delivered the opinion of the court:
This case involves three actions by three different plaintiffs, Lois Bicknell (Bicknell), Vernadine Thacker (Thacker), and Delores McClure (McClure), against defendant Owens Corning Fiberglas Corporation (Owens Corning). McClure also named Owens-Illinois, Inc. (Owens-Illinois), as a defendant in her suit. All three plaintiffs’ complaints were based on injuries allegedly resulting from their husbands’ exposure to asbestos while employed by the Union Asbestos and Rubber Company (Unarco) at its Bloomington, Illinois, plant. Neither Unarco nor Johns-Manville Corporation (Johns-Manville), a supplier of asbestos used in the Bloomington plant, is named as a defendant in plaintiffs’ complaints. In addition, plaintiffs alleged no exposure to Owens Corning or Owens-Illinois products and no employment relationship between defendants and their husbands. Nevertheless, plaintiffs claimed that defendants were responsible for the injuries alleged in their complaints because defendants engaged in a civil conspiracy with Unarco, JohnsManville, and other companies to suppress information concerning the harmful health effects of asbestos exposure and to falsely represent that it was safe for people to work in close proximity to asbestos-containing materials.
At plaintiffs’ request, these three cases were consolidated for trial before the circuit court of McLean County. After trial, the jury returned verdicts in favor of Bicknell and Thacker against Owens Corning. The jury also found in favor of McClure against both Owens Corning and Owens-Illinois. The appellate court affirmed the judgments entered on these verdicts. 298 Ill. App. 3d 591. Owens-Illinois then filed a petition for leave to appeal the judgment in the McClure case, and Owens Corning filed a petition for leave to appeal the judgments in all three cases. This court granted defendants’ petitions (177 Ill. 2d R. 315) and allowed plaintiffs’ motion to consolidate defendants’ appeals. In addition, we have permitted amicus curiae briefs to be filed by the Illinois Association of Defense Trial Counsel, the Product Liability Advisory Council, Inc., and the Illinois Manufacturers’ Association. 155 Ill. 2d R. 345. We reverse on the basis that the evidence was insufficient to show that either defendant engaged in the alleged civil conspiracy with Unarco or Johns-Manville.
BACKGROUND
In her complaint against Owens Corning and Owens-Illinois,1 Delores McClure, individually and as special administrator of the estate of Robert McClure, alleged that her husband, Robert McClure, worked in Unarco’s Bloomington plant during the period 1959-61. She asserted that, while her husband worked at the Blooming-ton plant, he was exposed to asbestos, including asbestos that Johns-Manville supplied to Unarco. According to McClure, as a result of this exposure, her husband contracted asbestosis and lung cancer, which caused his death.
McClure asserted that defendants, Unarco, and Johns-Manville knew of the health hazards associated with asbestos exposure and had a duty to warn their employees of these hazards and to provide a safe workplace. She further alleged that defendants conspired with Unarco, Johns-Manville, and other companies to “positively assert in a manner not warranted by the information possessed by the conspirators, that *** it was safe for people to work with and in close proximity to asbestos and asbestos containing materials” and to “suppress information about the harmful effects of asbestos *** causing asbestos workers to be and to remain ignorant of that information.” In furtherance of this alleged conspiracy, the conspirators, inter alla, sold asbestos without warning of its adverse health effects, refused to warn their own employees and Robert McClure about these hazards, and altered and suppressed published reports concerning these hazards. According to McClure, these acts were proximate causes of Robert McClure’s injuries and death.
Lois Bicknell’s complaint against Owens Corning (Owens-Illinois was a defendant only in McClure’s complaint)2 contained essentially the same conspiracy allegations stated by McClure. Bicknell alleged that, like Robert McClure, her husband, Hugh Bicknell, worked in Unarco’s Bloomington plant, although his period of employment was 1954 to 1955. According to Bicknell, as a result of her husband’s exposure to asbestos while working in the plant, he developed pulmonary fibrosis and lung cancer, which resulted in his death. Also like McClure, Bicknell alleged that her husband’s injuries and death were proximately caused by the acts taken in furtherance of the alleged conspiracy.
Vernadine Thacker’s complaint against Owens Corning3 also contained the same allegations of conspiracy as McClure’s complaint. Unlike McClure’s and Bicknell’s complaints, however, Thacker’s complaint was based on her own exposure to asbestos. Thacker alleged that her husband, Charles Thacker, was employed at Unarco’s Bloomington plant between 1952 and 1965. Thacker asserted that she was exposed to asbestos fibers carried home from the plant on her husband’s person and clothing. According to Thacker, as a result of her exposure to these fibers, she developed pulmonary fibrosis and lung cancer. Thacker claimed that the acts taken in furtheranee of the alleged conspiracy were a proximate cause of her injuries.
Prior to trial, the circuit court granted plaintiffs’ motions to consolidate these three cases. It denied Owens-Illinois’ motion requesting that it be tried separately from Owens Corning, although when evidence was admitted against only one of these parties during trial, the circuit court gave the jury a limiting instruction.
The trial was divided into two phases. All issues except for punitive damages were determined at the first phase. Defendants argue that the verdicts against them at the first phase must be overturned because there was insufficient evidence to support plaintiffs’ conspiracy claims and because certain trial errors require reversal. We agree with defendants that the verdicts finding them hable for conspiracy cannot stand because the evidence does not permit a conclusion that they agreed with Unarco or Johns-Manville to conceal the health hazards of asbestos from employees. The following is a description of the evidence relevant to our disposition.
I. Medical and Scientific Literature
As circumstantial evidence of defendants’ alleged knowledge of the health hazards of asbestos, plaintiffs presented evidence of the information about these hazards contained in medical and scientific publications during the relevant time periods. Prior to 1930, there were approximately 12 case reports linking asbestos and disease in Britain, the United States, and Europe. In 1930, the first epidemiological study relating asbestos exposure to disease was published. This study, which was published in Britain and the United States, showed that approximately 25% of the asbestos industry workers studied had lung scarring and that the scarring was more severe in individuals who worked in dustier areas and had been exposed to asbestos for longer amounts of time. To prevent such lung damage, the author recommended dust control, respiratory protection, periodic health examinations of the workers, and education of workers concerning the risks of asbestos.
In 1938, a bulletin issued by the United States Public Health Service reported the results of a study on asbestos textile manufacturing employees. The study connected asbestosis, which is irreversible lung scarring caused by asbestos fibers, to exposure to asbestos dust in concentrations greater than five million particles per cubic foot. The bulletin contained a recommendation that dust levels be kept below this threshold in order to prevent asbestosis.
In the 1940s and 1950s, the five million particles per cubic foot threshold limit value for asbestos was adopted by the American Conference of Governmental and Industrial Hygienists and by many states, including Ohio and New Jersey. In the 1950s, however, the five million particles per cubic foot asbestos dust limit was criticized in the medical and scientific literature because it was designed only to prevent asbestosis and did not account for the cancer risk associated with asbestos exposure. Nevertheless, the threshold limit value for asbestos remained the same until 1969, when it was lowered.
Beginning in the 1930s, case reports linking asbestos to lung cancer were published in the medical and scientific literature. In 1955, an epidemiological study involving asbestos and lung cancer was published in the British Journal of Industrial Medicine. This study found that asbestos manufacturing workers with asbestosis had 10 times the rate of lung cancer as the general population. In 1960, an epidemiological study linking asbestos exposure to mesothelioma, a cancer of the membrane surrounding the lungs, was published in the British Journal of Industrial Medicine.
In the 1960s, Dr. Irving Selikoff began a large scale epidemiological study of asbestos and disease. Selikoff became one of the leading authorities on asbestos and disease. Throughout the 1960s, 1970s, and 1980s, Selikoff published the results of an ongoing health survey of members of the International Association of Heat and Frost Insulators and Asbestos Workers Union. In a 1964 article published in the Journal of the American Medical Association, Selikoff stated that, although many researchers had questioned the link between lung cancer and asbestosis, the results of his study proved that there was a connection. He found that asbestos exposure put insulation workers at greater risk for diseases such as asbestosis, lung cancer, and mesothelioma. These diseases were most apparent in workers with more than 20 years since the date of their first exposure.
II. Efforts by Asbestos Product Manufacturers Other
Than Owens Corning and Owens-Illinois to Suppress Information Concerning the Health Hazards of Asbestos
As other circumstantial evidence of defendants’ involvement in the alleged conspiracy, plaintiffs presented evidence intended to demonstrate that defendants’ actions paralleled those of the other alleged conspirators. Plaintiffs’ theory at trial was that parallel conduct by the alleged conspirators demonstrated an agreement among "them to suppress and misrepresent the health hazards of asbestos. The majority of the evidence at trial related to similarities and differences in the activities of the alleged conspirators.
For example, plaintiffs presented evidence that, at the same time that information about the health hazards associated with asbestos exposure was appearing in medical and scientific publications, asbestos product manufacturers other than defendants were causing published reports of such information to be suppressed or altered. There was no evidence that defendants participated in these particular activities. Instead, evidence of the actions of these other companies was part of plaintiffs’ evidence of alleged parallel conduct.
A. The Lanza Study
There was evidence that attorneys for Johns-Manville and Raybestos-Manhattan, another asbestos product manufacturer, edited an article published in 1935 by Dr. Anthony Lanza, a physician employed by Metropolitan Life Insurance Company. Prior to publishing this article, which discussed the results of a study of workers at plants where asbestos-containing products were manufactured, Lanza permitted these attorneys to review the article and incorporated the changes they suggested. These were changes that were favorable to the asbestos-containing products industry or consistent with its legal positions. According to plaintiffs’ expert, Dr. Barry Castleman, the substance of these changes was inconsistent with information contained in medical and governmental literature at the time.
B. Asbestos Magazine
Plaintiffs also presented evidence that, around the same time, Raybestos-Manhattan and Johns-Manville prevented the publication of information concerning the health hazards of asbestos in Asbestos magazine, a monthly trade publication. In September 1935, the editor of Asbestos magazine wrote Sumner Simpson, the president of Raybestos-Manhattan, to ask whether Simpson had any objections to the magazine publishing information about asbestosis and efforts to control it. The editor observed: “Always you have requested that for certain obvious reasons we publish nothing, and, naturally your wishes have been respected.” Simpson consulted with Vandiver Brown, an attorney for Johns-Manville, about the editor’s request, and the two men agreed that it would be better if nothing were said about asbestosis.
C. Results of the Saranac Laboratory Research
Similarly, plaintiffs presented evidence that, due to the actions of certain asbestos product manufacturers other than defendants, information concerning the health hazards of asbestos exposure was omitted from a 1951 article discussing the results of a study conducted by the Saranac Laboratory. In 1936, a group of asbestos product manufacturers agreed to sponsor research on the health effects of asbestos dust. Members of this group included Johns-Manville and Unarco but not defendants. In their agreement with Saranac, the sponsoring companies required that the results of the research remain their property, that they maintain control over the disclosure of the results, and that any manuscript discussing the results be submitted to them for approval prior to publication.
Consistent with this agreement, Saranac submitted a manuscript to the sponsoring companies. At a meeting, these companies agreed that certain changes should be made to the manuscript, and Saranac incorporated these changes. For example, at the request of the sponsoring companies, references to cancer and tumors were removed from the manuscript before publication. The published article nevertheless stated that it is a “complete survey” of the research and did not acknowledge the input of the sponsoring companies.
III. Unarco’s and Johns-Manville’s Plant Operations
Additional evidence relating to plaintiffs’ parallel-conduct theory included testimony that Unarco and Johns-Manville failed to warn their employees of the health hazards of asbestos exposure and failed to protect their employees from these hazards. Employees who worked in Unarco’s Bloomington plant during the fifties and sixties testified that Unarco never told them or other employees that breathing asbestos dust posed any health risks. These employees also described conditions in the Bloomington plant before Owens Corning purchased it from Unarco in 1970. According to these employees, the plant used raw asbestos supplied by Johns-Manville to make insulation and other products. Dust from the plant operations was released into the air of the plant. The dust in the air was visible and covered surfaces in the plant. Ron Thacker, plaintiff Vernadine Thacker’s son, testified that he worked with his father in the plant and, when they left the plant, there was dust on their clothing and hair.
These employees further testified that Bloomington plant employees were not required to wear respirators. One employee stated that he had a respirator, but employees did not use them frequently because they had difficulty obtaining filters for the respirators, and the filters needed to be changed every day. There was some dust-collection equipment in the plant, but it did not collect all of the dust because there was always dust in the air. Unarco had no ventilation equipment and no industrial hygienist. Unarco also did not require employees to get annual chest X rays. According to the industrial hygiene survey Owens Corning conducted during the summer after it purchased the Unarco plant, atmospheric conditions in the Bloomington plant were “unbelievably bad,” and exposures to asbestos were “excessive.”
A Johns-Manville employee testified that, during the period 1946 to 1961, Johns-Manville did not give its employees warnings about the health hazards of asbestos. Johns-Manville did require employees to undergo periodic physical examinations, which included a chest X ray. These X rays were read, however, by Johns-Manville’s own doctors and treated as the property of the company. It was Johns-Manville’s policy not to inform an employee that his X ray showed asbestosis or another lung disease unless the disease became disabling. This policy did not change until the early 1970s.
IV Evidence of Owens-Illinois’ Activities
In an attempt to connect Owens-Illinois and Owens Corning to Unarco and Johns-Manville, plaintiffs presented evidence intended to show that defendants’ actions with respect to their own production and distribution of asbestos-containing products were similar to Unarco’s and Johns-Manville’s. The evidence relating to defendants focused on their manufacture of “Kayla,” a high-temperature hydrous calcium silicate insulation.
In 1948, Owens-Illinois began commercial production and distribution of Kaylo. Kaylo included 15% to 22% asbestos. In 1953, Owens-Illinois and Owens Corning entered into an agreement under which Owens-Illinois continued to manufacture Kaylo, but Owens Corning assumed the responsibility for distribution. In 1958, Owens-Illinois sold the Kaylo division to Owens Corning. Owens Corning also purchased plants where Owens-Illinois had manufactured Kaylo, including its plant in Berlin, New Jersey. Although Owens Corning manufactured other asbestos-containing products, Kaylo was the only asbestos-containing product manufactured by Owens-Illinois. Beginning in the late sixties, Owens Corning performed research to find a substitute for the asbestos in Kaylo. A substitute was found in 1972, after which Owens Corning also did not manufacture asbestos-containing Kaylo.
A. Knowledge and Communication of Health Hazards
With respect to Owens-Illinois’ knowledge of the health hazards associated with exposure to Kaylo, there was evidence that, in 1943, Owens-Illinois hired Saranac Laboratory to determine whether Kaylo presented an air hazard when mixed, sawed, or applied. In its initial correspondence with Owens-Illinois, Saranac expressed the desire to publish any results of this research, but stated that “nothing [would] be published without [Owens-Illinois’] authorization.”
In 1946, Saranac reported to Owens-Illinois that, with respect to the Kaylo experiments, “no serious results [had] developed ***. The dust alone [was] not causing anything suggestive of either silicosis or asbestosis.” According to the preliminary report Saranac provided in 1947, Kaylo dust was biologically inactive, did not cause silicosis or asbestosis, and would probably be harmless if inhaled in moderate amounts by humans over a long period of time. Just one year later, however, in November 1948, Saranac informed Owens-Illinois that further research had revealed that Kaylo dust did in fact cause asbestosis. Saranac reported: “Kayla *** is capable of producing asbestosis and should be handled as a hazardous industrial dust.”
Saranac submitted the final report on the Kaylo research to Owens-Illinois in February 1952. According to this final report, Kaylo was capable of causing a “peribronchiolar fibrosis typical of asbestosis,” as a result of which, “every precaution should be taken to protect workers against inhaling the dust.” Saranac indicated that it hoped to publish the results of the Kaylo research but that it would omit references to Kaylo and Owens-Illinois from any publication to protect the interests of the company. In addition, Saranac told Owens-Illinois that it would submit a manuscript to Owens-Illinois prior to publication and would welcome comments by Owens-Illinois.
Subsequent correspondence, however, indicates that, because of management changes at Saranac, Owens-Illinois was never given the opportunity to review the article discussing the results of the Saranac research before this article was published. The article contained no references to Kaylo or Owens-Illinois.
In an effort to show that Owens-Illinois failed to share the information it had about the health hazards of Kaylo with its employees or consumers, plaintiffs presented evidence that Owens-Illinois failed to place warning labels on Kaylo during the time it manufactured this product. Plaintiffs also showed that, in 1952, Owens-Illinois issued an advertising brochure in which it represented that Kaylo was nontoxic. There was conflicting evidence at trial as to the meaning of “toxic” and the applicability of this term to asbestos.
In addition, plaintiffs presented the testimony of Jerry Helser, who had worked for Owens Corning since 1961. Helser testified that, during the first few years of his employment at Owens Corning, he worked with individuals who had previously worked for Owens-Illinois at its Berlin plant and other locations. According to Helser, none of these former Owens-Illinois employees told him that the asbestos used in Kaylo caused lung scarring and cancer or gave him any indication that Owens-Illinois had informed them of these risks.
Contrary to Helser’s testimony, a former Owens-Illinois employee, Richard Grimmie, testified that Owens-Illinois employees at the Berlin plant did receive warnings about the health hazards associated with exposure to the asbestos in Kaylo dust. Grimmie testified that he began working for Owens-Illinois in 1945. Grimmie stated that, before beginning work at the Berlin plant, he received a physical examination. At the time of this examination, he was told about asbestos and asbestosis and was informed that respirators were required in certain areas of the plant because of the silica and asbestos content of Kaylo.
Grimmie testified that he informed job applicants about the hazards of asbestos when he worked as personnel manager for Owens-Illinois. He testified that he told applicants that respirators were required in certain areas of the plant and that the plant manufactured a product containing silica, which causes silicosis, and asbestos, which causes asbestosis.
B. Plant Conditions
In addition to evidence of Owens-Illinois’ communications concerning the health hazards of exposure to Kaylo dust, there was evidence relating to the conditions of the plants where Kaylo was manufactured. Grimmie testified that the Berlin plant employed a very large, powerful, and well-maintained dust collector with inlets near certain dusty plant operations, such as the saws used to shape the finished Kaylo product. In addition to the dust collector, Owens-Illinois had mechanical sweepers to vacuum dust from the floors and, before the end of each shift, workers were given time to clean their work areas.
According to Grimmie, Owens-Illinois also had a respirator program, under which employees were required to wear respirators in areas where dust could not be controlled. The plant nurse administered the program. Respirators with clean filters were given to workers by their supervisors at the beginning of each shift. At the end of the shift, the respirators were returned to the nurse, who would clean the filters. Workers often refused to wear their respirators and were disciplined for these refusals. Grimmie testified that Owens-Illinois also employed an industrial hygienist, Willis Hazard, who would frequently take dust samples throughout the plants.
Grimmie further testified that Owens-Illinois required every Owens-Illinois employee at the Berlin plant to have a preemployment X ray and annual X rays thereafter. Grimmie was aware of no asbestos-related disease showing on any employee’s X ray during the time Owens-Illinois owned the Berlin plant. Grimmie was also unaware of any asbestos-related workers’ compensation claim made during this time.
In addition to Grimmie’s testimony, there was evidence of an industrial hygiene survey taken at the Berlin plant a few days before and a few days after ownership of the Berlin plant was transferred from Owens-Illinois to Owens Corning in 1958. Consistent with Grimmie’s testimony, the report noted that employees working at certain operations were wearing respirators and that exhaust systems were available at certain locations. Nevertheless, the report stated that several air samples exceeded or closely approached acceptable limits for asbestos dust. One sample taken near a saw showed 91.8 million particles per cubic foot of air. The report also suggested improvements to the plant’s exhaust systems and respirator program.
V Evidence of Owens Coming’s Activities
Extensive evidence of Owens Coming’s activities with respect to its own products and own employees was presented. Again, this evidence was intended to support plaintiffs’ theory that similarities in the activities of Owens Corning and the other alleged conspirators demonstrated the alleged conspiracy.
A. Knowledge and Communication of Health Hazards
To show that, like the other alleged conspirators, Owens Corning was aware of the health hazards of asbestos and failed to communicate information about these hazards to its employees, plaintiffs relied on a 1942 internal memorandum. This memorandum indicated that Owens Corning planned to gather medical and scientific literature concerning asbestosis and use this information as a “weapon-in-reserve” during negotiations with the Asbestos Workers Union. The memorandum proposed that the information on asbestosis be disclosed to union locals only if the union leadership rejected the company’s offer. Other evidence showed that, in 1956, Owens Corning received information from Saranac Laboratory that asbestos had been “fairly well incriminated as a carcinogen.”
Plaintiffs presented other evidence intended to show that Owens Corning made efforts to conceal the information it had about the health risks related to asbestos. For example, according to a sales brochure published by Owens Corning in 1956, Kaylo was “[n] on-irritating to the skin and non-toxic.” In 1966, Owens Corning placed a label on Kaylo cartons concerning its asbestos content, but the label did not identify specific health hazards. The label provided: “This product contains asbestos fiber. If dust is created when this product is handled, avoid breathing the dust. If adequate ventilation control is not possible, wear respirator approved by U.S. Bureau of Mines.” In 1970, Owens Corning changed the label on cartons of Kaylo. The new label read: “Caution — Product contains asbestos fiber. Inhalation of dust in excessive quantities over long periods of time may be harmful. Avoid breathing dust. If adequate ventilation is not possible wear respirators approved by the U.S. Bureau of Mines for pneumoconiosis producing dust.”
In addition to this evidence, plaintiffs relied on numerous internal Owens Corning memoranda as evidence that Owens Corning endeavored to conceal the health risks of asbestos. A 1966 internal memorandum contained a reminder regarding Owens Coming’s “longstanding” policy that inquiries and complaints concerning the health hazards of Owens Corning products should be referred to certain corporate officers or the company’s legal department. A 1967 internal memorandum indicated that an Owens Corning employee had questioned plans to expand Kaylo manufacturing given that “the Government will probably blow the whistle relative to the use of asbestos in the not too distant future.”
In a 1968 internal memorandum, the author recognized the association between asbestos exposure and asbestosis, lung cancer, and mesothelioma, but observed that Owens Coming’s position had been to indicate that “all medical research to date indicates no hazard to health.” Another internal memorandum from that same year concerned the company’s participation in the Insulation Industry Hygiene Council. According to this memorandum, “much care and consideration went into developing the proposed draft for the constitution bylaws of this organization *** [in order to] limit the influence of Dr. Selikoff.” Selikoff was an authority on asbestos and disease who had worked to publicize and protect workers from the health hazards of asbestos. Similarly, internal correspondence from 1970 indicates that Owens Corning was reluctant to participate in an industrial hygiene course organized by Selikoff. Owens Corning believed it would be unwise to attend the course because, inter alla, it would give “tacit approval to Selikoff.”
The evidence presented at trial indicates that Owens Corning first began notifying employees of the health hazards associated with asbestos exposure in the 1970s. Some employees received this information in the early seventies. For example, internal Owens Corning memoranda show that, at meetings in 1971, Berlin employees were informed by their plant manager about the adverse health effects of asbestos exposure. In addition, an employee in the Bloomington plant testified that she first learned of these health risks when Owens Corning sent her and certain other employees to a meeting at Illinois State University in 1971.
The evidence showed that Owens Corning did not inform other employees of these health hazards until 1978. In April of that year, United States Department of Health, Education, and Welfare Secretary Joseph Califano issued a “broadly publicized statement bringing attention to the possible increased risk of death from an asbestos related disease many years after exposure had terminated.” Califano suggested that exposed workers should stop smoking cigarettes and receive a physical examination from a physician. Around the time of the Califano announcement, the Surgeon General issued a “Physicians Advisory” concerning the health effects of asbestos.
Following the Califano announcement, Owens Corning communicated these health hazards to all Blooming-ton plant employees. Before that time, “no formal education programs regarding asbestos hazards” had occurred at the Bloomington plant. After the Califano announcement, Owens Corning also notified former Bloomington plant employees of the diseases that could be caused by asbestos exposure.
There was also testimony, however, that not all Owens Corning employees received information from their employer about the health risks related to asbestos. According to Jerry Helser, Owens Corning never informed him of these risks. Helser testified that Owens Corning never told him that asbestos could cause asbestosis, lung cancer, or mesothelioma. He was never informed that Kaylo dust was a hazardous industrial dust or that he should wear a respirator to avoid inhaling Kaylo dust because the dust was harmful.
B. Plant Conditions
In addition to this evidence of Owens-Corning’s knowledge of and failure to communicate the health hazards of asbestos, the parties presented evidence of conditions at plants where Owens Corning produced asbestos-containing products. In 1951, the Saranac Laboratory conducted an industrial hygiene survey at Owens-Illinois’ Sayreville, New Jersey, plant, at which Kaylo was manufactured. The report concluded that “considerable attention had been given in the plant to the control of dust.” Given that some of the dust samples approached the “maximum allowable limits,” however, the report recommended certain improvements to the plant’s dust control measures.
In 1961, only two of the dust samples taken during an industrial hygiene survey of the Berlin plant exceeded the threshold limit value of five million particles per cubic foot. Of these two samples, one only slightly exceeded the limit and, with respect to the other, “exposure was intermittent.” The survey report made no recommendations for improvements.
Helser testified that, when he worked at the Berlin plant in the sixties, the plant had a ventilation and dust-collection system, which had intake vents at all the saws and sanders used to shape the final product. Some dust, however, escaped the system. In addition to the dust-collection system, the Berlin plant had a respirator program under which respirators were required in certain areas of the plant and a program under which employees received annual physical examinations and X rays. Throughout his career at Owens Corning, Helser received annual chest X rays.
One Bloomington plant employee testified that, after the sale of the Bloomington plant to Owens Corning in 1970, the plant continued to operate. Owens Corning installed dust-collection equipment in the plant, which helped control the dust, but it took approximately a year for this equipment to be installed. Owens Corning asked employees to wear respirators, but did not make this a requirement. It also held safety meetings with employees.
According to a 1978 internal Owens Corning memorandum, the Berlin plant was cleaned after asbestos production stopped there in 1972, and air samples showed that the concentrations of asbestos there were extremely low. Insulation containing asbestos was produced at the Bloomington plant only IV2 years after Owens Corning purchased the plant. Nevertheless, the memorandum acknowledged that “there were significant exposures” at both of these plants and that cases of asbestosis had arisen in workers from both plants.
VI. Evidence of Contacts Among Defendants and Other Asbestos Product Manufacturers
Although the majority of the evidence presented at trial concerned the separate, but allegedly parallel, activities of defendants and other asbestos-containing product manufacturers, plaintiffs did present some evidence of contacts among these companies.
A. Owens Coming’s Contacts With Owens-Illinois
Plaintiffs presented the following evidence of a relationship between Owens Corning and Owens-Illinois. Owens Corning was formed in 1938 by Owens-Illinois and Corning Glass. In addition, William Boeschenstein, Owens Coming’s chief executive officer, testified that Owens-Illinois “owned a large part of Owens-Corning,” although he did not indicate the time period of this ownership.
With respect to contacts between these companies, there was evidence that, in 1941, an Owens Corning attorney returned two published articles he had borrowed from Owens-Illinois’ industrial hygienist. These articles concerned the health effects of asbestos and dust-control methods in asbestos manufacturing plants.
Evidence of other contacts between these companies included Owens Coming’s agreement to distribute Kaylo in 1953 and to later purchase the Kaylo division from Owens-Illinois in 1958. In addition, the 1956 Owens Corning brochure advertising Kaylo as nontoxic contained the information that Kaylo was manufactured by Owens-Illinois and was a trademark of this company. Under the distribution agreement, Owens Corning was required to use Owens-Illinois’ trademark and trade name.
B. Labeling
Plaintiffs’ evidence of contacts between Owens Corning and Johns-Manville included a 1964 internal Owens Corning memorandum. This memorandum indicated that Johns-Manville’s medical director had informed Owens Corning that, in October of that year, Johns-Manville planned to label shipping cartons of asbestos-containing products with warnings of the “alleged hazards of asbestos.” The author of the Owens Corning memorandum observed:
“It is obvious that these warning labels will have some impact in the field and possibly upon Public Health and other government officials.
The question before us is whether or not Fiberglas Kaylo should protect itself against more stringent and punitive health laws and the possibility of third party actions by following the J-M lead.”
A 1965 internal Owens Corning memorandum indicated that, several months after receiving this information from Johns-Manville, Owens Corning was still deciding whether to place warning labels on its own asbestos-containing products. The author of the 1965 memorandum recommended that Owens Corning “continue to give serious consideration to the labeling of *** Kaylo products in a manner similar to that currently being used by Johns-Manville*** [because] the fact that Johns-Manville is labeling their preformed products is in itself a pressure on the whole industry to consider labeling.”
C. NIMA Pamphlet
Plaintiffs also presented evidence that Owens Corning and Johns-Manville employees were involved in the National Insulation Manufacturers Association, Inc. (NIMA), a trade organization. In 1968, Owens Coming’s medical director joined this organization as a representative for his company. There was also evidence that another Owens Corning employee and a Johns-Manville employee participated in drafting a pamphlet published by NIMA in 1969 or 1970.
This pamphlet, entitled “Recommended Health Safety Practices For Handling and Applying Thermal Insulation Products Containing Asbestos,” did not identify the specific health hazards involved with asbestos exposure, such as asbestosis, lung cancer, or mesothelioma. Instead, the pamphlet stated generally that there were certain health risks associated with asbestos insulation. For example, the introduction provided:
“The modern mass production of hundreds of natural and synthetic materials, such as asbestos, has brought the need to protect workers from certain health risks, known or suspected.
* * *
Based on the fact that all materials can be handled safely, the National Insulation Manufacturers Association recommends the following practices for handling and applying Thermal Insulation Products containing asbestos and/or other potentially injurious material to protect the worker.”
The recommendations contained in the pamphlet included taking precautions to minimize dust, wearing United States Bureau of Mines respirators, observing good housekeeping practices, and using water to clean material.
D. Owens Coming’s Purchase of the Bloomington Plant
Other evidence pertained to contacts between Owens Corning and Unarco. As evidence of the alleged conspiracy between Owens Corning and Unarco, plaintiffs rely on the indemnity clause in Owens Coming’s 1970 agreement to purchase the Bloomington plant from Unarco. Under this clause, Unarco agreed, inter alla, to indemnify Owens Corning for claims based on asbestosis or other respiratory ailments resulting from employees’ exposure to Unarco products prior to Owens Coming’s purchase of the plant. No indemnification was to be provided for pending litigation.
E. 1978 Konzen Memorandum
Other contacts between Owens Corning and the alleged conspirators were evidenced by a 1978 internal memorandum written by Dr. Jon Konzen, Owens Corning’s medical director. In this memorandum, Konzen noted that the Califano. announcement had caused Owens Corning to review its handling of former employees from the Berlin and Bloomington plants. He stated that, as part of this review, Owens Corning had contacted other companies who had similar operations, including “JohnsManville, CSG, GAF, Armstrong, Keene, Eagle-Picher, Pabco and O-I.”
Konzen described the information obtained from these companies with respect to their handling of asbestos-exposed workers. Johns-Manville employees exposed to asbestos were given educational material and presentations concerning the health risks. At retirement, Johns-Manville informally offered employees the opportunity to return for periodic physical examinations. Owens-Illinois indicated that it did not plan to take any action with respect to former Owens-Illinois employees who worked at the Berlin plant. Other companies did not plan to offer examinations to employees or to educate them concerning the health risks of asbestos.
Konzen recommended in the memorandum that retirees and former employees of Owens Corning be offered periodic physical examinations. In addition, he suggested that this group of individuals be notified by letter that they had been exposed to asbestos, that the exposure might place them at a higher risk of contracting an asbestos-related disease, that they should stop smoking, and that they should schedule physical examinations by their personal physicians.
F. 1979 Meeting
In addition, plaintiffs presented evidence of defendants’ involvement in two meetings with other asbestos-containing product manufacturing companies. One of these meetings was held in 1979 following the Califano announcement. The chief officers of asbestos-containing product manufacturers were invited to a meeting at Johns-Manville offices to discuss the “Congressional, media and regulatory forces being brought to bear [which] raise serious questions as to the vi[a]bility of the industry.” Owens Corning and Unarco attended this meeting, but Owens-Illinois did not.
G. 1983 Asbestos Claims Conference
The other meeting was held in 1983 and was entitled the “Asbestos Claims Conference.” Attendees included both defendants, Johns-Manville, other former asbestos-containing product manufacturers, and insurance companies. The agenda from the meeting indicated that among the topics discussed at the meeting were bankruptcy proceedings, legislation, and strategies for resolving asbestos litigation. Written material from the meeting also listed common goals of asbestos product manufacturers as establishing a “better, more equitable and efficient system to resolve the asbestos litigation”; developing a consortium to provide an umbrella of insurance; passing legislation to accomplish the companies’ goals; and establishing a “representative industry group” consisting of executives.
H. Castleman’s Opinion
In addition to this evidence of contacts between defendants and the other alleged conspirators, plaintiffs presented the opinion of their expert witness, Dr. Barry Castleman, that defendants were involved in a conspiracy with Johns-Manville, Unarco, and other companies to suppress information about the adverse health effects of asbestos exposure. Castleman testified that he holds a doctorate in occupational and environmental health policy and has researched and written extensively on the historical response of corporations to the asbestos health hazard. Castleman opined that Owens-Illinois’ and Owens Coming’s failure to educate workers and users of the health hazards of asbestos was part of an “implicit understanding among the companies that nobody would blow the whistle on asbestos and they would avoid the problems of labor unrest, strikes, other kinds of things.” He explained that there was no formal agreement but that, based on the failure to disclose the health risks of asbestos, he was “assuming” that there was an understanding that “[n]obody would raise the health issue and everybody would benefit thereby, at least in the short term financially who were selling asbestos-containing products, like Kaylo.” According to Castleman, the companies involved in the alleged conspiracy included Johns-Manville, Raybestos-Manhattan, Unarco, Owens Corning, and Owens-Illinois.
Following the presentation of this evidence at trial, the jury found in favor of Bicknell and Thacker against Owens Corning. The jury awarded Bicknell $225,000 and awarded Thacker $220,000 in damages. The jury also returned a verdict in favor of McClure against both Owens Corning and Owens-Illinois. It awarded McClure $400,000 in damages and determined that each defendant was responsible for one half of these damages. At the second phase of the trial, the jury rejected Thacker’s claim for punitive damages. The second phase was limited to Thacker’s claim against Owens Corning for punitive damages. The other two plaintiffs did not make requests for punitive damages.
The appellate court affirmed the judgments entered on these verdicts. The appellate court rejected defendants’ argument that the evidence was insufficient to prove them liable for civil conspiracy. The appellate court held: “If evidence of parallel conduct is sufficiently persuasive we see no reason why a jury could not rely upon it alone to find that a conspiracy existed.” 298 Ill. App. 3d at 598. Further, the appellate court found that, in addition to parallel conduct, plaintiffs had shown “contacts” between Owens Corning and other companies in the industiy, as well as “ties” between Owens Corning and Owens-Illinois. 298 Ill. App. 3d at 598. According to the appellate court, this evidence was sufficient to support the jury’s verdicts against defendants. With respect to the trial errors asserted by defendants, the appellate court held that there was no error or that any error was harmless. 298 Ill. App. 3d at 599-604. Both defendants filed petitions for leave to appeal the appellate court’s decision. We granted each petition and consolidated their appeals for review.
ANALYSIS
In this court, both defendants argue that the circuit court erred in denying their post-trial motions for judgments notwithstanding the verdict or, in the alternative, new trials. According to defendants, the jury verdicts against them must be overturned because the evidence was insufficient to support the jury’s determination that they participated in the alleged conspiracy. In addition, defendants assert numerous trial errors, including (1) the admission of extensive evidence under the co-conspirator exception to the hearsay rule, (2) the admission of Castleman’s opinion that defendants were involved in a conspiracy, (3) the circuit court’s exclusion of the deposition testimony of Willis Hazard, (4) the inclusion of facts not in the record and inflammatory statements in plaintiffs’ closing argument; (5) error and bias in the instructions the circuit court gave to the jury, and (6) the circuit court’s refusal of instructions and special interrogatories submitted by defendants. According to defendants, these errors, individually and collectively, denied them a fair trial.
We begin our review by addressing defendants’ argument that the insufficiency of the evidence required that the circuit court enter judgments notwithstanding the verdict in their favor or, in the alternative, grant them new trials. The standards for granting each of these two forms of relief differ. “Judgment notwithstanding the verdict should not be entered unless the evidence, when viewbd in the light most favorable to the opponent, so overwhelmingly favors the movant that no contrary verdict based on that evidence could ever stand.” Holton v. Memorial Hospital, 176 Ill. 2d 95, 109 (1997); Pedrick v. Peoria & Eastern R.R. Co., 37 Ill. 2d 494, 510 (1967). Judgment notwithstanding the verdict is not appropriate if “reasonable minds might differ as to inferences or conclusions to be drawn from the facts presented.” Pasquale v. Speed Products Engineering, 166 Ill. 2d 337, 351 (1995). As this court stated in Maple v. Gustafson, 151 Ill. 2d 445, 453 (1992):
“A trial court cannot reweigh the evidence and set aside a verdict merely because the jury could have drawn different inferences or conclusions, or because the court feels that other results are more reasonable. [Citations.] Likewise, the appellate court should not usurp the function of the jury and substitute its judgment on questions of fact fairly submitted, tried, and determined from the evidence which did not greatly preponderate either way. [Citations.]” Maple, 151 Ill. 2d at 452-53.
We apply a de nova standard to our review of decisions on motions for judgments notwithstanding the verdict. See, e.g., Gaffney v. City of Chicago, 302 Ill. App. 3d 41, 48 (1998).
By contrast, a circuit court will order a new trial if, after weighing the evidence, the court determines that the verdict is contrary to the manifest weight of the evidence. Maple, 151 Ill. 2d at 454. “ ‘A verdict is against the manifest weight of the evidence where the opposite conclusion is clearly evident or where the findings of the jury are unreasonable, arbitrary and not based upon any of the evidence.’ ” Maple, 151 Ill. 2d at 454, quoting Villa v. Crown Cork & Seal Co., 202 Ill. App. 3d 1082, 1089 (1990). A reviewing court will not reverse a circuit court’s decision with respect to a motion for a new trial unless it finds that the circuit court abused its discretion. Maple, 151 Ill. 2d at 455.
In addition to these principles, our review of the circuit court’s decision is governed by the elements of the cause of action alleged by plaintiffs in their complaints. Plaintiffs allege no employment relationship between their husbands and defendants, nor do they allege that they or their husbands were exposed to defendants’ products. Plaintiffs base their claims against defendants exclusively on a theory of civil conspiracy. According to plaintiffs, defendants are liable for the injuries alleged in their complaints because defendants conspired with Unarco and Johns-Manville to suppress information about the harmful effects of asbestos and to falsely represent that it was safe to work with asbestos-containing materials.
Civil conspiracy is defined as “a combination of two or more persons for the purpose of accomplishing by concerted action either an unlawful purpose or a lawful purpose by unlawful means.” Buckner v. Atlantic Plant Maintenance, Inc., 182 Ill. 2d 12, 23 (1998). In order to state a claim for civil conspiracy, a plaintiff must allege an agreement and a tortious act committed in furtherance of that agreement. Adcock v. Brakegate, Ltd., 164 Ill. 2d 54, 62-64 (1994). The agreement is “a necessary and important” element of this cause of action. Adcock, 164 Ill. 2d at 62. The civil conspiracy theory has the effect of extending liability for a tortious act beyond the active tortfeasor to individuals who have not acted but have only planned, assisted, or encouraged the act. Adcock, 164 Ill. 2d at 62-63.
Civil conspiracy is an intentional tort and requires proof that a defendant “knowingly and voluntarily participates in a common scheme to commit an unlawful act or a lawful act in an unlawful manner.” Adcock, 164 Ill. 2d at 64. Accidental, inadvertent, or negligent participation in a common scheme does not amount to conspiracy. Adcock, 164 Ill. 2d at 64. Mere knowledge of the fraudulent or illegal actions of another is also not enough to show a conspiracy. Tribune Co. v. Thompson, 342 Ill. 503, 530 (1930). Similarly, “[a] defendant who innocently performs an act which happens to fortuitously further the tortious purpose of another is not liable under the theory of civil conspiracy.” Adcock, 164 Ill. 2d at 64. However, “[a] defendant who understands the general objectives of the conspiratorial scheme, accepts them, and agrees, either explicitly or implicitly to do its part to further those objectives *** is liable as a conspirator.” Adcock, 164 Ill. 2d at 64.
A conspiracy is almost never susceptible to direct proof. Walsh v. Fanslow, 123 Ill. App. 3d 417, 422 (1984). Usually, it must be established “from circumstantial evidence and inferences drawn from evidence, coupled with common-sense knowledge of the behavior of persons in similar circumstances.” Adcock, 164 Ill. 2d at 66. If a civil conspiracy is shown by circumstantial evidence, however, that evidence must be clear and convincing. E.g., Majewski v. Gallina, 17 Ill. 2d 92, 99 (1959); Tribune Co., 342 Ill. at 529; Bosak v. McDonough, 192 Ill. App. 3d 799, 804 (1989).
In this case, defendants argue that the jury verdicts finding them liable for civil conspiracy must be overturned because the evidence showed no agreement between either of them and Unarco or Johns-Manville to conceal or misrepresent information concerning the health hazards of asbestos. They challenge the appellate court’s ruling that evidence of similarities between defendants’ conduct and the conduct of Unarco and Johns-Manville was sufficient to establish the required agreement for civil conspiracy. According to defendants, the evidence showed that they acted unilaterally, and mere parallel conduct is never enough to establish that there was an agreement for purposes of civil conspiracy.
In addition, defendants argue that the appellate court erred in concluding that there was any evidence other than parallel conduct that demonstrated an agreement between defendants and the alleged conspirators. Defendants assert that any contacts they may have had with each other failed to show their involvement in the alleged conspiracy with Unarco or Johns-Manville. In addition, defendants argue, the few contacts Owens Corning had with others in the industry were for legitimate business reasons and not due to the alleged conspiracy. Based on the lack of evidence of an agreement, defendants contend that the jury verdicts cannot stand.
Our review of the sufficiency of the evidence in this case, therefore, involves a threshold legal question: whether parallel conduct alone can suffice as proof of agreement for civil conspiracy. Prior to the appellate court decision in this case, no Illinois court had addressed this question. But see Smith v. Eli Lilly & Co., 173 Ill. App. 3d 1, 28-30 (1988) (holding that parallel activity by drug manufacturers in producing and marketing DES does not establish civil conspiracy or concerted action), rev’d on other grounds, 137 Ill. 2d 222 (1990). This case, therefore, presents us with an issue of first impression. After reviewing decisions on point from other jurisdictions, as well as related precedent in Illinois, we hold that parallel conduct may serve as circumstantial evidence of a civil conspiracy among manufacturers of the same or similar products but is insufficient proof, by itself, of the agreement element of this tort.
Our review of case law from other jurisdictions convinces us that the overwhelming weight of authority has refused to accept mere parallel action as proof of conspiracy. In the context of federal antitrust litigation, it is well-settled that mere consciously parallel behavior is insufficient to establish a conspiracy under the Sherman Act (15 U.S.C. § 1 (1994)). See, e.g., Petruzzi’s IGA Supermarkets, Inc. v. Darling-Delaware Co., 998 F.2d 1224, 1242 (3d Cir. 1993); Reserve Supply Corp. v. Owens-Corning Fiberglas Corp., 971 F.2d 37, 50-51 (7th Cir. 1992); Apex Oil Co. v. DiMauro, 822 F.2d 246, 252 (2d Cir. 1987). According to these courts, conscious parallelism is circumstantial evidence of a conspiracy, but additional evidence is required before the inference of an agreement is permitted. Some federal courts have held that “substantial additional evidence” is required. See Coleman v. Cannon Oil Co., 849 F. Supp. 1458, 1466 (M.D. Ala. 1993). Other courts have required only additional evidence that reasonably tends to exclude the possibility that the defendants were acting independently. See, e.g., Todorov v. DCH Healthcare Authority, 921 F.2d 1438, 1456 (11th Cir. 1991). Such additional evidence includes evidence of “(1) actions contrary to the defendants’ economic interests, and (2) a motivation to enter into such an agreement.” Petruzzi’s IGA, 998 F.2d at 1242. The reason federal courts require evidence in addition to mere parallel conduct before imposing liability for conspiracy is to avoid inadvertent condemnation of non-conspiratorial conduct. See Coleman, 849 F. Supp. at 1466. As the Coleman court explained, “The requirement is intended to help assure that there is a reasonable basis to conclude that persons exchanged assurances of common action or otherwise adopted a common plan, albeit not necessarily through meetings, conversations, or exchanged documents.” Coleman, 849 F. Supp. at 1466. The federal court rule with respect to evidence of parallel conduct has been adopted in one Illinois case involving a claim under the Illinois Antitrust Act (740 ILCS 10/1 et seq. (West 1994)), which is based on the Sherman Act. See Trowbridge Farm Supply Co. v. W.R. Grace & Co., 82 Ill. App. 3d 1140 (1980) (finding the plaintiffs evidence of parallel conduct insufficient to establish a conspiracy to restrain trade).
Similarly, in cases involving the tort of conspiracy, courts in other jurisdictions have held that proof of mere parallel conduct is insufficient. See, e.g., In re Asbestos School Litigation, 46 F.3d 1284, 1292 (3d Cir. 1994); Burnside v. Abbott Laboratories, 351 Pa. Super. 264, 280, 505 A.2d 973, 982 (1985); Collins v. Eli Lilly Co., 116 Wis. 2d 166, 188, 342 N.W.2d 37, 47-48 (1984). For example, in Collins v. Eli Lilly Co., the Wisconsin Supreme Court held that the plaintiffs proof of parallel conduct was insufficient to establish the defendants’ liability under a civil conspiracy theory. The plaintiff had sued the defendants, who were manufacturers of the drug diethylstilbestrol (DES), for injuries she allegedly suffered as a result of her mother’s ingestion of this drug while pregnant. According to the plaintiff, the defendant drug companies engaged in a conspiracy to obtain FDA approval in 1941 and 1947, to market DES as a treatment for pregnancy problems, and to misrepresent that it was safe to use the drug for this purpose. Collins, 116 Wis. 2d at 187, 342 N.W.2d at 47.
The Collins court rejected the plaintiffs conspiracy claim. According to the court:
“[T]he drug companies apparently engaged in parallel behavior in both 1941 and 1947, but parallel behavior alone cannot prove agreement. There is no indication in the record that the defendants either explicitly or tacitly collaborated to gain FDA approval so that they could in turn collaborate to misrepresent the safety and efficacy of DES for use in preventing miscarriages.” Collins, 116 Wis. 2d at 188, 342 N.W.2d at 47-48.
The Collins court, therefore, held that the summary judgment entered in favor of the defendants was appropriate.
Proof of parallel conduct was held insufficient in another DES case involving a civil conspiracy claim. In Burnside v. Abbott Laboratories, the Pennsylvania Supreme Court held that the plaintiffs’ proof of parallel conduct was insufficient to hold the defendant drug manufacturers liable for civil conspiracy. The plaintiffs alleged that each of the defendants marketed DES in a generic form as a miscarriage preventative; that they knew or should have known that it was potentially carcinogenic; that they failed to test the drug for carcinogenic or teratogenic effects; and that they marketed it without warnings. Burnside, 351 Pa. Super, at 280, 505 A.2d at 981-82.
The Burnside court held that these allegations of parallel conduct were insufficient to withstand the defendants’ motion for summary judgment. The court explained:
“[P]laintiffs in the instant case have failed to allege the manner in which a conspiratorial scheme was devised and carried out. The complaint contains no averments of meetings, conferences, telephone calls, joint filings, cooperation, consolidation, or joint licensing. The plaintiffs have alleged no more than a contemporaneous and negligent failure to act. This was insufficient to state either a conspiratorial agreement or the requisite intent to cause injury.” Burnside, 351 Pa. Super, at 280, 505 A.2d at 982.
For these reasons, the court found that the trial court had properly granted summary judgment in favor of the defendants on this claim.
Similarly, in In re Asbestos School Litigation, the United States Court of Appeals for the Third Circuit held that a former manufacturer of asbestos-containing products could not be held liable under a civil conspiracy theory based on conscious parallel activity with other asbestos manufacturers. The evidence of parallel conduct presented in that case was as follows: (1) the defendant began to sell Kilnoise, an asbestos-containing product, in 1964 without warnings; (2) in 1965, the defendant learned of the connection Dr. Selikoff had made between asbestos and cancer; (3) thereafter the defendant continued to sell Kilnoise; (4) the defendant and other asbestos-containing product manufacturers sold their products without warnings despite knowledge of the dangers of these products; and (5) the defendant and other asbestos-containing product manufacturers were aware that each was selling these products without warnings. In re Asbestos School Litigation, 46 F.3d at 1291. Although the court found that these actions indicated that the defendant and the other manufacturers engaged in parallel courses of conduct, it held that “conscious parallelism is not sufficient to establish *** a civil conspiracy.” In re Asbestos School Litigation, 46 F.3d at 1292.
In cases involving claims of concerted action, a tort very similar to civil conspiracy,4 evidence of parallel conduct has also been deemed insufficient by courts in other jurisdictions. See, e.g., Payton v. Abbott Labs, 512 F. Supp. 1031, 1037 (D. Mass. 1981); Ryan v. Eli Lilly & Co., 514 F. Supp. 1004, 1016 (D.S.C. 1981); Ford Motor Co. v. Wood, 119 Md. App. 1, 36, 703 A.2d 1315, 1332 (1998); Rastelli v. Goodyear Tire & Rubber Co., 79 N.Y.2d 289, 296, 591 N.E.2d 222, 224-25, 582 N.Y.S.2d 373, 375-76 (1992); Martin v. Abbott Laboratories, 102 Wash. 2d 581, 599, 689 P.2d 368, 379 (1984). In one of these cases, Rastelli, the Court of Appeals of New York held that the manufacturer of a multipiece tire rim could not be held liable under a concert of action theory for injuries caused by a rim manufactured by another company. The plaintiff had alleged that the defendant and other manufacturers had lobbied OSHA to make employers rather than manufacturers responsible for safe truck maintenance; had chosen not to issue warnings; had prevented a ban of the multipiece rims; and had refused to voluntarily recall the rim that injured the plaintiffs decedent. Rastelli, 79 N.Y.2d at 296, 591 N.E.2d at 224, 582 N.Y.S.2d at 375.
According to the Rastelli court, these allegations were insufficient to support the plaintiffs claim of concerted action. The court stated that the plaintiff had showed parallel activity by the manufacturers, but parallel activity by companies developing and marketing the same product is insufficient, in itself, to show an agreement under a concerted action theory. Rastelli, 79 N.Y.2d at 296, 591 N.E.2d at 224-25, 582 N.Y.S.2d at 375-76. The Rastelli court noted that it had refused to accept mere parallel activity as proof of concerted action because to do so would “expand the concept of concerted action beyond a rational or fair limit.” The Rastelli court explained that “because application of concerted action renders each manufacturer jointly liable for all damages stemming from any defective product of an entire industry, parallel activity by manufacturers is not sufficient justification for making one manufacturer responsible for the liability caused by the product of another manufacturer.” Rastelli, 79 N.Y.2d at 295-96, 591 N.E.2d at 224, 582 N.Y.S.2d at 375.
Like these other courts, we find that requiring more than proof of mere parallel conduct in civil conspiracy cases involving manufacturers of the same or similar products is necessary to make certain that there is a reasonable basis for inferring an agreement and to minimize the risk that liability will be imposed based on nonconspiratorial conduct. Our conclusion that parallel conduct alone is insufficient to establish civil conspiracy in such cases finds support in the clear and convincing standard of proof that applies to the elements of that tort when the evidence is circumstantial, as it is in the case before us. See, e.g., Bosak, 192 Ill. App. 3d at 804 (noting the standard of proof for civil conspiracy). Under this clear and convincing standard, “if the facts and circumstances relied upon are as consistent with innocence as with guilt it is the duty of the court to find that the conspiracy has not been proved.” Tribune Co., 342 Ill. at 529; see also Regan v. Garfield Ridge Trust & Savings Bank, 220 Ill. App. 3d 1078, 1091-92 (1991); ABC Trans National Transport, Inc. v. Aeronautics Forwarders, Inc., 90 Ill. App. 3d 817, 830 (1980). As defendants and the amici argue, there are many potential innocent explanations for parallel conduct by competitors. These include encountering the same business problems, the same consumer demands, and the same competitive pressures. As defendants observe, “[bjasic economic principles dictate that competitive companies will often act in a highly similar manner.” See also, e.g., Sindell v. Abbott Laboratories, 26 Cal. 3d 588, 606, 607 P.2d 924, 933, 163 Cal. Rptr. 132, 141 (1980) (it is a common practice in the industry for manufacturers to use the experience and methods of others making the same or similar products). Parallel conduct alone by manufacturers of the same or similar products is, therefore, as consistent with innocence as with guilt and cannot be considered, in itself, clear and convincing evidence of a conspiracy.
Not only does our rejection of mere parallel conduct as proof of civil conspiracy comport with the clear and convincing standard of proof Illinois courts have applied to this tort, it is consistent with this court’s previous descriptions of the scope of a manufacturer’s liability. In rejecting the market share theory of product liability in Smith v. Eli Lilly & Co., 137 Ill. 2d 222 (1990), this court reasoned that imposing liability under this theory violated the principles that manufacturers are not “insurers of their industry” and that “ ‘a logical limit must be placed on the scope of a manufacturer’s liability.’ ” Smith, 137 Ill. 2d at 266-67, quoting Woodill v. Parke Davis & Co., 79 Ill. 2d 26, 37 (1980) (requiring plaintiffs to prove that a defendant knew of or should have known of the danger in order to establish strict product liability for failure to warn).
Requiring proof of more than parallel action to establish civil conspiracy liability is necessary to protect manufacturers from becoming insurers of their industry. This case illustrates the potential for industrywide liability under the civil conspiracy theory. Plaintiffs were allegedly injured as a result of the actions and products of defendants’ competitors. Defendants have been found liable for the injuries plaintiffs allege in their complaints even though it is undisputed that neither plaintiffs nor their husbands were employed by defendants, worked at the Unarco plant after Owens Corning purchased it, or used defendants’ products. To permit conspiracy liability based on proof of parallel action alone, when competitors engage in similar conduct for many nonconspiratorial reasons, would expand the civil conspiracy theory “beyond a rational or fair limit” (Rastelli, 79 N.Y.2d at 295, 591 N.E.2d at 224, 582 N.Y.S.2d at 375). Requiring evidence in addition to parallel conduct ensures that a manufacturer’s responsibility for the actions of a competitor is based on more than speculation and conjecture. Accordingly, we hold that, while mere parallel conduct may serve as circumstantial evidence of an agreement under the civil conspiracy theory, it cannot, in itself, be considered clear and convincing evidence of such an agreement among manufacturers of the same or similar products.
Having decided that parallel action alone will not support liability under the civil conspiracy theory in cases such as the one before us, our review of the evidence becomes a two-step process. First, we must examine whether there is evidence to support a finding by the jury that defendants and the alleged conspirators engaged in parallel conduct. Next, we must determine whether any evidence, other than evidence of parallel conduct, was presented at trial and whether this evidence, considered with any evidence of parallel conduct, was sufficient to establish the existence of an agreement between defendants and Unarco or Johns-Manville to suppress or misrepresent information regarding the health hazards of asbestos.
Plaintiffs presented no direct evidence of an agreement. Instead, they relied entirely on circumstantial evidence to prove the alleged agreement. The majority of this evidence related to plaintiffs’ theory that parallel conduct by these companies demonstrated such an agreement. The evidence showed that, in the 1930s, 1940s, and 1950s, there were numerous reports in the medical and scientific literature linking asbestos exposure to asbestosis and cancer. Despite this information, defendants, Unarco, and Johns-Manville produced and sold asbestos-containing products during this time period. Before 1964, none of these companies placed warnings on their asbestos-containing products. Johns-Manville was the first to place a warning label on its products in 1964. In 1966, Owens Corning also added warning labels to its asbestos-containing products. These warnings did not, however, identify the specific diseases caused by asbestos exposure.
There was evidence that Unarco and Johns-Manville prevented information about the health hazards of asbestos from being published. At their request, Asbestos magazine refrained from publishing articles on this topic. In addition, these companies required Saranac Laboratory to omit references to cancer and tumors from the 1951 article it published concerning the results of asbestos research sponsored by Unarco, Johns-Manville, and other asbestos product manufacturers.
While Owens-Illinois did not interfere with Saranac’s publication of the results of the asbestos research Owens-Illinois sponsored, there was evidence that, like Unarco and Johns-Manville, Owens-Illinois caused inaccurate information about the health hazards of its asbestos-containing product to be published. Despite knowledge that Kaylo dust caused asbestosis, in 1952 Owens-Illinois published a brochure stating that Kaylo was “non-toxic.”
Other evidence indicated that, like Unarco and JohnsManville, Owens Corning failed to share information about the health hazards of asbestos with the public. In the 1950s, Owens Corning also published a brochure representing Kaylo as “non-toxic.” Internal company memoranda showed that (1) Owens Corning chose to use information on the health hazards of asbestos as a “weapon-in-reserve” during union negotiations rather than freely disclose this information; (2) the company had a policy that required complaints about health hazards of its products to be referred to certain corporate officers or its legal department; (3) the company was concerned that the government would “blow the whistle” on asbestos; (4) despite knowledge of the health hazards of asbestos, the company maintained a position that the medical research indicated no such hazards; and (5) the company tried to limit the influence of Dr. Selikoff, who had publicized and attempted to protect workers from the hazards of asbestos exposure.
There was also evidence that defendants, Unarco, and Johns-Manville failed to warn their employees of, and adequately protect them from, the health hazards of asbestos. Former Unarco employees who worked in the Bloomington plant testified that Unarco failed to warn them of these hazards, did not have a respirator program, had almost no dust-collection equipment, permitted plant conditions that were “unbelievably bad,” employed no industrial hygienist, and had no annual X-ray program for employees. Likewise, a former Johns-Manville employee testified that this company also did not tell employees of the adverse health effects of asbestos exposure and, although periodic X rays were required, JohnsManville did not tell employees of disease that appeared on the X rays unless the disease became disabling.
Owens Corning employees testified that, like Unarco and Johns-Manville, Owens Corning failed to warn its employees of the health hazards of asbestos. Some employees testified that Owens Corning did not warn them of these hazards until the 1970s. Another employee, Jerry Helser, testified that he was never warned of the hazards.
There was conflicting testimony with respect to Owens-Illinois’ efforts to warn its employees of the dangers of asbestos. Helser testified that former Owens-Illinois employees with whom he had contact did not communicate any such warnings to him. By contrast, Richard Grimmie testified that Owens-Illinois did warn its employees that asbestos exposure could cause asbestosis.
While there was evidence that the conditions and dust-control measures in Owens Coming’s and Owens-Illinois’ plants were better than those in Unarco’s Bloomington plant, there was also evidence that, like Unarco, these companies did not adequately control the dust in their plants. Some dust samples taken in these plants exceeded the threshold limit value for asbestos. After purchasing the Bloomington plant, Owens Corning did not install dust collection equipment immediately and did not require employees to wear respirators. In addition, by 1972, many employees from the Berlin plant had been diagnosed with asbestosis.
Defendants dispute that the evidence showed that their conduct paralleled that of Unarco or JohnsManville. For example, they assert, there was evidence that the warnings they gave their employees, the conditions in their plants, and their industrial hygiene programs were better than those of Unarco and JohnsManville. In reviewing a motion for judgment notwithstanding the verdict, however, a court may not resolve conflicts in the evidence, and the evidence must be considered in the light most favorable to the nonmoving party. See Maple, 151 Ill. 2d at 452-53. Accordingly, we find that the jury could have found parallel action based on the evidence that, like Unarco or Johns-Manville, defendants (1) knew that asbestos could cause disease at the time they sold asbestos-containing products; (2) sold these products without warning of these diseases; (3) failed to warn employees and consumers of these diseases; and (4) failed to adequately protect their employees from exposure to asbestos dust.
As stated previously, however, evidence of parallel conduct alone is insufficient to establish a civil conspiracy by clear and convincing evidence. Thus, we move to the second step of our review of the evidence. Under this step, we determine whether there was any evidence of agreement other than parallel conduct and whether this additional evidence, when considered along with the evidence of parallel conduct, permitted the jury to conclude that there was clear and convincing evidence of an agreement.
In addition to parallel conduct, plaintiffs rely on the following evidence as proof of an agreement between defendants and Unarco or Johns-Manville to suppress or misrepresent information concerning the health hazards of asbestos: (1) evidence of the relationship between Owens Corning and Owens-Illinois; (2) the fact that Owens Corning received information from JohnsManville about its plan to place warning labels on its products; (3) Owens Coming’s participation in the drafting of the NIMA pamphlet; (4) the indemnity clause contained in Owens Coming’s agreement to purchase the Bloomington plant from Unarco; (5) the fact that Owens Corning contacted other asbestos product manufacturers about their responses to the Califano announcement; (6) the 1979 meeting among asbestos-containing product manufacturers; (7) the 1983 meeting among asbestos-containing product manufacturers; and (8) Castleman’s opinion that defendants were involved in the alleged conspiracy with Unarco and Johns-Manville.
Even reviewing this evidence in the light most favorable to plaintiffs, we find that it does not permit a reasonable inference of the alleged agreement between defendants and Unarco or Johns-Manville. At most, these facts are as consistent with innocent as with guilty conduct. Thus, they do not support a finding by the jury that there was clear and convincing evidence of an agreement. See Tribune Co., 342 Ill. at 529; Regan, 220 Ill. App. 3d at 1091-92.
Much of plaintiffs’ additional evidence of the alleged agreement between defendants and Unarco or JohnsManville demonstrated only a sharing of information among these companies. Plaintiff showed that Owens-Illinois lent Owens Corning two published articles about the health effects of asbestos, that Owens Corning received information from Johns-Manville about its labeling decision, that Owens Corning sought information from other asbestos product manufacturers about their responses to the Califano announcement, and that asbestos product manufacturers held meetings in 1979 and 1983 to discuss litigation strategy, bankruptcy, insurance, and the impact of the Califano announcement. The mere exchange of information by manufacturers of the same or similar products is a common practice, however, and does not support an inference of an agreement. See Payton, 512 F Supp. at 1038 (membership in industry-wide trade organizations and participation in scientific conferences are common in most industries and do not support an inference of agreement); Sindell, 26 Cal. 3d at 606, 607 P.2d at 933, 163 Cal. Rptr. at 141 (it is a common practice in industry for manufacturers of the same or similar products to use the experience and methods of others).
Indeed, an inference of agreement based on these exchanges of information is undermined by the circumstances surrounding this conduct. For example, plaintiffs assert that the existence of the alleged agreement is the reason Johns-Manville would have shared information about its labeling decision with Owens Corning. The evidence that Johns-Manville was among the first to place a warning label on its product, however, suggests an innocent explanation for its communication of this decision to its competitors. If Johns-Manville were alone in placing warning labels on its products, consumers might perceive that its products were more dangerous than its competitors’ and choose to buy a competitor’s product. If it persuaded its competitors, such as Owens Corning, to also place warning labels on their products, JohnsManville could avoid this problem. Given this nonconspiratorial explanation for Johns-Manville’s communication of its labeling decision, this fact fails to support a finding of conspiracy.
In addition, evidence that asbestos product manufacturers acted differently with respect to the shared information also prohibits an inference of agreement. Despite the fact that Owens Corning learned that Johns-Manville was placing warning labels on its products in 1964, Owens Corning did not add warning labels to its products until 1966, and internal memoranda indicated that the company decided to add these labels only after considering whether to do so would be in its own best interest. Likewise, the evidence that Owens Corning contacted other asbestos product manufacturers about their responses to the Califano announcement also showed that Owens Corning acted independently. The actions it decided to take differed from those of Johns-Manville and the other companies it consulted.
The circumstances of the 1979 and 1983 meetings also prohibit a reasonable inference of the alleged agreement. As a preliminary matter, we note that Owens-Illinois did not attend the 1979 meeting. In addition, both meetings occurred after the Califano announcement, which publicized the health hazards of asbestos, and after Owens Corning itself had warned a large number of its employees and former employees about these hazards. Given that these disclosures had already occurred at the time of the meetings, it is highly unlikely that the purpose of the meetings was to suppress information about the health hazards of asbestos. Suppression of such information at that point would have been futile, as well as contrary to Owens Coming’s efforts to inform employees about the health risks of asbestos. Evidence that Owens Corning and Owens-Illinois met with other asbestos product manufacturers in 1979 and 1983 cannot, therefore, support a reasonable inference of the alleged agreement.
Evidence of Owens Coming’s participation in the drafting of the NIMA pamphlet and the indemnity clause in its agreement to purchase the Unarco plant also does not support an inference of agreement. According to plaintiffs, the fact that an Owens Corning and a JohnsManville employee were involved in drafting the NIMA pamphlet shows an agreement by these companies to conceal the health hazards of asbestos because, even though the pamphlet did state that asbestos was potentially injurious and had been associated with certain health hazards, it did not specifically identify the diseases associated with asbestos. The evidence does not support this inference. In other cases involving allegations of a civil conspiracy among manufacturers, courts have been unwilling to infer an agreement based on membership in industry trade organizations. See, e.g., In re Asbestos School Litigation, 46 F.3d at 1287-90; Payton, 512 F. Supp. at 1038. In this case, inference of an agreement is improper because, even though Owens Corning and Johns-Manville employees may have participated in drafting the pamphlet, there is no evidence indicating to what extent these companies controlled the content of the pamphlet. To conclude that the content of the pamphlet demonstrates an agreement between these companies, therefore, is unreasonable.
Likewise, no rational inference of agreement can be made based on the indemnity clause contained in Owens Coming’s agreement with Unarco to purchase the Bloomington plant. Plaintiff asserts that the fact that this indemnity clause discussed asbestos claims by employees is proof of an agreement by Owens Corning and Unarco to suppress or misrepresent information about the harmful effects of asbestos. To the contrary, the language of the clause itself demonstrates a legitimate reason for the discussion of asbestos claims: litigation concerning asbestos exposure of the Bloomington plant employees had already begun at the time of Owens Coming’s purchase of the plant. The clause does nothing more than identify each parties’ responsibilities with respect to that litigation, such as their obligations to pay judgments and to share relevant documents. There is no evidence that Owens Coming’s purchase of the Bloomington plant was anything other than an arm’s-length transaction between competitors (see Payton, 512 F. Supp. at 1038 (such transactions do not permit an inference of conspiracy)).
As proof of the alleged agreement, plaintiffs also presented evidence pertaining to the relationship between Owens Corning and Owens-Illinois. This evidence is only tangentially related to the essential question in this case, which is whether plaintiffs proved the existence of an agreement between defendants and Unarco or JohnsManville. Proof of a relationship between defendants themselves does not establish the required agreement with Unarco or Johns-Manville.
Castleman’s opinion that defendants were involved in the alleged conspiracy with Unarco and JohnsManville to suppress information regarding the health hazards of asbestos also did not permit the jury to conclude that an agreement existed. Castleman testified that he “assumed” that there was an agreement among these companies because none of them disclosed the health risks of asbestos. An expert’s opinion is only as valid as the bases and reasons for that opinion. State Bank v. City of Chicago, 287 Ill. App. 3d 904, 918 (1997). Castleman’s “assumption,” therefore, cannot be considered proof of conspiracy, especially when this assumption is based on mere parallel conduct, which we have explained is insufficient to establish a conspiracy. Even assuming the jury found Castleman credible and accepted his opinion, his testimony that defendants participated in the alleged conspiracy does not support the jury’s verdict. See, e.g., Kleiss v. Cassida, 297 Ill. App. 3d 165, 174 (1998) (affirming judgment notwithstanding the verdict after concluding that the plaintiff’s expert’s opinions were conclusory and unsupported); Aguilera v. Mount Sinai Hospital Medical Center, 293 Ill. App. 3d 967, 975-76 (1997) (same).
Although the scope of our review of jury verdicts is limited, we find that the evidence in this case so overwhelmingly favors defendants that judgment notwithstanding the verdict should have been granted. Plaintiffs’ evidence of parallel conduct is insufficient to establish the agreement required by the civil conspiracy theory. When plaintiffs’ evidence of contacts between defendants and Unarco or Johns-Manville is added to this parallel conduct, the evidence still cannot support the jury’s determination that plaintiffs proved agreement by clear and convincing evidence. The contacts between defendants, Unarco, and Johns-Manville were isolated, particularly with respect to Owens-Illinois, and an inference of agreement based on these contacts is not reasonable. Even when considered in the light most favorable to plaintiffs, evidence of these contacts was as consistent with innocence as with guilt. See Tribune Co., 342 Ill. at 529; Bosak, 192 Ill. App. 3d at 804. Plaintiffs showed separate acts by the alleged conspirators, but the evidence failed to show that these acts were connected by an agreement. See Bergeson v. Mullinix, 399 Ill. 470, 475 (1948) (finding that evidence of the separate acts of the alleged conspirators was insufficient to establish a conspiracy when the evidence showed no “connection or confederation” between them). To conclude, based on the evidence of record, that defendants engaged in a conspiracy requires speculation. Liability based on such speculation is contrary to tort principles in Illinois (see Smith, 137 Ill. 2d at 254, 259) and to the clear and convincing standard of proof applicable in civil conspiracy cases. Given the lack of evidence supporting this agreement element of plaintiffs’ conspiracy claims, the jury verdicts cannot stand, and judgment must be granted in favor of defendants.
In defense of the jury verdicts in this case, plaintiffs rely on our recent decision in Adcock v. Brakegate, Ltd., 164 Ill. 2d 54 (1994). While the facts involved in that case are similar to those before us, our holding in Adcock does not support a conclusion that the evidence of conspiracy is sufficient in this case. In Adcock, the executor of the estate of a deceased Unarco Bloomington plant employee filed suit against Owens Corning and other asbestos-containing product manufacturers. The plaintiff alleged that the decedent worked in the Bloomington plant from 1954 until the end of 1970. According to the plaintiff, the decedent was exposed to asbestos in the plant, as a result of which he developed asbestosis and mesothelioma, which resulted in his death. As in the cases before us, the Adcock plaintiff alleged that Owens Corning was liable for the decedent’s injuries based on a civil conspiracy theory. The plaintiff claimed that Owens Corning conspired with other asbestos manufacturers to suppress information about the hazards of asbestos and to falsely assert that it was safe to work with asbestos. Adcock, 164 Ill. 2d at 57.
After the circuit court denied Owens Coming’s motion to dismiss the complaint for failure to state a cause of action, Owens Corning answered the complaint. As a sanction for Owens Coming’s failure to produce certain witnesses, the circuit court entered judgment against Owens Corning as to liability. There was a trial on damages only. Adcock, 164 Ill. 2d at 58-60.
On appeal before this court, Owens Corning argued only that the circuit court erred in denying its motion to dismiss. Owens Corning contended that civil conspiracy is not actionable without underlying, intentional conduct. Adcock, 164 Ill. 2d at 60. In rejecting this argument, we explained that Owens Coming’s answer to the complaint waived any objection to the sufficiency of the allegations, provided that the complaint stated a recognized cause of action. After finding that civil conspiracy, based on either intentional or negligent tortious conduct, is a recognized cause of action, we held that Owens Corning was precluded from challenging the factual sufficiency of the complaint. We refused to express any opinion on the factual sufficiency of the complaint. Adcock, 164 Ill. 2d at 65-66.
Our holding in Adcock, therefore, does not support plaintiffs’ position that the evidence was sufficient in this case. Although the facts involved in Adcock were similar to those at issue in this case, Adcock was decided on the pleadings. There was no trial with respect to Owens Coming’s liability, and this court did not address the sufficiency of the evidence supporting the plaintiffs conspiracy allegations in that case. By contrast, the sufficiency of the evidence is the determinative issue in this case.
Given our decision in this case that judgment notwithstanding the verdict should have been granted based on the insufficiency of the evidence, we need not consider defendants’ request for a new trial based on the insufficiency of the evidence or the remaining arguments defendants raise in their briefs. In connection with their argument that the circuit court erred in excluding the testimony of Willis Hazard, Owens-Illinois filed a motion in this court to supplement the record with a transcript of that testimony. We took that motion with the case. We now deny Owens-Illinois’ motion to supplement the record because it involves an issue that is not relevant to our disposition of defendants’ appeal.
CONCLUSION
For the foregoing reasons, we hold that the judgment of the circuit court against Owens Corning and Owens-Illinois must be reversed and order that judgment be entered in defendants’ favor in each plaintiffs case.
Judgments reversed.
McClure also named the Illinois Central Railroad Company and the Metropolitan Life Insurance Company as defendants in her complaint, but these companies are not parties to this appeal.
Grefco, Inc., Great Lakes Carbon Corporation, General Refractories Company, Abex Corporation, Metropolitan Life Insurance Company, Cape Industries, PLC, and Illinois Central Railroad Company were also named as defendants in Bicknell’s complaint, but none of these companies were parties to this appeal.
Cape Industries, PLC, Abex Corporation, Metropolitan Life Insurance Company, and Illinois Central Railroad Company were also named as defendants in Thacker’s complaint, but none of these companies were parties to this appeal.
See Smith v. Eli Lilly & Co., 137 Ill. 2d 222, 235 (1990) (“Concert of action applies when a tortious act is done in concert with another or pursuant to a common design, or a party gives substantial assistance to another knowing that the other’s conduct constitutes a breach of duty”).