A new paper in Science, Differences Between Tight and Loose Cultures: A 33-Nation Study, is making the media rounds. Here’s NPR:
…The idea for this study really dates to the 1960s. Back then, an anthropologist decided to evaluate a few dozen obscure cultures and see if he could rank them on a scale from “tight” to “loose.” He defined tight cultures as having a lot of rules, which people violate at their peril. Loose cultures are more relaxed in their expectations, and more forgiving of people who deviate.
The Tightness Scale
“So for example, you might have been asked, how appropriate is it to curse in the bank or kiss in a public park, or eat or read a newspaper in a classroom? And we were able to derive scores of how constrained, in general situations, they are, versus how much they have latitude in different countries.”
“Some of the cultures that are quite tight in our sample include places like Singapore, Japan, Pakistan,” Gelfand says. “Whereas many loose societies include countries like New Zealand, the Netherlands, the United States.”
The abstract from the paper is a little harder to parse:
With data from 33 nations, we illustrate the differences between cultures that are tight (have many strong norms and a low tolerance of deviant behavior) versus loose (have weak social norms and a high tolerance of deviant behavior). Tightness-looseness is part of a complex, loosely integrated multilevel system that comprises distal ecological and historical threats (e.g., high population density, resource scarcity, a history of territorial conflict, and disease and environmental threats), broad versus narrow socialization in societal institutions (e.g., autocracy, media regulations), the strength of everyday recurring situations, and micro-level psychological affordances (e.g., prevention self-guides, high regulatory strength, need for structure). This research advances knowledge that can foster cross-cultural understanding in a world of increasing global interdependence and has implications for modeling cultural change.
This schematic from the paper illustrates the general model of how differences in “tightness” emerge:
Like many social science studies the authors relied a lot on survey data and conversion of rank ordered categorical responses into dependent variables. That’s a problem insofar as you need to take the quantities that are generated out of their statistical analyses with a grain of salt. They aren’t measuring someone’s height or temperature. Rather, they’re generating an aggregate measure from a range of concrete subcomponents. Granted, this measure has been shown to correlate well with individual questions in terms of how they vary cross-culturally. This is the “tightness,” the higher the score, the more tight the society. I do have some issues with this usage of a summary for a range of characters, but first let’s hit the raw results.
They are displayed in tabular format in the paper. That’s fine, but I decided to change it up a little for the purposes of presentation here. I took their table and focused on the “tightness” score, and added my own column which placed each national sample into a subjective broader region-cultural category.
|Korean||East Asian||South Korea||10|
|West European||West European||Germany (East)||7.5|
|West European||West European||Austria||6.8|
|West European||West European||Germany (West)||6.5|
|Chinese||East Asian||Hong Kong||6.3|
Tables leave something to be desired in gaining a gestalt understanding of relationships, so here’s a bar plot rank ordered by tightness score, with colors corresponding to region-culture. A lot of this is presumably not too surprising. Pakistan is the “tightest” nation which they sampled. But is Norway much tighter than Estonia? I picked this pair because Estonia is the most Nordic of the ex-Soviet Baltic nations, it traditionally being a Lutheran society due to German and Scandinavian influence and hegemony until absorption into the Russian Empire. Those who have visited both nations are probably the best to ask how this comports with their own experiences.
Backing up a bit, in the introduction to the paper they take a very broad historical view. They seem to imply that there is a gap between “small scale” hunter-gatherer societies and more dense agricultural ones in terms of the importance of social norms and conformity. There’s a plausible ecological rational for this: there are many more opportunities for “free riding” in dense and large scale societies. In contrast, inter-personal relationships are probably sufficient for cultures which exist mostly at the band level. The Code of Hammurabi is only necessary in cultures where personal relationships have diffused to the point where impersonal rules and heuristics need to be interposed between parties which are literally or de facto strangers. This is probably the difference between survival and extinction in a world which was predominantly at subsistence.
In the supplements there is a table of correlations between “tightness” and predictor variables, controlling for per capita GNP. I’ve selected out the most interesting (to me):
|Population density in 1500 (Log)||11||0.77||0.01||0.59|
|Population density (Log)||32||0.31||0.10||0.10|
|Rural Population density (Log)||30||0.59||0.01||0.35|
|Natural disaster vulnerability||30||0.47||0.01||0.22|
|Historical prevalence of pathogens||32||0.36||0.05||0.13|
|Death due to communicable diseases (Log)||31||0.59||0.01||0.35|
|Prevalence of tuberculosis (Log)||31||0.61||0.01||0.37|
|Infant mortality rate (Log)||32||0.42||0.02||0.18|
|Openness of media||29||-0.53||0.01||0.28|
|% attending religious services||31||0.54||0.01||0.29|
Notice the difference between population density in 1500 vs. population density today in terms of prediction! This may point us to the possibility that the long arm of cultural memory still reigns supreme to some extent. The effect size is the square of the correlation, and gives us a sense of how much of the variation in the dependent variable is predicted by the independent variable when you hold GNP per capita content. Of course it is important to observe that the N has dropped when you go back to 1500, probably because the individual data points are nations, and nations can’t always be projected back in time. All that being said I like predictor variables like population density and death due to communicable diseases best, because they’re a lot less clear and distinct than something like openness of media. Openness of media is a valid measure in my opinion, but since the statistic we’re predicting only comes out via a process of human directed calculation, having both ends of the line be open to disputation is not optimal.
As for the tightness measure itself, there’s some strangeness here. On the one hand, some if it makes sense. But scores for other nations surprising, as noted by the authors. For example, Israel. But that just leads to ad hoc explanations:
…Gelfand was surprised to find that Israel — which is under threat from its neighbors and its desert environment — is still culturally loose. Gelfand suspects that’s in part because lots of Israelis came from relatively loose cultures in Eastern Europe.
“It’s also a culture of argumentation, debate, dissent, that really is very much consistent with Judaism. And these things all promote looseness,” she says.
There are two points here. I’ll address the second first: the time depth of the culture of disputation in Judaism is something I’ll actually dispute. One can make the case that as a generality this is very much a feature of modern Ashkenazi Jewish culture, with the opening of the public debate to all sectors of society. Of course I grant that disputation between eminent rabbis occurred in the past, but pre-modern Jewry was run like most pre-modern societies, there were authorities on on high who dictated what was, and wasn’t, permissible. European Jewish communities were run as corporate subnational entities before their liberation in the wake of the Enlightenment. The expulsion of Baruch Spinoza from the Sephardic Jewish community of the Netherlands illustrates the nature of pre-modern Jewish, and gentile, society on the cusp (by this, I mean that the religiously plural Netherlands of the period exhibited a cohabitation between pre-modern exclusiveness and parochialism, and post-modern pluralism). Modern day stereotypes and generalizations are often very much the result of modern day conditions.
But the first point is of more concern to me: the aggregation of genuinely different societies into one sample. The idea that the European Jewry shared something with its Eastern European milieu is a questionable assertion. European Jews for much of the pre-modern era were in the West, but not of it. More accurately, Jews in the world of Islam and Christianity were suffered to exist, but lived in a parallel world unless they converted to the majority religion and left the Jewish community. The Yiddish (and later standard German) speaking Eastern European Jew had a strained and complex relationship with the nation-states of Eastern Europe which arose in the wake of the collapse of the old empires (Austria-Hungary, the Second Reich, the Ottoman Empire, and the Russian Empire). Can we speak of Hungarian or Romanian Jews who were distinctive from each other because of their association with the Hungarian or Romanian majority? As an illustration, Paul Erdos’ family had changed their name from Englander, as part of the process of de-Germanization and indigenization of Hungarian Jews.
This issue of the “nations” which were evaluated crops up elsewhere. The Indian sample was from west-central India, on the margins of the Hindi-Punjabi-Gujarati “cow belt.” It was very similar in “tightness” to Pakistan. But what would the “tightness” be in southern India? It may be very different. Additionally, comparing Iceland to China, as if they are comparable units, is obviously ridiculous (something the authors acknowledge). Despite my qualms with the “tightness ” statistic I would be very interested to see how this varies on a subnational scale. If it is measuring something informative and useful the correlations should start going up as you proceed down to a finer grain (“tightness” may be representative of only one region, while GNP per capita and the independent variables are drawn from the whole nation).
Though the top line of the research is focused on inter-cultural differences, the authors argue for the importance of cultural context to individual response and expectation. This is actually pretty obvious on the internet, and even among Americans. There are lots of cryptic subcultures and cultures which bubble up out of the woodwork when something of dispute comes to the fore. Prior to the issue which highlights the differences, one may not have been aware of implicit or background variation in norms.
The future direction of this sort of research will be in the direction of gene-culture coevolution and pathogen-culture coevolution, and their combinations. Pathogens are critical covariates of any shift toward dense living, and in the modern world tend to hit those from historic low density backgrounds much worse. The difference between high conformity and low conformity to me is well illustrated by the varied paths toward Christianization of the peoples of Oceania. In Polynesia the missionaries generally converted the chiefs, who then brought their people to the new faith en masse. Apparently this was just not feasible among Australian Aborigines, who were only predominantly Christianized by the 1970s. This development had to occur one individual at a time, because the “big men” in these societies simply had no ability or will to enforce conformity of religious belief.
Citation: Gelfand MJ, Raver JL, Nishii L, Leslie LM, Lun J, Lim BC, Duan L, Almaliach A, Ang S, Arnadottir J, Aycan Z, Boehnke K, Boski P, Cabecinhas R, Chan D, Chhokar J, D’Amato A, Ferrer M, Fischlmayr IC, Fischer R, Fülöp M, Georgas J, Kashima ES, Kashima Y, Kim K, Lempereur A, Marquez P, Othman R, Overlaet B, Panagiotopoulou P, Peltzer K, Perez-Florizno LR, Ponomarenko L, Realo A, Schei V, Schmitt M, Smith PB, Soomro N, Szabo E, Taveesin N, Toyama M, Van de Vliert E, Vohra N, Ward C, & Yamaguchi S (2011). Differences between tight and loose cultures: a 33-nation study. Science (New York, N.Y.), 332 (6033), 1100-4 PMID: 21617077