Over the past 16 months, Christopher St Cavish has visited 52 restaurants and ordered 7.243kg of the Nanxiang-style soup dumplings known as xiaolongbao at a cost of 712.5RMB. He stood in line at Yu Garden’s Cheng Huang Miao for 43 minutes, ate dumplings just 0.73mm thick at Zun Ke Lai, and exploded an 85.69g orange-sized dumpling at Taikang Tangbao. In The Shanghai Soup Dumpling Index, which St Cavish describes as a “quantitative interpretation of the colloquial standards for a well-constructed soup dumpling”, the numbers matter.
St Cavish took hairdresser’s scissors, a Kubei 200 gram/0.01 gram digital scale and a pair of Mitutoyo 0-150mm digital calipers (Model 500-196-20) to each of the restaurants to determine the size and weight of the dumplings and the ratio of soup to filling to dumpling skin. He then scored the dumplings according to the accepted colloquial criteria: a thin wrapper, plentiful soup, and abundant filling (皮薄, 汁多, 馅大). The result is an A3 poster explaining the methodology and scoring the dumplings, available here for 50RMB.
It’s a mad and wonderful project, a monomaniacal quest to find the perfect dumpling without using the one criterion you’d think would matter most: subjective taste. And yet quantitative and qualitative research are not irreconcilable; there are ways to quantify subjective assessments (eg scoring life satisfaction to compare between groups), and add qualitative perspectives on objective data (eg the income you require to ‘live comfortably’).
The story has already been picked up by Lucky Peach, the food magazine founded by Momofuku’s David Chang, but we had our own questions for Chris.
Why xiaolongbao?
They‘re a mascot of Shanghai, and they can easily measured with cheap tools. It helps that they are often delicious.
Why such a scientific approach to something that yields a subjective taste rather than an objective performance?
Looking at the technical angle of construction is important. It’s the foundation that “good” dumplings are built on. I wanted to recognise that.
Is it implicit in the project that you or someone else might want to use the report to replicate the top scorers?
I’d certainly hope that happens. Thirty five of the 50+ restaurants were not very good. They could learn a thing or two from those at the top. That said, I will leave it to them — I do not plan to open a shop of my own.
Alternatively, do you see this as being almost useless in its obsessive attention to detail, and thus aspiring to a kind of art?
Yes, there is that. I think there is humour in pushing this over the top. If I want to, I can stand on my soapbox and lecture about subjectivity, about the ultimate futility of comparing a subjective experience. But more often than not, I just chuckle to myself. In many ways, the Index misses the mark tremendously. That’s by design.
Any really strange interactions with people as you were carrying this out?
The strangest thing was that hardly anybody even looked twice at me.
Has there been any cringe or questioning about you doing this project as a foreigner (prior to this question)?
No. The Chinese response has been enthusiastic. It’s data, not taste.
How have people responded to the final index?
I’ve attracted the obsessives, so there might be some bias in my sample, but I’ve already had to do a second print run. Otherwise, I think people probably chuckle a bit to themselves at the ridiculousness of it all. That would suit me just fine.
This article was originally posted on Flamingo’s online magazine Lens