Last month the Army surgeon general announced that 30 percent of soldiers who returned from Iraq had developed mental health problems. A few days later, the House Veterans Affairs subcommittee on health heard testimony on how Department of Veterans Affairs clinics are preparing for a wave of veterans with post-traumatic stress disorder.

How many mentally ill veterans will there be? No one knows, but the testimony of Col. Charles Hoge, chief of psychiatry at Walter Reed Army Medical Center, was refreshingly optimistic. He acknowledged that some fraction of veterans would indeed become ill but also cautioned how important it is to "convey the message . . . that many of the reactions they experience after combat are common and expected."

Drawing the line between normal reactions to intense experience on the one hand and a pathological response on the other can be difficult for psychiatrists. The conundrum was starkly illustrated last June when the long-awaited National Comorbidity Study was published in the Archives of General Psychiatry. The study, funded by the National Institutes of Health at a cost of $20 million, found that a quarter of all adults in the United States -- 26 percent -- qualified as having a mental illness within a given year. Can a rate so high be true?

A closer look at the study reveals a less startling picture. First, the survey used in the study was based on the standard psychiatric handbook -- the Diagnostic and Statistical Manual, 4th Edition (DSM IV) -- which has a low threshold for calling a collection of symptoms a "mental disorder." For example, a balky, stubborn, aggressive child might well be diagnosed as having "Oppositional Defiant Disorder" (ODD), according to the DSM, and sent to a therapist. Yet a layman might simply regard him as spoiled and in need of a strict British nanny.

Harvard public health professor Ronald Kessler, the lead author on the study, acknowledges the problem of blurry boundaries that standard psychiatric definitions create. As he said on National Public Radio, the "vast majority" of reactions to life events are "self-limiting and mild and people get over them." He used moving from one town to another as an example. For a couple of weeks, he said, people may "get depressed because they miss all their friends and maybe can't focus on their work and so forth." According to the DSM, Kessler said, "that would meet minimally the criteria for a major depressive episode."

Technically, he is correct -- according to the DSM, individuals who report a certain set of symptoms with only mild dysfunction qualify for a diagnosis. What's missing from this practice, though, is the social context in which symptoms occur. A strong but temporary emotional reaction to moving -- a "psychiatric hangnail," in Kessler's words -- is not really psychopathology by common-sense standards. By contrast, a person who is persistently sad with no obvious cause might well qualify as having the mental disorder of depression -- perhaps, or perhaps not, requiring professional help.

The study classified disorders by severity. Six percent of the population had "serious cases" (and needed professional attention), while "moderate" and "mild" cases each affected 10 percent. Mild cases often represent garden-variety anxieties and despair associated with problems in living -- or moving. So, conservatively, 6 to 16 percent of us will suffer a mental condition this year, as defined by psychiatrists.

The other key finding of the study -- that half of all people diagnosed at some point in their lives with a disorder show signs of it by age 14 -- also warrants skepticism. True, many of the conditions examined in the study -- phobias, anxiety, obsessive-compulsive disorder, impulsivity -- are manifested in late childhood or early teens. But that does not mean that every child with these syndromes will have them as an adult.

This qualification is especially important because mental health advocates, such as the National Alliance for the Mentally Ill, are now citing this new study as proof that we should treat even trivial syndromes in childhood, because they could otherwise blossom into more serious forms of illness.

But while the catch-it-early approach might be the correct one for some number of youngsters who display minor distress or behavioral problems, we don't know which ones. If we treat all youths who display low-level symptoms, many risk being given medications or other treatments they don't really need. An already overburdened child mental health care system will buckle.

Furthermore, expectations are set up all around when we send the message to kids, their parents and teachers that little Johnny may not be ill now but if we don't send him to the psychiatrist for his minor problem he is destined for mental illness. If there is one thing that Freud keenly knew it was that the power of suggestion creates self-fulfilling prophecies.

Over the past 20 years or so there has been a massive and admirable effort to raise consciousness about mental health in this country. Increased awareness and improved treatments are saving millions from suffering. But we should not define mental illness down. Most of us -- and most veterans -- will never have a serious mental illness. Will we have periods of intense sadness, frustration, anxiety and insecurity? Sure. Not because we are ill, but because we are human. And being human is not a condition in need of a cure.

Sally Satel, a psychiatrist, and Christina Sommers are resident scholars at the American Enterprise Institute and co-authors of "One Nation Under Therapy."