The Pentagon study on “Don’t ask, don’t tell” was designed to study how to welcome openly gay Service members into 100% of military units. However, it elicited information that argues against such a one-size-fits-all approach. The study identifies military units in which welcoming gays would be problematic, and it sheds light on what privacy infrastructure will be best to accommodate gays more widely.
The survey, conducted by the contract research organization Westat, is the raw data. It is 324 pages long, and dwarfed by its large appendices of statistics and documents. It is accompanied by a report consisting of the Pentagon’s interpretation of the survey. But much can be learned by focusing on Question 71a in the survey:
If Don’t Ask, Don’t Tell is repealed and you are working with a Service member in your immediate unit who has said he or she is gay or lesbian, how, if at all, would it affect your immediate unit’s effectiveness at completing its mission… In a field environment or out to sea?
For the military as a whole, 44.3% responded that the effect of DADT repeal would be negative; 44.4% said neutral and 11.4% said positive (appendix E). Most striking is the difference in numbers for different units. For the Navy, negative was 35.3%, but for Army combat arms it was 57.6% (appendix J) and for Marines it was 59.4%, rising to 66.5% for combat Marines (appendix L). Furthermore, Marines were prepared to act on their concerns at twice the rate of other services (volume 1, p. 6 and table 4.24):
Among all Service members, Marine Corps members were most likely to say they will consider leaving sooner or will leave sooner than planned (38.1%) if repeal occurs.
From the Pentagon study it is clear that implementing DADT repeal overall will be far from trivial, and for Marines, implementation is likely to involve particular difficulty. Members of all service branches focused on the same underlying issue (volume 1, p. 141, volume 1, p. 135 and volume 2, p. 64):
Privacy concerns were the most frequently commented upon issue, regardless of a respondent’s Service.
… respondents appeared to be most concerned about the possibility of showering or rooming with someone who was known to the respondent to be gay or lesbian.
“If my roommate turns out to be homosexual, I feel like I am part of his target audience. It is a violation of a social norm; for example you wouldn’t have me room with a female.”
Why are the Marines so different from other service branches? Part of the difference may be that Marines are exempt from a Defense Department rule for troops to have private living quarters; Marines have two people in each room to promote a sense of unity. Marine commandant Gen. James Amos alluded to such concerns when he said of DADT repeal:
There’s risk involved … I’m trying to determine how to measure that risk … There is nothing more intimate than young men and young women — and when you talk of infantry, we’re talking our young men — laying out, sleeping alongside of one another and sharing death, fear and loss of brothers.
A second factor distinguishing Army and Marine combat units from other units may be the lack of privacy inherent in such combat situations.
Proponents of DADT repeal have not ignored such “opposition in small pockets of the force“. But they suggest that such problems will be transitory, and give two reasons why across-the-board DADT repeal will work better over time. However, both reasons seem unconvincing:
Generational change: Reform proponents suggest that “resistance within combat arms groups is likely higher because of older commanders” and therefore the situation will improve as older commanders retire. However, the age data in appendix H for Question 71a shows little effect of age.
Familiarity: One of the key implications being claimed from the study is that when straights and gays get familiar with one another, concerns about gays in the military will go away. However, information in appendix P for those who “Served With Gay or Lesbian Service Member” shows the familiarity effect to be tiny. For Question 71a, negative responses drop from 49.5% for those who haven’t served with gays to 43.0% for those with such familiarity.
The size of the familiarity effect is so small that it may be due not to familiarity at all, but instead due to ascertainment bias. As documented in appendix E Question 85, “Shared a room with a person believed to be homosexual” is 46.9% in the Navy and 26.8% for Marines. People in units in which gays are effectively “out” are more likely to be in units in which there is no problem with gays being “out”. Therefore, a sample weighted towards people in such inherently gay-friendly units would be biased towards underestimating the difficulties in other units such as combat Marines. (A simple example of such an ascertainment bias is the “full airplane” fallacy: if half of flights are 90% full and half are 10% full, the average load factor is 50%, but the average load factor experienced by fliers is (0.1 x 10%) + (0.9 x 90%) = 82%. Similarly, the posited familiarity effect may be a result of the lower percent of perceived homosexuality in the Marines.)
What does the Pentagon study tell us about how we can accommodate gays in the military?
The data suggest that a one-size-fits-all approach is not wise. We can open many units to gays quickly but we need to be careful not to require 100% of units to welcome gays quickly.
We shouldn’t be pessimistic, however, about changes in familiarity and generational attitudes. Rather than wait passively for familiarity and generational changes to influence behavior, we can improve the underlying problem with advances in privacy infrastructure and thereby lead to genuine change in familiarity and generational attitudes.
Infrastructure changes need to be done correctly, however (volume 1, p. 135 and volume 2, p. 102):
The need to address the housing and showering arrangements was also the most frequently raised implementation issue in this sample of comments.
“I recommend not trying to make segregated facilities. You’d end up with straight men’s heads, gay men’s heads, bisexual men’s heads, plus the female counterparts. My sub doesn’t have room for 6 different types of bunkrooms and heads.”
A better approach is to do what has been implemented at universities over the past few decades: giving everyone more privacy, with widespread availability of single rooms, separate shower stalls and so forth. Much infrastructure of this type has already been added in the military to accommodate women, who are now welcome in 80% of units (volume 1, p. 28). Adding such infrastructure will be practical in some situations and less practical in others, providing another argument against the one-size-fits-all approach to DADT reform.
We don’t need to choose between 0% or 100% of units being open to gays. Indeed, both extreme positions are demonstrably silly since there is no good reason to bar gay lawyers from the military and there are clear problems with opening submarines and some Marine units to gays, given current infrastructure.
Congress should give the Pentagon the authority to open up units to gays based on the Pentagon’s assessment of their infrastructure for sexual privacy. Using appropriate infrastructure changes we can continue to move forward, as we have for women in the military, to push openness closer and closer to 100%.
Jumping right to 100% of units welcoming gays is not the best strategy. It would be particularly tragic if such a push for 100% created a deadlock that left us stuck at 0%.
Posted: 1 Dec 2010
Photos: defense.gov, by U.S. Navy Petty Officer 1st Class Chad J. McNeeley