How many people sing shanties, really?

The Monitor Amateur Arts 2015 has appeared. Always interesting food for thought.

This year in particular, at least for me: one of the questions asked in the questionnaire on which the monitor is based is what kind of music people make - and this year 'shanty' was included in the list. My colleague dr. Teunis IJdens from LKCA kindly provided me with the precise question asked in the questionnaire. It is: "Which kind of music do you play or sing?", with the possibility to indicate one or more genres from a list of 10, including "shanty", and the possibility to add other genres.

The Monitor shows that shanty is actually quite popular. These are the figures:

6-11: 0% of people active in music are active in shanty
12-19: 0% etc.
20-34: 3%
35-49: 1%
50-64: 4%
65+: 7%
Overall: 2%

On a first glance I recognized my observation that shanty singing is done mainly by elderly people. But a second glance awoke lots of questions. Is really about one out of 15 active 65+ musicians a shanty musician? What about those 3% active shanty singers between 20 and 34? I never meet them (I am, at 51, mostly considered very much to be a youthful shanty singer).

And according to the Monitor, the percentages may be extrapolated from the survey sample to the total population - the claim is that 3 million Dutch people are active musicians. 2% of them would mean that we have 60.000 shanty singers in the Netherlands. And then I got really suspicious.

I have gotten to know the shanty world those past few years a little bit. In my observation, the vast majority of active shanty singers sing in male shanty choirs with on average approximately 30 singers. I know most of those choirs are a member of the Shanty Nederland organization, if only because Shanty Nederland takes care of the copyright dues of the choirs. Shanty Nederland boasts about 300 members. That would mean that about 300x30=9.000 active shanty singers are organized through Shanty Nederland. Which would count, if the monitor is right, for only 15% of active shanty singers.

Some possible explanations:

1. Shanty singing is done in choirs but 85% of active shanty choir singers are not organized through Shanty Nederland. But actually, nearly all the choirs I have encountered thus far are Shanty Nederland members. In the province of Groningen this would mean there would not be some 50 active choirs (the members of Shanty Nederland), but over 300. I simply don't believe this. It is true there will be smaller folk groups and maybe even some temporary choirs around outside of the organization, but not on this scale.

2. There is a lot of shanty singing going on outside the more formal shanty choirs. Maybe members of choirs who have one shanty on the repertoire have also indicated they are shanty singers. Possible, but for me again hard to believe: shanty singing in the Netherlands is quite well known across the Dutch population and is, in my observation, connected to the idea that shanty singing is being done in shanty choirs. I don't believe many ordinary choir members will register in this survey as shanty singers because they have one shanty on their repertoire.

3. Because shanty choir performances are rather participatory, audience members who sang along with a shanty choir have all indicated in the questionnaire that they have sung shanties recently. Possible, but again I don't believe that in a survey like this sing-along-audience-members will register as active singers - again, the idea that the performing is done by the choir and not by the audience singing along is, I believe, too engrained in our culture to make this plausible.

4. Extrapolation from the sample to the general population should not be done. The Monitor claims that the sample is representative for the Dutch population regarding the variables gender, age, level of education, household size and region. Maybe the wish to say things about 'the population' rather than about the sample has led the researchers to believe that shanty singing is in some mysterious way co-variant with gender, age, level of education, household size and region (connected to common thoughts amongst the more quantitative social scientists - looking for cause-and-effect-chains, thinking that behavior can be predicted if circumstances are identical [oh, fiction], believing that humans are determined by structure rather than having agency within structure). In my experience, the daily musical life of individuals is far too whimsical and messy for such easy inferences. The only way to know whether a sample is representative for the variable 'singing shanty' is to ask all 16 million inhabitants of the Netherlands whether or not they sing shanty. Because shanty singing has never been seriously studied in the Netherlands (or elsewhere, I believe) there is no other way yet to estimate the total number of shanty singers in the Netherlands.

And I must say that actually such an estimation is for me one of the least interesting things to know about shanty singing. What do people do, when singing shanty? What does it bring them? Those are the more interesting questions for me; not how many people sing shanties nation-wide. And I wonder who - apart maybe from Shanty Nederland, who may use the alleged 60.000 shanty singers as a support to get shanty singing recognized as Dutch intangible cultural heritage by UNESCO - is interested in collecting such numbers.

The example again made me aware why I do not really believe this type of large-scale quantitative survey research tells us much. People answer a question. But we do not know the meaning of their answer. We do not know what those individuals ticking boxes mean by "sing", or by "shanty". If we want to value their answers, that kind of knowledge is crucial.

All this adds up, eventually, to the bare fact that what we have found out through the Monitor is the following: in April 2015, of 5.134 persons living in the Netherlands (a sample representative for Dutch population as a whole when it comes to gender, age, level of education, household size and region) 2% ticked the box "shanty" when they had to answer the question "Which kind of music do you play or sing?" Crudely said: the Monitor has measured a very particular type of shanty-related behavior (ticking a box named 'shanty' in a questionnaire) of a group of 5.134 persons and has found that 103 people actually did tick the shanty box when asked which kind of music they play or sing.

In favor of the Monitor I must say that it has found out something about a very specific form of shanty-related behavior in a far larger sample than I do my research in. After five years of participant observation, I will be able to say something not about 5.134 persons, and not even about 103 people, but about approximately 40 - the members of one choir. And they will surely not be representative for Dutch population as a whole over gender, age, level of education, household size and region (if only because the average age of the choir is over 70).

In my favor must be said at least that I try to say something about more 'naturally occurring' behavior than 'ticking a box in a questionnaire' - indeed a peripheral sort of shanty-related behavior for most people actively engaged in shanty singing, far outshadowed in importance by such forms of shanty-related behavior as 'singing shanties during choir rehearsals on Tuesday nights' or 'giving performances with the shanty choir in residential homes for the elderly'.

We all know there is a trade-off between quantitative and qualitative research in the social sciences: either you know a very little bit about a large sample, or you know a lot about a small sample. And of course, the ideal situation is when you can make the combination. Which is what I started to do in this little blog: I tried to interpret the value of the Monitor-percentages on the basis of my more in-depth knowledge of shanty singing in the Netherlands AD 2015. But at present, I do not feel both types of knowledge have really informed each other.

Let's hope for better times.

Teunis IJdens. Kunstzinning en creatief in de vrije tijd. Monitor amateurkunst 2015. Utrecht: LKCA, 2015.