Hillary’s Homebrew Server and Lessons for Texas Open Government

Illuminating the Shadows in the Texas Public Information Act

The Texas Public Information Act (“TPIA”), subject to several enumerated exceptions, permits members of the public to request public information from governmental bodies. As the purpose of the Act reflects, it is Texas public policy that “each person is entitled, unless otherwise expressly provided by law, at all times to complete information about the affairs of government and the official acts of public officials and employees.” Tex. Gov’t Code § 552.001. Moreover, the law provides that “The people, in delegating authority, do not give their public servants the right to decide what is good for the people to know and what is not good for them to know.” Tex. Gov’t Code § 552.001. That is, government employees do not have unfettered discretion to decide the information to which the general public is entitled.

If Hillary Clinton had been the Texas Secretary of State when she used a private server to send and receive emails about government business, would the TPIA have prohibited her use of the server? Would her office have been able to force her to disclose the emails if they were requested by the public? In a thought-provoking law review article, Bill Aleshire explains that the answer to both of those questions is “no.” Bill Aleshire, “Hillary” Emails and the Unenforceable Texas Public Information Act, 17 Tex. Tech Admin. L.J. 175 (Summer 2016).

He argues that the TPIA’s failure to prohibit government officials and employees from using personal emails to conduct public business—and its omission of a means to force them to disclose those emails on request—are important problems. These gaps in the TPIA prevent the public from being able to access the information that, by law, is actually theirs.

To address these failures, Aleshire suggests three ways that the TPIA should be amended: (1) the state should prohibit government officials and employees from using private email accounts to conduct public business; (2) there should be a private cause of action to compel anyone who possesses a government record to disclose it upon request; and (3) the only people who should be allowed to destroy government records are those who are formally trained to comply with records retention requirements in the State Records Act and Local Government Records Act. Each of the proposed amendments is sensible and effective, bringing Texas open government law closer to fulfilling its goal of allowing the public to know what their government is really doing. Nevertheless, by filling gaps in existing state law, the amendments point to other, much more important gaps that are left unfilled.

First, though, consider Aleshire’s suggested improvement to the state’s open government laws. He begins by arguing that the state should prohibit government officials and employees from using private email accounts to conduct public business. The prohibition, he says, should be enforced by criminal penalties. It is difficult to find anything objectionable about that; state law provides criminal penalties for other violations of public information laws, and Aleshire’s proposal would serve the useful public purpose of preventing government officials and employees from hiding public documents on personal accounts.

Aleshire then suggests that there should be a private cause of action to compel anyone who possesses a government record to disclose it upon request. As he notes, all emails held by government officials and employees are subject to required public disclosure under Texas law, Tex. Gov’t Code § 552.002(a)(3), (b)(3), but a recent Texas appeals court opinion ruled that courts cannot force the officials and employees to disclose emails held in their personal accounts. City of El Paso v. Abbott, 444 S.W.3d 315 (Tex. App—Austin 2014, pet. denied). If Texas law were amended to allow a court to issue a writ of mandamus requiring individuals—not just governmental bodies—to disclose whatever public information they may have, that information would not be merely subject to required disclosure, it would be required to be disclosed.

Third, Aleshire advises that the only people who should be allowed to destroy government records are those who are formally trained to comply with records retention requirements in the State Records Act and Local Government Records Act. Limiting destruction of governmental records to those formally trained to comply with record retention laws would reduce the likelihood that records maintained on private or public servers would be deleted before they are scheduled, under law, for destruction.

All those amendments are sensible and practical ways to prevent the concealment and destruction of records that are supposed to be available to the public. The proposed amendments fill a gap that the Legislature presumably did not intend to leave open, and bring the state closer to its stated goal of making government officials servants, rather than masters, of the people. Tex. Gov’t Code § 552.001.

But consider a gap that the amendments do not fill—it is a gap that the Legislature seemingly intended to leave open. If a Texas Secretary of State Clinton had spoken exactly the same words she wrote in her emails, her words would not be public information subject to required public disclosure under current state law, because the law does not apply to information; it applies only to documents (or recorded conversations). Id. § 552.002. She could also witness an action related to the official business of state government, and she would not have to disclose anything about the event, if it were not written in a document (either tangible or electronic) or recorded. Id. A state official or employee can spend all day talking about public business, and not a single word is public information subject to required public disclosure. Officials could watch their colleagues do various things that implicate official business, and never have to say a word about it, even when asked to do so by the public they serve. Similarly, officials and employees are allowed to exclude members of the public from almost all meetings and conversations they have about public business, and they are not required to maintain any record of those conversations. Id. §§ 551.001(4), 551.002, 551.021, 551.022.

We tend to accept this peculiar result because the alternative is impractical: state officials and employees cannot work in offices that are truly, fully, open to the public; and no one thinks that all discussions of public business should be recorded. Nor do we think that government officials and employees should have to write a summary of everything they witnessed at the office, or during lunch out with colleagues. Our acceptance of the status quo, however, is not a defense of it. No, we shouldn’t put a microphone and a digital recorder in every office, or force state employees to keep notes when they talk shop at a happy hour, but that is only because doing so would be logistically difficult, as well as a little creepy.

But if there were a way for state open records and open meetings laws to treat the spoken word, or the witnessed act, with as much interest and seriousness as those laws treat the written word, they should do so. Not only would it be fair and intellectually consistent, it would allow the public to know what their servants are really doing. Can that be done with practical, common-sense measures that would not be overly intrusive? We think it can.

To fill the vast, inequitable, and intellectually-indefensible gap between the law’s treatment of written words on the one hand, and spoken words or witnessed conduct on the other, the Texas Government Code could be amended to allow the public to submit interrogatories to government agencies.

An agency’s public information officer would receive the set of interrogatories and, as is the case with requests for written information now, the officer could ask people in the agency whether they have heard or seen anything that would be responsive to the interrogatories. Only those agency officers and employees with responsive information would have to answer. The number of interrogatories would be limited, as would the permissible length of responses and the amount of time that agency officers and employees would be expected to spend on them. The agency could charge requestors for time spent responding to the interrogatories, and would be allowed to refuse to respond to any inquiry that is not about the transaction of official business, or that satisfies any of the current exceptions to required public disclosure.

Permitting the public to submit interrogatories would fill the gaping maw between what the state’s public information laws are supposed to do, and what they actually do. In fact, the public would truly have access to information, not merely artifacts. Aleshire’s suggestions are good and should be implemented, but there is no reason to stop there. The law should not be allowed to be obsessed with the written word, and abjectly insouciant about things that are seen and heard. Open government is an essential and beautiful thing; government employees and officials must not be allowed to prevent the public from reading their thoughts on public matters. The law’s attempt to shower government with sunshine will always be inconsistent and inadequate, but dappled light is better than shadow, and the Legislature should continue to let a little more light in.