In a conversation about the intractable problems of the Middle East, one of David Grossman's characters in his compelling novel, "To the End of the Land," says, "Who could possibly come up with a new, decisive argument that hasn't been heard?" An equal sense of frustration must lurk in the efforts to construct a reasonable, fair way to assess the performances of public school teachers.

I have no "Eureka!" solution for unknotting the academic tangle, but I do, because of a long career as a classroom teacher in a large, diverse high school, have some insights that may help to explain why establishing a set of standards for judging teachers is so complicated.

First, and touchiest, is the matter of tenure. Of course the critics who oppose it are correct in saying the system allows tired, bored and boring deadwood to stay on while creative and idealistic young teachers who enliven and inspire their classes are the first to be let go. So, eliminating tenure seems an easy choice.

But what about the gifted, experienced teachers who, without tenure, could be let go because they irritated administrators or annoyed colleagues with complaints like their disdain for a contracted school vacation that must include the first day of hunting season?

Or the inventive instructor who chooses to enrich a Vietnam history lesson by inviting an embittered veteran to speak about the way some civilians treated him after his discharge? Not necessarily a popular patriotic response, but a valuable learning lesson. Without tenure, would he be forced out because he doesn't always adhere to conventional lesson plans or alter his out-of-class attitudes to please the majority of associates?

If we think about it, eliminating tenure is more complex than it may seem on the surface.

Another major element in evaluating teacher performance is classroom observations. In the district where I taught that meant an administrator -- usually the principal or one of the vice principals -- would visit each classroom once or twice a year and note the subject content, student-teacher involvement, and (often weighing heavily) the decorum and general atmosphere in the room.

In my own English department we sometimes wondered how our principal, a good and supportive man we all agreed, with a background in social studies and years devoted to coaching, could know much about expository writing, critical analysis or literary motifs.

One such observation stands out in my memory. My students were studying "The Great Gatsby" and, in the midst of a lively discussion of what was meant by the American dream and whether it had changed since Fitzgerald's time, the principal arrived to observe the class. He sat quietly as students often disagreed and sometimes interrupted each other with their thoughts, citing examples from the novel, until the class ended with a reminder from me of the next day's assignment.

I had no idea what my observer had gleaned from the visit, but later in the day he stopped me in the hall to ask me what happened next in the novel. At least, I thought, he apparently was caught up in the story.

But what did it demonstrate about classroom evaluations, no matter how well-intended, by observers unfamiliar with the materials studied? Would evaluations from other English teachers be more meaningful? Would the criteria applied in observing math or chemistry teachers differ from those in history or psychology classes? And if teachers were to do the assessments, should they be colleagues or visiting observers from other districts with no collegial relationships? How complicated is that!

Finally, there is the option of student involvement in teacher evaluation. All educators like to think they've established a rapport with their pupils, but that connection may be different with an A student thoroughly involved in the subject and a disgruntled C student who is certain she deserved better. The latter student has disregarded her after-school meetings with the teacher, depending on the subject, to clarify math confusion or to get individual help with writing skills or to improve her French pronunciation. All she remembers at evaluation time is her dissatisfaction with the grade "the teacher gave her," not the grade she earned.

There is also the matter of teaching styles. It's likely the easy-going instructor who gently teases his students and encourages a "buddy" atmosphere will be more favorably judged by a teenager than an equally skilled teacher who believes in a more formal environment, with a respectful distance between class and instructor.

Will these teaching styles be part of the students' evaluations? Most of us remember at least one classroom martinet who pushed us harder than we liked -- and taught us far more than we realized at the time. Quantifying all these variables in student responses has to be a daunting task.

My own observations are meant to show how overwhelming, subjective and difficult is the matter of teacher evaluations. We can only hope that, along with the best of intentions, the people who decide how to make the process viable demonstrate patience, resourcefulness and, most of all, the wisdom to make the plan succeed.