Jeff Chiu/AP Images

Medical marijuana patient Kevin Brown at the Apothecarium, a medical cannabis dispensary in San Francisco, December 2011

In the summer of 2006, a young scientist from Israel joined my laboratory. He came to learn how viruses attack cells, a major focus of my research program. And I looked forward to drawing on his expertise in an emerging area of science that intrigued me: the biological effects of cannabinoids, the active chemical compounds in the marijuana plant. The Israeli researcher had trained at Jerusalem’s Hebrew University with Professor Raphael Mechoulam, a chemist credited with the discovery in 1964 of delta-9-tetrahydrocannabinol (THC), the primary psychoactive compound in marijuana. Mechoulam later characterized cannabidiol (CBD), a related substance plentiful in the plant, as distinct from THC in that it had no discernible effects on mood, perception, wakefulness, or appetite.1

The work of the young scientist proved productive. In short order, he tested the effects of several cannabinoids on a herpes virus that promotes the development of Kaposi’s sarcoma, a disfiguring and sometimes fatal tumor among people with impaired immunity, like those with AIDS. It turned out that CBD, the plentiful, nonpsychoactive compound, could switch off the malignant effects of the virus.2 Scientists in my department also found that cannabinoids could alter how white blood cells migrated in response to physiological stimuli, a key aspect of immune defense; other research teams found that THC inhibited the growth and spread of lung cancer and CBD of breast cancer in laboratory models.3 Clearly, chemicals in the plant could have diverse and potent effects on normal and malignant cells.

But what I found most fascinating was that we have a natural or “endogenous” cannabinoid system. In 1988, researchers identified a specific docking site, or receptor, on the surface of cells in the brain that bound THC. This first receptor was termed cannabinoid receptor 1, or CB1.4 Five years later, a second receptor for cannabinoids, CB2, was found.5 This latter docking protein was less plentiful in the central nervous system but richly present on white blood cells. Again, it was Raphael Mechoulam who discovered the first endogenous cannabinoid, a fatty acid in the brain, which he termed “anandamide.” (The name is derived from the Sanskrit word ananda, which means “bliss.”) When anandamide attached to CB1 it triggered a cascade of biochemical changes within our neurons.6

Other endogenous cannabinoids were later identified. This makes evolutionary sense, since the CB1 and CB2 receptors would not be present on our cells if we did not normally make molecules to dock on them. The physiological ramifications of endogenous cannabinoids appeared quite broad; their most impressive effects were related to perception of and response to pain.

Cannabis is one of the oldest psychotropic drugs in continuous use. Archaeologists have discovered it in digs in Asia that date to the Neolithic period, around 4000 BCE. The most common species of the plant is Cannabis sativa, found in both tropical and temperate climates. Marijuana is a Mexican term that first referred to cheap tobacco and now denotes the dried leaves and flowers of the hemp plant. Hashish is Arabic for Indian hemp and refers to its viscous resin. An emperor of China, Shen Nung, also the discoverer of tea and ephedrine, is held to be among the first to report on therapeutic uses of cannabis in a medicinal compendium that dates to 2737 BCE. In 1839, William O’Shaughnessy, a British doctor working in India, published a paper on cannabis as an analgesic and appetite stimulant that also tempered nausea, relaxed muscles, and might ameliorate epileptic seizures. His observations led to widespread medical use of cannabis in the United Kingdom; it was prescribed to Queen Victoria for relief of menstrual discomfort.7

The cannabis plant contains some 460 compounds, including more than 60 cannabinoids. THC, the key psychoactive substance in marijuana, has increased from about 1–5 percent to as much as 10–15 percent in cultivated plants since the 1960s. When herbal cannabis is smoked, some 20 to 50 percent of the THC is absorbed via the lungs. When herbal cannabis is eaten, less THC reaches the brain because it is metabolized as it passes from the gut through the liver. THC accumulates in fatty tissues, from which it is slowly released, and acts primarily on CB1 receptors in the brain’s mesolimbic dopamine system, which is believed to contribute to the positive reinforcing and rewarding effects of the drug.8

While smoking or eating cannabis typically results in the user’s feeling “high,” with a relaxed, euphoric sense as anxiety and alertness decrease, some first-time users, as well as individuals who have psychological problems, can experience dysphoria, fear, and panic. Typically, when high on marijuana, there is an increased sense of sociability, although among those who have a dysphoric reaction, there can be sharp social withdrawal. Perception of time is altered, generally with perceived time faster than clock time; spatial perception also may change, and colors may seem brighter and music more resonant. High doses of cannabis can result in hallucinations, which may account for its religious use in some cultures. Yet unlike opioids, there are no reported cases of death due to a THC overdose, probably because cannabinoids do not inhibit our respiratory drive, which would result in asphyxiation. Among regular users, abstinence from marijuana can cause an uncomfortable or distressing withdrawal syndrome.


In 2008 the World Health Organization published a Mental Health Survey of 54,068 persons age sixteen and older in seventeen nations. On the basis of this survey, cannabis was found to have been used at least once by some 160 million people between the ages of fifteen and sixty-five; reported use was lowest in the People’s Republic of China, 0.3 percent, and highest in the United States, 42.4 percent, with New Zealand close behind.9

Despite such widespread use, cannabis is illegal in most countries. Harry J. Anslinger, a prominent prohibitionist, successfully lobbied Congress to pass the Marihuana Tax Act in 1937, making access to the plant costly. Anslinger was the head of the Federal Bureau of Narcotics and presented cannabis use to the public as an unalloyed danger, resulting in “reefer madness.” The American Medical Association opposed the Marihuana Tax Act, fearing that it would limit medicinal study and potential prescription of the plant. Long a part of the United States Pharmacopeia, a compendium that set standards for medicines and foods, cannabis was removed in 1942.

In 1970, Congress enacted the Controlled Substances Act, classifying marijuana along with heroin as a Schedule I drug. Drugs in this category have a proven potential for abuse and no medical value. (Opium, the source of morphine, and amphetamines are Schedule II drugs, classified as less dangerous despite their potent addictive properties.) Soon thereafter, President Nixon launched the “war on drugs,” and in 1986, President Reagan signed the Anti-Drug Abuse Act, which mandated prison sentences without parole for offenders convicted of possession and sale of all illegal drugs, including marijuana.

The study of cannabinoids, both those derived from plant sources as well as the endocannabinoids that exist naturally within our body, is now an extensive enterprise that spans the globe and links numerous scientists in both academic centers and pharmaceutical companies.

Mitch Earleywine, a prominent researcher on drugs and addiction at SUNY Albany, observed how results from current studies on marijuana are akin to Rorschach blots. “People purportedly see these ambiguous pictures in a way that reveals more about them than the ink.” Many who make public policy or are associated with interest groups, he contends, may respond to marijuana research according to the views of these groups: their interpretations say more about their own biases than about the actual data. For example, prohibitionists contend that THC often appears in the blood of people involved in auto accidents; yet they omit the fact that most of these people also had been drinking alcohol. Antiprohibitionists cite research that showed no sign of memory problems in chronic marijuana smokers; but they do not mention that the cognitive tests were so easy that even an impaired person could perform them.

Two recent reviews avoid such biases and critically examine data from more than a hundred randomized placebo-controlled clinical trials involving some 6,100 patients with a variety of medical conditions.10 Marijuana appears useful in treating anorexia, nausea and vomiting, glaucoma, irritable bowel disease, muscle spasticity, multiple sclerosis, symptoms of amyotropic lateral sclerosis (Lou Gehrig’s disease), epilepsy, and Tourette’s syndrome. (Recent clinical trials confirm many of the claims of Emperor Shen Nung and Dr. O’Shaughnessy.) Despite findings from experiments in my laboratory and others, its anticancer effects in patients are more uncertain and neither THC nor CBD is a proven antineoplastic agent, i.e., effective in treating abnormal growth of tissue.

Judy Foreman, an accomplished medical journalist, devotes a chapter to marijuana in her recent book A Nation in Pain: Healing Our Biggest Health Problem.11 She judiciously reviews the data on the risks and benefits of marijuana as a therapy for medical conditions marked by pain, highlighting where it appears ameliorative, where it falls short, and where there is lack of clarity about its value. Foreman writes:

To put it bluntly, marijuana works. Not dazzlingly, but about as well as opioids. That is, it can reduce chronic pain by more than 30 percent. And with fewer serious side effects. To be sure, some researchers think it’s too soon to declare marijuana and synthetic cannabinoids a first-line treatment for pain, arguing that other drugs should be tried first. But that may be too cautious a view.

Ultimately, marijuana may be used in conjunction with opioids like morphine to allow for lower doses and fewer of the side effects of the opioid family of analgesics. While chronic pain seems amenable to amelioration by marijuana, its impact on reducing acute pain, such as after surgery, is minimal.


How do cannabinoids reduce pain? Some of the benefit appears to result from cognitive dissociation: you realize that pain is present, but don’t respond to it emotionally. If you are able to detach yourself from pain in that way, there is less suffering.

Every therapy, whether a drug or a procedure, involves a tradeoff of benefits versus risks. Perhaps the most controversial and important concern around cannabinoids is whether they increase the risk of psychoses like schizophrenia. This question is most germane for adolescents and young adults. A number of studies reviewed the health records of young people in Sweden, New Zealand, and Holland who reported cannabis use, as compared to the records of those who did not. A combined or metaanalysis of results from nearly three dozen such studies linked cannabis use to later development of schizophrenia and other psychosis.12

The limitation of such observational studies is that they may suggest an association but in no way prove a causal link. Indeed, the medical literature is littered with observational studies that were taken as meaningful but later overturned when randomized placebo-controlled trials were conducted. Here the Women’s Health Initiative comes to mind. This was a randomized study, using placebos as controls, that reversed some four decades of thinking about the alleged benefits of hormonal replacement therapy among postmenopausal women in preventing dementia and heart disease. No one is likely to conduct a randomized controlled trial of thousands of teenagers, assigning one group to smoke or ingest cannabis and the other group to receive placebos. The issue of marijuana as a cofactor in the development of schizophrenia and other psychosis will therefore remain unresolved.

What is clear is that cannabis impairs cognition and psychomotor responses. Numerous studies show that it lengthens a person’s reaction time and impairs his or her attention, concentration, short-term memory, and assessment of risks. These changes in psychomotor performance can last longer than the feeling of being high. Trials with licensed pilots found that marijuana impaired performance on a flight simulator for up to twenty-four hours.13 Further, most of the pilots were unaware that their performance was still impaired a day later. Several studies demonstrate associations between cannabis and collisions: drivers who use it are estimated to be some two to seven times more likely to be responsible for accidents compared to drivers not using drugs or alcohol.14

The American Psychiatric Association, in the new DSM-5, has defined a diagnosis of “cannabis use disorder.” These people had a repeated pattern of use with harmful consequences, such as inability to fulfill major responsibilities at work and persistent social problems at home. Both the DSM-5 and the World Health Organization’s International Classification of Diseases 10th edition (ICD-10) also include a list of possible symptoms of withdrawal from using cannabis: significant fatigue, sleepiness, psychomotor retardation, anxiety, and depression.15 Yet there is fierce argument about whether marijuana is addictive. Proponents of cannabis doubt that it can cause true addiction, a physiological condition with compulsive craving and use despite harm; they argue that any dependence is less significant than that seen with alcohol. Opponents of cannabis use, particularly those from the National Institutes of Health, affirm both dependence and addiction as real risks, although at a much lower percentage than that seen with cocaine or heroin.16

A New Leaf is a detailed account of the history of the regulation of cannabis, presenting in a blow-by-blow manner the legal and political battles around its prohibition. It opens on a celebratory note, with the legalization of marijuana for recreational use in two states:

Another prohibition is ending. On November 6, 2012, voters in Colorado and Washington were the first in the world to successfully challenge nearly a century of bad policy and misconceptions about cannabis.

In downtown Seattle, the Hotel Ändra was dressed white and blue, the team colors of Washington State’s…campaign….

Around 7 p.m., the owner of one of the largest and most successful medical cannabis dispensaries in the country arrived. Steve DeAngelo was unmistakable even in a crowd, with his signature long, tight pigtail braids and dark fedora…. Earlier that year, he was the star of his own Discovery Channel show, Weed Wars. His two Harborside Health Centers are in the Bay Area, but he had a soft spot for Seattle. Just a few months before, he had spoken at Seattle’s well-known Hempfest, attended by tens of thousands each year. “I’ve been working on this issue for my entire life…. And I know tonight…that there’s going to be a whole lot of angels dancing in heaven,” DeAngelo said, his eyes flooding.

The authors describe a similar scene in Denver:

Brian Vincente, a lawyer who advocated for medical cannabis in Colorado for nearly a decade,… took the stage. “Tonight we made history. This is something you’re going to tell your kids about,” Vincente said. “Marijuana prohibition started in 1937. The first person arrested was in Colorado.” The crowd booed. “Colorado fucking turned this thing around tonight.” And with the f-word came gaiety.

These successes resulted from a unique effort joining groups from the ends of the political spectrum:

The support of conservative Republicans and Libertarians was as important to the Colorado…campaign as that of Democrats and liberals…. The swing state of Colorado, birthplace of the Libertarian party, is decidedly purple. The Libertarian Party of Colorado emphatically endorsed Amendment 64 in May, for example, while the Colorado Democratic Party offered support but stopped short of an endorsement. The Republican Liberty Caucus of Colorado also endorsed the amendment because prohibition is “inconsistent with Republican values,” which call for more “personal responsibility” and less “federal overreach.”

Recent articles in The New Yorker17 and The Nation18 describe in a succinct and focused way the political terrain around cannabis legalization for medicinal or recreational use in the United States. The New Yorker article features Professor Mark Kleiman, a drug policy expert at the University of California, Los Angeles, who sees legalization through the perspective of a scientist, who regards it as a kind of ongoing experiment. Legalization will test a group of hypotheses about public policy, and he suspends conclusions until more data are available.

As with every social initiative, there could be negative effects and Kleiman advocates close monitoring of excessive use among adolescents and of driving under the influence when cannabis is legal for recreational use. He “appears,” according to the New Yorker article, “to derive grim pleasure from informing politicians that they have underestimated the complexity of a problem.” One major concern is that when legal marijuana goes on sale in Washington State this spring, the current black market will not disappear; rather, legal over-the-counter marijuana will be competing with illicit sources. Kleiman argues that to support the legal market, there should be even greater law enforcement pressure on those who do not respect the rules. In Washington, few in government wanted to hear such a proposal.

Similarly, Kleiman is not confident that alcohol will become less appealing as marijuana is made available. While he acknowledges that alcohol is the greater danger of the two, he raises the possibility that cannabis will be used to complement drinking. Finally, he says that in the “Manichaean world of politics,” the pendulum may swing from marijuana as illegal—with sale or use of it causing imprisonment—to “going all the way to ‘We should sell it like cornflakes.’”

Unlike the cautious New Yorker piece, the articles in The Nation offer a robust endorsement of legalization. The cover of the magazine displays a photograph of a young Barack Obama flashing the V for victory sign with friends in high school clustered around the logo of the “Choom Gang.” An accompanying editorial by Katrina vanden Heuvel notes that recent presidents, including Bill Clinton, George W. Bush, and Barack Obama, all “have more or less owned up to breaking America’s drug laws” through possession or use of cannabis; if they had been observed by the police, they might well have been incarcerated, with no hope of a career leading to the White House. A New Leaf emphasizes the risks of arrest for possession. Racial discrimination, with disproportionate numbers of African-Americans arrested, is one ugly reality of prohibition:

While cannabis users who are arrested are not often sent to prison, there are still more than twenty thousand people incarcerated for mere possession. According to a comprehensive 2013 report released by the ACLU, between 2001 and 2010 more than 8 million cannabis arrests were made in the United States (88 percent for possession), and the possession enforcement alone cost more than $3.6 million in 2010.

Across the country, blacks are nearly four times more likely than whites to be arrested for cannabis possession, despite comparable rates of use; in some counties that number increases from four to thirty. Finally, 62 percent of those arrested are twenty-four or younger, which means their arrest records will follow them throughout adulthood.

All of these wasted hours, dollars, and arrests are a distraction from hard drug use and trafficking:

Again, when cannabis—which accounts for 80 percent of all illegal substance use in the United States—is removed from the drug war picture, the country can more effectively discuss and implement a new and more fitting public health approach for the remaining hard drugs.

Several years ago, I consulted on the case of a young woman with anemia. Her internist had made an exhaustive evaluation of her condition but had found no cause for it. The patient had been under a great deal of stress at work, and when I asked how she dealt with this, she said she had been smoking marijuana every night. A bone marrow examination showed reduced numbers of cells, not severe enough to be classified as aplastic anemia, but certainly abnormal in a woman in her twenties. The numerous components of cannabis are not known to be toxic to blood cells; marijuana smoking has not been reported as a cause of anemia. But I recalled that some of the illicit crops had been sprayed with toxins that might have deleterious effects on blood cell development.

So together we decided that she would suspend smoking, and over a period of months her anemia was resolved. A subsequent bone marrow examination showed full restoration of normal blood cell numbers. This was not definitive proof, but it certainly suggested that something in the grass she got from a dealer was the potential culprit. If there is not adequate oversight of the marijuana on sale, those seeking street cannabis could be exposed to dangerous contaminants.

In a forthcoming book, Weed Land, Peter Hecht, a journalist at The Sacramento Bee, charts the evolution of California’s medical marijuana law, the first in the nation.19 Much of the momentum behind its passage came from a joining of forces between AIDS activists and academic physicians like Donald Abrams at San Francisco General Hospital, who demonstrated the clinical benefits of augmented appetite and relief of pain in patients with cachexia from HIV. Medical marijuana, now legal in twenty states and the District of Columbia, is regulated like a supplement rather than a drug. There is no standardization of optimal amounts of psychoactive THC and nonpsychoactive CBD, although they must be free of toxins. (A British company, GW Pharmaceuticals, makes Sativex, an oral spray containing extracts of two standardized cannabis strains that are mixed to give exact doses of THC and CBD. Sativex was approved in several countries, but not in the United States.)

For a physician like myself prescribing a therapy, this is an uncomfortable situation, because a prescription should be exact in specifying how much drug is delivered. Further, side effects may occur in patients taking multiple other medications, due to so-called “drug-drug interactions.” Such interactions have not been well studied with THC and CBD, in part because of the restriction of access to the plant for the clinical research community. Scientists in my laboratory studied pure chemicals, THC and CBD, under strict federal oversight; we purchased the cannabinoids from chemical companies that used quality control. As Martin and Rashidian note, clinical study of the plant itself, with its scores of active chemicals, is another matter:

The federal government has imposed additional and unique restrictions on cannabis research, with little rationale—beyond politics. The federal government has enabled only one institution, the University of Mississippi, to legally grow cannabis for research on its behalf, although it is free to award additional and alternative contracts. And cannabis is the only research substance for which the government is the sole supplier. For a scientist to receive cannabis from the federal farm at the University of Mississippi, a trifecta of approvals…must be obtained from the FDA, DEA, and a Public Health Service panel.

Perhaps as states legalize marijuana, this barrier to research will be lowered, as it was for stem cell research, once restricted by federal law. And as more studies are conducted on marijuana for medical or recreational uses, opponents and enthusiasts may both discover that they were neither entirely right nor entirely wrong.