As we pointed out in our article on drug policy [1], the important question involves not so much the details of a particular policy but whether the policy would result in more or less drug use. We do not support any drug policy that would result in an increase in drug abuse. Legal drugs cost society far more than illegal drugs in terms of medical care, lost productivity, death, crime, and the acquired immunodeficiency syndrome (AIDS) ($66.9 billion compared with $170 billion). We have never seen anyone addicted to legal or illegal drugs who was "happier, more free to achieve their dreams," because of his or her drug use, as Dr. Wilbur suggests. As we also noted, some of the advocates of legalization support such policy to reduce the pressure on their own admitted drug use, and many of the supporters of legalization have no experience or understanding of addiction.

The negative responses to our article advance the notion that legalization or decriminalization of drugs would drive out the criminal element and thus would result in lower overall harm to society. Any action that increases availability of drugs risks increases in drug-associated crime. Fifty-six percent of violent crimes are committed under the influence of drugs [2]. Crime has drastically increased in the Netherlands since the decriminalization of marijuana. In 1988, only 3 organized crime groups were listed in Holland; this number had increased to 93 in 1993 [3]. Because of associated crime and drug dealing, actions have been taken in the Netherlands to tighten controls on marijuana coffee shops. It is naive to suggest that criminals would throw down their weapons and take up clean living simply because drugs were made available to addicts.

What substances would the government distribute if currently illegal drugs were legalized? All drugs, including tobacco and alcohol? Only certain strengths? Attempts to standardize or control the types or strengths of drugs would result in illegal drug markets.

Although the comments of Des Jarlais and colleagues are sophisticated, they miss our point. We do not simply support restrictive drug policy—we want to discourage drug use. We support greater treatment availability, but not among programs that cannot show clear reductions in drug use. Needle exchanges address only a tiny fragment of the overall drug-using community, have a questionable effect on the reduction of the incidence of AIDS, and do nothing to reduce the underlying problem of illicit intravenous drug use. Also, no evidence of crime reduction has been associated with needle exchanges. The Montreal experience (not yet published) may even show an increased number of cases of AIDS in association with needle exchanges.

Finally, no one has yet shown any system that reduces the legal use of alcohol or tobacco or their associated harms while simultaneously reducing constraints or controls on those substances. Because these agents cause the most harm to society, this would be a good place to begin true harm reduction.