About 130 years ago, French psychologist Gustave Le Bon published his book “Psychologie des foules” (The Crowd: A Study of the Popular Mind). The gist of his book is that individuals, when part of a larger crowd, succumb their will to the crowd. They stop acting as individuals and begin acting as one large mass. Le Bon’s book was published in a time where especially the elites of Europe feared mass uprisings. At the height of industrialization in many European countries, strikes and riots became a common occurrence, so that powerful elites had to think about ways to mitigate this threat (Clover 2016).
Le Bon’s language is full of despise for the working class and lacks evidence, which makes it an easy target for academic deconstruction (Erz 2019). Mid-century, then, a new approach to society became popular in sociology: individualism. Suddenly, books were full of individualist notions of society; and theories began focusing more on a micro-sociological perspective.
Crowd Behavior: Individualist or Collective?
Of course, both “mass psychological” as well as individual factors play a role in the behavior of crowds, as the famous Iroquois theater fire of 1903 shows (Coleman 1990, 204). Mark Granovetter (1978), who takes up the theater fire example, shows how autonomous, individual behavior can amount to seemingly “irrational” macro-behavior, in other words: crowd behavior.
This is the reason why Analytical Sociology tries to stick to so-called “middle range theories” (Merton 2007) which theoretically describe a single set of phenomena – not more, not less – in order to maximize explanatory power without sacrificing theoretical rigor. In other words: since macro-sociological theories are bound to fail in the face of the complexity of society, restricting yourself to a certain phenomenon increases your chances that what you’re interpreting into your data actually makes sense.
When I was dealing with crowd science and tried to estimate which approach was more sensible given the real world, I couldn’t really arrive at a conclusion. Both individualist and collectivist approaches did have something valuable to offer.
As sociologists, we are bound by the “laws” of societal interaction, just as physicists are bound by the laws of thermodynamics. This means that, at the same time we are researching something, we are also affected by it. However, while gravity didn’t change simply because Isaac Newton described it, the ways society work are much less stable: First, the functioning of society changes constantly; second, it changes also in response to sociological inquiry; and third, how we describe the functioning of society has a massive impact on our interpretation of society.
Remember the theater fire example: If you frame the crowd’s behavior solely in mass-psychological terms, it will make just as much sense as if you were to frame it solely in individualist terms, or anything in between.
Sociologists Participating in Society
So which perspective is the correct one? Obviously, all of them have their merits. While it is impossible to find “the” correct theoretical frame for the example, there is a way to at least gauge which one makes most sense (depending on the spatial-temporal context, of course). And how do we find it out? By participating in society.
There are these stories from sociologists actively participating in the sites of their research. Wayne Zachary, for example, studied a Karate club for three years to deeply understand its dynamics. In the end, he was able to predict the following split just by looking at his collected data, which is now a foundational example of network science. Loïc Wacquant became a boxer while studying the Chicago underclass to gain experience no theoretical framework could’ve given him. Others again, such as Guy Debord or Walter Benjamin, have taken random strolls through cities to get a feeling for the city’s structure, an ethnographic method called “go-alongs”.
The practice of sociologists participating in society seems to have fallen out of favor, however. Maybe it was out of fear that it could taint our judgment because in those moments we are part of the structures we have to analyze, or because we didn’t want to tarnish the data we are collecting. Or maybe it just became unfavorable. I don’t know for sure, I only know that I myself never did this on purpose.
But a recent experience showed me how incredibly useful this participation can be in order to get a better understanding of your own research topic.
How Volunteering Can Help
Ten years ago, in the summer of 2012 – my first university summer break –, I was bored and wanted to do something. So I searched for activities, and found a call for participation by a local, back then relatively unknown music festival. The organizers were searching for people to help them build the stage and the festival area, and to help sell beers and merchandise on the day of the festival itself. I liked it and continued to help out. Today, the festival itself attracts many more people, has grown to three instead of just one day, and I have gained more responsibilities. Specifically, my responsibilities now include keeping an overview over the visitor’s area, fix small hiccups here and there, and make sure everyone has a good time.
This year, the organizers ordered water-flushed toilets for the first time. This is great as it gives the festival its edge regarding comfort, but unlike those shabby plastic boxes, more comfortable toilets also require power and water. On the second day of the festival, the former suddenly cut out. No big deal, you may think: It was warm and during daytime, so we just didn’t have power for the main stage for a few minutes until our technicians were able to restore power. Definitely not ideal, but nothing too bad.
However, power cutting out also meant that the power for the toilets went out, rendering them unusable. But how are the visitors supposed to know? Right. So my job was to close the toilets while we had no power. This meant positioning myself and a few helpers in the entrance of the area and telling people they had to wait until they could enter again.
This was a dreadful experience, however. Because if the main stage is suddenly quiet, this signals to people that they may now do something else until the festival resumes, which unsurprisingly includes visiting the toilet. So for a few minutes until the helpers arrived, I was the only person standing between a constantly growing crowd of hundreds of people and the toilets. I’m an academic, and not a security guard with crowd control training, so I had no idea how to keep the situation under control.
I did what I thought was best in the situation: be transparent, tell the people we’re actively working on the problem and that it’s only a matter of time until the people could resume. But then I made a mistake: I told the people that if we’re lucky it might be only 30 seconds until the toilets are open again. Immediately, a choir of several dozen people started to count to 30. What would happen when they finished? I couldn’t tell. The people all had a very good mood, and so nothing happened. Everyone remained calm and happy until we could reopen the area again.
In that moment, I experienced what sociologists before me had described for the first time in my life: that rational individuals may do something stupid just because they’re part of a larger crowd — even if they understand the situation.
To me, this situation was an eye-opener: As a very left-wing academic and given how I approached crowds in the past (see again my article on Le Bon, Erz 2019), my default assumption is always that people are smarter than we acknowledge, and I always attribute explanations drawing on irrationality to academic elitism. This makes it sometimes very hard for me to understand why certain patterns emerge in my data, patterns I can not explain with rational individualism, only with crowd behavior.
This little episode showed me how valuable it is to actively participate in those areas you are researching, to get a better feeling for what is a correct interpretation of the data you retrieve. For the last nine years, I always saw helping out on the festival as a welcoming interruption of my computer-heavy day-job: Do one week of heavy manual labor, see many bands for free, and have fun. Now I see that this is not just a welcome interruption, but also incredibly helpful for becoming a better sociologist.
I do not work on riots/strikes/crowds at the moment, so the experience does not help me with my current research project. But what it showed is how valuable it can be to perform a job that has nothing to do with sociology at all. It certainly doesn’t free you from reading and understanding theory, but it gives you this edge, this experience that can show you whether or not your theoretical ideas about how society works actually make sense. It is a literal reality check for your studies.
Participant Observation is Insufficient
A few days after the festival ended I talked to a friend over the phone, and told them about this experience of mine. They agreed, but added an important point: It is important to not just simply perform “participant observation”, since in that case you’re still performing your role as a sociologist. No, it is very important to actually fulfill a different role whose primary reason is not to gain sociological insight. Nobody stops you from later on reflecting on your experiences (as I do in this article), but it is important that the work one does is not just a pretense for collecting data.
As academics we are frequently exposed to the accusation that we live in an “ivory tower”, detached from the lived realities of the people around us. And yes, if we only focus on theory and perform scientific data analysis, this doesn’t give us experience how the world works for non-academics; those we frequently analyze. Only by actively and sincerely working the jobs of those people we wish to understand have we any chance of percolating our object of investigation.
My lesson learned from this experience is that I will continue to be on the lookout for jobs that I can work which can help me understand the lived reality of the people I’m interested in much better than any reading could ever do. And I urge all of my fellow social scientists to do the same. It’s absolutely worth it.
- Clover, Joshua. 2016. Riot. Strike. Riot: The New Era of Uprisings. London: Verso.
- Coleman, James S. 1990. Foundations of Social Theory. Cambridge, Mass: Belknap Press of Harvard University Press.
- Erz, Hendrik. 2019. ‘Der lange Schatten von Gustave Le Bon. Zum sprachlichen Einfluss der Crowd Science auf die Soziologie der Gewalt’. Soziologiemagazin 2019 (2): 71–88. https://doi.org/10.3224/soz.v12i2.06.
- Granovetter, Mark. 1978. ‘Threshold Models of Collective Behavior’. The American Journal of Sociology 83 (6): 1420–43. https://sociology.stanford.edu/sites/g/files/sbiybj9501/f/publications/threshold_models_ajs_1978.pdf.
- Merton, Robert K. 2007. ‘Theories of the Middle Range’. In Classical Sociological Theory, 448–59. Blackwell Publishing. http://www.csun.edu/~snk1966/Robert%20K%20Merton%20-%20On%20Sociological%20Theories%20of%20the%20Middle%20Range.pdf.