Officially, the state of Maryland does not hold any of the 22,000 inmates in its prison system in what is called "solitary confinement," a cruel form of extreme punishment that isolates certain prisoners from any contact with other human beings, sometimes for months, years or even decades at a time. In fact, the term "solitary confinement" doesn't even appear in the state regulations governing prisoner treatment, nor is it anywhere mentioned in guidelines issued by theU.S. Department of Justicefor the federal Bureau of Prisons.

That silence is understandable. For years, civil liberties groups and inmate advocates have argued the widespread use of solitary confinement in the nation's prisons is a form of mental and physical torture that should be outlawed under the Constitution's ban against cruel and unusual punishment. Critics charge it grievously harms those subjected to it, is exorbitantly expensive despite the fact that it does nothing to increase the safety of prison staff and other inmates, and that it actually increases the likelihood inmates will re-offend when they get out.

Moreover, the United Nations Human Rights Council in Geneva has unequivocally condemned solitary confinement as a degrading, inhuman form of abuse that violates international norms. And tellingly, it recently singled out the U.S. as one of the worst offenders. With some 2.3 million people behind bars in its prisons and jails, the U.S. not only incarcerates more of its citizens than any other country, it also holds the world's largest number of people in solitary confinement.

Rather than call it solitary confinement, however, the abuses are cloaked behind such bureaucratic euphemisms as "administrative segregation," "disciplinary segregation" and "protective custody." For the inmates subjected to such abusive confinement, the terms amount to little more than torture by another name.

While exact numbers are difficult to obtain, The American Civil Liberties Union estimates at least 20,000 inmates in the U.S. are in solitary confinement at any given time, and possibly many more. The uncertainty stems from the difficulty of interpreting federal and state data compiled using different methodologies and employing different terms for categorizing inmates. This resulting lack of transparency also serves to obscure the full extent of the problem.

In Maryland, for example, figures released by the Department of Public Safety and Correctional Services suggest that about 8 percent of inmates in the state, or about 1,760 people, are in some form of administrative or disciplinary segregation. But the department doesn't keep records of how long the average inmate stays in segregation, what proportion of inmates are juveniles or adults suffering from mental illnesses, or whether segregated inmates are more likely than other prisoners to re-offend when they are released.

Although the state guidelines regarding how prisoners are treated closely mirror their federal counterparts, the lack of adequate records means in practice there's no way for the public to know whether they protect inmates from abuse or even if the rules are being followed. For example, while state regulations prohibit prison officials from confining inmates in disciplinary segregation for more than 365 days for a single infraction of the rules, in principle there's no reason an inmate couldn't be penalized for longer periods if a hearing officer found him to have committed multiple infractions stemming from the same incident.

Another example is the department's claim that most inmates confined to administrative or disciplinary segregation are not held in isolation but share a cell with another inmate on the prison's segregated wing, where they can participate in activities such as meals, exercise and bathing alongside other prisoners on that tier.

But psychologists have found the negative impact of being housed with an anti-social or violent cellmate in the segregated wing can be at least as devastating to an inmate's mental well being as isolation in solitary confinement, with a concomitant rise in the risk of inmate-on-inmate violence. In both cases conditions are such that healthy prisoners are more likely to develop a mental illness, while those already suffering mental problems are likely to get worse.

Prisoners generally are not people whose rights are foremost in the public mind. Politicians ignore them because they can't vote, and no public official running for re-election wants to be branded as soft on crime for sympathizing with their plight. But Maryland can't afford to just lock them up and throw away the key, if for no other reason that eventually most of them will be released back into the community and bring the problems they had in prison with them.

That's why the legislature needs to require that prison officials publish more data about the conditions under which inmates in administrative and disciplinary segregation are held and for how long. The current lack of transparency also makes it nearly impossible for those outside the system to judge whether all the inmates in segregation really need to be there. Other states, including Colorado, Illinois, Maine, Ohio, Washington and Mississippi, have already begun re-evaluating how many inmates really require long-term isolation, if only because of the enormous costs involved. Maryland should join them, and the first step is getting a much more complete picture of the problem.