Discriminatory Diplomacy

• In Event: International Collaboration Virtual Poster Session

Abstract

How do individual diplomats affect foreign policy outcomes? Existing literature has highlighted the importance of their “presence” in the host country in maintaining peace and cooperation between the sending country and the host country. However, it has focused on the sending country’s need to invest in diplomacy; it has yet to investigate the dynamic at the host country’s end. I argue that systemic discrimination against minority diplomats both by the host and sending governments hinders cooperation between the two.

I test my arguments using new datasets of over 3 million previously classified diplomatic documents spanning the period of 1861--1985 introduced by Connelly, Hicks, Jervis, Spirling, and Suong (2019) and public documents on public diplomacy as well as data on diplomatic representation, visits, statements, and appointments.

The new datasets consist of 8 corpora of government documents: six corpora of U government documents and one corpus each from the UK and Brazil. US corpora include the documents from the State Department Central Foreign Policy Files, the Foreign Relations of the United States series, the President’s Daily Briefs, the Henry Kissinger Telephone Conversation Collection, the Hillary Clinton Email Collection, and the Declassified Documents Online collection.

In particular, our Central Foreign Policy File corpus includes full text and associated metadata of the 2,081,276 cables exchanged between and among US diplomatic posts and the State Department’s headquarters in Washington, DC from 1973 to 1979. Our data also includes 209,046 documents from the Foreign Relations of the United States series and their metadata, a published collection chosen by State Department historians as the most important records from across the federal government that spans the period of 1861--1985.

This paper deals with key questions in IR about outcomes of diplomatic actions and the context in which they can be effective. This paper also utilizes new datasets of historical documents with state-of-the-art Natural Language Processing tools and text-as-data approach.

Author

• Clara H. Suong, Virginia Tech