The Nuclear Regulatory Commission's own investigation of the nuclear accident at Three Mile Island was released last week, and it was a disappointment. The NRC set out to prepare the authoritative technical study of the detailed sequence of events during the accident, the licensing and operating history of the plant and its utility owner, and its own emergency planning and response. This was a reasonable decision. But the carrying out left much to be desired.

Leaning over backward to avoid any appearance of conflict of interest, the NRC turned to the Washington law firm of Rogovin, Stern and Huge to run the study, using NRC technical staff to do the actual work. The NRC thereby got the worst of both worlds: its hired investigators had little or no independent knowledge of the issues they had to investigate, or of the institutional and political history that had laid the foundation for current mistakes, while virtually all of the technical staff on whom they had to depend were employed by the very institution they were investigating. It is therefore perhaps not surprising that the report they produced is the least useful of those that have emerged in the wake of the accident.

The report itself begins with a narrative of the accident. Written in the present tense ("It is shortly before 10 a.m., and the day is only beginning"), and illustrated with cartoons, it is apparently designed to be an entertaining account for the casual reader. There is little new in it. The important unanswered questions -- for example, whether TMI's operators and owners wrongfully withheld information from the NRC during the early hours of the accident -- are left unanswered. The unusually gripping style is attributable to the hiring of a professional writer at a cost of $40,000. If that seems high, it comports with the rest of the affair. The cost of the entire project was an astounding $3 million.

The second half on the Rogovin report is a collection of conclusions and recommendations, some of which are familiar and widely supported (for example, better selection and better training of reactor operators are called for), while others are new but poorly thought out. Still others, such as the recommendation that the NRC should no longer play a role in licensing nuclear exports, are unsupported by any documentation or analysis in the body of the report.

In its authors' opinion, the "central theme" of the report is that "the principal deficiencies in commercial reactor safety today are not hardware problems, they are management problems." While no one could defend the confusion and mismanagement that have reigned at the NRC, one has only to look at the record of its predecessor, the Atomic Energy Commission -- which, as Mr. Rogovin himself asserted in presenting his report, was quite well managed -- to know that safety is not necessarily a byproduct of good management. A focus on streamlining the organization and procedures of the Nuclear Regulatory Commission therefore misses the point, and in fact could send exactly the wrong message to an already confused NRC. What's needed, as the Kemeny Commission made clear, is not a change in organizational tables or bureaucratic procedures, but a change in attitude, a recognition that though there has not yet been a devastating accident, nuclear reactors are far from safe enough.