Use, Abuse, Dependence - Who Decides?

I often hear people refer to the distinction between drug "use" and drug "abuse" as if it were an immutable, medical fact. A review of recent history suggests otherwise. Just as the American Psychiatric Association once viewed homosexuality as a mental disorder, the meaning of "substance use disorder" has evolved over time. The definitions we use today are partly a social construction, subject to changing mores, values, and even our political culture.
As readers of this blog certainly know, social service professionals use a variety of screening and assessment instruments to detect drug problems. Many tools still rely on the underlying logic of the American Psychiatric Association's Diagnostic and Statistical Manual, or DSM.
The DSM is widely endorsed, but few policymakers and other nonclinicians appreciate how much it has evolved. Consider this passage from the DSM 1980 edition:

"In our society, use of certain substances to modify mood or behavior under certain circumstances is generally regarded as normal and appropriate. Such use includes recreational drinking of alcohol, in which a majority of adult Americans participate, and the use of caffeine as a stimulant in the form of coffee. On the other hand, there are wide subcultural variations. In some groups even the recreational use of alcohol is frowned upon, while in other groups the use of various illegal substances for recreational purposes is widely accepted. In addition, certain substances are used medically for the alleviation of pain, relief of tension, or to suppress appetite.
This diagnostic class deals with behavioral changes associated with more or less regular use of substances that affect the central nervous system. These behavioral changes in almost all subcultures would be viewed as extremely undesirable. Examples of such behavioral changes include impairment in social or occupational functioning as a consequence of substance use, inability to control use of or to stop taking the substance, and the development of serious withdrawal symptoms after cessation of or reduction in substance use. These conditions are here conceptualized as mental disorders and are therefore to be distinguished from nonpathological substance use for recreational or medical purposes." 
- American Psychiatric Association 1980, p. 163 
This language was removed from more recent editions of the DSM. It is difficult today even to imagine an official body that would endorse a concept like “nonpathological substance use" for "recreational" purposes (even though this is essentially why state and federal policies continue to permit the socially sanctioned, regulated use of tobacco and alcohol). 
The DSM committee members of 1980 understood the cultural foundations of drug use and the social dynamics underpinning drug laws. Their anthropological approach, however, did not last.
By 1994, drug-related behaviors didn't have to be pathological to be considered "drug abuse." They merely had to be illegal. The DSM of 1994 allowed legal problems alone to qualify an individual for an abuse disorder, even in the absence of tolerance and withdrawal, and without any indications of school problems, occupational troubles, or family strife. Clinicians needed only to define a person’s legal problems as "recurrent" to diagnose drug behavior as abuse rather than use. 
In effect, the DSM of the 1990s created a new policy environment in which a person’s chances of being defined as a drug abuser instead of a drug user were partly determined by the person's behavior and partly by the odds that their behavior would be detected by police and pursued by prosecutors. The social consequences of such a change are not hard to imagine.
No matter how much they may improve, our procedures for identifying use, abuse, and dependence are not based entirely on medicine and science. They are derived from an evolving practitioner consensus about drug use and its harms. As such, diagnostic procedures reflect social attitudes toward psychoactive substances and toward the people who use them.