“It's a den of noise
Filled with fidgety boys
Our home beneath the firmament
And I'm glad it's a phase
Just lasting two days
Rather than something that's permanent.”
On July 31st, 1959, two young parents and their three children entered an 8-foot-by-9-foot soundproof room in the basement of a laboratory at Princeton University. They stayed inside for the next fourteen days.[Ve][MN]
This was the first shelter occupancy experiment in the United States. There had been previous stays in fallout shelters, but these had been essentially publicity stunts by shelter manufacturers. Princeton's “Project Hideaway” was the first effort to gather empirical data on how untrained people would react to being confined underground for long periods of time – but it was not the last.
The early 1960s were the high-water mark of the US public civil defense program. President Kennedy strongly supported civil defense, and two foreign crises in quick succession led Congress to give substantial funds to the Office of Civil Defense (OCD) for the first time. Kennedy's shelter push is still visible in the rusting yellow-black fallout signs on public buildings across the country, and many experts and advocates hoped that it would lead to a large-scale public shelter program.[Bl] However, if war did break out, no one knew how the average citizen would react to living in a fallout shelter.
To find out, government contractors recruited volunteers from the general population and locked them into simulated shelters for up to two weeks. The studies included volunteers as young as three months and as old as 79 years[SESP][HA2], including one study of 28 children supervised by only two adults[Ha]. Groups ranged in size from 5 to 1,046.[HA2][Ve]
Early studies were tentative and exploratory, intended just to determine whether or not people could stand being cooped up underground for so long. They tended to use relatively spacious, luxurious simulators with amenities such as beds and kitchens.[Ca] A typical example was the American Institute for Research's shelter simulator. The AIR's simulator had three-tier bunk beds, a “radio” playing CONELRAD messages, canned food, a stove, a moveable wall to adjust the space available to the occupants, and ubiquitous microphones and one-way mirrors.[ASM]
Figure 1: Interior of AIR Shelter[ASM]
Figure 2: Map of AIR Shelter[ASM]
(Used with Permission)
As research progressed, and it became clear that the average American could tolerate more austere conditions than they had been given credit for, the shelters became more spartan.[Ca] The University of Georgia built several shelters of this type, in one of the longest-running research programs the OCD sponsored. The Georgia researchers at times seemed to be trying to test how awful they could make conditions in the shelter before occupants decided they would prefer a lingering death to remaining inside. In their most extreme study, shelterees slept on bare concrete floors and lived on just 315 calories of OCD “biscuits” per day. Eight of the study's volunteers defected before it had finished – but twenty-two stayed inside.[HO]
Shelter leaders tried to follow a daily schedule, with mixed success. A typical schedule had shelterees waking at about 7:30. After group exercises and shelter clean-up, breakfast was served at 8:30.[He2]
Figure 3: Feeding an Infant[SESP]
Figure 4: Group Exercises[SESP]
From 9:00 to 11:00 the children had “school” while adults filled out their study diaries and had other quiet activities. Group games were held from 11:00 to 12:00, followed by lunch and then a rest period until 2:00. A second exercise and game session was held from 2:00 to 2:30, then a training lecture and discussion session on civil defense topics.[Ha2]
Figure 5: Lecture on CD Topics[SESP]
An afternoon snack was served at 3:30, followed by a second round of schoolwork for the children. Dinner was at 6:00, followed by group singing and whatever special events the shelterees could improvise. A second round of study diaries were filled out at 9:00, and lights out scheduled for 11:00.[Ha2]
Most studies followed a similar arc. After an initial period of disorganization, occupants settled into some sort of regular schedule. Morale and energy would initially be quite high. Over time, stress and boredom would take their toll and sprits would drop. Occupants never acted out violently, but became irritable, withdrawn, and apathetic. Then, some time after the halfway point, people would begin to perk up again, returning to normal by the time they were scheduled to leave.
Life in the shelters was difficult. Common complaints included the noise of ventilation equipment, heat, the stench from the group toilet, the inability to bathe or wash clothes, and the quality of shelter rations and water. Many shelterees complained about the bad taste of stored water and the dryness and flavorlessness of the OCD “biscuits”.
Shelterees also often had trouble sleeping. Shelter sleeping accommodations varied between hard bunk beds and bare concrete, and with thirty or more people crammed into a small room, a single noisy person could disturb everyone else. In the most extreme case, the shelter manager – an unemployed 35-year-old youth instructor, who had not received any OCD training – became convinced the other shelterees were planning to remove him from leadership. He stopped sleeping altogether, apparently to ensure no one could plan against him while he was unconscious. After three days he became convinced the scientists were irradiating the shelterees through the one-way mirrors and started carrying a screwdriver for “self-protection”. He was finally coaxed out after several mothers in the shelter slipped notes under the door demanding his removal for the safety of the children. He reportedly recovered fully.[ASM]
Despite the hardships of living in a concrete box, most shelterees reported a strong sense of espirit de corps, especially in the smaller groups. In some cases people with legitimate medical or personal reasons for leaving insisted on remaining inside so as to not let the group down. Being trapped in a small room with complete strangers for days was an intense bonding experience; many shelterees continued to socialize with their fellows for some time after the experiment finished.
Shelterees proved impressively adaptable and innovative in finding ways to stay occupied while under cover. Card games and singing were very popular, especially since they could be done as a group. Religion was a particular source of strength; most groups improvised some form of non-denominational Sunday service.[ASM]
In fact, despite the odors, the bad food, the lack of privacy, and the general discomfort, most shelterees described their experience as a positive one that they would be willing to repeat – and no shelteree reported any long-lasting negative consequences of their stay.[ASM]
Interestingly, at least four large studies at the University of Georgia were racially integrated; between 10% and 20% of the shelter populations of 150 to 504 were black. No racial conflicts were reported in the surviving documents.[HA]
OCD researchers were particularly interested in shelter leadership. In most early studies leaders were appointed and trained by the researchers, although several studies were conducted with “emergent” leadership. Pre-shelter training, even for just a few hours, dramatically improved results; shelters with trained leaders had stronger morale, fewer mid-experiment defections, and higher post-study evaluations than shelters without. However, in a war, there would probably not be enough trained personnel for more than a handful of shelters; the shelterees would have to organize themselves, with the aid of whatever instructional material was stocked in the shelter.
The University of Georgia conducted several experiments between 1964 and 1968 where leaders were determined in the shelter through a byzantine system of cards and pamphlets. The first three shelterees inside were greeted by a sign appointing them “Temporary Shelter Managers”, with leaflets to explain what that meant. They were then supposed to select a (rather sizable) temporary staff, who operated the shelter for several hours while passing out information cards for the other shelterees to fill out. The cards were then used to select a permanent staff on the basis of instructions in the temporary shelter manager's leaflet, who were then provided with still more pamphlets and manuals to explain their duties.
Despite repeated tinkering by the researchers in between studies, this system never worked very well. Staff members, particularly among the temporary staff, often didn't read or follow instructions properly, or decided that this was more than they had signed up for and just ignored their appointment. In one of the University's largest studies, the pamphlets for the permanent staff were accidentally distributed to the temporary staff, leading to much confusion. Eight members of the shelter staff defected from the experiment.[HA] Still, in every study the permanent staff was eventually formed and began to operate, and things settled into some sort of routine.
Obviously, these experiments could never truly replicate the conditions of nuclear war, and the scientists running them understood that. Their test subjects knew that, no matter how smelly and cramped the shelter was, civilization still continued outside, and that after a fixed period of time they could leave and return to their normal lives. A few studies tried to make the experiments more realistic: in one study in West Virginia, an actor with faked leg wounds was included among the shelterees as a simulated casualty. A second actor, “contaminated” by fallout and armed with a hatchet, attempted to force his way into the shelter, screaming “Let me in! Let me in! I'm dying! You god-damned bunch of Communists!”, and had to be forced back out the door. Later in the exercise the shelterees improvised a defense plan with pen knives after they were told via “radio” that a band of looters was heading for their shelter.[Fl] Even more dramatic approaches to realism were considered: the use of hypnosis to convince people that an attack was imminent or had taken place was discussed in a 1963 study. But the OCD decided that, while informative, such extreme measures could not be morally justified – not in peacetime, anyway.[Da]
Despite the limitations, research pressed on through the 1960s. Bad data was, after all, better than no data at all. By 1970, over 7,100 people had participated in 82 simulated shelter studies.[Le] However, after the early 60s, the projects gradually dried up. OCD researchers tried to find other sponsors for these studies, but were unsuccessful. Cresson Kearny at Oak Ridge National Laboratory conducted experiments using his famous “expedient shelters” as late as 1976[Ke], but his work was unique.
I have not found any documents specifically addressing why the studies ended, but it is not hard to guess. The shelter occupancy experiments followed the same arc as the civil defense program in general. The funding that flowed into the OCD dried up once the crises of the early 60s receded and Kennedy died, and it never came back.[Bl] With little money to do anything, research of this sort must have seemed like a luxury. Except for Kearny's work, I have found no records of similar studies in the United States after 1970.
[ASM]: Altman, James W., Smith, Robert W., et al. Psychological and Social Adjustment in a Simulated Shelter: A Research Report. American Institute for Research, 1961. www.dtic.mil/cgi-bin/GetTRDoc?AD=AD0270163
[Ba]: Barnes, Dick. “Nation's Fallout Shelters Provide Only Barest Needs.” Robesonian, Monday, May 5 1969. http://newspaperarchive.com/robesonian/1969-05-05/page-9
[Bl]: Blanchard, B. Wayne. “American Civil Defense, 1945 – 1984: The Evolution of Programs and Policies.” Federal Emergency Management Agency, 1985. http://www.training.fema.gov/emiweb/edu/docs/Blanchard%20-%20American%20Civil%20Defense%201945-1984.pdf
[Ca]: Carr, Fred. “State of the Art Shelter Management Research, Volume 1.” Second Edition (Revised). Defense Civil Preparedness Agency, 1976.
[COA]: Hearing on Independent Offices and Department of Housing and Urban Development Appropriations. Subcommittee on Independent Offices Appropriations, Senate Committee on Appropriations. April 21-23, 28-29, May 12-14, 19-21, 27, and June 3, 1970. US Government Printing Office, 1970.
[Da]: Davis, Tracy. Stages of Emergency: Cold War Nuclear Civil Defense. Duke University Press, 2007.
[Fl]: Fleming, Helen Parr. “The Unprepared.” Sunday Gazette-Mail, September 26, 1965. http://conelrad.blogspot.com/2011/07/survival-verite-fallout-shelter-test.html
[Ha]: Hammes, John A. Final Report: Shelter Occupancy Studies at the University of Georgia, 1965. Office of Civil Defense, 1966. http://www.dtic.mil/get-tr-doc/pdf?AD=AD0653881
[Ha2]: Hammes, John A. Final Report: Shelter Occupancy Studies at the University of Georgia, 1964 Appendix. Office of Civil Defense, 1964. http://www.dtic.mil/get-tr-doc/pdf?AD=AD0615003
[HA]: Hammes, John A., and Ahearn, Thomas R. Final Report: Shelter Occupancy Studies at the University of Georgia, 1966. Office of Civil Defense, 1966. http://www.dtic.mil/dtic/tr/fulltext/u2/653881.pdf
[HA2]: Abstract of Hammes, John A., and Ahearn, Thomas R. Final Report: Shelter Occupancy Studies at the University of Georgia, 1967. Office of Civil Defense, 1967. http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=AD0673778
[HO]: Hammes, John A., and Osborne, R. Travis. Final Report: Shelter Occupancy Studies at the University of Georgia, 1962-1963. Office of Civil Defense, 1963. http://www.dtic.mil/dtic/tr/fulltext/u2/439396.pdf
[Ke]: Kearny, Cresson. Expedient Shelter Construction and Occupancy Experiments. Oak Ridge National Laboratory, 1976. ORNL-5039. http://alwaysprepared.info/index.php?topic=1267.0
[Le]: Levit, R. A. “Behavioral Aspects of Fallout Shelter Stay.” Defense Nuclear Agency, 1979. http://www.dtic.mil/cgi-bin/GetTRDoc?Location=U2&doc=GetTRDoc.pdf&AD=ADA064144
[MN]: “Family of Five Survives 14 Days in 8x9 Shelter.” Miami News, August 16 1959.
[SESP]: Strope, W. E., Etter, H. S., Schultze, H. P., and Pond, J. J. The Family Occupancy Test, 4-6 November 1960. US Naval Radiological Defense Laboratory, 1960. www.dtic.mil/dtic/tr/fulltext/u2/288228.pdf
[Ve]: Vernon, Jack A. “Project Hideaway: A Pilot Feasibility Study of Fallout Shelters for Families.” Office of Civil Defense, 1959. http://www.dtic.mil/dtic/tr/fulltext/u2/270225.pdf