May 03 2021

The Science of Feel-Good Storytelling

From one perspective art is mostly a science that we understand better at an intuitive rather than analytical level. This does not reduce the creative elements off artistic expression, but it does mean there is an underlying empirical phenomenon to be understood. Storytelling, for example, has a basic structure, with elements that serve a specific purpose. One of the more famous attempts at breaking down the structure of certain types of stories is the Hero’s Journey by Joseph Campbell, in which he explains the common elements of epic quests in literature, that hold true even in more modern storytelling like Star Wars.

A recent study by German authors takes a similar look at the “feel good film”, which is not really a specific genre but more of a vague category. The “feel good film” is meant, as the name implies, to elevate the mood and be pleasant to watch. The authors point out that this is often a point of criticism by serious film critics, but simultaneously a point of praise from viewers. There is just as much of an art and science behind making a good feel-good film as any other type, so I think the criticism is unfair. I would focus more on the quality of any specific movie.

After extensive surveying, the authors found that the best formula for a feel good effect is the romantic comedy. The authors write;

“Often these involve outsiders in search of true love, who have to prove themselves and fight against adverse circumstances, and who eventually find their role in the community.”

Further, the authors found that the introduction of a “fairy tale” element served well to lighten the movie and enhance the feel-good effect. The movie that immediately came to mind when I read this was Enchanted, which seems to deliberately follow this formula (and it worked, extremely well). The authors also point out that such films require genuine drama and conflict. There appears to be a sweet spot – where we feel a real threat but we know the good guys are going to pull it out in the end. It’s like a rollercoaster – it’s simulated danger, but we know we are safe. In Enchanted the evil queen needs to seem genuinely menacing, for example.

One way filmmakers think about this (the science of storytelling through film) is that there is an understood language between the filmmaker and the audience. This language can evolve over time, but there are many elements that are extremely stable, even going back to the early plays of ancient Greece. This language allows the filmmaker to communicate information on many levels very efficiently to the audience. If the camera lingers for half a second on a glance from an actor, we know that means they are hiding something, for example. But every aspect of a film from the music to the camera angles and editing style communicates to the audience. Horror films are perhaps the most obvious example of this – we know when we are supposed to be scared, everything about the film tells us so.

The real trick for filmmakers, and a powerful determinant of the overall quality of the film, is how subtly and artfully this communication is achieved. If you are too obvious and unoriginal, then a film can descend into tropes and cliches, and the film becomes too predictable. You have to follow a formula without looking like you are following a formula. There is a word for this as well – formulaic. It’s a criticism pointed at stories that too obviously follow a formula.

The other real art of filmmaking and storytelling is knowing when to break the rules. The rules are important, and you need to understand them as a writer and it’s a good default to abide by them. In fact, some experts refer to these rules as a “contract” with the audience – break the contract at your own peril. I don’t think Enchanted would have worked as a movie if the hero died a the end. That would have broken the contract with the audience.

But if you very deliberately and carefully break the rules in order to achieve a specific effect, this can be powerful. It can “deconstruct” a genre is a very interesting way. I think the books and series Game of Thrones was successful partly because Martin occasionally broke the rules. When he (spoiler alert) killed off the main character at the end of the first book/season, the reader was disoriented, questioned the genre of story they had gotten themselves into, and knew they were in for a wild ride. Creating and then breaking expectations, however, is just another form of storytelling.

I am discussing all this to lead to a profound question – how far can we go in this reductionist scientific approach to storytelling? This is part of a broader question – how much can we have a science of art? Can we really reduce the sublime emotion of a masterful painting or classic piece of literature to a formula? Is it just a matter of understanding that formula in a sufficiently complex, thorough, and reductionist way? Or is there something irreducible about art that will never be expressed as a set of rules?

I confess I think the answer is closer to the former than the latter. We perceive the incredible complexity of art as being irreducible, but how is that any different than interpreting the incredible complexity of biology as being due to a magical life force? I think one big test of this question will come as we are trying to develop creative computer algorithms. There already is storytelling AI. I suspect that like playing chess and pretty much everything else, such narrow AI will eventually get better than humans at their task. At some point we will cross the threshold where a movie script by an AI will be indistinguishable from a human-written script. And soon after it will blow past that point and get as good and then better than the best human writers. I don’t see any reason why this won’t happen. If you think I’m wrong, great, in the near future we will likely have our answer.

It’s also likely that for a long time such creative AIs will be used as tools for humans, rather than stand-alone creators. If you have a story to tell you can put the story elements you want into the computer and it will then use them to skillfully craft the end result. You could then alter it until you get the result you want. The end will be a human-computer collaboration, allowing people to express themselves without spending year mastering a craft. Whether this is ultimately a good thing or not remains to be seen, but it will happen.

No responses yet