Administration officials note that there is a tradition for this kind of biological risk assessment, one that extends at least to the Clinton administration. In the late 1990s, for example, a clandestine project run by the Defense Department re-created a genetically modified, drug-resistant strain of the anthrax bacteria believed to have been made by Soviet bioweaponeers. Such research helped the government anticipate and prepare for emerging threats, according to officials familiar with the anthrax study.
Some arms-control experts see the comparison as troubling. They argued, then and now, that the work was a possible breach of a U.S.-negotiated international law.
The Bush administration argues that its biodefense research complies with the Biological and Toxin Weapons Convention, the 1972 treaty outlawing the manufacture of biological weapons, because U.S. motives are pure.
"All the programs we do are defensive in nature," said Maureen McCarthy, Homeland Security's director of research and development, who oversees NBACC. "Our job is to ensure that the civilian population of the country is protected, and that we know what the threats are."
Current and former administration officials say that compliance with the treaty hinges on intent, and that making small amounts of biowarfare pathogens for study is permitted under a broad interpretation of the treaty. Some also argue that the need for a strong biodefense in an age of genetic engineering trumps concerns over what they see as legal hair-splitting.
"How can I go to the people of this country and say, 'I can't do this important research because some arms-control advocate told me I can't'?" asked Albright, the former Homeland Security assistant secretary.
But some experts in international law believe that certain experiments envisioned for the lab could violate the treaty's ban on developing, stockpiling, acquiring or retaining microbes "of types and in quantities that have no justification" for peaceful purposes.
"The main problem with the 'defensive intent' test is that it does not reflect what the treaty actually says," said David Fidler, an Indiana University School of Law professor and expert on the bioweapons convention. The treaty, largely a U.S. creation, does not make a distinction between defensive and offensive activities, Fidler said.
More practically, arms experts say, future U.S. governments may find it harder to object if other countries test genetically engineered pathogens and novel delivery systems, invoking the same need for biodefense.
Already, they say, there is evidence abroad of what some are calling a "global biodefense boom." In the past five years, numerous governments, including some in the developing world -- India, China and Cuba among them -- have begun building high-security labs for studying the most lethal bacteria and viruses.
"These labs have become a status symbol, a prestige item," said Alan Pearson, a biologist at the Center for Arms Control and Non-Proliferation. "A big question is: Will these labs have transparency?"
When it opens in two years, the NBACC lab will house an impressive collection of deadly germs and teams of scientists in full-body "spacesuits" to work with them. It will also have large aerosol-test chambers where animals will be exposed to deadly microbes. But the lab's most controversial feature may be its secrecy.
Homeland Security officials disclosed plans to contractors and other government agencies to classify the entire lab as a Sensitive Compartmented Information Facility, or SCIF.
In common practice, a SCIF (pronounced "skiff") is a secure room where highly sensitive information is stored and discussed. Access to SCIFs is severely limited, and all of the activity and conversation inside is presumed to be restricted from public disclosure. There are SCIFs in the U.S. Capitol, where members of Congress are briefed on military secrets. In U.S. nuclear labs, computers that store weapons data are housed inside SCIFs.
Homeland Security officials plan to operate all 160,000 square feet of NBACC as a SCIF. Because of the building's physical security features -- intended to prevent the accidental release of dangerous pathogens -- it was logical to operate it as a SCIF, McCarthy said.
"We need to protect information at a level that is appropriate," McCarthy added, saying she expects much of the lab's less-sensitive work to be made public eventually.
But some biodefense experts, including some from past administrations, viewed the decision as a mistake.