Harmony and Discord In Berklee’s Songwriting MOOC
In March, Berklee College of Music in Boston in the U.S. launched a five-week course on Songwriting, taught by Pat Pattison. You may not have heard of Pattison – in fact, you probably haven’t, as he isn’t a particularly notable songwriter. It would be easy in our celebrity-obsessed era to dismiss him on these grounds, but to do that would be a mistake, as his academic view on the craft of songwriting helped make this a truly insightful and enjoyable course.
It has been reviewed here previously by Rosalia Ziebecki, but recently I received a notification from the course team that Songwriting was being taken offline in preparation for a new version to go online starting July 19, and I wanted to present my take on the course. This was well on its way to being my first properly completed MOOC, until end-of-term marking and exam prep caused me to fall at the final hurdle, not submitting anything for the final assignment.
The class group was mixed – really mixed. Everyone from professional musicians to simple dabblers were enrolled, but the most notable difference from most MOOCs was the number of high-school aged children taking the course. To begin with, this led to some trouble on the forums as people jumped on “stupid questions”, but the atmosphere changed remarkably quickly once we began to realize just how young some of our peers were. That usual fear on the internet – of coming across as patronizing when trying to explain something in simple terms – was gone. We were dealing with younger people who needed more support, after all. In many cases this led to more fruitful, illuminating exchanges, even among those of us a bit longer in the tooth!
However, there was another group who tended to sow a bit more discord: the “music has no rules” crowd. Why would someone who objects to rules even take a songwriting course? What is there to be learnt if there are no rules? [pullquote position=”right”]One individual in particular sticks in my memory – he appeared on the forums to complain about people marking him down on his “dope” track. The song was about six minutes long and was mostly instrumental. [/pullquote]The lyrics did not address the task description, which was to write a single verse and chorus applying the lyrical devices taught during the week’s lessons. I had several assignments to mark of this sort, where the submitter made no attempt to complete the task. As these submissions were invariably longer than those that did try to complete the given task, it was just out-and-out rude, asking us to do more work, simply because they didn’t want to take part in the class.
(To those of you who share these concerns about “rules” in songwriting, I’ll say this. Songwriting is both a craft and an art. Learning the tools of the craft don’t detract from the art, but rather adds to it. Once you know the rules you can apply them or ignore them as you wish; but if you don’t know them, you can never choose to apply them. This course is ideal.)
The course was delivered on the Coursera platform and showed some of the potential of the technology while also highlighting some of the major restrictions. It used a mix of multiple choice quizzes and peer-reviewed assignments to assess progress.
The first and most obvious problem was in the assignment submission page. Unlike in technical subjects, the individual sections of each assignment were not independent, meaning that in order to answer fully, you would frequently have to refer back up to something you had previously written or an earlier part of the task description. Mainstream web browsers do nothing to assist you in viewing two different parts of a webpage at once, making this a fairly onerous task, and I personally found it broke into my “flow” during the tasks. This was compounded by the fact that the text entry boxes were of a fixed size, usually smaller than truly suited the question. This meant that looking for something you had previously written often meant scrolling twice – first the page, then the textbox. When marking peer assignments, scrolling was still a huge distraction, but thankfully the textboxes were replaced with the students’ answer embedded into the page.
Pattison also made an unfortunate slip-up with the first assignment that led to a great deal of confusion. He started the course by discussing the development of the narrative in a song, using a device he called “three boxes”. The first question in the assignment was to describe what goes in those “three boxes”, but as there were three questions in the assignment, there were three boxes for answers. That a large proportion of students got mixed up here should not be surprising, and I would be very surprised if this assignment wasn’t radically restructured in the next presentation of the course.
The peer review system was very apt to this type of course, as it required us as students to deal with a wider interpretation of the course material and reflect quite deeply on the range of options available to us. However, the structuring of deadlines made it unnecessarily difficult in an “open” course. Assignment deadlines were on a Monday, meaning we always had a weekend to work on them, but we couldn’t start reviewing before that deadline, and they were due before the end of the week. [pullquote position=”right”]Several participants said that they were effectively forced to drop out due to the lack of time or internet access during the week. As this was widely discussed on the forums, I hope that the next sitting will offer more flexible deadlines.[/pullquote]
Another problem that arose with the peer review system is that while some of us got five full reviews for our assignments, some got none – Coursera’s method of handing out assignments for review does not seem to guarantee fairness. It appears that when the submission deadline is reached, Coursera’s system essentially builds a deck of cards containing five cards for each assignment, then deals out a hand of five cards to each person who submitted. A more intelligent task scheduler would not deal out all the cards at once, instead dealing when a student starts reviewing. It would also not keep back the fourth and fifth copies of each assignment until every student had received two orthree pieces of peer feedback. It surprises me greatly, given the research interests of the founders, that Coursera didn’t think about this when first designing the peer review algorithms.
So that’s mostly about the mechanics behind the Songwriting MOOC, but what about the material and the teaching?
As you might expect from a college dedicated to performance, the quality of delivery was exemplary. PowerPoint was nowhere to be seen. No screencasts, no pencasts. Just one man and three cameras. Pattison was practically never off the screen, and if they needed to show us a word on screen, it would be added as an overlay in post-production, just as in any other educational video. He didn’t have to slow down or stop talking while he wrote, he didn’t have to do two things at once, he just talked to us. The director switched between the three cameras without any clear justification, and while several people commented negatively on this in the forums, it was actually very clever. Most MOOCs stick with one fixed camera, with the presenter talking straight to the camera, and you know when they’ve cut and done a retake. Not in this MOOC, you don’t – it just looks like any other cut between cameras.
(MOOC providers take note: there is a reason that mainstream educational video abandoned trying to capture the chalk-and-talk experience on video and moved to trying to incorporate standard television techniques into education instead. It just works better.)
Pattison also had the interesting quirk of wearing T-shirts with keywords from the course written on them. He wore a leather jacket over these, partially obscuring the word, and then he would reveal them at an appropriate time. It’s one of those things that made the course seem more “human”, while at the same time reassuring the student that the whole lesson was properly planned in advance.
The one problem with Pattison’s delivery is that he near constantly used tone-of-voice to highlight and exaggerate the effects of word choice, timing and rhythm, giving little opportunity for the students to listen to the effect without the interference of tone-of-voice. This led to some confusion that was discussed on the forum, particularly among students whose native language wasn’t English.
As Rozalia noted in her review, it was stated that no knowledge of music was required, and many of the course participants didn’t have any such knowledge, so it was a bit of a surprise to many when later in the course, some musical notation was used. One of the unit quizzes relied on being able to read the timing from standard notation, which obviously made it impossible for a large part of the cohort. He also included a section on the meaning of musical notes within the scale, which for me is an essential part of the course, but it was impenetrable to many.
(It seems to be a common problem on Coursera that the published prerequisites are inaccurate or unspecific – for example, the Natural Language Processing course by Columbia University states that some programming experience is required, but fails to advise of the specific languages that are used in the course, and the programming examples were presented using features of Python that would not be readily understood by someone not familiar with the language.)
The course for me was of the ideal length. By the beginning of the fourth week I was starting to weary of it a little, and if I hadn’t known there was only two weeks left, I would probably have stopped then and there. In the end, I didn’t submit the final assignment and therefore didn’t officially complete the course, but I did follow the videos right through to the end.
The course was very, very rewarding and interesting, and if I wasn’t going to be on holiday when the course relaunches, I’d probably actually go through the whole thing again, and I’m sure there are many people from the first sitting who feel the same. The assignments can be addressed in a different way each time, and there’s plenty to learn from repeating the whole process.
The MOOC that came back . . . .
One of the most interesting things about the Songwriting MOOC, though, is that it is being rerun so soon. Many of the first courses on Coursera have never had a second presentation, and I now have a long watchlist of items with a start date of “TBS”. Why has this course proved different?
Quite simply, this course is an advert. At this length, the course simply cannot cover the songwriting process in all its detail. At several points throughout the course, Pat comments that there is more detail on the topic in his books. [pullquote position=”right”]As the course continued, the advertising ramped up. The course included, as a bonus, an extract from one of his masterclasses . . . masterclasses that are available for purchase on DVD. [/pullquote]Then there was the emails from the course team advertising Berklee’s distance courses . . . available at a special discount to Coursera students, naturally; the prize draw for a free course for two lucky course completers; the college’s summer programs and their Master’s scheme.
It’s a good course, so it’s a good advert, but at the same time it serves as a timely reminder that in the business of modern education, there ain’t no such thing as a free lunch.