If you arrive a bit early for an appointment with Prof. Dorn McGrath, chairman of the department of urban and regional planning at George Washington University, chances are you will be seated in a little partitioned alcove where there is nothing much to look at other than a large painting of a cluster of men in 18th-century costume going rather stiffly about their business on a bright, beautiful hillside. One of the men holds a surveyor's level. Another gazes toward a river and ring of hills in the distance.

When McGrath arrives, he explains the painting. It was made as a master's thesis in mural painting by Garnet W. Jex in 1931. The subject is an imagined great moment from the past, one of those days when George Washington, Charles Pierre L'Enfant, Andrew Ellicott, Benjamin Banneker and others picked their way through the bushes along a high ridge on the Maryland side of the Potomac. In this still wild place, in their minds and on their maps, the city of Washington is beginning to take its unpredicted shape.

The irony, you feel, is intended, for what McGrath talks more and more about is the breakdown of planning in the planned federal city. His conversation, his speeches and, increasingly, the reports he and his students turn out semester by semester are studded with references to "a nightmare of a planning problem," a "frantic project-by-project" effort to "chase the rabbit of development over the next hill." It adds up, McGrath concludes ironically, "to an invitation to solve some very serious future problems."

McGrath, 51, a tall, former officer in the Naval Corps of Engineers, does not seem the type to panic. His demeanor is reassuring -- he is, after all, a rationalist, a manager who believes that urban problems, no matter how gritty or how complex, exist in order to be solved. But he turns on his message as if it were recorded on a baleful cassette, as if he were some droning Cassandra. "It's a hard subject to bring up because it seems so critical of the city," he says. "We don't want to be just the dog in the manger in this department, but the ball has been in the city's court for a long time."

In the Home Rule Act of 1973 the city government was charged with the responsibility of preparing a comprehensive plan for the "District elements" in the capital city. (Another body, the National Capital Planning Commission, retains jurisdiction over the "federal elements.") McGrath's basic charge, the one he repeats, is that the city government, both mayor and council, has dallied far too long on this issue.

There is, as yet, no comprehensive plan. The District's Comprehensive Plan Goals and Policies Act of 1978 is "so vague and broad that it pleases everyone and no one at the same time," in the judgment of a GWU report of two years ago. As a result, McGrath says, confusion abounds. Citizens, neighborhoods, preservationists, bankers, developers and businessmen do not know what to expect, and the effect is acrimonious debate. Decisions are made on a day-to-day basis; the city's understaffed Office of Planning and Development is kept busy "putting out fires." What's worse, decisions are often contradictory.

Citing "one example of many," McGrath says "the convention center is a very myopic kind of project -- everybody has high hopes for what it will generate but no one has a clear idea what activity it should generate, and where and when and how." Its effect on nearby Chinatown has been pernicious, he points out. "We've already watched Chinatown virtually disappear. A couple of street signs lettered in Chinese just won't do. The whole objective of planning is to break out of this habituated single-site mentality, to take into account the variables such as a neighborhood like Chinatown, the automobile circulation system, the needs of pedestrians and so on."

Georgetown is another of his favorite examples. "In Georgetown you have an 18th-century street pattern with late 20th-century traffic," he says. "The super-saturation of those streets with traffic would seem to be a certainty." Like everyone else, McGrath has watched the ineffective efforts of Georgetown's "affluent and articulate citizens," who "lost the kind of service that would have done away with a lot of cars" when they successfully opposed the subway. "Georgetown now reflects a hundred lessons in mindless, if occasionally elegant, intensification of land use . . . Having no plan, residents are reduced to wrangling with the developers -- and each other -- in the unfriendly forum of the zoning commission."

Planner that he is, McGrath tends to take an aerial view of the development process in the District -- that is, he sees the "desperate" need to relate developments in one part of the city to those in another place: commercial and office uses in Georgetown, for instance, to those in the West End, and these to the current office building boom in the old downtown. These developments tend to compete with one another, he says, and together they pose a threat to the city's already heavy burden of services. "There should be some knowledge of the limits to which the city can grow without requiring heroic engineering efforts to provide services," he says. "That limit may already have been reached in D.C. at the Blue Plains sewage treatment plant."

McGrath keeps a close nationwide watch on the planning process. All too often, in his view, the city comes up comparatively short. In California, he says, state law requires that "every locality have a comprehensive plan, and you get in serious trouble if your zoning committee goes riding off on its own in contradiction to the plan."

In the District, this situation is reversed. Not only is the city divided, as he says, into a number of planning "fiefdoms," but here the zoning cart is put before the planning horse: lacking a comprehensive land-use guide, the zoning is, in effect, the land-use plan. And yet, as McGrath says, "zoning doesn't address the broad qualitative questions of where, when and why. It simply says what." The "what" these days is office buildings, but McGrath warns that "if you keep settling for whatever is easiest to build, you're going to end up with big problems. You can't just leave it to the marketplace."

McGrath's professorial, book-filled office is a repository of the planner's art. As he talks, he'll select documents to support his points. "Look here," he says, stretching to retrieve what looks like a copy of Time magazine from a stack of books and manuscripts behind his desk. It turns out to be an imaginative, fact-filled promotional booklet in the form of the magazine. "This is what Springfield, Mass., is doing to attract enlightened industries and developers," McGrath explains. The comparison between the nation's capital and a city of some 170,000 in Massachusetts is eloquent, if unstated.

The professor gets wistfully explicit when he talks about Baltimore. "In Baltimore they'll latch onto any tool available," he says, "and they'll probably do pretty well with it. They're a pretty creative bunch." In Baltimore, he points out, there are some 70 full-time professionals in the city's planning department, compared with some 15 in the District's Office of Planning and Development, and in Baltimore there has been consistent leadership from the executive and legislative branches of the city government. Public money has been targeted to attract specific kinds of private development.

In the nearly 14 years since McGrath arrived at GWU to establish the planning department there, his students have turned out more than 35 detailed studies of planning problems in the city. These range from large-scale issues such as the 1980 study, "Towards a Comprehensive Plan," to neighborhood projects in Takoma Park, Dupont Circle and Congress Heights, to highly critical studies of the university's own development policies. Such studies have provided McGrath an unusual vantage point on the development of Washington. Furthermore, he points out, these studies and many others like them from universities, private consultants and government planning agencies provide much of the foundation upon which a plan for the District could be constructed.

Not unmindful of the District's tremendous set of budgetary and social problems, and not uncritical of the urban policies of the Reagan administration, McGrath still believes that the city can and should expand its planning office and "get on speed" with the business of making a comprehensive plan. "When your funds are most restricted, when your problems are the worst," he asks, "is that the time to get out of planning? Or is that the time to confront the problems, to lead the way?" The answer the city gives to this question will have more than a little to do with the kind of city it becomes.