When a dispute does arise, so does the matter of deciding which technique will best solve the problem. There are no simple answers, but anyone faced with this decision should ask certain questions:

* Is this a dispute about facts or about law? If facts predominate, as in a patent case involving questions of product design, it is less likely to require the courts to resolve it. Indeed, a minitrial may be the most appropriate approach. On the other hand, if it is mostly a question of law, such as a novel antitrust question, there may be no choice but to go to court.

* Is secrecy important? If a serious dispute with a high-level executive involves compensation or another potentially embarrassing matter, it may be preferable to keep the dispute, and all its details, out of court. But occasionally there are advantages in providing a more public message. For example, if an employee steals trade secrets, perhaps the case ought to be publicized to discourage anyone else from making the same attempt.

* How unusual is the problem, and will it recur? If it is a factual dispute about a new kind of technology, it may be more appropriate to use a neutral technical expert, such as a mediator, than to go to court. But if the problem is a common one, as with a special clause in a new standard-form contract that would otherwise be used in the future, consider the advantages of a court precedent.

* What will emerge from a court battle? Even if the ruling is favorable, weigh it against the costs of attorneys and the management time diverted from more productive endeavors.

* What is the stage of the dispute? If it has just begun, it is too early for trial by management. Instead try negotiations and a settlement discussion.

* How flexible is the other side and how much hostility is there? If the two of you are not even on speaking terms, a mediator may be the only way.