Dear Diary: What Analyzing a Century of Russophone Journaling Reveals About the Diary as a Genre

What do a Soviet soldier scribbling on scrap paper during wartime and a 1990s diarist typing on a computer have in common? According to research from Carnegie Mellon University scholars, far more than you might think.

Using computational humanities methods and traditional close reading, the researchers  analyzed more than a thousand personal diaries written in Russian from the early 20^th century through the end of the Soviet era. They found that across vastly different historical, social and material contexts, diarists consistently returned to a familiar mix of themes, including introspection, routine cataloging and reflection on the diary as a literary object.

“I think it suggests that the diary has stronger generic constraints than we might assume,” Tatyana Gershkovich, lead author of the study. “So when you sit down to write a diary, you might think that, you know, ‘This is me. I'm pouring myself out on the page,’ that it's kind of the most unvarnished form of writing. But in fact, even in this act that seems quite intimate, there are already built-in expectations of what the text will look like.”

Gershkovich, an associate professor of Russian studies in CMU’s Department of Languages, Cultures and Applied Linguistics; co-authored the study with Madeline Kehl, a 2019 graduate of the University of Pittsburgh; and Simon DeDeo, a professor in CMU’s Department of Social and Decision Sciences. Their paper, published in The Russian Review, explored the vast Prozhito archive, a St. Petersburg-based project that has collected a mass of diaries written by individuals from all walks of life.

Studying Diaries of Others

The researchers did find two noticeable categories of diarists: some were very concerned with the literary aspects of their writing, and others were not.

“Because I'm a literature scholar, it did stand out to me how many practitioners of diaristic writing looked to other diarists,” she said. “They’d read diaries that were circulating in various publications, journals and so forth. So, the idea is that the diary is a literary kind of genre like any other, that it circulates and exists in this literary ecosystem in a full-fledged way.”

The researchers found the people who looked to those examples were more traditionally educated. They explain these writers may have conformed to norms of the genre, even unconsciously, guided by a sense of what a diary “ought” to contain.

A Computational Approach to Understanding Diary Keeping

Through their study, the scholars sought to clear up important misconceptions about diary-keeping, particularly in the Soviet era. There had been a commonly held view that diary writing had nearly disappeared in the 1930s and 1940s, often for fear of arrest. Prominent literary scholar Nikolai Bogomolov was among those who revised this view as the Prozhito archive grew. Bogomolov ultimately urged a “big data” approach to the study of diaries.

“Our study responds to Bogomolov’s exhortation by demonstrating how a bird’s-eye view of a vast archive, enabled by digital analysis, can amend misconceptions about diary-keeping and the literary forces that shape it,” the co-authors wrote.

An interdisciplinary effort, the CMU analysis used a novel computational reading technique — Semantic Collocational Clustering — to identify all mentions of the Russian words for “diary” (“dnevnik” / “zapiski”). This helped the researchers to pinpoint where each author consciously reflected on the practice of diary-keeping. 

“The digital approach can be like a metal detector on a beach,” Gershkovich said. “It identifies where to look for something that seems interesting. And then the deeper digging is pursued with more traditional methods of close reading and interpretation.” 

For DeDeo, a study like this is crucial for the psychological sciences.

“Scientists love laboratory experiments — tests we do on subjects who visit our lab for a few minutes on a weekday afternoon. This is very different: a window onto how humans live and record their lives over tumultuous decades. It was a great surprise, to me, to find this stability, which seems to be telling us something fundamental about how the mind handles memory and how writing changes and channels it. I’ll be trying to understand it for a while to come.”

“Public Patterns in Private Writing: Computational Insights into Russophone Diaries” was supported, in part, by funding from Carnegie Mellon’s Dietrich College of Humanities and Social Sciences.