73 years of age, began smoking cigarettes in the late 1940s. In the 1950s she became a regular smoker, preferring Camels, manufactured by R.J. Reynolds. She soon began smoking at least a pack a day. In the 1960s, she began smoking Pall Malls, manufactured by the American Tobacco Company, now nonexistent by reason of merger with Brown & Williamson Tobacco Corp.
She continued smoking Pall Malls until 1973 or 1974, when she began smoking filtered cigarettes. Briefly, she used two low tar brands, Merit and Vantage, but did not like their taste. She then began smoking Benson & Hedges, manufactured by defendant Philip Morris USA. Norma Rose finally quit smoking in 1993 after approximately 15 failed attempts to do so.
In 1995, she was diagnosed with lung cancer and a related neurological condition, paraneoplastic cerebellar degeneration. The lung cancer is in remission, but she still suffers from the neurological condition, which has caused permanent brain damage.
This lawsuit was brought against six cigarette manufacturers and two tobacco industry research organizations. Many defendants were dismissed from the action, and only Philip Morris, R.J. Reynolds, and Brown & Williamson were remaining at the time of trial.
Ultimately, the trial proceeded on a single theory of liability with all other claims being either dismissed or withdrawn before trial. The plaintiffs alleged that the defendants negligently designed and marketed cigarettes containing addictive levels of nicotine which was a substantial factor causing plaintiffs injury because it significantly increased her aggregate exposure to cancer-causing cigarette tar. The plaintiffs’ theory was that an ultra light cigarette, containing very low tar, or very little nicotine so as to be nonaddictive was the reasonable feasible alternative design to the cigarettes manufactured by defendants, and thát defendants should have ceased manufacturing regular cigarettes with more than .4 milligram of nicotine per cigarette.
The trial proceeded in three phases. During Phase I, the jury determined liability and compensatory damages. At trial, the defendants conceded that cigarettes cause cancer. The plaintiffs submitted evidence that one in every six deaths in the United States results from cigarette smoking. A 1979 Surgeon General’s Report documented cigarette smoking as the single most important preventable environmental factor contributing to illness, disability and death in the United States. American *93Tobacco was aware as early as the 1930s that cigarette smoke contained carcinogens, and Philip Morris knew as of the 1950s that there were carcinogens and cancer promoters in cigarettes. In 1961, during a Philip Morris presentation, it was indicated that at least 40 compounds in cigarette smoke were carcinogens.
The plaintiffs proffered evidence that the greater a smoker’s cumulative exposure to tar, the greater the risk of developing cancer. Conversely, less aggregate exposure to tar results in lower cancer risks. Exposure to reduced levels of tar, such as cigarettes with 5 to 6 milligrams of tar, in comparison with 16 or 17 milligrams of tar, results in a 20% to 50% reduction in lung cancer risk.
There was also substantial evidence as to the addictive properties of nicotine, of which the defendants were aware as of 1959. According to the plaintiffs’ expert psychopharmacologist, Dr. Classman, Mrs. Rose was “severely addicted” to nicotine.
Another plaintiffs’ expert, Dr. Wigand, testified that .4 milligram of nicotine represents a key threshold, below which cigarettes would not initiate a new addiction, and would not sustain an existing addiction. The defendants were able to design and manufacture cigarettes with less than .4 milligram of nicotine in the 1950s and 1960s.
Notably, American Tobacco released its Carlton brand in 1964, containing .3 milligram of nicotine. Philip Morris introduced Benson & Hedges’ lowest-nicotine brand, with .1 milligram of nicotine, in 1978. Both companies simultaneously continued production of brands containing higher nicotine levels.
As early as 1968, American Tobacco conducted experiments to increase nicotine content in cigarettes. Dr. Wigand testified that cigarette manufacturers added sugars and other additives to enhance the effects and release of nicotine. A 1978 Philip Morris document referred to research aimed at ensuring the total nicotine in the system remains at or near the nicotine need threshold, maximizing the proportion of the day’s cigarette consumption which is smoked out of need.
After seven weeks of trial, the court denied the defendants’ motion for a directed verdict. The court held that the plaintiffs had established a prima facie claim because the defendants knew that cigarettes caused lung cancer and addiction; that at the time they had the knowledge and technical feasibility to manufacture a safer product; that they chose to continue marketing the regular (defectively designed) cigarettes; and *94that the defectively designed product was a substantial factor in the plaintiffs injuries. Subsequently, the jury returned a verdict for the plaintiffs, assessing damages in the amount of $3,420,000 to be divided equally between Philip Morris and Brown & Williamson. The jury found no liability against R.J. Reynolds. The defendants moved for judgment notwithstanding the verdict. The court denied this motion also finding that the defendants had failed to sustain their burden of proving either that there was no valid line of reasoning or permissible inferences which could possibly lead rational persons to the conclusion reached by the jury, or that there was no fair interpretation of the credible evidence to support the jury’s verdict.
During Phase II, the jury considered punitive liability, and in Phase III, the amount of punitive damages. At the end of Phase II, the jury found that only Philip Morris was liable for punitive damages. After hearing evidence in Phase III of Philip Morris’s financial structure and resources, the jury awarded the plaintiffs $17,100,000 in punitive damages.
On appeal, the defendants-appellants, Brown & Williamson and Philip Morris, contend that the plaintiffs failed to establish the essential elements of their claim. They also argue that the trial court excluded key, relevant evidence including that of consumer awareness, preferences and acceptability, and improperly charged the jury during the liability phase of the trial. The defendants further submit that the plaintiffs’ negligence claim, as it was sent to the jury, was preempted by federal law. Finally, defendant Philip Morris argues that the punitive damages award is not sustainable.
For the reasons set forth below, I disagree with the majority and believe we should affirm the jury verdict of liability.
In a defective design cause of action, a claim for negligent design defect is functionally synonymous with a claim for strict products liability with respect to the manufacturer. (Denny v Ford Motor Co., 87 NY2d 248, 258 [1995].) In order to establish a prima facie case of negligent design defect, the plaintiff must prove that the manufacturer failed to exercise reasonable care in designing the product, and that it knew or should have known of the dangerous condition of the product. (Giunta v Delta Intl. Mach., 300 AD2d 350, 352 [2d Dept 2002].) To prevail in strict products liability, a plaintiff must prove that the product contained an unreasonably dangerous design defect. (Id.) Thus, the plaintiff in a design defect action “must show that the manufacturer breached its duty to market safe products when it *95marketed a product designed so that it was not reasonably safe and that the defective design, was a substantial factor in causing plaintiffs injury.” (Voss v Black & Decker Mfg. Co., 59 NY2d 102, 107 [1983].)
It is the plaintiffs obligation to present evidence that “the product, as designed, was not reasonably safe because there was a substantial likelihood of harm and it was feasible to design the product in a safer manner.” (Id. at 108.)
The defendant manufacturer, on the other hand, may present evidence in opposition seeking to show that the product is a safe product—that is, “one whose utility outweighs its risks when the product has been designed so that the risks are reduced to the greatest extent possible while retaining the product’s inherent usefulness at an acceptable cost.” (Id.)
The defendants conceded at trial that tar and nicotine are dangerous and that it is technologically feasible to produce a cigarette that has safer levels of tar and nicotine. Indeed, evidence showed that the defendants had manufactured and marketed safer alternatives including ultra lights with markedly lower levels of nicotine and tar.
The defendants, however, assert that it was not feasible to cease manufacturing and marketing regular cigarettes because ultra lights do not have the same consumer acceptability. Further, they argue that they are insulated from liability because they offered consumers the safer alternatives including ultra lights alongside the regular cigarettes. For the latter assertion, the defendants rely on Scarangella v Thomas Built Buses (93 NY2d 655 [1999]).
In Scarangella, an employee of a school bus company who was injured when she was struck by a bus operated in reverse, argued that a backup alarm, which was an optional safety feature, should have been a standard feature for the bus. The Court found, as a matter of law, that where evidence and reasonable inferences show that the buyer, not the manufacturer, is in the superior position to make the risk-utility assessment, and there is a well-considered decision by the buyer to dispense with the optional safety equipment this would excuse the manufacturer from liability. (See also Pigliavento v Tyler Equip. Corp., 248 AD2d 840 [3d Dept 1998], lv dismissed and denied 92 NY2d 868 [1998] [no negligent design where platform guardrail was available as optional equipment and 85% of concrete truck purchasers declined this option primarily because of its tendency to catch trees]; see also Jackson v Bomag GmbH, 225 AD2d 879 [1996], lv denied 88 NY2d 805 [1996].)
*96Scarangella applies the fundamental precept of products liability jurisprudence holding a manufacturer liable for selling a defectively designed product “because the manufacturer is in the superior position to discover any design defects and alter the design before making the product available to the public.” (93 NY2d at 659 [internal quotation marks and citation omitted].) Nevertheless, the Court found that if three essential elements are all met, a moderately priced safety feature need not be incorporated into the product design as standard equipment, but can be offered to consumers as an add-on option. The three elements are:
“(1) the buyer is thoroughly knowledgeable regarding the product and its use and is actually aware that the safety feature is available; (2) there exist normal circumstances of use in which the product is not unreasonably dangerous without the optional equipment; and (3) the buyer is in a position, given the range of uses of the product, to balance the benefits and the risks of not having the safety device in the specifically contemplated circumstances of the buyer’s use of the product.” (Id. at 661 [some emphasis added].)
In my view, to the extent the defendants cite Scarangella in order to assert that they satisfied their duty of care by offering safer ultra lights as an option, they clearly fail. First, the optional safety features in Scarangella were safety devices that were added on to the standard product. In this case, nothing can be added by the consumer onto or into regular cigarettes to render them safer. Further, it was the defendants’ burden to show compliance with all three of the required Scarangella elements to excuse the manufacturer from liability.
The record on appeal indicates that not one of the three essential Scarangella elements was satisfied. First, the defendants could not show that the plaintiff was “thoroughly knowledgeable” regarding the product and “in a position to balance the benefits and the risks” of smoking high-yield cigarettes, so that she, not the manufacturer (of cigarettes) was in the superior position to make the risk-utility assessment to smoke regular cigarettes. Indeed, the evidence is directly to the contrary.
Specifically, the defendants knew by 1959 that cigarettes were addictive. At least by 1972, the industry knew that nicotine was the active addictive constituent of cigarettes. As set forth above, the defendants knew in the 1960s that cigarette smoke *97contained at least 40 carcinogenic compounds, possibly including nitrosamines, “the most potent carcinogens known.” The industry confirmed in 1982 that nitrosamines were present in “significant amounts.” The plaintiff, on the other hand, was a consumer who typically would have received information about the hazards of smoking from the news outlets of the time, and from the warnings on cigarette packs. The warning labels were not mandatory until 1969.
The defendants argue that they were erroneously precluded from submitting evidence of consumers’ knowledge of the health risks of smoking. In my opinion, the evidence properly precluded by the court included, inter alia, the defendants’ “string-and-flash” 40-minute video showing every newspaper headline and TV show that purported to have any information whatsoever about tobacco at the time. Moreover, there was no attendant offer of proof that the plaintiff or any other rational consumer had actually read or seen any of this information. Second, the defendants’ burden pursuant to a Scarangella analysis was to show that the plaintiff had superior knowledge and was in a superior position as a buyer to make the relevant decision as to which cigarette to purchase and smoke. Since the record clearly establishes that it was the defendants, not the plaintiff, who were actively and extensively researching tobacco and smoking habits since the early 1930s, any conceivable suggestion that the defendants could satisfy the first and third Scarangella elements with regard to the plaintiff is patently absurd.
It is, however, the second Scarangella element that, in my opinion, ultimately defeats any defense that the defendants have, since they cannot show that there are any “normal circumstances of use” under which regular cigarettes are “not unreasonably dangerous.” The plaintiffs assert that there simply are no circumstances, normal or otherwise, when regular cigarettes are “not unreasonably dangerous.” They point to the statistics in the 1979 Surgeon General’s Report that show, for example, that one in every six deaths in the U.S. is the result of smoking and that smoking causes 87% of lung cancer deaths.
The plaintiffs, however, cannot rest on that assertion. It is not sufficient to show that a product is dangerous. As the defendants correctly contend, manufacturing and marketing dangerous products is not per se negligent. (See Forni v Ferguson, 232 AD2d 176 [1st Dept 1996]; see also Robinson v Reed-Prentice Div. of Package Mach. Co., 49 NY2d 471, 479 [1980] [some products, for example, knives, must by their very nature *98be dangerous to be functional].) Instead, a defective design cause of action must establish that a product is unreasonably dangerous or “not reasonably safe”—that is, there is a substantial likelihood of harm and a feasible safer alternative design. (Voss, 59 NY2d at 108.)
Conversely, for the defendants to succeed in their claim that there are circumstances when regular cigarettes are “not unreasonably dangerous,” they must show that there are normal circumstances when the utility of regular cigarettes outweighs the risk inherent in them. (Id.; see also Rainbow v Elia Bldg. Co., 79 AD2d 287, 294 [1981] [balancing process weighs the benefits of a particular manufacturing design against the risks of using it].)
When pressed at oral argument, however, the defendants had no rejoinder to the question of what circumstances might render regular cigarettes “not unreasonably dangerous” or, in other words, when they would be “reasonably safe.” Understandably so. As one federal court recently observed, defendants in tobacco cases “wish to avoid having to make the awkward argument that their product’s ‘utility’ outweighs its risk, when their product is known to ‘sicken and kill hundreds of thousands of Americans each year for the ‘benefit’ of satisfying an addiction.’ ” (See Clinton v Brown & Williamson Holdings, Inc., 498 F Supp 2d 639, 647 [SD NY 2007], quoting David G. Owen, Inherent Product Hazards, 93 Ky LJ 377, 381 [2004-2005].)
The defendants’ desire to avoid that particular line of reasoning is abundantly clear in the instant case. Indeed, the defendants construct an entire defense—120 pages of briefs and reply briefs—around the proverbial “800-pound gorilla” without even alluding to its presence in the room. The defendants simply ignore the fact that the overwhelming majority of smokers smoke because they are addicted to nicotine.
Instead, the defendants’ risk-utility assessment proceeds along the following lines:
“[P]eople smoke cigarettes because they like the taste that tar creates in the smoke,” and they “desire the pharmacological effect of nicotine”;
Because a sensory experience is subjective, the utility of regular cigarettes can only be demonstrated by consumer choice;
In this case, utility is unquestionably demonstrated by the fact that “many people choose to smoke [regular cigarettes] in the face of Congressional warnings and with the full knowledge of the risks including addiction”;
*99The lesson of Scarangella ... as well as the long line of New York cases holding that it is not negligent simply to sell dangerous products—is that consumer choice is a critical component of the reasonableness of a product’s risks;
Further, smokers’ choice of regular cigarettes shows that while safer ultra lights are technologically feasible, it is not feasible to cease manufacturing and marketing regular cigarettes because ultra lights are not the same product;
Sales and marketing data would have provided proof of all of the above, had the trial court not erroneously excluded that evidence.
Thus, the defendants persuade the majority that regular cigarettes Eire “not unreasonably dangerous” because the majority of smokers, knowing they risk sickness, cancer, disability and death, nevertheless choose to smoke regular cigarettes simply because they are enjoyable, tasty and relaxing, and they cannot get the same “benefits” from ultra lights.
First, in my view, the foregoing circumstances cannot be characterized as in any way “normal.” The conduct of smokers as described is in direct contravention to normal consumer reaction. As noted, for example, in the apple/Alar scare of 1989, generally mass rejection, if not panic, follows when a product is identified as containing carcinogens or ingredients likely to cause sickness and death. (See Elizabeth Whelan, The Chemical Scare: Are Politics Driving the Fear?, Heritage Lecture No. 295, The Heritage Foundation [1990], available at http:// www.heritage.org/Research/SocialSecurity/HL295.cfm.)
Second, the defendants’ argument that because a cigarette’s function is to provide a sensory experience its utility can only be measured by consumer choice and acceptability is, in my view, circular, self-serving, and without legal merit. While smoking can indeed be described as a sensory experience, the focus of the argument in this case—the relative levels of nicotine with their attendant Eimounts of tar—is a factor that does not foreclose Em objective assessment as to utility or feasibility.
Both the plaintiffs and the defendants agree that the essential function of a cigarette is as a delivery system for nicotine, and that cigarettes provide “pleasure and relaxation.” Both sides agree that both ultra lights and regular cigarettes deliver nicotine as well as tar. Although there was some testimony about the differences in taste, along with the dubious assertion by the defendants that smokers “like” the taste of tar in regular *100cigarettes, there is no evidence in the record to suggest that an ultra light looks any different or feels any different to hold between the lips or between fingers. Nor is there any evidence that an ultra light either costs more to manufacture or to purchase. As such, an ultra light cannot be characterized as a product that is different from a regular cigarette. In this regard, the defendants’ and the majority’s reliance on Felix v Akzo Nobel Coatings (262 AD2d 447 [2d Dept 1999]) is misplaced. In Felix, the Court found that a water-based wood sealer offered as an alternative to a petroleum-based highly flammable product was a different product. Setting aside the fact that the plaintiffs expert testified that the alternative was a different product, in that case, in addition to the difference between the ingredients contained in the product (petroleum-based ingredients versus water), there were several other considerable differences between them including the time taken for the drying process, the look of the finished product, and a prohibitive cost difference. Additionally, the danger of explosions by using the petroleum-based wood sealer was limited to a negligible number of accidents among a very limited population. In this case, not even the ingredients are different; the difference lies simply in the levels of nicotine and the amount of tar, that is, an ultra light contains less than .4 milligram of nicotine per cigarette while a regular cigarette contains .4 milligram or more of nicotine. The record shows .4 milligram is indisputably the level that creates and sustains addiction, and so increases exposure to tar. Consequently, consumer acceptability cannot be a measure for utility or feasibility in this case. To proffer sales and marketing data as evidence that given a choice, a smoker who is addicted to nicotine will choose a product that satisfies the addiction is, in my opinion, a tautological exercise and therefore meaningless. Second, the concept of choice is itself suspect in this situation. As the trial court aptly observed, “one who is addicted has lost the ability to make free choices concerning the substance of the addiction.” (Rose v Brown & Williamson Tobacco Corp., 10 Misc 3d 680, 690 [Sup Ct, NY County 2005].)
“By definition one who suffers from addiction is compelled to continue an activity even though he or she knows the activity is detrimental [because] even though a cigarette smoker may be made fully aware of the elevated risks, if he or she is addicted to smoking, he or she may not be making a free choice to participate in the activity of continued smoking.” (Id.)
*101Third, case law is replete with products liability actions where consumers have found safer alternatives “unacceptable” and have modified safety features without manufacturers or courts declaring these safer alternatives unfeasible. (See Montufar v Shiva Automation Serv., 256 AD2d 607 [2d Dept 1998]; Mack-ney v Ford Motor Co., 251 AD2d 298 [2d Dept 1998]; Wyda v Makita Elec. Works, 232 AD2d 407 [2d Dept 1996].)
In this case, while the issue of taste (the amount of tar in the cigarette) was raised in testimony as one factor that made ultra lights unacceptable, more significant was the testimony of Dr. Blackie, an expert witness for the defendants, that smokers did not like ultra lights because of the problem with “draw.” In the sense that smokers complained about the effort involved in getting nicotine into their system, their “lack of acceptance” is analogous to that of workers or employers in the foregoing cases discarding or removing safety features because they slow down the work process. The decisions in those cases, however, do not reflect any support for the argument that lack of acceptance by consumers absolves the manufacturer from manufacturing the technologically feasible safer alternative, or even from warning a consumer about the danger of modifying a safety feature. (See Montufar, 256 AD2d at 607-608.)
In any event, any sales or marketing data the defendants may have submitted to the jury on the issue of consumer acceptability or preferences would merely have shown that when ultra lights and regular cigarettes were offered side by side, and incidentally at a time when consumers had little idea of the health effects, consumers preferred to smoke the high-yield cigarettes. Demonstrating consumer preference when options are available is a far cry from the unequivocal statement that given only one type of product, smokers would have rejected it.
In my opinion, the plaintiffs are correct in their assertion that traditional products liability jurisprudence does not require consumer acceptability to be an element proving feasibility. (See Rypkema v Time Mfg. Co., 263 F Supp 2d 687, 692 [SD NY 2003] [a federal court applying New York law to a design defect case held that plaintiff may prove a feasible safer alternative design with evidence that such design is “within the realm of practical engineering feasibility”]; see also Micallef v Miehle Co., Div. of Miehle-Goss Dexter, 39 NY2d 376, 386 [1976] [alternative is not feasible if product was “unworkable” or so expensive as to be priced out of the market].)
According to those principles, I believe the plaintiffs in this case established that there was a feasible safer alternative. The *102defendants agreed that a safer alternative was technologically feasible. They argued, however, that it was not feasible to manufacture and market only the safer alternative because the safer alternative did not have the same utility as the regular cigarettes. It therefore became the defendants’ burden to show that the safer alternative had not retained its inherent usefulness. (See Voss, 59 NY2d at 108 [“The plaintiff ... is under an obligation to present evidence that ... it was feasible to design the product in a safer manner. The defendant manufacturer, on the other hand, may present evidence in opposition”].) The defendants then argued that “inherent usefulness” or utility could only be demonstrated by evidence of consumer acceptability—an assertion with which I disagree for the reasons already noted.
The defendants further maintained that they had acted reasonably in manufacturing and marketing both regular and ultra light cigarettes because consumers rejected ultra lights and showed a clear preference for regular cigarettes. On appeal, the defendants asserted that a new trial is warranted on the grounds that the trial court excluded evidence of consumer preferences.
In my opinion, the proffered evidence which the trial court ruled out did not establish that smokers had rejected ultra lights. There was no evidence or testimony from the defendants that established that smokers would not purchase ultra lights. Rather, the evidence proffered by the defendants established that there were characteristics that smokers did not like about ultra lights. Surveys conducted by the defendants found that safer levels of nicotine and tar “adversely affected taste.” There was abundant testimony as to consumer complaints about “draw” and the inability to “get any flavor through.” There was also testimony that the modifications used to make cigarettes safer “result in significant changes in the flavor and the sensory characteristics of the product . . . That’s always been a problem.”
At no time, however, did the defendants effectively connect the dots and offer evidence to show that smokers who were faced with purchasing either ultra lights or no cigarette at all had refused to purchase the ultra lights. In fact, as one line of questioning established, consumer acceptance, preferences and taste have little to do with feasibility where cigarettes are concerned. Upon cross-examination of Dr. Blackie, an expert witness for the defendants, the plaintiffs established that dur*103ing the 1960s, most smokers preferred unfiltered cigarettes to cigarettes with filters which reduced carcinogens. Dr. Blackie testified that therefore the unfiltered cigarettes were “considerably more feasible.” She then stated, “consumers wanted to buy [unfiltered cigarettes] and . . . nobody was telling us to stop making them.” The cross-examination also established that over a period of time the market “dramatically switched” and that there was “a significant swing from non-filtered cigarettes to filtered cigarettes.” The following Q and A then ensued:
“[plaintiff]: That’s proof that it [the filtered cigarette] was feasible, right?
“Dr. Blackie: No, it’s not. Because if you ask consumers to change tastes dramatically from one day to the next, what other countries experience has shown you end up with a black market for the products that they want. And, so they tend to go and buy overseas or they have cigarettes shipped in from another country . . . Whether that would have happened in this particular case . . . nobody can say . . . But, it’s entirely possible.”
Based upon the foregoing, it is evident that had tobacco litigation existed in the 1960s, manufacturer defendants would have argued that filtered cigarettes were not a feasible alternative because consumers “wanted” to buy nonfiltered. Today, not even the defendants in the instant case attempt to claim that a filtered regular Marlboro is not functionally interchangeable with a nonfiltered Camel.
What then should one effectively make of their argument that cigarettes manufactured with reduced carcinogen levels and lower nicotine yields are likewise not feasible alternatives because the overwhelming majority of smokers “want” regular cigarettes? Specifically, what should be made of such an argument when the further testimony of Dr. Blackie (“[y]ou train the consumer to accept new products”) indicates that consumer tastes are nothing more than sensory experiences that can be influenced and molded by manufacturers?
I would therefore find “there is no justification for departure from the accepted rationale imposing strict liability upon the manufacturer because it ‘is in the superior position to discover any design defects.’ ” (See Scarangella, 93 NY2d at 661, quoting Voss, 59 NY2d at 107 [emphasis added].)
*104If there is any argument at all to be made about ultra lights lacking the inherent usefulness of regular cigarettes or lacking the “ingredient critical to performance,” then it must be made in context. The sole utility of regular cigarettes is to satisfy an addiction which, albeit an addiction to a legal substance, nevertheless increases exposure to cancer-causing tar.
However, that should not help the defendants in this negligent design defect action where the focus must move to the conduct of the manufacturer, and the question of whether the manufacturer acted unreasonably in designing the product. (See Gonzalez v Morflo Indus., Inc., 931 F Supp 159, 165 [ED NY 1996] [“the focus shifts from whether the product, as designed, was not reasonably safe to whether the manufacturer was aware of that condition and chose to market the product anyway”].)
The record shows that the defendants possessed information about the reasons for smoking as far back as the 1950s. One exhibit, a 1959 report aimed at discovering why people smoke or do not smoke, showed that available data suggested the following physiological reasons why people smoked: gratification of senses such as oral and digital satisfaction, and stimulation or relaxation; and it suggested the following psychological reasons: conformity, sociability, sophistication, ritual, mimicry, and boredom.
The defendants conceded that their research established that nicotine was a “major pharmacological substance in tobacco smoke”; that nicotine satisfaction was the dominant desire of smokers; and that .4 milligram was the tolerance level below which the addiction process does not start.
Further, evidence showed that Philip Morris had aimed its research at establishing that level of nicotine that was required to cause addiction. Testimony was also elicited as to the defendants’ manipulation of nicotine levels after determining the level needed to keep smokers addicted. An expert for the plaintiffs, Dr. Wigand, testified that, “nicotine is the product that sells cigarettes and unless you keep people addicted, you cannot sell cigarettes.” (Emphasis added.)
Thus, it is clear from the record that a cigarette could be manufactured to deliver more or less nicotine with greater or lesser attendant levels of cancer-causing tar depending on the purpose for which it was to be smoked. However, after the defendants discovered that .4 milligram was the threshold level to sustain addiction, they chose to manufacture not the lower level product, but one that delivered nicotine at addictive levels thereby creating for themselves a captive commercial market.
*105If, as defendants now assert, an ultra light does not serve the same function as a regular cigarette, it is because the defendants manipulated the product so that the function became almost exclusively to satisfy addiction, and it is clear that the defendants were not interested in those consumers who would smoke a cigarette “for other aspects” that would allow them to take or leave the smoking habit.
Proximate cause
The defendants further argue that the plaintiffs failed to prove that the failure to adopt the proffered alternative design caused the plaintiff Norma Rose’s injuries. The defendants assert that the plaintiffs were required to show that the design caused her cancer rather than the cigarettes themselves. The defendants further state that there was no proof that the difference in the tar and nicotine yields between the brands was a substantial contributing cause. The defendants point to the plaintiffs’ own expert acknowledging that ultra lights are unsafe, and contend that evidence was adduced that smokers of ultra lights sometimes compensate by smoking more cigarettes.
It is true that proximate cause in a products liability case serves a different role than in a case sounding in negligence because the cause of action seeks to impute liability to the manufacturer not on the basis of its negligence but because the product is not reasonably safe as it was designed. (Voss, 59 NY2d at 110.) Thus, in order to impose liability on the manufacturer, the plaintiffs burden is to tie the design defect of the product to the injury—that is, the plaintiff must show that the design defect in the product was a substantial factor in causing his or her injury. (Id.)
In this case, however, I believe the defendants miss the point because the plaintiffs’ theory is not that the safer design would have delivered less tar and nicotine, and thus less exposure to carcinogens. Rather, the plaintiffs’ theory is that because the design was defective in maintaining the level of nicotine at an addictive level it caused Norma Rose to continue smoking and thus continued her exposure to cancer-causing carcinogens which, in fact, caused her cancer.
Additionally, the plaintiffs assert that the defendants should have ceased manufacturing and marketing the defectively designed cigarettes. Consequently, the plaintiffs did not have to establish that Norma Rose would have smoked the safer cigarette (there would have been no other kind). Nor were they ob*106ligated to present expert evidence to exclude the possibility that she could have maintained her addiction on ultra lights by smoking more given that the evidence indisputably established that Norma Rose wanted to quit smoking, and attempted to do so 15 times. Her testimony further established that when she tried to quit she exhibited all the characteristics of a severely addicted smoker including being “very nervous,” “veiy nasty” and preoccupied with smoking. In my opinion, the evidence was sufficient to demonstrate that Norma Rose’s undisputed genuine, recurring desire to quit would not have been thwarted by her addiction if her addiction was nonexistent or not sustainable on ultra lights. In turn, had she been able to quit when she desired, her exposure to cancer-causing tar would have been terminated.
“Proximate cause is a question of fact for the jury where varying inferences are possible.” (Mirand v City of New York, 84 NY2d 44, 51 [1994], citing O’Neill v City of Port Jervis, 253 NY 423, 433 [1930]; see Nowlin v City of New York, 81 NY2d 81, 89 [1993]; Mercado v Vega, 77 NY2d 918, 920 [1991].) Given that causation is an inherently factual issue, “[a]s a general rule, the question of proximate cause is to be decided by the finder of fact, once negligence has been shown.” (Equitable Life Assur. Socy. of U.S. v Nico Constr. Co., 245 AD2d 194, 196 [1st Dept 1997]; see Mortensen v Memorial Hosp., 105 AD2d 151, 157 [1st Dept 1984]; McKinnon v Bell Sec., 268 AD2d 220, 221 [1st Dept 2000].) Further, a jury’s causation finding should stand unless there is simply no valid line of reasoning and permissible inferences which could possibly lead rational people to the conclusion reached by the jury on the basis of evidence presented at trial. (See Cohen v Hallmark Cards, 45 NY2d 493, 499 [1978].)
In my view, the factual record here was more than sufficient to support a rational conclusion that the negligent design at issue—high-yield cigarettes containing .4 milligram or more of nicotine—was a substantial factor in causing Norma Rose’s exposure to the tar in the cigarettes she smoked for more than four decades. This in turn caused her lung cancer and her brain injuries. It is undisputed that cigarette tar contains especially powerful carcinogens and the more cumulative exposure to tar, the greater the cancer risk. Conversely, less aggregate exposure to tar results in substantially lower cancer risk.
There is no dispute that the cigarettes the plaintiff smoked contained levels of nicotine well above the addictive threshold, and that that nicotine caused her to become severely addicted. *107Both the strength of her addiction and her motivation to stop smoking were demonstrated by her approximately 15 failed attempts to stop smoking in the late 1970s and early 1980s.
Thus, the record contains evidence from which rational inferences could be drawn that the unreasonably unsafe design at issue here—cigarettes with .4 milligram or more of nicotine—(1) resulted in the maintenance of Norma Rose’s nicotine addiction, which (2) compelled her to continue smoking high-yield cigarettes despite her motivation and serious attempts to quit, and/or to smoke more cigarettes than she otherwise would have in order to satisfy the addiction, and (3) thus significantly increased her exposure to the cigarette tar which, it is undisputed, caused her cancer. Conversely, the jury could rationally conclude that if the cigarettes Norma Rose smoked had contained less than .4 milligram of nicotine, her addiction would not have been sustained and either her efforts to quit smoking would have succeeded or she would have smoked far fewer cigarettes, thereby substantially lowering her cancer risk.
Federal preemption
The defendants claim, and the majority supports their view, that the trial court’s finding of liability on the negligent design defect claim effectively bans the sale of all cigarettes currently on the market in New York State, and that such ban is preempted by federal law. (See e.g. Geier v American Honda Motor Co., 529 US 861 [2000] [common-law tort action alleging that automobile manufacturer was negligent in failing to equip automobile with driver’s side airbag was preempted in that it actually conflicted with National Traffic and Motor Vehicle Safety Act, requiring manufacturers to place driver’s side airbags in some but not all 1987 automobiles].)
The defendants assert that Congress has foreclosed a cigarette ban (see FDA v Brown & Williamson Tobacco Corp., 529 US 120 [2000]), instead regulating cigarette sales by imposing requirements for labeling and advertising. (See Federal Cigarette Labeling and Advertising Act of 1965, codified at 15 USC § 1331 et seq.) The defendants claim that allowing the plaintiffs to recover damages based upon the dangers inherent in tobacco products would preclude the defendants from selling virtually all cigarettes currently on the market. (See e.g. Conley v R.J. Reynolds Tobacco Co., 286 F Supp 2d 1097, 1107-1108 [ND Cal 2002] [claims were preempted to the extent that they relied on design defect so inherent in tobacco products that its removal *108was not scientifically or commercially feasible]; Insolia v Philip Morris Inc., 128 F Supp 2d 1220, 1224 [WD Wis 2000] [Congress’s considered decision that sale of cigarettes was not illegal and was part of market that government supported preempted state law negligence claim against tobacco manufacturers based on their continuing to manufacture and sell cigarettes once they realized danger that cigarettes posed, even though no statute or regulation explicitly preempted such claims].)
I find the defendants’ contentions to be unpersuasive.
“Under the Supremacy Clause of the United States Constitution (US Const, art VI, [2]), state law may be preempted in three circumstances: first, through express statutory language; second, when it regulates conduct in a field that Congress intended the Federal Government to occupy exclusively; and third, when it actually conflicts with federal law.” (Feldman v CSX Transp., Inc., 31 AD3d 698, 701 [2d Dept 2006].)
The defendants do not point to express statutory language of preemption. Further, although there is federal regulation of labeling and advertising (15 USC § 1331 et seq.), there is no federal regulation of tar and nicotine content in cigarettes. (Cf. Cipollone v Liggett Group, Inc., 505 US 504 [1992] [state failure to warn claims against cigarette manufacturers were preempted by Federal Cigarette Labeling and Advertising Act, as amended by the Public Health Cigarette Smoking Act of 1969].)
In my opinion, contrary to the defendants’ contentions, Congress’s omission in not wholly banning cigarettes or regulating tar and nicotine levels cannot serve to preempt state law tort claims premised on their danger to the public. Finally, there is no actual conflict with federal law, since there is no law on the subject matter serving as the basis of this lawsuit.
In FDA v Brown & Williamson Tobacco Corp. (529 US 120 [2000]) the Supreme Court held that the FDA lacked authority under the Federal Food, Drug & Cosmetic Act to regulate tobacco products as customarily marketed, primarily because FDA regulation would result in a cigarette ban given the FDA’s mandate under the statute. The Court found that Congress, by regulating advertising and labeling under a separate legislative directive, had not intended the FDA to regulate this subject matter under the Federal Food, Drug & Cosmetic Act. Clearly, this case did not foreclose state tort claims on defective products.
Further, in my view, the damages award in this case is not synonymous with a ban on cigarettes. Instead, it holds the *109defendants liable for a failure to limit dangerous ingredients in their products, i.e., failure to sell a safer product. (See Conley v R.J. Reynolds Tobacco Co., 286 F Supp 2d 1097, 1107-1108 [ND Cal 2002] [to the extent that claims targeted design defect in manufacturers’ cigarettes which was scientifically and commercially feasible to remove from products used by smoker before his death, state law design defect claims against cigarette manufacturers were not preempted, through conflict preemption, based on congressional policy precluding complete ban of tobacco products, inasmuch as successful claims would result in ban only as to those products which suffered from defective design, and thus involved only selective regulation of tobacco products].) Thus, I would find that negligent design cases like this one against cigarette manufacturers are not preempted by federal law.
Punitive damages
Finally, Philip Morris contends that its conduct in marketing different cigarette brands with a range of tar and nicotine yields cannot subject it to punishment in New York. Philip Morris argues that, as a matter of due process, an award of punitive damages cannot be based upon conduct that it could reasonably have believed to be lawful. On this issue, I agree with the defendant.
To warrant an award of punitive damages, there must be proof of recklessness, or a conscious disregard of the rights of others. (See Hartford Ace. & Indem. Co. v Village of Hempstead, 48 NY2d 218 [1979].) It is also well settled that punitive damages may not be premised upon mere negligence. (See Everett v Loretto Adult Community, Inc., 32 AD3d 1273 [4th Dept 2006]; Morton v Brookhaven Mem. Hosp., 32 AD3d 381 [2d Dept 2006] [“(p)unitive damages are recoverable where the conduct in question evidences a high degree of moral culpability, or the conduct is so flagrant as to transcend mere carelessness, or the conduct constitutes willful or wanton negligence or recklessness” (internal quotation marks and citations omitted)].)
In this case, the record established that Philip Morris was aware that cigarettes caused cancer, and were addictive. It knew the threshold for a smoker’s need, and could control the precise levels of nicotine in cigarettes. Further, Philip Morris continued to sell cigarettes with nicotine above the addictive threshold, and the plaintiffs presented documentation where Philip Morris set this as its goal. Thus, there was evidence to suggest that the *110defendant consciously disregarded the health risks posed to billions of consumers.
Nevertheless, Philip Morris’s conduct in marketing different cigarette brands with a range of tar and nicotine yields cannot subject it to punishment under New York law. As a matter of due process, an award of punitive damages cannot be based upon conduct—such as that at issue here—that the defendant could reasonably have believed to be lawful. In BMW of North America, Inc. v Gore (517 US 559, 574 [1996]), the Supreme Court explained that “[e]lementary notions , of fairness enshrined in our constitutional jurisprudence dictate that a person receive fair notice ... of the conduct that will subject him to punishment.” (See also Bouie v City of Columbia, 378 US 347, 355 [1964] [when punishment is imposed based on novel construction of statute, “the effect is to deprive (defendant) of due process of law in the sense of fair warning that his contemplated conduct constitutes a crime”].)
Philip Morris did not have “fair notice” that the conduct at issue in this case might result in severe punishment. Indeed, the verdict in this case is novel. Congress not only has made a purposeful choice to regulate sales and advertising rather than to bar the sales of regular cigarettes, but has also blocked attempts to regulate tar and nicotine levels. The Surgeon General has never recommended removing regular-yield cigarettes from the market. No state or federal legislator or regulator has ever adopted a rule banning or restricting full-flavored cigarettes. And until this case, no court had ever held any tobacco manufacturer liable simply for continuing to sell regular brands, much less suggested that such conduct was punishable. In my view, punitive damages may not be imposed under such circumstances.
Tom, J.P., and Malone, J., concur with Friedman, J. Nardelli and Catterson, JJ., dissent in a separate opinion by Catterson, J.
Judgment, Supreme Court, New York County, entered July 18, 2005, reversed, on the law, without costs, defendants-appellants’ motions for a directed verdict and for judgment notwithstanding the verdict granted, and the second amended verified complaint dismissed. The Clerk is directed to enter judgment accordingly.