Introducing Wagnerism

If Richard Wagner were to have known before he died that some scientists in the future would create a psychometric tool to measure the "Wagnerism" in a given individual, would he be flattered or appalled? It's a strange question, but it's a question you could ask since that's what we have been working on here on the Transforming Musicology project.

Now it wasn't our initial intention to create such a Wagner tool, but rather it was a necessary step to come closer to one of the initial questions we asked at Transforming Musicology: "How do people hear leitmotifs?"

If you haven't read my earlier blog post on hearing leitmotifs, I can save you a bit of time by saying that we have finished collecting data on an experiment we have been running at Goldsmiths (and which has taken me all around London to meet and test Wagnerians) that required participants to listen to a ten-minute passage from Siegfried, then to complete a ten-minute memory test where they had to recall which leitmotifs they remembered hearing in the passage. Some people did very well on the memory test, others not so well.

So what leads to a person's ability to do well on this musical listening task?

As you might expect, we thought that how musical someone is would contribute to their recognition rate, but we also postulated that other variables may be at play. In addition to asking people about their musical background, we also surveyed participants' affinity for Richard Wagner and put them to the test by cross checking their self-reported Wagnerism with an fiendishly difficult objective Wagner quiz that was constructed in collaboration with Laurence Dreyfus.

The rationale behind the survey and the quiz was to attempt to get at what psychologists and statisticians refer to as a latent trait. A latent trait is something that cannot necessarily be measured directly like height, weight, or age, but rather one that is presumed to exist in the real world such as intelligence, anxiety, or neuroticism. In our case the latent variable we sought to measure was Wagnerism. We think that this trait actually exists in the population and its presence can be inferred from behavioural manifestations such as finding out the answer to questions like how many Der Ring des Nibelungen recordings an individual owns.

After collecting the results from 100 different participants, we attempted to use these scores from the surveys and quiz to attempt to predict how well someone would perform on a task such as the one presented to all of our subjects.

The numbers were crunched and what came out the other end of the analysis was quite surprising. We found that our Wagner survey and quiz actually served as a better tool to predict an individual's ability to identify leitmotifs than any other factor, including musical training! Our Wagner tool could actually predict leitmotif recognition ability at a rate at least twice that of an individual's musical training.

I should also mention here that an individual's German speaking abilities came in at a completely insignificant rate, thus providing (to my knowledge) the first empirical proof that the Germans can no longer claim any sort of richer experience of the leitmotifs than other nations. I also don't know how Wagner would have felt about that...

So essentially we have developed the first statistically backed psychometric tool for measuring a latent trait we're calling Wagnerism. So what? Though pretty useless on its own (unless you want a great party trick for your Ring screenings), this test suggests that factors other than someone's musical ability contribute to how an individual perceives music, which could help break down some of those inaccessibility complaints people seem to have about Wagner.

Our initial findings would suggest that you don't have to be the next Pierre Boulez to have a rich listening experience with Wagner's music, but rather you would stand a better chance if you just locked yourself in a room and listened to and read about his music until Walhalla burns down.

Some of the results of this work have been published in a paper given at the European Conference on Data Analysis in Bremen in July 2014.

If you would like to participate in a Transforming Musicology experiment at Goldsmiths, please contact David Baker at ps301db@gold.ac.uk or @DavidJohnBaker.