مقاله سال ۲۰۱۷

On his first full day in office President Obama signed a memorandum concerning “Transparency and Open Government” committing his administration to “an unprecedentedlevel of openness” (Obama, 2009). Later,the Administration launched the Open GovernmentInitiative,1 widely viewed as part of a global trend to provide more information to citizens. “[T]here are few more important struggles in the world today than the battle over who gets to know what” (Florini, 2007, p. 7). Obama’s premise, that transparency isunambiguouslypositive,is onemany scholars doubt. In fact, some argue, under certain conditions transparency may harm the deliberation so essential to democracy. In this paper we examine the effect of sunshine on deliberation. We study a natural experiment involving an abrupt and substantial increase in transparency in a powerful policy-making venue, the US Federal Reserve’s Federal Open Market Committee (FOMC). The FOMC is the key policy-making institution for US monetary policy. In October 1993, members of the FOMC learned transcripts of past meetings existed. They shortly thereafter decided to continue keeping transcripts and to publish all existingtranscripts with a five-year lag. This presents a remarkable opportunity to look at the effects of a sudden and unanticipated change in transparency. Monetary policy is of great intrinsic importance, butitis not our primary interest here. Rather, the shift at the FOMC provides a rare experiment allowing us to study the consequences of increased transparency in real policy-making.

To examine the effects of transparency on deliberation, we must first address another issue: How can deliberation be measured? While empirical work on deliberation has grown substantially (Black, Burkhalter, Gastil, & StromerGalley, 2013), little of that work has produced measures of the central feature of deliberation—reasoning occurring between persons. In the terms of Bohman and Rehg (1999), it would be fair to say that existing empirical work has focused mostly on assessing conditions that facilitate deliberation and whether the goals of deliberation have been achieved, but not on the actual deliberative process or action. There are few studies that systematically measure the things that make discussions “deliberative”: processes of reasoning, arguing, and persuading.

In light of this, we develop a measure of the use of language reflecting a necessary element of deliberation, reasoning.We assemble a ‘dictionary’ of words and phrases reflecting reasoned exchanges among participants in a dialog. We assessed the validity of this measure by applying our method to dozens of transcripts of dialogic exchanges with varying levels of deliberation, ranging from conversations used for English instruction to public policy debates andmeetings of a federalfinancial regulatory agency. These tests indicate that our method succeeds in distinguishing deliberative exchanges from non-deliberative ones and show that our approach could be used to study deliberation in policy areas beyond monetary policy. We use this measure to assess the effects of transparency on FOMC deliberation. In this case, increased transparency did not harm deliberation, although it clearly did affect behavior in several ways. This finding speaks to deliberative scholars concerned with the relative values of public versus secret deliberation. Also of interest for deliberative theorists, we found that leaders powerfully shape deliberation. The role of a leader or facilitator is under-theorized and this suggests an important avenue for future work. Consistent with the views of theorists, our estimates show that both increasing equality and disagreement are drivers of reasoned deliberation.