"You can't conduct detailed negotiationsbetween two people sending handwritten notes backand forth," the official adds.

Talks halted for at least a month.

A Dec. 6, 1998 meeting between members of theboard and the corporation proved unsatisfying.Sources say both sides were too polite, too vagueto make progress. Another evasive meeting inFebruary 1999 was even more discouraging, a sourcesays.

Part of the problem was that Rudenstine hadlong insisted that discussions remain as informalas possible, a high-level source says. He reasonedthat once Harvard publicly acknowledged that itwas in talks with Radcliffe, any breakdown wouldbe seen as Harvard bullying its weak sister onceagain. But without a larger negotiating team, thepresident was getting exhausted, squandering hisprecious time on minutiae.

In short, the source says, Rudenstine andSheerr had gone as far as they could together.They needed help.

To the Rescue

Enter Harvey Fineberg and Susan Wallach, theprovost and the trustee.

Wallach was a newcomer to the Radcliffe Boardof Trustees, having joined in June of 1997. She isa successful New York divorce attorney whospecializes in negotiating the terms ofpre-nuptial agreements--an irony not lost onobservers. Wallach's easygoing personality won herquick respect from both sides.

"She talks fast, moves fast--a dynamo kind ofperson...She thought it was going too slowly andso she inserted herself," says one Radcliffeofficial.

Negotiation lore has it that Wallach andFineberg--who barely knew each otherbeforehand--ran into one another coincidentally atthe American Museum of Natural History in NewYork. Away from the scrutiny of Cambridge, the twofound they had a powerful common interest:speeding up the discussions.

Fineberg had already been tapped to relieveRudenstine from the crush of details. More thananyone at Harvard, the practical and levelheadedprovost is intended to be the president'sright-hand man. As Rudenstine's point-person oninterfaculty collaboration, a Harvard-Radcliffemerger was an obvious problem for him to handle.

The pressure was on as Fineberg and Wallach gotto work in early 1999. A steady drip ofadministrative departures at Radcliffe hadembarrassed the administration, which claimed theturnover was just normal institutional attrition.

Then there was the Wilson factor. By alongstanding arrangement, Wilson and the board hadagreed she would depart at the end of theRadcliffe capital campaign in 2000, says a sourceclose to the Radcliffe administration.

But another source says that earlier this year,Radcliffe officials began to warn that Wilsonmight resign-deal or no deal. Radcliffe could notafford to announce its president's departure insuch an overheated atmosphere, and they toldHarvard they needed a solution soon.

"That put pressure on the whole process...butit didn't work to [Radcliffe's] advantage," theofficial says.