The project
This project is part of a larger research conflomerate called the NCCR - on the move, financed by the Swiss National Science Foundation, comprised of 11 projects, all studying topics related to crises and/or migration. Unlike most of the other projects, we are primarily interesting in how people experiences crises, and how this experience develops over time. You can see our project page here.
We are a team of five sociocultural psychologists (Maeva Perrin, Oliver Clifford Pedersen, Alex Gillespie, Nathalie Mirza Muller, and Tania Zittoun), which means that we believe that human development and all psychological processes happens in specific social, cultural, political, and natural environmental. People are both shaped by these contexts and are responsible for maintaining and generating them. So, people living in comparable circumstances do not necessarily experience the same events as crises, and it is this difference in how people experience and navigate crisis we explore. Simultaneously, we also try to break with conventional understandings of what constitute a crisis - namely that is a temporary state of exception. Instead, we also try to account for the crises that are slow-burners (e.g. ecological crisis) and occurs out-of-sight (e.g. pollution). To explore how people experience and imagine in and through various crises, we turn to the diaries.
What is special about diaries?
Research have long used diaries as windows into people’s everyday lives or as time-capsule into specific sociopolitical periods - e.g. pandemic or wars. For example, two members of our team have analysed diaries written during the Second World Ward, shared at the Mass Observatory, detailing how the life of a young woman, June, changed over the course of the war. To extend this line of research, we searched for more contemporary diary-writing and stumbled upon Prosebox, among others.
However, this kind of analysis is not too common. A more conventional approach would be to interview a person about the past, and while these can be very illuminating, they are always weaved into a coherent narrative with the benefit of hindsight. The person recollecting past events often know where the story “ends” and tend to leave out mundane details not deemed important for the narration. Yet, reality is often messier than we remember. And this is where diaries come into the picture because make it possible for us to follow people’s experiences in close to real time and reveals some of the more “mundane” and overlooked twists and turns that are otherwise left out, but which can be developmentally significant. Unlike backward-looking accounts, diaries are written towards the future and are therefore tainted by uncertainty, revealing how people make sense of their life and the world around them ongoingly. Importantly, the diaries are also written without us prompting the content, showing what, at any given point in time, is significant to the person writing.
Combined, these qualities of diary writing makes them ideal for exploring questions such as when (and how) crises manifest and for whom, how they “develop” over time, and how they might impact people later in life. So, when we want to study how people experiences and lives through diverse crises, diaries allow us to look at where they may manifest in people’s everyday life, what events are more salient than others, and how these crises unfold differently depending on the person.
How we study the diaries?
So far, 41 diarists - across different platforms - have agreed to participate in this study. All have been writing for at least one decade and have combined written more than 80.000 entries and 35 million words.
One of the issues we faced was how to process so many words since the research project is limited to four years. To put things into perspective, 35 million words is roughly equivalent to 350 standard length books, and if we were to spend half of our working time reading (approximately 4 hours a day), then it would take roughly one year and 7 months just to read the diaries. And this does not account for the time-consuming process of analysing diaries and finding themes or patterns, often requiring close and repeated reading.
Considering the number of words and time-restrictions, we turned to natural language processing technique to assist the analytical procedure and to generate an interactive visualisation of the diaries (as shown below). Essentially, this is a script that allow us to detect patterns in the diaries, for example, by searching through the text for a group of words that are semantically linked - such as “climate change”, “global warming”, “Greta Thunberg”, and “Paris Agreement”. Words that all fall under the ecological crisis umbrella. We use this process across a number of pre-established and more personalised word groupings, developed computationally and through reading.
This visualisation displays a map out a single diary, spanning more than two decades. Each of the dots represent a journal entry and the size of the dot reflects the length of the entry. The yellow line cutting across is a rolling sentiment line, indicating whether the writing is positive or negative in terms of words used and the intensifiers (e.g. “very”). This algorithm use words like “amazing” or “fantastic” to score positively and words like “awful” or “terrible” to score negatively. To minimise large and sudden fluctuations, the sentiment line is constructed by taking the average of 120-days, so at any given point, the sentiment captures 60 days before and after the date.
The pastel coloured columns indicate societal crises in yellow and personal ones in blue. We have defined the so-called societal crises, while the personal crises are inferred from reading the diaries. We have anonymised the personal crises here. In practice, we use the peaks and drops in the sentiment line to orient our reading towards periods where significant events might happen for the individual. We can also use the boxes to assess whether a societal event correspond to a change in the sentiment. Once we have established a list of significant events or periods, we then proceed with a deeper reading around those, gathering relevant extracts from the diaries.
While the diaries are massive documents, they are nevertheless “incomplete” and capture just a fragmented part of the diarist’s experience. As some of you might know, we then decided to visit a handful of diarists across the United States to form a fuller picture of the environments in which they live, their routines and daily life, and to interview them about their diary and some of the events that we noted down in the process of reading.
The future
Over the next months, we will publish short summaries of the articles already published, for example, papers that explore how the act of journalling can be transformative in and of itself.
If you have any questions or suggestions, we are of course happy to answer them, either here on Prosebox or via email ([email protected]).
Loading comments...